* [Qemu-devel] [Minios-devel] [PATCH v4 0/<VARIOUS>] Begin to disentangle libxenctrl and provide some stable libraries
@ 2015-11-09 11:59 Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (3 more replies)
0 siblings, 4 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 11:59 UTC (permalink / raw)
To: xen-devel
Cc: Wei Liu, Stefano Stabellini, Ian Jackson, qemu-devel,
minios-devel, samuel.thibault, Roger Pau Monne
In <1431963008.4944.80.camel@citrix.com> I proposed stabilising some
parts of the libxenctrl API/ABI by disaggregating into separate
libraries.
This is v5 of that set of series against:
xen
qemu-xen
qemu-xen-traditional
mini-os
NB: Samuel+minios-devel will only get the mini-os side and Stefano+qemu
-devel the qemu-xen side.
The code for all repos can be found in:
git://xenbits.xen.org/people/ianc/libxenctrl-split/xen.git v5
git://xenbits.xen.org/people/ianc/libxenctrl-split/qemu-xen.git v5
git://xenbits.xen.org/people/ianc/libxenctrl-split/qemu-xen-traditional.git v5
git://xenbits.xen.org/people/ianc/libxenctrl-split/mini-os.git v5
The tip of the xen.git branch contains an extra patch hacking Config.mk
to point to all the others above, which should get the correct things for
the HEAD of the branch, but not further back in time.
The new libraries here are:
* libxentoollog: Common logging infrastructure
* libxenevtchn: Userspace access to evtchns (via /dev/xen/evtchn etc)
* libxengnttab: Userspace access to grant tables (via /dev/xen/gnt??? etc)
* libxencall: Making hypercalls (i.e. the IOCTL_PRIVCMD_HYPERCALL type
functionality)
* libxenforeignmemory: Privileged mappings of foreign memory
(IOCTL_PRIVCMD_MMAP et al)
The first three were actually pretty distinct within libxenctrl already and
have not changed in quite some time.
Although the other two are somewhat new they are based on top of long
standing stable ioctls, which gives me some confidence.
Nonetheless I would appreciate extra review of at least the interface
headers of all of these with a particular eye to the suitability of these
interfaces being maintained in an ABI (_B_, not _P_) stable way going
forward.
Still to come would be libraries for specific out of tree purposes
(device model, kexec), which would be adding new library at the same
level as libxc I think, rather than underneath, i.e. also using the
libraries split out here, but hopefully not libxenctrl itself.
The new libraries use linker version-scripts to hopefully make future
ABI changes be possible in a compatible way.
Since last time:
* xen.git now has a substantial number of acks
* mini-os side is now completely acked
* qemu-xen is little over half acked/reviewed, and the rest are updated
based on feedback.
* qemu-xen-traditional one patch acked
The whole thing has been build tested on Linux (incl stubdoms), and on
FreeBSD. I have runtime on Linux with qemu-xen, qemu-xen-trad and stubdoms.
Neither NetBSD nor Solaris have been tested at all. It's certainly not
impossible that I've not got the #includes in the new files quite right.
http://xenbits.xen.org/people/ianc/libxenctrl-split/v5.html is the document
I've been using to try and track what I'm doing. It may not be all that
useful. The history of it is in the v5-with-doc branch of the xen.git
linked to above.
I propose that these precursors (and the corresponding qemu-xen-traditional
+ mini-os patches) should go in now:
tools/Rules.mk: Properly handle libraries with recursive dependencies.
tools: Refactor "xentoollog" into its own library
I believe all the relevant/related patches are acked and this would make a
(small) dent in the set of things to keep in the air.
Ian.
_______________________________________________
Minios-devel mailing list
Minios-devel@lists.xenproject.org
http://lists.xenproject.org/cgi-bin/mailman/listinfo/minios-devel
_______________________________________________
Minios-devel mailing list
Minios-devel@lists.xenproject.org
http://lists.xenproject.org/cgi-bin/mailman/listinfo/minios-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* [PATCH XEN v5 00/23] Begin to disentangle libxenctrl and provide some stable libraries
2015-11-09 11:59 [Qemu-devel] [Minios-devel] [PATCH v4 0/<VARIOUS>] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 01/23] tools/Rules.mk: Properly handle libraries with recursive dependencies Ian Campbell
` (23 more replies)
2015-11-09 12:01 ` Ian Campbell
` (2 subsequent siblings)
3 siblings, 24 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
We intend to stabilise some parts of the libxenctrl interface by
splitting out some functionality into separate stable libraries.
This is the xen part of the first phase of that change.
This mail is (or is intended to be) a reply to a "0/<VARIOUS>"
super-intro mail covering all of the related patch series and which
contains more details.
Ian Campbell (23):
tools/Rules.mk: Properly handle libraries with recursive dependencies.
tools: Refactor "xentoollog" into its own library
tools/libxc: Remove osdep indirection for xc_evtchn
tools: Refactor /dev/xen/evtchn wrappers into libxenevtchn.
tools: Arrange to check public headers for ANSI compatiblity
tools/libxc: Remove osdep indirection for xc_gnt{shr,tab}
tools: Refactor /dev/xen/gnt{dev,shr} wrappers into libxengnttab.
tools/libxc: Remove osdep indirection for privcmd
tools: Refactor hypercall calling wrappers into libxencall.
tools/libxc: drop xc_map_foreign_bulk_compat wrappers
tools: Remove xc_map_foreign_batch
tools: Implement xc_map_foreign_range(s) in terms of common helper
tools: Refactor foreign memory mapping into libxenforeignmemory
tools: foreignmemory: provide xenforeignmemory_unmap.
foreignmemory: use size_t for size arguments.
tools/libs/evtchn: Review and update doc comments.
tools/libs: Clean up hard tabs.
tools/libs/gnttab: Review and update doc comments.
tools/libs/call: Update some log messages to not refer to xc.
tools/libs/call: Avoid xc_memalign in netbsd and solaris backends
tools/libs/foreignmemory: Mention restrictions on fork in docs.
tools: Update CFLAGS for qemu-xen to allow it to use new libraries
HACK: Update Config.mk to pull all the right bits from my xenbits
trees
.gitignore | 10 +
Config.mk | 20 +-
config/FreeBSD.mk | 2 -
config/NetBSD.mk | 3 -
config/NetBSDRump.mk | 1 -
config/StdGNU.mk | 1 -
config/SunOS.mk | 1 -
stubdom/Makefile | 92 ++-
stubdom/grub/Makefile | 1 +
tools/Makefile | 18 +-
tools/Rules.mk | 124 ++-
tools/console/Makefile | 3 +-
tools/console/daemon/io.c | 64 +-
tools/console/daemon/utils.c | 1 -
tools/console/daemon/utils.h | 1 +
tools/libs/Makefile | 11 +
tools/libs/call/Makefile | 67 ++
tools/libs/call/buffer.c | 192 +++++
tools/libs/call/core.c | 147 ++++
tools/libs/call/freebsd.c | 140 ++++
tools/libs/call/include/xencall.h | 84 ++
tools/libs/call/libxencall.map | 19 +
tools/libs/call/linux.c | 132 +++
tools/libs/call/minios.c | 81 ++
tools/libs/call/netbsd.c | 121 +++
tools/libs/call/private.h | 68 ++
tools/libs/call/solaris.c | 97 +++
tools/libs/evtchn/Makefile | 67 ++
tools/libs/evtchn/core.c | 72 ++
tools/libs/evtchn/freebsd.c | 138 ++++
tools/libs/evtchn/include/xenevtchn.h | 148 ++++
tools/libs/evtchn/libxenevtchn.map | 14 +
tools/libs/evtchn/linux.c | 136 ++++
tools/libs/evtchn/minios.c | 266 ++++++
tools/libs/evtchn/netbsd.c | 147 ++++
tools/libs/evtchn/private.h | 25 +
tools/libs/evtchn/solaris.c | 135 ++++
tools/libs/foreignmemory/Makefile | 67 ++
tools/libs/foreignmemory/compat.c | 72 ++
tools/libs/foreignmemory/core.c | 71 ++
tools/libs/foreignmemory/freebsd.c | 135 ++++
.../libs/foreignmemory/include/xenforeignmemory.h | 79 ++
tools/libs/foreignmemory/libxenforeignmemory.map | 8 +
tools/libs/foreignmemory/linux.c | 294 +++++++
tools/libs/foreignmemory/minios.c | 68 ++
tools/libs/foreignmemory/netbsd.c | 111 +++
tools/libs/foreignmemory/private.h | 47 ++
tools/libs/foreignmemory/solaris.c | 109 +++
tools/libs/gnttab/Makefile | 69 ++
tools/libs/gnttab/gntshr_core.c | 95 +++
tools/libs/gnttab/gntshr_unimp.c | 62 ++
tools/libs/gnttab/gnttab_core.c | 124 +++
tools/libs/gnttab/gnttab_unimp.c | 89 ++
tools/libs/gnttab/include/xengnttab.h | 219 +++++
tools/libs/gnttab/libxengnttab.map | 23 +
tools/libs/gnttab/linux.c | 329 ++++++++
tools/libs/gnttab/minios.c | 117 +++
tools/libs/gnttab/private.h | 47 ++
tools/libs/toollog/Makefile | 62 ++
tools/{libxc => libs/toollog}/include/xentoollog.h | 10 +
tools/libs/toollog/libxentoollog.map | 12 +
tools/{libxc => libs/toollog}/xtl_core.c | 10 +
tools/{libxc => libs/toollog}/xtl_logger_stdio.c | 12 +-
tools/libvchan/Makefile | 8 +-
tools/libvchan/init.c | 44 +-
tools/libvchan/io.c | 20 +-
tools/libvchan/libxenvchan.h | 9 +-
tools/libxc/Makefile | 44 +-
tools/libxc/include/xenctrl.h | 312 +------
tools/libxc/include/xenctrl_compat.h | 132 +++
tools/libxc/include/xenctrlosdep.h | 172 ----
tools/libxc/include/xenguest.h | 14 +-
tools/libxc/xc_altp2m.c | 64 +-
tools/libxc/xc_domain.c | 105 +--
tools/libxc/xc_evtchn.c | 54 +-
tools/libxc/xc_evtchn_compat.c | 75 ++
tools/libxc/xc_flask.c | 8 +-
tools/libxc/xc_foreign_memory.c | 99 +--
tools/libxc/xc_freebsd_osdep.c | 407 ----------
tools/libxc/xc_gnttab.c | 93 +--
tools/libxc/xc_gnttab_compat.c | 111 +++
tools/libxc/xc_hcall_buf.c | 138 +---
tools/libxc/xc_kexec.c | 36 +-
tools/libxc/xc_linux_osdep.c | 898 ---------------------
tools/libxc/xc_minios.c | 482 -----------
tools/libxc/xc_misc.c | 79 +-
tools/libxc/xc_netbsd.c | 345 --------
tools/libxc/xc_private.c | 277 +------
tools/libxc/xc_private.h | 105 +--
tools/libxc/xc_solaris.c | 305 -------
tools/libxc/xc_sr_restore.c | 6 +-
tools/libxc/xc_sr_save.c | 6 +-
tools/libxc/xc_suspend.c | 18 +-
tools/libxc/xc_tmem.c | 7 +-
tools/libxc/xc_vm_event.c | 29 +-
tools/libxc/xenctrl_osdep_ENOSYS.c | 210 -----
tools/libxc/xg_private.h | 3 +-
tools/libxl/Makefile | 18 +-
tools/libxl/libxl.c | 2 +-
tools/libxl/libxl_dom_suspend.c | 4 +-
tools/libxl/libxl_event.c | 14 +-
tools/libxl/libxl_internal.h | 6 +-
tools/misc/Makefile | 11 +-
tools/misc/xen-hptool.c | 13 +-
tools/misc/xen-lowmemd.c | 15 +-
tools/misc/xen-mfndump.c | 1 +
tools/ocaml/Makefile.rules | 26 +-
tools/ocaml/libs/eventchn/Makefile | 4 +-
tools/ocaml/libs/eventchn/xeneventchn_stubs.c | 20 +-
tools/ocaml/libs/xc/xenctrl.ml | 2 -
tools/ocaml/libs/xc/xenctrl.mli | 1 -
tools/ocaml/libs/xc/xenctrl_stubs.c | 8 +-
tools/ocaml/libs/xentoollog/Makefile | 6 +-
tools/ocaml/libs/xentoollog/genlevels.py | 2 +-
tools/ocaml/xenstored/domains.ml | 6 +-
tools/ocaml/xenstored/xenstored.ml | 5 +-
tools/python/setup.py | 10 +-
tools/python/xen/lowlevel/xc/xc.c | 2 +-
tools/tests/xen-access/xen-access.c | 2 +-
tools/xcutils/Makefile | 4 +-
tools/xenmon/Makefile | 2 +
tools/xenmon/xenbaked.c | 33 +-
tools/xenpaging/Makefile | 4 +-
tools/xenpaging/pagein.c | 5 +-
tools/xenpaging/xenpaging.c | 31 +-
tools/xenpaging/xenpaging.h | 5 +-
tools/xenstore/Makefile | 7 +-
tools/xenstore/xenstored_core.c | 7 +-
tools/xenstore/xenstored_core.h | 5 +-
tools/xenstore/xenstored_domain.c | 49 +-
tools/xenstore/xenstored_minios.c | 5 +-
tools/xentrace/Makefile | 5 +-
tools/xentrace/xenctx.c | 3 +-
tools/xentrace/xentrace.c | 24 +-
134 files changed, 5777 insertions(+), 4295 deletions(-)
create mode 100644 tools/libs/Makefile
create mode 100644 tools/libs/call/Makefile
create mode 100644 tools/libs/call/buffer.c
create mode 100644 tools/libs/call/core.c
create mode 100644 tools/libs/call/freebsd.c
create mode 100644 tools/libs/call/include/xencall.h
create mode 100644 tools/libs/call/libxencall.map
create mode 100644 tools/libs/call/linux.c
create mode 100644 tools/libs/call/minios.c
create mode 100644 tools/libs/call/netbsd.c
create mode 100644 tools/libs/call/private.h
create mode 100644 tools/libs/call/solaris.c
create mode 100644 tools/libs/evtchn/Makefile
create mode 100644 tools/libs/evtchn/core.c
create mode 100644 tools/libs/evtchn/freebsd.c
create mode 100644 tools/libs/evtchn/include/xenevtchn.h
create mode 100644 tools/libs/evtchn/libxenevtchn.map
create mode 100644 tools/libs/evtchn/linux.c
create mode 100644 tools/libs/evtchn/minios.c
create mode 100644 tools/libs/evtchn/netbsd.c
create mode 100644 tools/libs/evtchn/private.h
create mode 100644 tools/libs/evtchn/solaris.c
create mode 100644 tools/libs/foreignmemory/Makefile
create mode 100644 tools/libs/foreignmemory/compat.c
create mode 100644 tools/libs/foreignmemory/core.c
create mode 100644 tools/libs/foreignmemory/freebsd.c
create mode 100644 tools/libs/foreignmemory/include/xenforeignmemory.h
create mode 100644 tools/libs/foreignmemory/libxenforeignmemory.map
create mode 100644 tools/libs/foreignmemory/linux.c
create mode 100644 tools/libs/foreignmemory/minios.c
create mode 100644 tools/libs/foreignmemory/netbsd.c
create mode 100644 tools/libs/foreignmemory/private.h
create mode 100644 tools/libs/foreignmemory/solaris.c
create mode 100644 tools/libs/gnttab/Makefile
create mode 100644 tools/libs/gnttab/gntshr_core.c
create mode 100644 tools/libs/gnttab/gntshr_unimp.c
create mode 100644 tools/libs/gnttab/gnttab_core.c
create mode 100644 tools/libs/gnttab/gnttab_unimp.c
create mode 100644 tools/libs/gnttab/include/xengnttab.h
create mode 100644 tools/libs/gnttab/libxengnttab.map
create mode 100644 tools/libs/gnttab/linux.c
create mode 100644 tools/libs/gnttab/minios.c
create mode 100644 tools/libs/gnttab/private.h
create mode 100644 tools/libs/toollog/Makefile
rename tools/{libxc => libs/toollog}/include/xentoollog.h (97%)
create mode 100644 tools/libs/toollog/libxentoollog.map
rename tools/{libxc => libs/toollog}/xtl_core.c (95%)
rename tools/{libxc => libs/toollog}/xtl_logger_stdio.c (97%)
create mode 100644 tools/libxc/include/xenctrl_compat.h
delete mode 100644 tools/libxc/include/xenctrlosdep.h
create mode 100644 tools/libxc/xc_evtchn_compat.c
delete mode 100644 tools/libxc/xc_freebsd_osdep.c
create mode 100644 tools/libxc/xc_gnttab_compat.c
delete mode 100644 tools/libxc/xc_linux_osdep.c
delete mode 100644 tools/libxc/xenctrl_osdep_ENOSYS.c
--
2.1.4
^ permalink raw reply [flat|nested] 111+ messages in thread
* [PATCH XEN v5 01/23] tools/Rules.mk: Properly handle libraries with recursive dependencies.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 02/23] tools: Refactor "xentoollog" into its own library Ian Campbell
` (22 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
In tree libraries which link against other in tree libraries in a way
which is opaque to their callers need special handling, specifically
correct use of -Wl,-rpath-link for the recusively used libraries.
Currently this is rather simple, but up coming changes are going to
introduce transitive dependencies more than 1 step deep.
Introduce a SHDEPS idiom to contain all the recursive deps for a
library and include those in both LDLIBS (for linking) and SHLIB (for
recursive uses).
Try and document the whole thing.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
v5: Handled rebase over removal of stray L in rpath-link.
v4: Unmanged description of libbaz.
---
tools/Rules.mk | 74 ++++++++++++++++++++++++++++++++++++++++++++++++----------
1 file changed, 62 insertions(+), 12 deletions(-)
diff --git a/tools/Rules.mk b/tools/Rules.mk
index ffa5fd3..37b0aaf 100644
--- a/tools/Rules.mk
+++ b/tools/Rules.mk
@@ -34,25 +34,72 @@ INSTALL_SHLIB = : install-shlib-unsupported-fail
SYMLINK_SHLIB = : symlink-shlib-unsupported-fail
endif
+# Compiling and linking against in tree libraries.
+#
+# In order to compile and link against an in-tree library various
+# cpp/compiler/linker options are required.
+#
+# For example consider a library "libfoo" which itself uses two other
+# libraries:
+# libbar - whose use is entirely internal to libfoo and not exposed
+# to users of libfoo at all.
+# libbaz - whose use is entirely internal to libfoo but libfoo's
+# public headers include one or more of libbaz's
+# public headers. Users of libfoo are therefore transitively
+# using libbaz's header but not linking against libbaz.
+#
+# SHDEPS_libfoo: Flags for linking recursive dependencies of
+# libfoo. Must contain SHLIB for every library which
+# libfoo links against. So must contain both
+# $(SHLIB_libbar) and $(SHLIB_libbaz).
+#
+# SHLIB_libfoo: Flags for recursively linking against libfoo. Must
+# contains SHDEPS_libfoo and:
+# -Wl,-rpath-link=<directory containing libfoo.so>
+#
+# CFLAGS_libfoo: Flags for compiling against libfoo. Must add the
+# directories containing libfoo's headers to the
+# include path. Must recursively include
+# $(CFLAGS_libbaz), to satisfy the transitive inclusion
+# of the headers but not $(CFLAGS_libbar) since none of
+# libbar's headers are required to build against
+# libfoo.
+#
+# LDLIBS_libfoo: Flags for linking against libfoo. Must contain
+# $(SHDEPS_libfoo) and the path to libfoo.so
+#
+# Consumers of libfoo should include $(CFLAGS_libfoo) and
+# $(LDLIBS_libfoo) in their appropriate directories. They should not
+# include any CFLAGS or LDLIBS relating to libbar or libbaz unless
+# they use those libraries directly (not via libfoo) too.
+#
+# Consumers of libfoo should not directly use $(SHDEPS_libfoo) or
+# $(SHLIB_libfoo)
+
CFLAGS_libxenctrl = -I$(XEN_LIBXC)/include $(CFLAGS_xeninclude)
+SHDEPS_libxenctrl =
LDLIBS_libxenctrl = $(XEN_LIBXC)/libxenctrl$(libextension)
SHLIB_libxenctrl = -Wl,-rpath-link=$(XEN_LIBXC)
CFLAGS_libxenguest = -I$(XEN_LIBXC)/include $(CFLAGS_xeninclude)
-LDLIBS_libxenguest = $(XEN_LIBXC)/libxenguest$(libextension)
-SHLIB_libxenguest = -Wl,-rpath-link=$(XEN_LIBXC)
+SHDEPS_libxenguest =
+LDLIBS_libxenguest = $(SHDEPS_libxenguest) $(XEN_LIBXC)/libxenguest$(libextension)
+SHLIB_libxenguest = $(SHDEPS_libxenguest) -Wl,-rpath-link=$(XEN_LIBXC)
CFLAGS_libxenstore = -I$(XEN_XENSTORE)/include $(CFLAGS_xeninclude)
-LDLIBS_libxenstore = $(XEN_XENSTORE)/libxenstore$(libextension)
-SHLIB_libxenstore = -Wl,-rpath-link=$(XEN_XENSTORE)
+SHDEPS_libxenstore =
+LDLIBS_libxenstore = $(SHDEPS_libxenguest) $(XEN_XENSTORE)/libxenstore$(libextension)
+SHLIB_libxenstore = $(SHDEPS_libxenguest) -Wl,-rpath-link=$(XEN_XENSTORE)
CFLAGS_libxenstat = -I$(XEN_LIBXENSTAT)
-LDLIBS_libxenstat = $(SHLIB_libxenctrl) $(SHLIB_libxenstore) $(XEN_LIBXENSTAT)/libxenstat$(libextension)
-SHLIB_libxenstat = -Wl,-rpath-link=$(XEN_LIBXENSTAT)
+SHDEPS_libxenstat = $(SHLIB_libxenctrl) $(SHLIB_libxenstore)
+LDLIBS_libxenstat = $(SHDEPS_libxenstat) $(XEN_LIBXENSTAT)/libxenstat$(libextension)
+SHLIB_libxenstat = $(SHDEPS_libxenstat) -Wl,-rpath-link=$(XEN_LIBXENSTAT)
CFLAGS_libxenvchan = -I$(XEN_LIBVCHAN)
-LDLIBS_libxenvchan = $(SHLIB_libxenctrl) $(SHLIB_libxenstore) $(XEN_LIBVCHAN)/libxenvchan$(libextension)
-SHLIB_libxenvchan = -Wl,-rpath-link=$(XEN_LIBVCHAN)
+SHDEPS_libxenvchan = $(SHLIB_libxenctrl) $(SHLIB_libxenstore)
+LDLIBS_libxenvchan = $(SHDEPS_libxenvchan) $(XEN_LIBVCHAN)/libxenvchan$(libextension)
+SHLIB_libxenvchan = $(SHDEPS_libxenvchan) -Wl,-rpath-link=$(XEN_LIBVCHAN)
ifeq ($(debug),y)
# Disable optimizations and enable debugging information for macros
@@ -65,17 +112,20 @@ LIBXL_BLKTAP ?= $(CONFIG_BLKTAP2)
ifeq ($(LIBXL_BLKTAP),y)
CFLAGS_libblktapctl = -I$(XEN_BLKTAP2)/control -I$(XEN_BLKTAP2)/include $(CFLAGS_xeninclude)
-LDLIBS_libblktapctl = $(XEN_BLKTAP2)/control/libblktapctl$(libextension)
-SHLIB_libblktapctl = -Wl,-rpath-link=$(XEN_BLKTAP2)/control
+SHDEPS_libblktapctl =
+LDLIBS_libblktapctl = $(SHDEPS_libblktapctl) $(XEN_BLKTAP2)/control/libblktapctl$(libextension)
+SHLIB_libblktapctl = $(SHDEPS_libblktapctl) -Wl,-rpath-link=$(XEN_BLKTAP2)/control
else
CFLAGS_libblktapctl =
+SHDEPS_libblktapctl =
LDLIBS_libblktapctl =
SHLIB_libblktapctl =
endif
CFLAGS_libxenlight = -I$(XEN_XENLIGHT) $(CFLAGS_libxenctrl) $(CFLAGS_xeninclude)
-LDLIBS_libxenlight = $(XEN_XENLIGHT)/libxenlight$(libextension) $(SHLIB_libxenctrl) $(SHLIB_libxenstore) $(SHLIB_libblktapctl)
-SHLIB_libxenlight = -Wl,-rpath-link=$(XEN_XENLIGHT)
+SHDEPS_libxenlight = $(SHLIB_libxenctrl) $(SHLIB_libxenstore) $(SHLIB_libblktapctl)
+LDLIBS_libxenlight = $(SHDEPS_libxenlight) $(XEN_XENLIGHT)/libxenlight$(libextension)
+SHLIB_libxenlight = $(SHDEPS_libxenlight) -Wl,-rpath-link=$(XEN_XENLIGHT)
CFLAGS += -D__XEN_TOOLS__
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 02/23] tools: Refactor "xentoollog" into its own library
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 01/23] tools/Rules.mk: Properly handle libraries with recursive dependencies Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-13 15:05 ` Andrew Cooper
2015-11-09 12:00 ` [PATCH XEN v5 03/23] tools/libxc: Remove osdep indirection for xc_evtchn Ian Campbell
` (21 subsequent siblings)
23 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
In attempting to disaggregate libxenctrl I found that many of the
pieces were going to want access to this library, so split it out (as
it probably should always have been).
Various build adjustments are needed. In particular things which use
xtl_* themselves now need to explicity link against the library.
This has a nice side effect which is that users of libxl no longer
need to link against libxenctrl just to create a logger, which was
counter to the principal that applications using libxl shouldn't be
required to look behind the curtain. This means that xl no longer
links against libxenctrl.
The new library uses a version script to ensure that only expected
symbols are exported and to version them such that ABI guarantees can
be kept in the future.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
Must be applied with:
- "qemu-xen-traditional: Use xentoollog as a separate library" and a
corresponding QEMU_TAG update folded here.
- "mini-os: Include libxentoollog with libxc" and a corresponding bump
to MINIOS_UPSTREAM_REVISION folded in here.
v2: Update doc at same time
v3: Move into tools/libs/toollog.
---
.gitignore | 1 +
stubdom/Makefile | 24 ++++++++-
stubdom/grub/Makefile | 1 +
tools/Makefile | 3 ++
tools/Rules.mk | 14 +++--
tools/libs/Makefile | 7 +++
tools/libs/toollog/Makefile | 59 ++++++++++++++++++++++
tools/{libxc => libs/toollog}/include/xentoollog.h | 0
tools/libs/toollog/libxentoollog.map | 12 +++++
tools/{libxc => libs/toollog}/xtl_core.c | 0
tools/{libxc => libs/toollog}/xtl_logger_stdio.c | 0
tools/libxc/Makefile | 7 ++-
tools/libxl/Makefile | 15 +++---
tools/ocaml/Makefile.rules | 26 +++++-----
tools/ocaml/libs/xentoollog/Makefile | 6 +--
tools/ocaml/libs/xentoollog/genlevels.py | 2 +-
tools/python/setup.py | 5 +-
tools/xenpaging/Makefile | 2 +-
18 files changed, 148 insertions(+), 36 deletions(-)
create mode 100644 tools/libs/Makefile
create mode 100644 tools/libs/toollog/Makefile
rename tools/{libxc => libs/toollog}/include/xentoollog.h (100%)
create mode 100644 tools/libs/toollog/libxentoollog.map
rename tools/{libxc => libs/toollog}/xtl_core.c (100%)
rename tools/{libxc => libs/toollog}/xtl_logger_stdio.c (100%)
diff --git a/.gitignore b/.gitignore
index 91e1430..a2c85e1 100644
--- a/.gitignore
+++ b/.gitignore
@@ -58,6 +58,7 @@ stubdom/gcc-*
stubdom/include
stubdom/ioemu
stubdom/xenstore
+stubdom/libxentoollog-*
stubdom/libxc-*
stubdom/lwip-*
stubdom/mini-os-*
diff --git a/stubdom/Makefile b/stubdom/Makefile
index e1359cf..9c923dd 100644
--- a/stubdom/Makefile
+++ b/stubdom/Makefile
@@ -313,6 +313,11 @@ mk-headers-$(XEN_TARGET_ARCH): $(IOEMU_LINKFARM_TARGET)
ln -sf $(wildcard $(XEN_ROOT)/tools/include/xen-foreign/*) include/xen-foreign/ && \
$(MAKE) DESTDIR= -C include/xen-foreign/ && \
( [ -h include/xen/foreign ] || ln -sf ../xen-foreign include/xen/foreign )
+ mkdir -p libs-$(XEN_TARGET_ARCH)/toollog
+ [ -h libs-$(XEN_TARGET_ARCH)/toollog/Makefile ] || ( cd libs-$(XEN_TARGET_ARCH)/toollog && \
+ ln -sf $(XEN_ROOT)/tools/libs/toollog/include/*.h . && \
+ ln -sf $(XEN_ROOT)/tools/libs/toollog/*.c . && \
+ ln -sf $(XEN_ROOT)/tools/libs/toollog/Makefile . )
mkdir -p libxc-$(XEN_TARGET_ARCH)
[ -h libxc-$(XEN_TARGET_ARCH)/Makefile ] || ( cd libxc-$(XEN_TARGET_ARCH) && \
ln -sf $(XEN_ROOT)/tools/libxc/*.h . && \
@@ -336,12 +341,23 @@ $(TARGETS_MINIOS): mini-os-%:
done
#######
+# libxentoollog
+#######
+
+.PHONY: libxentoollog
+libxentoollog: libs-$(XEN_TARGET_ARCH)/toollog/libxentoollog.a
+libs-$(XEN_TARGET_ARCH)/toollog/libxentoollog.a: $(NEWLIB_STAMPFILE)
+ $(MAKE) -C $(XEN_ROOT)/tools/include
+ $(MAKE) DESTDIR= -C $(MINI_OS) links
+ CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/toollog
+
+#######
# libxc
#######
.PHONY: libxc
libxc: libxc-$(XEN_TARGET_ARCH)/libxenctrl.a libxc-$(XEN_TARGET_ARCH)/libxenguest.a
-libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: cross-zlib
+libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog cross-zlib
$(MAKE) -C $(XEN_ROOT)/tools/include
$(MAKE) DESTDIR= -C $(MINI_OS) links
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= CONFIG_LIBXC_MINIOS=y -C libxc-$(XEN_TARGET_ARCH)
@@ -515,6 +531,11 @@ clean:
$(MAKE) -C vtpmmgr clean
rm -fr grub-$(XEN_TARGET_ARCH)
rm -f $(STUBDOMPATH)
+ [ ! -e libs-$(XEN_TARGET_ARCH)/toollog/Makefile ] || $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/toollog clean
+ [ ! -e libs-$(XEN_TARGET_ARCH)/evtchn/Makefile ] || $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/evtchn clean
+ [ ! -e libs-$(XEN_TARGET_ARCH)/gnttab/Makefile ] || $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/gnttab clean
+ [ ! -e libs-$(XEN_TARGET_ARCH)/call/Makefile ] || $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/call clean
+ [ ! -e libs-$(XEN_TARGET_ARCH)/foreignmemory/Makefile ] || $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/foreignmemory clean
[ ! -e libxc-$(XEN_TARGET_ARCH)/Makefile ] || $(MAKE) DESTDIR= -C libxc-$(XEN_TARGET_ARCH) clean
-[ ! -d ioemu ] || $(MAKE) DESTDIR= -C ioemu clean
-[ ! -d xenstore ] || $(MAKE) DESTDIR= -C xenstore clean
@@ -525,6 +546,7 @@ crossclean: clean
rm -fr $(CROSS_ROOT)
rm -fr newlib-$(XEN_TARGET_ARCH)
rm -fr zlib-$(XEN_TARGET_ARCH) pciutils-$(XEN_TARGET_ARCH)
+ rm -fr libs-$(XEN_TARGET_ARCH)
rm -fr libxc-$(XEN_TARGET_ARCH) ioemu xenstore
rm -fr gmp-$(XEN_TARGET_ARCH)
rm -fr polarssl-$(XEN_TARGET_ARCH)
diff --git a/stubdom/grub/Makefile b/stubdom/grub/Makefile
index 934cc4c..26dff45 100644
--- a/stubdom/grub/Makefile
+++ b/stubdom/grub/Makefile
@@ -5,6 +5,7 @@ vpath %.c ../grub-upstream
BOOT=$(OBJ_DIR)/boot-$(XEN_TARGET_ARCH).o
+DEF_CPPFLAGS += -I$(XEN_ROOT)/tools/libs/toollog/include
DEF_CPPFLAGS += -I$(XEN_ROOT)/tools/libxc/include -I$(XEN_ROOT)/tools/include -I.
DEF_CPPFLAGS += -I../grub-upstream/stage1
DEF_CPPFLAGS += -I../grub-upstream/stage2
diff --git a/tools/Makefile b/tools/Makefile
index 820ca40..9f74ac7 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -3,6 +3,7 @@ include $(XEN_ROOT)/tools/Rules.mk
SUBDIRS-y :=
SUBDIRS-y += include
+SUBDIRS-y += libs
SUBDIRS-y += libxc
SUBDIRS-$(FLASK_ENABLE) += flask
SUBDIRS-y += xenstore
@@ -248,12 +249,14 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
--includedir=$(LIBEXEC_INC) \
--source-path=$$source \
--extra-cflags="-I$(XEN_ROOT)/tools/include \
+ -I$(XEN_ROOT)/tools/libs/toollog/include \
-I$(XEN_ROOT)/tools/libxc/include \
-I$(XEN_ROOT)/tools/xenstore/include \
-I$(XEN_ROOT)/tools/xenstore/compat/include \
$(EXTRA_CFLAGS_QEMU_XEN)" \
--extra-ldflags="-L$(XEN_ROOT)/tools/libxc \
-L$(XEN_ROOT)/tools/xenstore \
+ -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog \
$(QEMU_UPSTREAM_RPATH)" \
--bindir=$(LIBEXEC_BIN) \
--datadir=$(SHAREDIR)/qemu-xen \
diff --git a/tools/Rules.mk b/tools/Rules.mk
index 37b0aaf..7516ddd 100644
--- a/tools/Rules.mk
+++ b/tools/Rules.mk
@@ -10,6 +10,7 @@ export _INSTALL := $(INSTALL)
INSTALL = $(XEN_ROOT)/tools/cross-install
XEN_INCLUDE = $(XEN_ROOT)/tools/include
+XEN_LIBXENTOOLLOG = $(XEN_ROOT)/tools/libs/toollog
XEN_LIBXC = $(XEN_ROOT)/tools/libxc
XEN_XENLIGHT = $(XEN_ROOT)/tools/libxl
XEN_XENSTORE = $(XEN_ROOT)/tools/xenstore
@@ -76,10 +77,15 @@ endif
# Consumers of libfoo should not directly use $(SHDEPS_libfoo) or
# $(SHLIB_libfoo)
-CFLAGS_libxenctrl = -I$(XEN_LIBXC)/include $(CFLAGS_xeninclude)
-SHDEPS_libxenctrl =
-LDLIBS_libxenctrl = $(XEN_LIBXC)/libxenctrl$(libextension)
-SHLIB_libxenctrl = -Wl,-rpath-link=$(XEN_LIBXC)
+CFLAGS_libxentoollog = -I$(XEN_LIBXENTOOLLOG)/include $(CFLAGS_xeninclude)
+SHDEPS_libxentoollog =
+LDLIBS_libxentoollog = $(XEN_LIBXENTOOLLOG)/libxentoollog$(libextension)
+SHLIB_libxentoollog = -Wl,-rpath-link=$(XEN_LIBXENTOOLLOG)
+
+CFLAGS_libxenctrl = -I$(XEN_LIBXC)/include $(CFLAGS_libxentoollog) $(CFLAGS_xeninclude)
+SHDEPS_libxenctrl = $(SHLIB_libxentoollog)
+LDLIBS_libxenctrl = $(SHDEPS_libxenctrl) $(XEN_LIBXC)/libxenctrl$(libextension)
+SHLIB_libxenctrl = $(SHDEPS_libxenctrl) -Wl,-rpath-link=$(XEN_LIBXC)
CFLAGS_libxenguest = -I$(XEN_LIBXC)/include $(CFLAGS_xeninclude)
SHDEPS_libxenguest =
diff --git a/tools/libs/Makefile b/tools/libs/Makefile
new file mode 100644
index 0000000..73be500
--- /dev/null
+++ b/tools/libs/Makefile
@@ -0,0 +1,7 @@
+XEN_ROOT = $(CURDIR)/../..
+include $(XEN_ROOT)/tools/Rules.mk
+
+SUBDIRS-y :=
+SUBDIRS-y += toollog
+
+all clean install distclean: %: subdirs-%
diff --git a/tools/libs/toollog/Makefile b/tools/libs/toollog/Makefile
new file mode 100644
index 0000000..bd12403
--- /dev/null
+++ b/tools/libs/toollog/Makefile
@@ -0,0 +1,59 @@
+XEN_ROOT = $(CURDIR)/../../..
+include $(XEN_ROOT)/tools/Rules.mk
+
+MAJOR = 1
+MINOR = 0
+SHLIB_LDFLAGS += -Wl,--version-script=libxentoollog.map
+
+CFLAGS += -Werror -Wmissing-prototypes
+CFLAGS += -I./include
+
+SRCS-y += xtl_core.c
+SRCS-y += xtl_logger_stdio.c
+
+LIB_OBJS := $(patsubst %.c,%.o,$(SRCS-y))
+PIC_OBJS := $(patsubst %.c,%.opic,$(SRCS-y))
+
+LIB := libxentoollog.a
+ifneq ($(nosharedlibs),y)
+LIB += libxentoollog.so
+endif
+
+.PHONY: all
+all: build
+
+.PHONY: build
+build:
+ $(MAKE) libs
+
+.PHONY: libs
+libs: $(LIB)
+
+libxentoollog.a: $(LIB_OBJS)
+ $(AR) rc $@ $^
+
+libxentoollog.so: libxentoollog.so.$(MAJOR)
+ $(SYMLINK_SHLIB) $< $@
+libxentoollog.so.$(MAJOR): libxentoollog.so.$(MAJOR).$(MINOR)
+ $(SYMLINK_SHLIB) $< $@
+
+libxentoollog.so.$(MAJOR).$(MINOR): $(PIC_OBJS) libxentoollog.map
+ $(CC) $(LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxentoollog.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $(PIC_OBJS) $(APPEND_LDFLAGS)
+
+.PHONY: install
+install: build
+ $(INSTALL_DIR) $(DESTDIR)$(libdir)
+ $(INSTALL_DIR) $(DESTDIR)$(includedir)
+ $(INSTALL_SHLIB) libxentoollog.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)
+ $(INSTALL_DATA) libxentoollog.a $(DESTDIR)$(libdir)
+ $(SYMLINK_SHLIB) libxentoollog.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxentoollog.so.$(MAJOR)
+ $(SYMLINK_SHLIB) libxentoollog.so.$(MAJOR) $(DESTDIR)$(libdir)/libxentoollog.so
+ $(INSTALL_DATA) include/xentoollog.h $(DESTDIR)$(includedir)
+
+.PHONY: TAGS
+TAGS:
+ etags -t *.c *.h
+
+.PHONY: clean
+clean:
+ rm -rf *.rpm $(LIB) *~ $(DEPS) $(LIB_OBJS) $(PIC_OBJS)
diff --git a/tools/libxc/include/xentoollog.h b/tools/libs/toollog/include/xentoollog.h
similarity index 100%
rename from tools/libxc/include/xentoollog.h
rename to tools/libs/toollog/include/xentoollog.h
diff --git a/tools/libs/toollog/libxentoollog.map b/tools/libs/toollog/libxentoollog.map
new file mode 100644
index 0000000..c183cf5
--- /dev/null
+++ b/tools/libs/toollog/libxentoollog.map
@@ -0,0 +1,12 @@
+VERS_1.0 {
+ global:
+ xtl_createlogger_stdiostream;
+ xtl_level_to_string;
+ xtl_log;
+ xtl_logger_destroy;
+ xtl_logv;
+ xtl_progress;
+ xtl_stdiostream_adjust_flags;
+ xtl_stdiostream_set_minlevel;
+ local: *; /* Do not expose anything by default */
+};
diff --git a/tools/libxc/xtl_core.c b/tools/libs/toollog/xtl_core.c
similarity index 100%
rename from tools/libxc/xtl_core.c
rename to tools/libs/toollog/xtl_core.c
diff --git a/tools/libxc/xtl_logger_stdio.c b/tools/libs/toollog/xtl_logger_stdio.c
similarity index 100%
rename from tools/libxc/xtl_logger_stdio.c
rename to tools/libs/toollog/xtl_logger_stdio.c
diff --git a/tools/libxc/Makefile b/tools/libxc/Makefile
index 818f2e4..940708f 100644
--- a/tools/libxc/Makefile
+++ b/tools/libxc/Makefile
@@ -39,8 +39,6 @@ CTRL_SRCS-y += xc_memshr.c
CTRL_SRCS-y += xc_hcall_buf.c
CTRL_SRCS-y += xc_foreign_memory.c
CTRL_SRCS-y += xc_kexec.c
-CTRL_SRCS-y += xtl_core.c
-CTRL_SRCS-y += xtl_logger_stdio.c
CTRL_SRCS-y += xc_resource.c
CTRL_SRCS-$(CONFIG_X86) += xc_psr.c
CTRL_SRCS-$(CONFIG_X86) += xc_pagetab.c
@@ -112,6 +110,7 @@ CFLAGS += -I. -I./include $(CFLAGS_xeninclude)
CFLAGS-$(CONFIG_Linux) += -D_GNU_SOURCE
CFLAGS += $(PTHREAD_CFLAGS)
+CFLAGS += $(CFLAGS_libxentoollog)
CTRL_LIB_OBJS := $(patsubst %.c,%.o,$(CTRL_SRCS-y))
CTRL_PIC_OBJS := $(patsubst %.c,%.opic,$(CTRL_SRCS-y))
@@ -165,7 +164,7 @@ install: build
$(INSTALL_DATA) libxenctrl.a $(DESTDIR)$(libdir)
$(SYMLINK_SHLIB) libxenctrl.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxenctrl.so.$(MAJOR)
$(SYMLINK_SHLIB) libxenctrl.so.$(MAJOR) $(DESTDIR)$(libdir)/libxenctrl.so
- $(INSTALL_DATA) include/xenctrl.h include/xenctrlosdep.h include/xentoollog.h $(DESTDIR)$(includedir)
+ $(INSTALL_DATA) include/xenctrl.h include/xenctrlosdep.h $(DESTDIR)$(includedir)
$(INSTALL_SHLIB) libxenguest.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)
$(INSTALL_DATA) libxenguest.a $(DESTDIR)$(libdir)
$(SYMLINK_SHLIB) libxenguest.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxenguest.so.$(MAJOR)
@@ -208,7 +207,7 @@ libxenctrl.so.$(MAJOR): libxenctrl.so.$(MAJOR).$(MINOR)
$(SYMLINK_SHLIB) $< $@
libxenctrl.so.$(MAJOR).$(MINOR): $(CTRL_PIC_OBJS)
- $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(DLOPEN_LIBS) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(DLOPEN_LIBS) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
# libxenguest
diff --git a/tools/libxl/Makefile b/tools/libxl/Makefile
index 6ff5bee..2abae0c 100644
--- a/tools/libxl/Makefile
+++ b/tools/libxl/Makefile
@@ -20,11 +20,12 @@ LIBUUID_LIBS += -luuid
endif
LIBXL_LIBS =
-LIBXL_LIBS = $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) $(LDLIBS_libblktapctl) $(PTYFUNCS_LIBS) $(LIBUUID_LIBS)
+LIBXL_LIBS = $(LDLIBS_libxentoollog) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) $(LDLIBS_libblktapctl) $(PTYFUNCS_LIBS) $(LIBUUID_LIBS)
ifeq ($(CONFIG_REMUS_NETBUF),y)
LIBXL_LIBS += $(LIBNL3_LIBS)
endif
+CFLAGS_LIBXL += $(CFLAGS_libxentoollog)
CFLAGS_LIBXL += $(CFLAGS_libxenctrl)
CFLAGS_LIBXL += $(CFLAGS_libxenguest)
CFLAGS_LIBXL += $(CFLAGS_libxenstore)
@@ -149,7 +150,7 @@ CFLAGS_XL += -Wshadow
XL_OBJS = xl.o xl_cmdimpl.o xl_cmdtable.o xl_sxp.o
$(XL_OBJS) $(TEST_PROG_OBJS) _libxl.api-for-check: \
- CFLAGS += $(CFLAGS_libxenctrl) # For xentoollog.h
+ CFLAGS += $(CFLAGS_libxentoollog)
$(XL_OBJS): CFLAGS += $(CFLAGS_XL)
$(XL_OBJS): CFLAGS += -include $(XEN_ROOT)/tools/config.h # libxl_json.h needs it.
@@ -253,19 +254,19 @@ libxlutil.a: $(LIBXLU_OBJS)
$(AR) rcs libxlutil.a $^
xl: $(XL_OBJS) libxlutil.so libxenlight.so
- $(CC) $(LDFLAGS) -o $@ $(XL_OBJS) libxlutil.so $(LDLIBS_libxenlight) $(LDLIBS_libxenctrl) -lyajl $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -o $@ $(XL_OBJS) libxlutil.so $(LDLIBS_libxenlight) $(LDLIBS_libxentoollog) -lyajl $(APPEND_LDFLAGS)
xen-init-dom0: $(XEN_INIT_DOM0_OBJS) libxenlight.so
- $(CC) $(LDFLAGS) -o $@ $(XEN_INIT_DOM0_OBJS) $(LDLIBS_libxenstore) $(LDLIBS_libxenlight) $(LDLIBS_libxenctrl) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -o $@ $(XEN_INIT_DOM0_OBJS) $(LDLIBS_libxenstore) $(LDLIBS_libxenlight) $(LDLIBS_libxentoollog) $(APPEND_LDFLAGS)
test_%: test_%.o test_common.o libxlutil.so libxenlight_test.so
- $(CC) $(LDFLAGS) -o $@ $^ $(filter-out %libxenlight.so, $(LDLIBS_libxenlight)) $(LDLIBS_libxenctrl) -lyajl $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -o $@ $^ $(filter-out %libxenlight.so, $(LDLIBS_libxenlight)) $(LDLIBS_libxentoollog) -lyajl $(APPEND_LDFLAGS)
libxl-save-helper: $(SAVE_HELPER_OBJS) libxenlight.so
- $(CC) $(LDFLAGS) -o $@ $(SAVE_HELPER_OBJS) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -o $@ $(SAVE_HELPER_OBJS) $(LDLIBS_libxentoollog) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(APPEND_LDFLAGS)
testidl: testidl.o libxlutil.so libxenlight.so
- $(CC) $(LDFLAGS) -o $@ testidl.o libxlutil.so $(LDLIBS_libxenlight) $(LDLIBS_libxenctrl) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -o $@ testidl.o libxlutil.so $(LDLIBS_libxenlight) $(LDLIBS_libxentoollog) $(APPEND_LDFLAGS)
$(PKG_CONFIG): % : %.in Makefile
@sed -e 's/@@version@@/$(MAJOR).$(MINOR)/g' < $< > $@.new
diff --git a/tools/ocaml/Makefile.rules b/tools/ocaml/Makefile.rules
index 1796060..45e71f2 100644
--- a/tools/ocaml/Makefile.rules
+++ b/tools/ocaml/Makefile.rules
@@ -1,18 +1,18 @@
-ifdef V
- ifeq ("$(origin V)", "command line")
- BUILD_VERBOSE = $(V)
- endif
-endif
-ifndef BUILD_VERBOSE
- BUILD_VERBOSE = 0
-endif
-ifeq ($(BUILD_VERBOSE),1)
+#ifdef V
+# ifeq ("$(origin V)", "command line")
+# BUILD_VERBOSE = $(V)
+# endif
+#endif
+#ifndef BUILD_VERBOSE
+# BUILD_VERBOSE = 0
+#endif
+#ifeq ($(BUILD_VERBOSE),1)
E = @true
Q =
-else
- E = @echo
- Q = @
-endif
+#else
+# E = @echo
+# Q = @
+#endif
.NOTPARALLEL:
diff --git a/tools/ocaml/libs/xentoollog/Makefile b/tools/ocaml/libs/xentoollog/Makefile
index 666eb66..8ae0a78 100644
--- a/tools/ocaml/libs/xentoollog/Makefile
+++ b/tools/ocaml/libs/xentoollog/Makefile
@@ -5,7 +5,7 @@ include $(TOPLEVEL)/common.make
# allow mixed declarations and code
CFLAGS += -Wno-declaration-after-statement
-CFLAGS += $(CFLAGS_libxenctrl) $(CFLAGS_libxenguest)
+CFLAGS += $(CFLAGS_libxentoollog)
CFLAGS += $(APPEND_CFLAGS)
OCAMLINCLUDE +=
@@ -13,7 +13,7 @@ OBJS = xentoollog
INTF = xentoollog.cmi
LIBS = xentoollog.cma xentoollog.cmxa
-LIBS_xentoollog = $(LDLIBS_libxenctrl)
+LIBS_xentoollog = $(LDLIBS_libxentoollog)
xentoollog_OBJS = $(OBJS)
xentoollog_C_OBJS = xentoollog_stubs
@@ -49,7 +49,7 @@ xentoollog.mli: xentoollog.mli.in _xtl_levels.mli.in
libs: $(LIBS)
-_xtl_levels.ml.in _xtl_levels.mli.in _xtl_levels.inc: genlevels.py $(XEN_ROOT)/tools/libxc/include/xentoollog.h
+_xtl_levels.ml.in _xtl_levels.mli.in _xtl_levels.inc: genlevels.py $(XEN_ROOT)/tools/libs/toollog/include/xentoollog.h
$(PYTHON) genlevels.py _xtl_levels.mli.in _xtl_levels.ml.in _xtl_levels.inc
.PHONY: install
diff --git a/tools/ocaml/libs/xentoollog/genlevels.py b/tools/ocaml/libs/xentoollog/genlevels.py
index 65d334f..8c233c5 100755
--- a/tools/ocaml/libs/xentoollog/genlevels.py
+++ b/tools/ocaml/libs/xentoollog/genlevels.py
@@ -3,7 +3,7 @@
import sys
def read_levels():
- f = open('../../../libxc/include/xentoollog.h', 'r')
+ f = open('../../../libs/toollog/include/xentoollog.h', 'r')
levels = []
record = False
diff --git a/tools/python/setup.py b/tools/python/setup.py
index fdba866..9771cc4 100644
--- a/tools/python/setup.py
+++ b/tools/python/setup.py
@@ -7,16 +7,17 @@ XEN_ROOT = "../.."
extra_compile_args = [ "-fno-strict-aliasing", "-Werror" ]
PATH_XEN = XEN_ROOT + "/tools/include"
+PATH_LIBXENTOOLLOG = XEN_ROOT + "/tools/libs/toollog"
PATH_LIBXC = XEN_ROOT + "/tools/libxc"
PATH_LIBXL = XEN_ROOT + "/tools/libxl"
PATH_XENSTORE = XEN_ROOT + "/tools/xenstore"
xc = Extension("xc",
extra_compile_args = extra_compile_args,
- include_dirs = [ PATH_XEN, PATH_LIBXC + "/include", "xen/lowlevel/xc" ],
+ include_dirs = [ PATH_XEN, PATH_LIBXENTOOLLOG + "/include", PATH_LIBXC + "/include", "xen/lowlevel/xc" ],
library_dirs = [ PATH_LIBXC ],
libraries = [ "xenctrl", "xenguest" ],
- depends = [ PATH_LIBXC + "/libxenctrl.so", PATH_LIBXC + "/libxenguest.so" ],
+ depends = [ PATH_LIBXC + "/libxenctrl.so", PATH_LIBXC + "/libxenguest.so", "-Wl,-rpath-link="+PATH_LIBXENTOOLLOG ],
sources = [ "xen/lowlevel/xc/xc.c" ])
xs = Extension("xs",
diff --git a/tools/xenpaging/Makefile b/tools/xenpaging/Makefile
index 2407a30..e63d894 100644
--- a/tools/xenpaging/Makefile
+++ b/tools/xenpaging/Makefile
@@ -3,7 +3,7 @@ include $(XEN_ROOT)/tools/Rules.mk
# xenpaging.c and file_ops.c incorrectly use libxc internals
CFLAGS += $(CFLAGS_libxenctrl) $(CFLAGS_libxenstore) $(PTHREAD_CFLAGS) -I$(XEN_ROOT)/tools/libxc
-LDLIBS += $(LDLIBS_libxenctrl) $(LDLIBS_libxenstore) $(PTHREAD_LIBS)
+LDLIBS += $(LDLIBS_libxentoollog) $(LDLIBS_libxenctrl) $(LDLIBS_libxenstore) $(PTHREAD_LIBS)
LDFLAGS += $(PTHREAD_LDFLAGS)
POLICY = default
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 03/23] tools/libxc: Remove osdep indirection for xc_evtchn
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 01/23] tools/Rules.mk: Properly handle libraries with recursive dependencies Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 02/23] tools: Refactor "xentoollog" into its own library Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 04/23] tools: Refactor /dev/xen/evtchn wrappers into libxenevtchn Ian Campbell
` (20 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
The alternative backend (a xen-api/xapi shim) is no longer around and
so this stuff is now just baggage which is getting in the way of
refactoring libxenctrl.
Note that the intention is to move this into a separate library
shortly.
Nested virt probably suffices for this use case now.
One incorrect instance of using xc_interface where xc_evtchn (in ocaml
stubs) is removed, this used to work because they were typedefs to the
same struct, but is no longer permitted.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libxc/include/xenctrl.h | 2 +-
tools/libxc/include/xenctrlosdep.h | 16 -----
tools/libxc/xc_evtchn.c | 45 --------------
tools/libxc/xc_freebsd_osdep.c | 79 ++++++++----------------
tools/libxc/xc_linux_osdep.c | 76 +++++++++--------------
tools/libxc/xc_minios.c | 63 ++++++++-----------
tools/libxc/xc_netbsd.c | 72 ++++++++--------------
tools/libxc/xc_private.c | 37 +++++++++---
tools/libxc/xc_private.h | 17 ++++++
tools/libxc/xc_solaris.c | 71 ++++++++--------------
tools/libxc/xenctrl_osdep_ENOSYS.c | 87 ---------------------------
tools/ocaml/libs/eventchn/xeneventchn_stubs.c | 4 +-
12 files changed, 180 insertions(+), 389 deletions(-)
diff --git a/tools/libxc/include/xenctrl.h b/tools/libxc/include/xenctrl.h
index 2fec1fb..29d55ad 100644
--- a/tools/libxc/include/xenctrl.h
+++ b/tools/libxc/include/xenctrl.h
@@ -117,7 +117,7 @@
*/
typedef struct xc_interface_core xc_interface;
-typedef struct xc_interface_core xc_evtchn;
+typedef struct xenevtchn_handle xc_evtchn;
typedef struct xc_interface_core xc_gnttab;
typedef struct xc_interface_core xc_gntshr;
diff --git a/tools/libxc/include/xenctrlosdep.h b/tools/libxc/include/xenctrlosdep.h
index 5121d9b..89564e1 100644
--- a/tools/libxc/include/xenctrlosdep.h
+++ b/tools/libxc/include/xenctrlosdep.h
@@ -51,7 +51,6 @@
enum xc_osdep_type {
XC_OSDEP_PRIVCMD,
- XC_OSDEP_EVTCHN,
XC_OSDEP_GNTTAB,
XC_OSDEP_GNTSHR,
};
@@ -90,21 +89,6 @@ struct xc_osdep_ops
int nentries);
} privcmd;
struct {
- int (*fd)(xc_evtchn *xce, xc_osdep_handle h);
-
- int (*notify)(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port);
-
- evtchn_port_or_error_t (*bind_unbound_port)(xc_evtchn *xce, xc_osdep_handle h, int domid);
- evtchn_port_or_error_t (*bind_interdomain)(xc_evtchn *xce, xc_osdep_handle h, int domid,
- evtchn_port_t remote_port);
- evtchn_port_or_error_t (*bind_virq)(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq);
-
- int (*unbind)(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port);
-
- evtchn_port_or_error_t (*pending)(xc_evtchn *xce, xc_osdep_handle h);
- int (*unmask)(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port);
- } evtchn;
- struct {
#define XC_GRANT_MAP_SINGLE_DOMAIN 0x1
void *(*grant_map)(xc_gnttab *xcg, xc_osdep_handle h,
uint32_t count, int flags, int prot,
diff --git a/tools/libxc/xc_evtchn.c b/tools/libxc/xc_evtchn.c
index 15f0580..ae2fe1a 100644
--- a/tools/libxc/xc_evtchn.c
+++ b/tools/libxc/xc_evtchn.c
@@ -77,51 +77,6 @@ int xc_evtchn_status(xc_interface *xch, xc_evtchn_status_t *status)
sizeof(*status), 1);
}
-int xc_evtchn_fd(xc_evtchn *xce)
-{
- return xce->ops->u.evtchn.fd(xce, xce->ops_handle);
-}
-
-int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
-{
- return xce->ops->u.evtchn.notify(xce, xce->ops_handle, port);
-}
-
-evtchn_port_or_error_t
-xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
-{
- return xce->ops->u.evtchn.bind_unbound_port(xce, xce->ops_handle, domid);
-}
-
-evtchn_port_or_error_t
-xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
- evtchn_port_t remote_port)
-{
- return xce->ops->u.evtchn.bind_interdomain(xce, xce->ops_handle, domid, remote_port);
-}
-
-evtchn_port_or_error_t
-xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
-{
- return xce->ops->u.evtchn.bind_virq(xce, xce->ops_handle, virq);
-}
-
-int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
-{
- return xce->ops->u.evtchn.unbind(xce, xce->ops_handle, port);
-}
-
-evtchn_port_or_error_t
-xc_evtchn_pending(xc_evtchn *xce)
-{
- return xce->ops->u.evtchn.pending(xce, xce->ops_handle);
-}
-
-int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
-{
- return xce->ops->u.evtchn.unmask(xce, xce->ops_handle, port);
-}
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xc_freebsd_osdep.c b/tools/libxc/xc_freebsd_osdep.c
index 4d31a1e..4323e16 100644
--- a/tools/libxc/xc_freebsd_osdep.c
+++ b/tools/libxc/xc_freebsd_osdep.c
@@ -252,34 +252,32 @@ static struct xc_osdep_ops freebsd_privcmd_ops = {
};
/*-------------------------- Evtchn device interface -------------------------*/
-static xc_osdep_handle
-freebsd_evtchn_open(xc_evtchn *xce)
+int osdep_evtchn_open(xc_evtchn *xce)
{
int fd = open(EVTCHN_DEV, O_RDWR);
if ( fd == -1 )
- return XC_OSDEP_OPEN_ERROR;
-
- return (xc_osdep_handle)fd;
+ return -1;
+ xce->fd = fd;
+ return 0;
}
-static int
-freebsd_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
+int osdep_evtchn_close(xc_evtchn *xce)
{
- int fd = (int)h;
- return close(fd);
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
}
-static int
-freebsd_evtchn_fd(xc_evtchn *xce, xc_osdep_handle h)
+int xc_evtchn_fd(xc_evtchn *xce)
{
- return (int)h;
+ return xce->fd;
}
/*------------------------------ Evtchn interface ----------------------------*/
-static int
-freebsd_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_notify notify;
notify.port = port;
@@ -287,10 +285,9 @@ freebsd_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
}
-static evtchn_port_or_error_t
-freebsd_evtchn_bind_unbound_port(xc_evtchn *xce, xc_osdep_handle h, int domid)
+evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
{
- int ret, fd = (int)h;
+ int ret, fd = xce->fd;
struct ioctl_evtchn_bind_unbound_port bind;
bind.remote_domain = domid;
@@ -299,11 +296,10 @@ freebsd_evtchn_bind_unbound_port(xc_evtchn *xce, xc_osdep_handle h, int domid)
return ( ret == 0 ) ? bind.port : ret;
}
-static evtchn_port_or_error_t
-freebsd_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
- evtchn_port_t remote_port)
+evtchn_port_or_error_t
+xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid, evtchn_port_t remote_port)
{
- int ret, fd = (int)h;
+ int ret, fd = xce->fd;
struct ioctl_evtchn_bind_interdomain bind;
bind.remote_domain = domid;
@@ -313,10 +309,9 @@ freebsd_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
return ( ret == 0 ) ? bind.port : ret;
}
-static evtchn_port_or_error_t
-freebsd_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
+evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
{
- int ret, fd = (int)h;
+ int ret, fd = xce->fd;
struct ioctl_evtchn_bind_virq bind;
bind.virq = virq;
@@ -325,10 +320,9 @@ freebsd_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
return ( ret == 0 ) ? bind.port : ret;
}
-static int
-freebsd_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_unbind unbind;
unbind.port = port;
@@ -336,10 +330,9 @@ freebsd_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
}
-static evtchn_port_or_error_t
-freebsd_evtchn_pending(xc_evtchn *xce, xc_osdep_handle h)
+evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
{
- int fd = (int)h;
+ int fd = xce->fd;
evtchn_port_t port;
if ( read(fd, &port, sizeof(port)) != sizeof(port) )
@@ -348,33 +341,15 @@ freebsd_evtchn_pending(xc_evtchn *xce, xc_osdep_handle h)
return port;
}
-static int
-freebsd_evtchn_unmask(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
if ( write(fd, &port, sizeof(port)) != sizeof(port) )
return -1;
return 0;
}
-/*----------------------------- Evtchn handlers ------------------------------*/
-static struct xc_osdep_ops freebsd_evtchn_ops = {
- .open = &freebsd_evtchn_open,
- .close = &freebsd_evtchn_close,
-
- .u.evtchn = {
- .fd = &freebsd_evtchn_fd,
- .notify = &freebsd_evtchn_notify,
- .bind_unbound_port = &freebsd_evtchn_bind_unbound_port,
- .bind_interdomain = &freebsd_evtchn_bind_interdomain,
- .bind_virq = &freebsd_evtchn_bind_virq,
- .unbind = &freebsd_evtchn_unbind,
- .pending = &freebsd_evtchn_pending,
- .unmask = &freebsd_evtchn_unmask,
- },
-};
-
/*---------------------------- FreeBSD interface -----------------------------*/
static struct xc_osdep_ops *
freebsd_osdep_init(xc_interface *xch, enum xc_osdep_type type)
@@ -383,8 +358,6 @@ freebsd_osdep_init(xc_interface *xch, enum xc_osdep_type type)
{
case XC_OSDEP_PRIVCMD:
return &freebsd_privcmd_ops;
- case XC_OSDEP_EVTCHN:
- return &freebsd_evtchn_ops;
default:
return NULL;
}
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index 76c55ff..c1f34de 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -38,10 +38,9 @@
#include "xenctrl.h"
#include "xenctrlosdep.h"
+#include "xc_private.h"
+
#define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & ~((1UL<<(_w))-1))
-#define ERROR(_m, _a...) xc_osdep_log(xch,XTL_ERROR,XC_INTERNAL_ERROR,_m , ## _a )
-#define PERROR(_m, _a...) xc_osdep_log(xch,XTL_ERROR,XC_INTERNAL_ERROR,_m \
- " (%d = %s)", ## _a , errno, xc_strerror(xch, errno))
static xc_osdep_handle linux_privcmd_open(xc_interface *xch)
{
@@ -456,29 +455,31 @@ static struct xc_osdep_ops linux_privcmd_ops = {
#define DEVXEN "/dev/xen/"
-static xc_osdep_handle linux_evtchn_open(xc_evtchn *xce)
+int osdep_evtchn_open(xc_evtchn *xce)
{
int fd = open(DEVXEN "evtchn", O_RDWR);
if ( fd == -1 )
- return XC_OSDEP_OPEN_ERROR;
-
- return (xc_osdep_handle)fd;
+ return -1;
+ xce->fd = fd;
+ return 0;
}
-static int linux_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
+int osdep_evtchn_close(xc_evtchn *xce)
{
- int fd = (int)h;
- return close(fd);
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
}
-static int linux_evtchn_fd(xc_evtchn *xce, xc_osdep_handle h)
+int xc_evtchn_fd(xc_evtchn *xce)
{
- return (int)h;
+ return xce->fd;
}
-static int linux_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_notify notify;
notify.port = port;
@@ -486,10 +487,9 @@ static int linux_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t
return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
}
-static evtchn_port_or_error_t
-linux_evtchn_bind_unbound_port(xc_evtchn *xce, xc_osdep_handle h, int domid)
+evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_unbound_port bind;
bind.remote_domain = domid;
@@ -497,11 +497,10 @@ linux_evtchn_bind_unbound_port(xc_evtchn *xce, xc_osdep_handle h, int domid)
return ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
}
-static evtchn_port_or_error_t
-linux_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
- evtchn_port_t remote_port)
+evtchn_port_or_error_t xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
+ evtchn_port_t remote_port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_interdomain bind;
bind.remote_domain = domid;
@@ -510,10 +509,9 @@ linux_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
return ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
}
-static evtchn_port_or_error_t
-linux_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
+evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_virq bind;
bind.virq = virq;
@@ -521,9 +519,9 @@ linux_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
return ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
}
-static int linux_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_unbind unbind;
unbind.port = port;
@@ -531,9 +529,9 @@ static int linux_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t
return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
}
-static evtchn_port_or_error_t linux_evtchn_pending(xc_evtchn *xce, xc_osdep_handle h)
+evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
{
- int fd = (int)h;
+ int fd = xce->fd;
evtchn_port_t port;
if ( read(fd, &port, sizeof(port)) != sizeof(port) )
@@ -542,31 +540,15 @@ static evtchn_port_or_error_t linux_evtchn_pending(xc_evtchn *xce, xc_osdep_hand
return port;
}
-static int linux_evtchn_unmask(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
if ( write(fd, &port, sizeof(port)) != sizeof(port) )
return -1;
return 0;
}
-static struct xc_osdep_ops linux_evtchn_ops = {
- .open = &linux_evtchn_open,
- .close = &linux_evtchn_close,
-
- .u.evtchn = {
- .fd = &linux_evtchn_fd,
- .notify = &linux_evtchn_notify,
- .bind_unbound_port = &linux_evtchn_bind_unbound_port,
- .bind_interdomain = &linux_evtchn_bind_interdomain,
- .bind_virq = &linux_evtchn_bind_virq,
- .unbind = &linux_evtchn_unbind,
- .pending = &linux_evtchn_pending,
- .unmask = &linux_evtchn_unmask,
- },
-};
-
static xc_osdep_handle linux_gnttab_open(xc_gnttab *xcg)
{
int fd = open(DEVXEN "gntdev", O_RDWR);
@@ -870,8 +852,6 @@ static struct xc_osdep_ops *linux_osdep_init(xc_interface *xch, enum xc_osdep_ty
{
case XC_OSDEP_PRIVCMD:
return &linux_privcmd_ops;
- case XC_OSDEP_EVTCHN:
- return &linux_evtchn_ops;
case XC_OSDEP_GNTTAB:
return &linux_gnttab_ops;
case XC_OSDEP_GNTSHR:
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index 4f6498f..4ace1b5 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -219,20 +219,23 @@ static void port_dealloc(struct evtchn_port_info *port_info) {
free(port_info);
}
-static xc_osdep_handle minios_evtchn_open(xc_evtchn *xce)
+int osdep_evtchn_open(xc_evtchn *xce)
{
int fd = alloc_fd(FTYPE_EVTCHN);
if ( fd == -1 )
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
LIST_INIT(&files[fd].evtchn.ports);
+ xce->fd = fd;
printf("evtchn_open() -> %d\n", fd);
- return (xc_osdep_handle)fd;
+ return 0;
}
-static int minios_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
+int osdep_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
{
- int fd = (int)h;
- return close(fd);
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
}
void minios_evtchn_close_fd(int fd)
@@ -244,12 +247,12 @@ void minios_evtchn_close_fd(int fd)
files[fd].type = FTYPE_NONE;
}
-static int minios_evtchn_fd(xc_evtchn *xce, xc_osdep_handle h)
+int xc_evtchn_fd(xc_evtchn *xce)
{
- return (int)h;
+ return xce->fd;
}
-static int minios_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
{
int ret;
@@ -281,9 +284,9 @@ static void evtchn_handler(evtchn_port_t port, struct pt_regs *regs, void *data)
wake_up(&event_queue);
}
-static evtchn_port_or_error_t minios_evtchn_bind_unbound_port(xc_evtchn *xce, xc_osdep_handle h, int domid)
+evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct evtchn_port_info *port_info;
int ret;
evtchn_port_t port;
@@ -308,10 +311,10 @@ static evtchn_port_or_error_t minios_evtchn_bind_unbound_port(xc_evtchn *xce, xc
return port;
}
-static evtchn_port_or_error_t minios_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
- evtchn_port_t remote_port)
+evtchn_port_or_error_t xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
+ evtchn_port_t remote_port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct evtchn_port_info *port_info;
evtchn_port_t local_port;
int ret;
@@ -336,9 +339,9 @@ static evtchn_port_or_error_t minios_evtchn_bind_interdomain(xc_evtchn *xce, xc_
return local_port;
}
-static int minios_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct evtchn_port_info *port_info;
LIST_FOREACH(port_info, &files[fd].evtchn.ports, list) {
@@ -352,9 +355,9 @@ static int minios_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t
return -1;
}
-static evtchn_port_or_error_t minios_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
+evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct evtchn_port_info *port_info;
evtchn_port_t port;
@@ -377,9 +380,9 @@ static evtchn_port_or_error_t minios_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_h
return port;
}
-static evtchn_port_or_error_t minios_evtchn_pending(xc_evtchn *xce, xc_osdep_handle h)
+evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct evtchn_port_info *port_info;
unsigned long flags;
evtchn_port_t ret = -1;
@@ -402,28 +405,12 @@ static evtchn_port_or_error_t minios_evtchn_pending(xc_evtchn *xce, xc_osdep_han
return ret;
}
-static int minios_evtchn_unmask(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
{
unmask_evtchn(port);
return 0;
}
-static struct xc_osdep_ops minios_evtchn_ops = {
- .open = &minios_evtchn_open,
- .close = &minios_evtchn_close,
-
- .u.evtchn = {
- .fd = &minios_evtchn_fd,
- .notify = &minios_evtchn_notify,
- .bind_unbound_port = &minios_evtchn_bind_unbound_port,
- .bind_interdomain = &minios_evtchn_bind_interdomain,
- .bind_virq = &minios_evtchn_bind_virq,
- .unbind = &minios_evtchn_unbind,
- .pending = &minios_evtchn_pending,
- .unmask = &minios_evtchn_unmask,
- },
-};
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
@@ -523,8 +510,6 @@ static struct xc_osdep_ops *minios_osdep_init(xc_interface *xch, enum xc_osdep_t
{
case XC_OSDEP_PRIVCMD:
return &minios_privcmd_ops;
- case XC_OSDEP_EVTCHN:
- return &minios_evtchn_ops;
case XC_OSDEP_GNTTAB:
return &minios_gnttab_ops;
default:
diff --git a/tools/libxc/xc_netbsd.c b/tools/libxc/xc_netbsd.c
index 54f1d7b..b1b828f 100644
--- a/tools/libxc/xc_netbsd.c
+++ b/tools/libxc/xc_netbsd.c
@@ -225,29 +225,31 @@ static struct xc_osdep_ops netbsd_privcmd_ops = {
#define EVTCHN_DEV_NAME "/dev/xenevt"
-static xc_osdep_handle netbsd_evtchn_open(xc_evtchn *xce)
+int osdep_evtchn_open(xc_evtchn *xce)
{
int fd = open(EVTCHN_DEV_NAME, O_NONBLOCK|O_RDWR);
if ( fd == -1 )
- return XC_OSDEP_OPEN_ERROR;
-
- return (xc_osdep_handle)fd;
+ return -1;
+ xce->fd = fd;
+ return 0;
}
-static int netbsd_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
+int osdep_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
{
- int fd = (int)h;
- return close(fd);
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
}
-static int netbsd_evtchn_fd(xc_evtchn *xce, xc_osdep_handle h)
+int xc_evtchn_fd(xc_evtchn *xce)
{
- return (int)h;
+ return xce->fd;
}
-static int netbsd_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_notify notify;
notify.port = port;
@@ -255,10 +257,9 @@ static int netbsd_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t
return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
}
-static evtchn_port_or_error_t
-netbsd_evtchn_bind_unbound_port(xc_evtchn * xce, xc_osdep_handle h, int domid)
+evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn * xce, int domid)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_unbound_port bind;
int ret;
@@ -271,11 +272,10 @@ netbsd_evtchn_bind_unbound_port(xc_evtchn * xce, xc_osdep_handle h, int domid)
return -1;
}
-static evtchn_port_or_error_t
-netbsd_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
- evtchn_port_t remote_port)
+evtchn_port_or_error_t xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
+ evtchn_port_t remote_port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_interdomain bind;
int ret;
@@ -289,9 +289,9 @@ netbsd_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
return -1;
}
-static int netbsd_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_unbind unbind;
unbind.port = port;
@@ -299,10 +299,9 @@ static int netbsd_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t
return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
}
-static evtchn_port_or_error_t
-netbsd_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
+evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_virq bind;
int err;
@@ -315,10 +314,9 @@ netbsd_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
return bind.port;
}
-static evtchn_port_or_error_t
-netbsd_evtchn_pending(xc_evtchn *xce, xc_osdep_handle h)
+evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
{
- int fd = (int)h;
+ int fd = xce->fd;
evtchn_port_t port;
if ( read_exact(fd, (char *)&port, sizeof(port)) == -1 )
@@ -327,28 +325,12 @@ netbsd_evtchn_pending(xc_evtchn *xce, xc_osdep_handle h)
return port;
}
-static int netbsd_evtchn_unmask(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
return write_exact(fd, (char *)&port, sizeof(port));
}
-static struct xc_osdep_ops netbsd_evtchn_ops = {
- .open = &netbsd_evtchn_open,
- .close = &netbsd_evtchn_close,
-
- .u.evtchn = {
- .fd = &netbsd_evtchn_fd,
- .notify = &netbsd_evtchn_notify,
- .bind_unbound_port = &netbsd_evtchn_bind_unbound_port,
- .bind_interdomain = &netbsd_evtchn_bind_interdomain,
- .bind_virq = &netbsd_evtchn_bind_virq,
- .unbind = &netbsd_evtchn_unbind,
- .pending = &netbsd_evtchn_pending,
- .unmask = &netbsd_evtchn_unmask,
- },
-};
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
@@ -395,8 +377,6 @@ static struct xc_osdep_ops *netbsd_osdep_init(xc_interface *xch, enum xc_osdep_t
{
case XC_OSDEP_PRIVCMD:
return &netbsd_privcmd_ops;
- case XC_OSDEP_EVTCHN:
- return &netbsd_evtchn_ops;
default:
return NULL;
}
diff --git a/tools/libxc/xc_private.c b/tools/libxc/xc_private.c
index 7c39897..154e553 100644
--- a/tools/libxc/xc_private.c
+++ b/tools/libxc/xc_private.c
@@ -119,7 +119,6 @@ static const char *xc_osdep_type_name(enum xc_osdep_type type)
switch ( type )
{
case XC_OSDEP_PRIVCMD: return "privcmd";
- case XC_OSDEP_EVTCHN: return "evtchn";
case XC_OSDEP_GNTTAB: return "gnttab";
case XC_OSDEP_GNTSHR: return "gntshr";
}
@@ -252,20 +251,44 @@ int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
return xch->ops->u.privcmd.hypercall(xch, xch->ops_handle, hypercall);
}
-xc_evtchn *xc_evtchn_open(xentoollog_logger *logger,
- unsigned open_flags)
+xc_evtchn *xc_evtchn_open(xentoollog_logger *logger, unsigned open_flags)
{
- xc_evtchn *xce;
+ xc_evtchn *xce = malloc(sizeof(*xce));
+ int rc;
+
+ if (!xce) return NULL;
+
+ xce->fd = -1;
+ xce->logger = logger;
+ xce->logger_tofree = NULL;
- xce = xc_interface_open_common(logger, NULL, open_flags,
- XC_OSDEP_EVTCHN);
+ if (!xce->logger) {
+ xce->logger = xce->logger_tofree =
+ (xentoollog_logger*)
+ xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
+ if (!xce->logger) goto err;
+ }
+
+ rc = osdep_evtchn_open(xce);
+ if ( rc < 0 ) goto err;
return xce;
+
+err:
+ osdep_evtchn_close(xce);
+ xtl_logger_destroy(xce->logger_tofree);
+ free(xce);
+ return NULL;
}
int xc_evtchn_close(xc_evtchn *xce)
{
- return xc_interface_close_common(xce);
+ int rc;
+
+ rc = osdep_evtchn_close(xce);
+ xtl_logger_destroy(xce->logger_tofree);
+ free(xce);
+ return rc;
}
xc_gnttab *xc_gnttab_open(xentoollog_logger *logger,
diff --git a/tools/libxc/xc_private.h b/tools/libxc/xc_private.h
index 2df1d59..a5bf79a 100644
--- a/tools/libxc/xc_private.h
+++ b/tools/libxc/xc_private.h
@@ -123,6 +123,13 @@ struct xc_interface_core {
xc_osdep_handle ops_handle; /* opaque data for xc_osdep_ops */
};
+struct xenevtchn_handle {
+ xentoollog_logger *logger, *logger_tofree;
+ int fd;
+};
+int osdep_evtchn_open(xc_evtchn *xce);
+int osdep_evtchn_close(xc_evtchn *xce);
+
void xc_report_error(xc_interface *xch, int code, const char *fmt, ...)
__attribute__((format(printf,3,4)));
void xc_reportv(xc_interface *xch, xentoollog_logger *lg, xentoollog_level,
@@ -446,3 +453,13 @@ void *xc_vm_event_enable(xc_interface *xch, domid_t domain_id, int param,
uint32_t *port);
#endif /* __XC_PRIVATE_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libxc/xc_solaris.c b/tools/libxc/xc_solaris.c
index 7e5d847..088228a 100644
--- a/tools/libxc/xc_solaris.c
+++ b/tools/libxc/xc_solaris.c
@@ -194,33 +194,36 @@ static struct xc_osdep_ops solaris_privcmd_ops = {
},
};
-static xc_osdep_handle solaris_evtchn_open(xc_evtchn *xce)
+int osdep_evtchn_open(xc_evtchn *xce)
{
int fd;
if ( (fd = open("/dev/xen/evtchn", O_RDWR)) == -1 )
{
PERROR("Could not open event channel interface");
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
}
- return (xc_osdep_handle)fd;
+ xce->fd = fd;
+ return 0;
}
-static int solaris_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
+int osdep_evtchn_close(xc_evtchn *xce)
{
- int fd = (int)h;
- return close(fd);
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
}
-static int solaris_evtchn_fd(xc_evtchn *xce, xc_osdep_handle h)
+int xc_evtchn_fd(xc_evtchn *xce)
{
- return (int)h;
+ return xce->fd;
}
-static int solaris_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_notify notify;
notify.port = port;
@@ -228,10 +231,9 @@ static int solaris_evtchn_notify(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_
return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
}
-static evtchn_port_or_error_t
-solaris_evtchn_bind_unbound_port(xc_evtchn *xce, xc_osdep_handle h, int domid)
+evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_unbound_port bind;
bind.remote_domain = domid;
@@ -239,11 +241,10 @@ solaris_evtchn_bind_unbound_port(xc_evtchn *xce, xc_osdep_handle h, int domid)
return ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
}
-evtchn_port_or_error_t
-solaris_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
- evtchn_port_t remote_port)
+evtchn_port_or_error_t xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
+ evtchn_port_t remote_port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_interdomain bind;
bind.remote_domain = domid;
@@ -252,10 +253,9 @@ solaris_evtchn_bind_interdomain(xc_evtchn *xce, xc_osdep_handle h, int domid,
return ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
}
-static evtchn_port_or_error_t
-solaris_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
+evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_bind_virq bind;
bind.virq = virq;
@@ -263,9 +263,9 @@ solaris_evtchn_bind_virq(xc_evtchn *xce, xc_osdep_handle h, unsigned int virq)
return ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
}
-static int solaris_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_t port)
+int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
struct ioctl_evtchn_unbind unbind;
unbind.port = port;
@@ -273,10 +273,9 @@ static int solaris_evtchn_unbind(xc_evtchn *xce, xc_osdep_handle h, evtchn_port_
return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
}
-static evtchn_port_or_error_t
-solaris_evtchn_pending(xc_evtchn *xce, xc_osdep_handle h)
+evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
{
- int fd = (int)h;
+ int fd = xce->fd;
evtchn_port_t port;
if ( read_exact(fd, (char *)&port, sizeof(port)) == -1 )
@@ -285,28 +284,12 @@ solaris_evtchn_pending(xc_evtchn *xce, xc_osdep_handle h)
return port;
}
-static int solaris_evtchn_unmask(xc_evtchn *xce, xc_osdep_handle h,evtchn_port_t port)
+int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
{
- int fd = (int)h;
+ int fd = xce->fd;
return write_exact(fd, (char *)&port, sizeof(port));
}
-static struct xc_osdep_ops solaris_evtchn_ops = {
- .open = &solaris_evtchn_open,
- .close = &solaris_evtchn_close,
-
- .u.evtchn = {
- .fd = &solaris_evtchn_fd,
- .notify = &solaris_evtchn_notify,
- .bind_unbound_port = &solaris_evtchn_bind_unbound_port,
- .bind_interdomain = &solaris_evtchn_bind_interdomain,
- .bind_virq = &solaris_evtchn_bind_virq,
- .unbind = &solaris_evtchn_unbind,
- .pending = &solaris_evtchn_pending,
- .unmask = &solaris_evtchn_unmask,
- },
-};
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
@@ -324,8 +307,6 @@ static struct xc_osdep_ops *solaris_osdep_init(xc_interface *xch, enum xc_osdep_
{
case XC_OSDEP_PRIVCMD:
return &solaris_privcmd_ops;
- case XC_OSDEP_EVTCHN:
- return &solaris_evtchn_ops;
default:
return NULL;
}
diff --git a/tools/libxc/xenctrl_osdep_ENOSYS.c b/tools/libxc/xenctrl_osdep_ENOSYS.c
index dd59dcd..5182532 100644
--- a/tools/libxc/xenctrl_osdep_ENOSYS.c
+++ b/tools/libxc/xenctrl_osdep_ENOSYS.c
@@ -81,90 +81,6 @@ static struct xc_osdep_ops ENOSYS_privcmd_ops =
}
};
-static xc_osdep_handle ENOSYS_evtchn_open(xc_interface *xce)
-{
- IPRINTF(xce, "ENOSYS_evtchn: opening handle %p\n", (void *)1);
- return (xc_osdep_handle)2; /*dummy*/
-}
-
-static int ENOSYS_evtchn_close(xc_interface *xce, xc_osdep_handle h)
-{
- IPRINTF(xce, "ENOSYS_evtchn: closing handle %lx\n", h);
- return 0;
-}
-
-static int ENOSYS_evtchn_fd(xc_interface *xce, xc_osdep_handle h)
-{
- IPRINTF(xce, "ENOSYS_fd %lx fd\n", h);
- return (int)h;
-}
-
-static int ENOSYS_evtchn_notify(xc_interface *xce, xc_osdep_handle h, evtchn_port_t port)
-{
- IPRINTF(xce, "ENOSYS_evtchn %lx notify: %d\n", h, port);
- return -ENOSYS;
-}
-
-static int ENOSYS_evtchn_bind_unbound_port(xc_interface *xce, xc_osdep_handle h, int domid)
-{
- IPRINTF(xce, "ENOSYS_evtchn %lx bind_unbound_port: dom%d\n", h, domid);
- return -ENOSYS;
-}
-
-
-static int ENOSYS_evtchn_bind_interdomain(xc_interface *xce, xc_osdep_handle h, int domid, evtchn_port_t remote_port)
-{
- IPRINTF(xce, "ENOSYS_evtchn %lx bind_interdomain: dmo%d %d\n", h, domid, remote_port);
- return -ENOSYS;
-}
-
-
-static int ENOSYS_evtchn_bind_virq(xc_interface *xce, xc_osdep_handle h, unsigned int virq)
-{
- IPRINTF(xce, "ENOSYS_evtchn %lx bind_virq: %d\n", h, virq);
- return -ENOSYS;
-}
-
-
-static int ENOSYS_evtchn_unbind(xc_interface *xce, xc_osdep_handle h, evtchn_port_t port)
-{
- IPRINTF(xce, "ENOSYS_evtchn %lx unbind: %d\n", h, port);
- return -ENOSYS;
-}
-
-
-static evtchn_port_or_error_t ENOSYS_evtchn_pending(xc_interface *xce, xc_osdep_handle h)
-{
- IPRINTF(xce, "ENOSYS_evtchn %lx pending\n", h);
- return -ENOSYS;
-}
-
-static int ENOSYS_evtchn_unmask(xc_interface *xce, xc_osdep_handle h, evtchn_port_t port)
-{
- IPRINTF(xce, "ENOSYS_evtchn %lx unmask: %d\n", h, port);
- return -ENOSYS;
-}
-
-static struct xc_osdep_ops ENOSYS_evtchn_ops = {
- .open = &ENOSYS_evtchn_open,
- .close = &ENOSYS_evtchn_close,
-
- .u.evtchn = {
- .fd = &ENOSYS_evtchn_fd,
-
- .notify = &ENOSYS_evtchn_notify,
-
- .bind_unbound_port = &ENOSYS_evtchn_bind_unbound_port,
- .bind_interdomain = &ENOSYS_evtchn_bind_interdomain,
- .bind_virq = &ENOSYS_evtchn_bind_virq,
-
- .unbind = &ENOSYS_evtchn_unbind,
-
- .pending = &ENOSYS_evtchn_pending,
- .unmask = &ENOSYS_evtchn_unmask,
- },
-};
-
static struct xc_osdep_ops * ENOSYS_osdep_init(xc_interface *xch, enum xc_osdep_type type)
{
struct xc_osdep_ops *ops;
@@ -180,9 +96,6 @@ static struct xc_osdep_ops * ENOSYS_osdep_init(xc_interface *xch, enum xc_osdep_
case XC_OSDEP_PRIVCMD:
ops = &ENOSYS_privcmd_ops;
break;
- case XC_OSDEP_EVTCHN:
- ops = &ENOSYS_evtchn_ops;
- break;
default:
ops = NULL;
break;
diff --git a/tools/ocaml/libs/eventchn/xeneventchn_stubs.c b/tools/ocaml/libs/eventchn/xeneventchn_stubs.c
index abefd6b..5939e7c 100644
--- a/tools/ocaml/libs/eventchn/xeneventchn_stubs.c
+++ b/tools/ocaml/libs/eventchn/xeneventchn_stubs.c
@@ -34,14 +34,14 @@
#include <caml/callback.h>
#include <caml/fail.h>
-#define _H(__h) ((xc_interface *)(__h))
+#define _H(__h) ((xc_evtchn *)(__h))
CAMLprim value stub_eventchn_init(void)
{
CAMLparam0();
CAMLlocal1(result);
- xc_interface *xce = xc_evtchn_open(NULL, XC_OPENFLAG_NON_REENTRANT);
+ xc_evtchn *xce = xc_evtchn_open(NULL, XC_OPENFLAG_NON_REENTRANT);
if (xce == NULL)
caml_failwith("open failed");
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 04/23] tools: Refactor /dev/xen/evtchn wrappers into libxenevtchn.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (2 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 03/23] tools/libxc: Remove osdep indirection for xc_evtchn Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 14:50 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 05/23] tools: Arrange to check public headers for ANSI compatiblity Ian Campbell
` (19 subsequent siblings)
23 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
libxenevtchn will provide a stable API and ABI for accessing the
evtchn device.
The functions are moved into the xenevtchn namespace to make a clean
break from libxc and avoid ambiguity regarding which interfaces are
stable.
All in-tree users are updated to use the new names.
Upon request (via #define XC_WANT_COMPAT_EVTCHN_API) libxenctrl will
provide a compat API for the old names. This is used by qemu-xen for
the time being. qemu-xen-traditional is updated in lockstep.
This leaves a few event channel related functions which go via privcmd
(EVTCHNOP) rather than ioctls on the /dev/xen/evtchn device in
libxenctrl. Specifically:
- xc_evtchn_alloc_unbound
- xc_evtchn_reset
- xc_evtchn_status
Note that xc_evtchn_alloc_unbound's functionality is also provided by
xenevtchn_bind_unbound_port() (née xc_evtchn_bind_unbound_port) and is
probably redundant.
These functions do not appear to be needed by qemu-dm, qemu-pv
(provision of device model to HVM guests and PV backends respectively)
or by libvchan suggesting they are not needed by non-toolstack uses of
event channels. QEMU does use these in hw/xenpv/xen_domainbuild.c but
that is a "toolstack use".
The new library uses a version script to ensure that only expected
symbols are exported and to version them such that ABI guarantees can
be kept in the future.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
Must be applied with:
- "qemu-xen-traditional: Use libxenevtchn" and a corresponding
QEMU_TAG update folded here.
- "mini-os: Include libxenevtchn with libxc"" and a corresponding
bump to MINIOS_UPSTREAM_REVISION folded in here.
v2: Update doc at same time
Removed some stray compat thing
v3: Moved to tools/libs/evtchn
v4: typedef the xenevtchn_handle in xenguest.h instead of #include, to
avoid leaking the libxenevtchn namespace into callers who may not
want it. (And also to avoid adding lots of -I to places like
stubdom builds)
v5: Handle NULL passed to _close()
---
.gitignore | 1 +
stubdom/Makefile | 19 +-
tools/Makefile | 5 +-
tools/Rules.mk | 14 +-
tools/console/Makefile | 2 +
tools/console/daemon/io.c | 43 ++--
tools/libs/Makefile | 1 +
tools/libs/evtchn/Makefile | 65 +++++++
tools/libs/evtchn/core.c | 72 +++++++
tools/libs/evtchn/freebsd.c | 138 +++++++++++++
tools/libs/evtchn/include/xenevtchn.h | 150 ++++++++++++++
tools/libs/evtchn/libxenevtchn.map | 14 ++
tools/libs/evtchn/linux.c | 136 +++++++++++++
tools/libs/evtchn/minios.c | 269 ++++++++++++++++++++++++++
tools/libs/evtchn/netbsd.c | 147 ++++++++++++++
tools/libs/evtchn/private.h | 25 +++
tools/libs/evtchn/solaris.c | 135 +++++++++++++
tools/libs/toollog/include/xentoollog.h | 10 +
tools/libs/toollog/xtl_core.c | 10 +
tools/libs/toollog/xtl_logger_stdio.c | 10 +
tools/libvchan/Makefile | 6 +-
tools/libvchan/init.c | 20 +-
tools/libvchan/io.c | 12 +-
tools/libvchan/libxenvchan.h | 3 +-
tools/libxc/Makefile | 8 +-
tools/libxc/include/xenctrl.h | 99 +---------
tools/libxc/include/xenctrl_compat.h | 48 +++++
tools/libxc/include/xenguest.h | 14 +-
tools/libxc/xc_evtchn_compat.c | 75 +++++++
tools/libxc/xc_freebsd_osdep.c | 101 ----------
tools/libxc/xc_linux_osdep.c | 95 ---------
tools/libxc/xc_minios.c | 221 ---------------------
tools/libxc/xc_netbsd.c | 109 -----------
tools/libxc/xc_private.c | 40 ----
tools/libxc/xc_private.h | 7 -
tools/libxc/xc_solaris.c | 97 ----------
tools/libxc/xc_suspend.c | 18 +-
tools/libxl/Makefile | 5 +-
tools/libxl/libxl.c | 2 +-
tools/libxl/libxl_dom_suspend.c | 4 +-
tools/libxl/libxl_event.c | 14 +-
tools/libxl/libxl_internal.h | 5 +-
tools/misc/Makefile | 7 +-
tools/misc/xen-hptool.c | 13 +-
tools/misc/xen-lowmemd.c | 15 +-
tools/ocaml/libs/eventchn/Makefile | 4 +-
tools/ocaml/libs/eventchn/xeneventchn_stubs.c | 20 +-
tools/python/setup.py | 7 +-
tools/tests/xen-access/xen-access.c | 2 +-
tools/xcutils/Makefile | 4 +-
tools/xenmon/Makefile | 2 +
tools/xenmon/xenbaked.c | 13 +-
tools/xenpaging/Makefile | 4 +-
tools/xenpaging/xenpaging.c | 18 +-
tools/xenpaging/xenpaging.h | 3 +-
tools/xenstore/Makefile | 5 +-
tools/xenstore/xenstored_core.c | 6 +-
tools/xenstore/xenstored_domain.c | 25 +--
tools/xentrace/Makefile | 5 +-
tools/xentrace/xentrace.c | 13 +-
60 files changed, 1520 insertions(+), 915 deletions(-)
create mode 100644 tools/libs/evtchn/Makefile
create mode 100644 tools/libs/evtchn/core.c
create mode 100644 tools/libs/evtchn/freebsd.c
create mode 100644 tools/libs/evtchn/include/xenevtchn.h
create mode 100644 tools/libs/evtchn/libxenevtchn.map
create mode 100644 tools/libs/evtchn/linux.c
create mode 100644 tools/libs/evtchn/minios.c
create mode 100644 tools/libs/evtchn/netbsd.c
create mode 100644 tools/libs/evtchn/private.h
create mode 100644 tools/libs/evtchn/solaris.c
create mode 100644 tools/libxc/include/xenctrl_compat.h
create mode 100644 tools/libxc/xc_evtchn_compat.c
diff --git a/.gitignore b/.gitignore
index a2c85e1..bf382e5 100644
--- a/.gitignore
+++ b/.gitignore
@@ -59,6 +59,7 @@ stubdom/include
stubdom/ioemu
stubdom/xenstore
stubdom/libxentoollog-*
+stubdom/libxenevtchn-*
stubdom/libxc-*
stubdom/lwip-*
stubdom/mini-os-*
diff --git a/stubdom/Makefile b/stubdom/Makefile
index 9c923dd..3b3bfc9 100644
--- a/stubdom/Makefile
+++ b/stubdom/Makefile
@@ -318,6 +318,12 @@ mk-headers-$(XEN_TARGET_ARCH): $(IOEMU_LINKFARM_TARGET)
ln -sf $(XEN_ROOT)/tools/libs/toollog/include/*.h . && \
ln -sf $(XEN_ROOT)/tools/libs/toollog/*.c . && \
ln -sf $(XEN_ROOT)/tools/libs/toollog/Makefile . )
+ mkdir -p libs-$(XEN_TARGET_ARCH)/evtchn/include
+ [ -h libs-$(XEN_TARGET_ARCH)/evtchn/Makefile ] || ( cd libs-$(XEN_TARGET_ARCH)/evtchn && \
+ ln -sf $(XEN_ROOT)/tools/libs/evtchn/*.h . && \
+ ln -sf $(XEN_ROOT)/tools/libs/evtchn/include/*.h include/ && \
+ ln -sf $(XEN_ROOT)/tools/libs/evtchn/*.c . && \
+ ln -sf $(XEN_ROOT)/tools/libs/evtchn/Makefile . )
mkdir -p libxc-$(XEN_TARGET_ARCH)
[ -h libxc-$(XEN_TARGET_ARCH)/Makefile ] || ( cd libxc-$(XEN_TARGET_ARCH) && \
ln -sf $(XEN_ROOT)/tools/libxc/*.h . && \
@@ -352,12 +358,23 @@ libs-$(XEN_TARGET_ARCH)/toollog/libxentoollog.a: $(NEWLIB_STAMPFILE)
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/toollog
#######
+# libxenevtchn
+#######
+
+.PHONY: libxenevtchn
+libxenevtchn: libs-$(XEN_TARGET_ARCH)/evtchn/libxenevtchn.a
+libs-$(XEN_TARGET_ARCH)/evtchn/libxenevtchn.a: $(NEWLIB_STAMPFILE)
+ $(MAKE) -C $(XEN_ROOT)/tools/include
+ $(MAKE) DESTDIR= -C $(MINI_OS) links
+ CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/evtchn
+
+#######
# libxc
#######
.PHONY: libxc
libxc: libxc-$(XEN_TARGET_ARCH)/libxenctrl.a libxc-$(XEN_TARGET_ARCH)/libxenguest.a
-libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog cross-zlib
+libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog libxenevtchn cross-zlib
$(MAKE) -C $(XEN_ROOT)/tools/include
$(MAKE) DESTDIR= -C $(MINI_OS) links
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= CONFIG_LIBXC_MINIOS=y -C libxc-$(XEN_TARGET_ARCH)
diff --git a/tools/Makefile b/tools/Makefile
index 9f74ac7..99e016a 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -248,8 +248,10 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
--libdir=$(LIBEXEC_LIB) \
--includedir=$(LIBEXEC_INC) \
--source-path=$$source \
- --extra-cflags="-I$(XEN_ROOT)/tools/include \
+ --extra-cflags="-DXC_WANT_COMPAT_EVTCHN_API=1 \
+ -I$(XEN_ROOT)/tools/include \
-I$(XEN_ROOT)/tools/libs/toollog/include \
+ -I$(XEN_ROOT)/tools/libs/evtchn/include \
-I$(XEN_ROOT)/tools/libxc/include \
-I$(XEN_ROOT)/tools/xenstore/include \
-I$(XEN_ROOT)/tools/xenstore/compat/include \
@@ -257,6 +259,7 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
--extra-ldflags="-L$(XEN_ROOT)/tools/libxc \
-L$(XEN_ROOT)/tools/xenstore \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog \
+ -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/evtchn \
$(QEMU_UPSTREAM_RPATH)" \
--bindir=$(LIBEXEC_BIN) \
--datadir=$(SHAREDIR)/qemu-xen \
diff --git a/tools/Rules.mk b/tools/Rules.mk
index 7516ddd..75d02c4 100644
--- a/tools/Rules.mk
+++ b/tools/Rules.mk
@@ -11,6 +11,7 @@ INSTALL = $(XEN_ROOT)/tools/cross-install
XEN_INCLUDE = $(XEN_ROOT)/tools/include
XEN_LIBXENTOOLLOG = $(XEN_ROOT)/tools/libs/toollog
+XEN_LIBXENEVTCHN = $(XEN_ROOT)/tools/libs/evtchn
XEN_LIBXC = $(XEN_ROOT)/tools/libxc
XEN_XENLIGHT = $(XEN_ROOT)/tools/libxl
XEN_XENSTORE = $(XEN_ROOT)/tools/xenstore
@@ -82,13 +83,18 @@ SHDEPS_libxentoollog =
LDLIBS_libxentoollog = $(XEN_LIBXENTOOLLOG)/libxentoollog$(libextension)
SHLIB_libxentoollog = -Wl,-rpath-link=$(XEN_LIBXENTOOLLOG)
+CFLAGS_libxenevtchn = -I$(XEN_LIBXENEVTCHN)/include $(CFLAGS_xeninclude)
+SHDEPS_libxenevtchn =
+LDLIBS_libxenevtchn = $(XEN_LIBXENEVTCHN)/libxenevtchn$(libextension)
+SHLIB_libxenevtchn = -Wl,-rpath-link=$(XEN_LIBXENEVTCHN)
+
CFLAGS_libxenctrl = -I$(XEN_LIBXC)/include $(CFLAGS_libxentoollog) $(CFLAGS_xeninclude)
-SHDEPS_libxenctrl = $(SHLIB_libxentoollog)
+SHDEPS_libxenctrl = $(SHLIB_libxentoollog) $(SHLIB_libxenevtchn)
LDLIBS_libxenctrl = $(SHDEPS_libxenctrl) $(XEN_LIBXC)/libxenctrl$(libextension)
SHLIB_libxenctrl = $(SHDEPS_libxenctrl) -Wl,-rpath-link=$(XEN_LIBXC)
-CFLAGS_libxenguest = -I$(XEN_LIBXC)/include $(CFLAGS_xeninclude)
-SHDEPS_libxenguest =
+CFLAGS_libxenguest = -I$(XEN_LIBXC)/include $(CFLAGS_libxenevtchn) $(CFLAGS_xeninclude)
+SHDEPS_libxenguest = $(SHLIB_libxenevtchn)
LDLIBS_libxenguest = $(SHDEPS_libxenguest) $(XEN_LIBXC)/libxenguest$(libextension)
SHLIB_libxenguest = $(SHDEPS_libxenguest) -Wl,-rpath-link=$(XEN_LIBXC)
@@ -103,7 +109,7 @@ LDLIBS_libxenstat = $(SHDEPS_libxenstat) $(XEN_LIBXENSTAT)/libxenstat$(libexten
SHLIB_libxenstat = $(SHDEPS_libxenstat) -Wl,-rpath-link=$(XEN_LIBXENSTAT)
CFLAGS_libxenvchan = -I$(XEN_LIBVCHAN)
-SHDEPS_libxenvchan = $(SHLIB_libxenctrl) $(SHLIB_libxenstore)
+SHDEPS_libxenvchan = $(SHLIB_libxenctrl) $(SHLIB_libxenstore) $(SHLIB_libxenevtchn)
LDLIBS_libxenvchan = $(SHDEPS_libxenvchan) $(XEN_LIBVCHAN)/libxenvchan$(libextension)
SHLIB_libxenvchan = $(SHDEPS_libxenvchan) -Wl,-rpath-link=$(XEN_LIBVCHAN)
diff --git a/tools/console/Makefile b/tools/console/Makefile
index 77e8f29..4b3a492 100644
--- a/tools/console/Makefile
+++ b/tools/console/Makefile
@@ -3,8 +3,10 @@ include $(XEN_ROOT)/tools/Rules.mk
CFLAGS += -Werror
+CFLAGS += $(CFLAGS_libxenevtchn)
CFLAGS += $(CFLAGS_libxenctrl)
CFLAGS += $(CFLAGS_libxenstore)
+LDLIBS += $(LDLIBS_libxenevtchn)
LDLIBS += $(LDLIBS_libxenctrl)
LDLIBS += $(LDLIBS_libxenstore)
LDLIBS += $(SOCKET_LIBS)
diff --git a/tools/console/daemon/io.c b/tools/console/daemon/io.c
index cafc7b7..2f2e9c5 100644
--- a/tools/console/daemon/io.c
+++ b/tools/console/daemon/io.c
@@ -21,6 +21,7 @@
#include "utils.h"
#include "io.h"
+#include <xenevtchn.h>
#include <xenstore.h>
#include <xen/io/console.h>
#include <xen/grant_table.h>
@@ -101,7 +102,7 @@ struct domain {
int ring_ref;
evtchn_port_or_error_t local_port;
evtchn_port_or_error_t remote_port;
- xc_evtchn *xce_handle;
+ xenevtchn_handle *xce_handle;
int xce_pollfd_idx;
struct xencons_interface *interface;
int event_count;
@@ -185,7 +186,7 @@ static void buffer_append(struct domain *dom)
xen_mb();
intf->out_cons = cons;
- xc_evtchn_notify(dom->xce_handle, dom->local_port);
+ xenevtchn_notify(dom->xce_handle, dom->local_port);
/* Get the data to the logfile as early as possible because if
* no one is listening on the console pty then it will fill up
@@ -584,22 +585,22 @@ static int domain_create_ring(struct domain *dom)
dom->local_port = -1;
dom->remote_port = -1;
if (dom->xce_handle != NULL)
- xc_evtchn_close(dom->xce_handle);
+ xenevtchn_close(dom->xce_handle);
/* Opening evtchn independently for each console is a bit
* wasteful, but that's how the code is structured... */
- dom->xce_handle = xc_evtchn_open(NULL, 0);
+ dom->xce_handle = xenevtchn_open(NULL, 0);
if (dom->xce_handle == NULL) {
err = errno;
goto out;
}
- rc = xc_evtchn_bind_interdomain(dom->xce_handle,
+ rc = xenevtchn_bind_interdomain(dom->xce_handle,
dom->domid, remote_port);
if (rc == -1) {
err = errno;
- xc_evtchn_close(dom->xce_handle);
+ xenevtchn_close(dom->xce_handle);
dom->xce_handle = NULL;
goto out;
}
@@ -609,7 +610,7 @@ static int domain_create_ring(struct domain *dom)
if (dom->master_fd == -1) {
if (!domain_create_tty(dom)) {
err = errno;
- xc_evtchn_close(dom->xce_handle);
+ xenevtchn_close(dom->xce_handle);
dom->xce_handle = NULL;
dom->local_port = -1;
dom->remote_port = -1;
@@ -749,7 +750,7 @@ static void shutdown_domain(struct domain *d)
watch_domain(d, false);
domain_unmap_interface(d);
if (d->xce_handle != NULL)
- xc_evtchn_close(d->xce_handle);
+ xenevtchn_close(d->xce_handle);
d->xce_handle = NULL;
}
@@ -839,7 +840,7 @@ static void handle_tty_read(struct domain *dom)
}
xen_wmb();
intf->in_prod = prod;
- xc_evtchn_notify(dom->xce_handle, dom->local_port);
+ xenevtchn_notify(dom->xce_handle, dom->local_port);
} else {
domain_close_tty(dom);
shutdown_domain(dom);
@@ -871,7 +872,7 @@ static void handle_ring_read(struct domain *dom)
if (dom->is_dead)
return;
- if ((port = xc_evtchn_pending(dom->xce_handle)) == -1)
+ if ((port = xenevtchn_pending(dom->xce_handle)) == -1)
return;
dom->event_count++;
@@ -879,7 +880,7 @@ static void handle_ring_read(struct domain *dom)
buffer_append(dom);
if (dom->event_count < RATE_LIMIT_ALLOWANCE)
- (void)xc_evtchn_unmask(dom->xce_handle, port);
+ (void)xenevtchn_unmask(dom->xce_handle, port);
}
static void handle_xs(void)
@@ -906,7 +907,7 @@ static void handle_xs(void)
free(vec);
}
-static void handle_hv_logs(xc_evtchn *xce_handle, bool force)
+static void handle_hv_logs(xenevtchn_handle *xce_handle, bool force)
{
static char buffer[1024*16];
char *bufptr = buffer;
@@ -914,7 +915,7 @@ static void handle_hv_logs(xc_evtchn *xce_handle, bool force)
static uint32_t index = 0;
evtchn_port_or_error_t port = -1;
- if (!force && ((port = xc_evtchn_pending(xce_handle)) == -1))
+ if (!force && ((port = xenevtchn_pending(xce_handle)) == -1))
return;
do
@@ -938,7 +939,7 @@ static void handle_hv_logs(xc_evtchn *xce_handle, bool force)
} while (size == sizeof(buffer));
if (port != -1)
- (void)xc_evtchn_unmask(xce_handle, port);
+ (void)xenevtchn_unmask(xce_handle, port);
}
static void handle_log_reload(void)
@@ -1006,10 +1007,10 @@ void handle_io(void)
evtchn_port_or_error_t log_hv_evtchn = -1;
int xce_pollfd_idx = -1;
int xs_pollfd_idx = -1;
- xc_evtchn *xce_handle = NULL;
+ xenevtchn_handle *xce_handle = NULL;
if (log_hv) {
- xce_handle = xc_evtchn_open(NULL, 0);
+ xce_handle = xenevtchn_open(NULL, 0);
if (xce_handle == NULL) {
dolog(LOG_ERR, "Failed to open xce handle: %d (%s)",
errno, strerror(errno));
@@ -1018,7 +1019,7 @@ void handle_io(void)
log_hv_fd = create_hv_log();
if (log_hv_fd == -1)
goto out;
- log_hv_evtchn = xc_evtchn_bind_virq(xce_handle, VIRQ_CON_RING);
+ log_hv_evtchn = xenevtchn_bind_virq(xce_handle, VIRQ_CON_RING);
if (log_hv_evtchn == -1) {
dolog(LOG_ERR, "Failed to bind to VIRQ_CON_RING: "
"%d (%s)", errno, strerror(errno));
@@ -1047,7 +1048,7 @@ void handle_io(void)
xs_pollfd_idx = set_fds(xs_fileno(xs), POLLIN|POLLPRI);
if (log_hv)
- xce_pollfd_idx = set_fds(xc_evtchn_fd(xce_handle),
+ xce_pollfd_idx = set_fds(xenevtchn_fd(xce_handle),
POLLIN|POLLPRI);
if (clock_gettime(CLOCK_MONOTONIC, &ts) < 0)
@@ -1066,7 +1067,7 @@ void handle_io(void)
if ((now+5) > d->next_period) {
d->next_period = now + RATE_LIMIT_PERIOD;
if (d->event_count >= RATE_LIMIT_ALLOWANCE) {
- (void)xc_evtchn_unmask(d->xce_handle, d->local_port);
+ (void)xenevtchn_unmask(d->xce_handle, d->local_port);
}
d->event_count = 0;
}
@@ -1082,7 +1083,7 @@ void handle_io(void)
if (discard_overflowed_data ||
!d->buffer.max_capacity ||
d->buffer.size < d->buffer.max_capacity) {
- int evtchn_fd = xc_evtchn_fd(d->xce_handle);
+ int evtchn_fd = xenevtchn_fd(d->xce_handle);
d->xce_pollfd_idx = set_fds(evtchn_fd,
POLLIN|POLLPRI);
}
@@ -1202,7 +1203,7 @@ void handle_io(void)
log_hv_fd = -1;
}
if (xce_handle != NULL) {
- xc_evtchn_close(xce_handle);
+ xenevtchn_close(xce_handle);
xce_handle = NULL;
}
if (xcg_handle != NULL) {
diff --git a/tools/libs/Makefile b/tools/libs/Makefile
index 73be500..0e3f523 100644
--- a/tools/libs/Makefile
+++ b/tools/libs/Makefile
@@ -3,5 +3,6 @@ include $(XEN_ROOT)/tools/Rules.mk
SUBDIRS-y :=
SUBDIRS-y += toollog
+SUBDIRS-y += evtchn
all clean install distclean: %: subdirs-%
diff --git a/tools/libs/evtchn/Makefile b/tools/libs/evtchn/Makefile
new file mode 100644
index 0000000..85ed6dc
--- /dev/null
+++ b/tools/libs/evtchn/Makefile
@@ -0,0 +1,65 @@
+XEN_ROOT = $(CURDIR)/../../..
+include $(XEN_ROOT)/tools/Rules.mk
+
+MAJOR = 1
+MINOR = 0
+SHLIB_LDFLAGS += -Wl,--version-script=libxenevtchn.map
+
+CFLAGS += -Werror -Wmissing-prototypes
+CFLAGS += -I./include $(CFLAGS_xeninclude)
+CFLAGS += $(CFLAGS_libxentoollog)
+
+SRCS-y += core.c
+SRCS-$(CONFIG_Linux) += linux.c
+SRCS-$(CONFIG_FreeBSD) += freebsd.c
+SRCS-$(CONFIG_SunOS) += solaris.c
+SRCS-$(CONFIG_NetBSD) += netbsd.c
+SRCS-$(CONFIG_MiniOS) += minios.c
+
+LIB_OBJS := $(patsubst %.c,%.o,$(SRCS-y))
+PIC_OBJS := $(patsubst %.c,%.opic,$(SRCS-y))
+
+LIB := libxenevtchn.a
+ifneq ($(nosharedlibs),y)
+LIB += libxenevtchn.so
+endif
+
+.PHONY: all
+all: build
+
+.PHONY: build
+build:
+ $(MAKE) libs
+
+.PHONY: libs
+libs: $(LIB)
+
+
+libxenevtchn.a: $(LIB_OBJS)
+ $(AR) rc $@ $^
+
+libxenevtchn.so: libxenevtchn.so.$(MAJOR)
+ $(SYMLINK_SHLIB) $< $@
+libxenevtchn.so.$(MAJOR): libxenevtchn.so.$(MAJOR).$(MINOR)
+ $(SYMLINK_SHLIB) $< $@
+
+libxenevtchn.so.$(MAJOR).$(MINOR): $(PIC_OBJS) libxenevtchn.map
+ $(CC) $(LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenevtchn.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $(PIC_OBJS) $(LDLIBS_libxentoollog) $(APPEND_LDFLAGS)
+
+.PHONY: install
+install: build
+ $(INSTALL_DIR) $(DESTDIR)$(libdir)
+ $(INSTALL_DIR) $(DESTDIR)$(includedir)
+ $(INSTALL_SHLIB) libxenevtchn.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)
+ $(INSTALL_DATA) libxenevtchn.a $(DESTDIR)$(libdir)
+ $(SYMLINK_SHLIB) libxenevtchn.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxenevtchn.so.$(MAJOR)
+ $(SYMLINK_SHLIB) libxenevtchn.so.$(MAJOR) $(DESTDIR)$(libdir)/libxenevtchn.so
+ $(INSTALL_DATA) include/xenevtchn.h $(DESTDIR)$(includedir)
+
+.PHONY: TAGS
+TAGS:
+ etags -t *.c *.h
+
+.PHONY: clean
+clean:
+ rm -rf *.rpm $(LIB) *~ $(DEPS) $(LIB_OBJS) $(PIC_OBJS)
diff --git a/tools/libs/evtchn/core.c b/tools/libs/evtchn/core.c
new file mode 100644
index 0000000..c31e08c
--- /dev/null
+++ b/tools/libs/evtchn/core.c
@@ -0,0 +1,72 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <unistd.h>
+#include <stdlib.h>
+
+#include "private.h"
+
+xenevtchn_handle *xenevtchn_open(xentoollog_logger *logger, unsigned open_flags)
+{
+ xenevtchn_handle *xce = malloc(sizeof(*xce));
+ int rc;
+
+ if (!xce) return NULL;
+
+ xce->fd = -1;
+ xce->logger = logger;
+ xce->logger_tofree = NULL;
+
+ if (!xce->logger) {
+ xce->logger = xce->logger_tofree =
+ (xentoollog_logger*)
+ xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
+ if (!xce->logger) goto err;
+ }
+
+ rc = osdep_evtchn_open(xce);
+ if ( rc < 0 ) goto err;
+
+ return xce;
+
+err:
+ osdep_evtchn_close(xce);
+ xtl_logger_destroy(xce->logger_tofree);
+ free(xce);
+ return NULL;
+}
+
+int xenevtchn_close(xenevtchn_handle *xce)
+{
+ int rc;
+
+ if ( !xce )
+ return 0;
+
+ rc = osdep_evtchn_close(xce);
+ xtl_logger_destroy(xce->logger_tofree);
+ free(xce);
+ return rc;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/evtchn/freebsd.c b/tools/libs/evtchn/freebsd.c
new file mode 100644
index 0000000..636f052
--- /dev/null
+++ b/tools/libs/evtchn/freebsd.c
@@ -0,0 +1,138 @@
+ /******************************************************************************
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split off from xc_freebsd_osdep.c
+ */
+
+#include <fcntl.h>
+#include <unistd.h>
+
+#include <sys/ioctl.h>
+
+#include <xen/sys/evtchn.h>
+
+#include "private.h"
+
+#define EVTCHN_DEV "/dev/xen/evtchn"
+
+int osdep_evtchn_open(xenevtchn_handle *xce)
+{
+ int fd = open(EVTCHN_DEV, O_RDWR);
+ if ( fd == -1 )
+ return -1;
+ xce->fd = fd;
+ return 0;
+}
+
+int osdep_evtchn_close(xenevtchn_handle *xce)
+{
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
+}
+
+int xenevtchn_fd(xenevtchn_handle *xce)
+{
+ return xce->fd;
+}
+
+int xenevtchn_notify(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_notify notify;
+
+ notify.port = port;
+
+ return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_unbound_port(xenevtchn_handle *xce, int domid)
+{
+ int ret, fd = xce->fd;
+ struct ioctl_evtchn_bind_unbound_port bind;
+
+ bind.remote_domain = domid;
+
+ ret = ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
+ return ( ret == 0 ) ? bind.port : ret;
+}
+
+evtchn_port_or_error_t
+xenevtchn_bind_interdomain(xenevtchn_handle *xce, int domid, evtchn_port_t remote_port)
+{
+ int ret, fd = xce->fd;
+ struct ioctl_evtchn_bind_interdomain bind;
+
+ bind.remote_domain = domid;
+ bind.remote_port = remote_port;
+
+ ret = ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
+ return ( ret == 0 ) ? bind.port : ret;
+}
+
+evtchn_port_or_error_t xenevtchn_bind_virq(xenevtchn_handle *xce, unsigned int virq)
+{
+ int ret, fd = xce->fd;
+ struct ioctl_evtchn_bind_virq bind;
+
+ bind.virq = virq;
+
+ ret = ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
+ return ( ret == 0 ) ? bind.port : ret;
+}
+
+int xenevtchn_unbind(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_unbind unbind;
+
+ unbind.port = port;
+
+ return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
+}
+
+evtchn_port_or_error_t xenevtchn_pending(xenevtchn_handle *xce)
+{
+ int fd = xce->fd;
+ evtchn_port_t port;
+
+ if ( read(fd, &port, sizeof(port)) != sizeof(port) )
+ return -1;
+
+ return port;
+}
+
+int xenevtchn_unmask(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+
+ if ( write(fd, &port, sizeof(port)) != sizeof(port) )
+ return -1;
+ return 0;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/evtchn/include/xenevtchn.h b/tools/libs/evtchn/include/xenevtchn.h
new file mode 100644
index 0000000..3380fa3
--- /dev/null
+++ b/tools/libs/evtchn/include/xenevtchn.h
@@ -0,0 +1,150 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split off from:
+ * xenctrl.h
+ *
+ * A library for low-level access to the Xen control interfaces.
+ *
+ * Copyright (c) 2003-2004, K A Fraser.
+ */
+
+#ifndef XENEVTCHN_H
+#define XENEVTCHN_H
+
+#include <stdint.h>
+
+#include <xen/event_channel.h>
+
+/* A port identifier is guaranteed to fit in 31 bits. */
+typedef int evtchn_port_or_error_t;
+
+typedef struct xenevtchn_handle xenevtchn_handle;
+
+/* Callers who don't care don't need to #include <xentoollog.h> */
+typedef struct xentoollog_logger xentoollog_logger;
+
+/*
+ * EVENT CHANNEL FUNCTIONS
+ *
+ * None of these do any logging.
+ */
+
+/*
+ * Return a handle to the event channel driver, or NULL on failure, in
+ * which case errno will be set appropriately.
+ *
+ * Note:
+ * After fork a child process must not use any opened xc evtchn
+ * handle inherited from their parent. They must open a new handle if
+ * they want to interact with xc.
+ *
+ * Before Xen pre-4.1 this function would sometimes report errors with perror.
+ */
+/* Currently no flags are defined */
+xenevtchn_handle *xenevtchn_open(xentoollog_logger *logger, unsigned open_flags);
+
+/*
+ * Close a handle previously allocated with xenevtchn_open().
+ */
+int xenevtchn_close(xenevtchn_handle *xce);
+
+/*
+ * Return an fd that can be select()ed on.
+ *
+ * Note that due to bugs, setting this fd to non blocking may not
+ * work: you would hope that it would result in xenevtchn_pending
+ * failing with EWOULDBLOCK if there are no events signaled, but in
+ * fact it may block. (Bug is present in at least Linux 3.12, and
+ * perhaps on other platforms or later version.)
+ *
+ * To be safe, you must use poll() or select() before each call to
+ * xenevtchn_pending. If you have multiple threads (or processes)
+ * sharing a single xce handle this will not work, and there is no
+ * straightforward workaround. Please design your program some other
+ * way.
+ */
+int xenevtchn_fd(xenevtchn_handle *xce);
+
+/*
+ * Notify the given event channel. Returns -1 on failure, in which case
+ * errno will be set appropriately.
+ */
+int xenevtchn_notify(xenevtchn_handle *xce, evtchn_port_t port);
+
+/*
+ * Returns a new event port awaiting interdomain connection from the given
+ * domain ID, or -1 on failure, in which case errno will be set appropriately.
+ */
+evtchn_port_or_error_t
+xenevtchn_bind_unbound_port(xenevtchn_handle *xce, int domid);
+
+/*
+ * Returns a new event port bound to the remote port for the given domain ID,
+ * or -1 on failure, in which case errno will be set appropriately.
+ */
+evtchn_port_or_error_t
+xenevtchn_bind_interdomain(xenevtchn_handle *xce, int domid,
+ evtchn_port_t remote_port);
+
+/*
+ * Bind an event channel to the given VIRQ. Returns the event channel bound to
+ * the VIRQ, or -1 on failure, in which case errno will be set appropriately.
+ */
+evtchn_port_or_error_t
+xenevtchn_bind_virq(xenevtchn_handle *xce, unsigned int virq);
+
+/*
+ * Unbind the given event channel. Returns -1 on failure, in which case errno
+ * will be set appropriately.
+ */
+int xenevtchn_unbind(xenevtchn_handle *xce, evtchn_port_t port);
+
+/*
+ * Return the next event channel to become pending, or -1 on failure, in which
+ * case errno will be set appropriately.
+ *
+ * At the hypervisor level the event channel will have been masked,
+ * and then cleared, by the underlying machinery (evtchn kernel
+ * driver, or equivalent). So if the event channel is signaled again
+ * after it is returned here, it will be queued up, and delivered
+ * again after you unmask it. (See the documentation in the Xen
+ * public header event_channel.h.)
+ *
+ * On receiving the notification from xenevtchn_pending, you should
+ * normally: check (by other means) what work needs doing; do the
+ * necessary work (if any); unmask the event channel with
+ * xenevtchn_unmask (if you want to receive any further
+ * notifications).
+ */
+evtchn_port_or_error_t
+xenevtchn_pending(xenevtchn_handle *xce);
+
+/*
+ * Unmask the given event channel. Returns -1 on failure, in which case errno
+ * will be set appropriately.
+ */
+int xenevtchn_unmask(xenevtchn_handle *xce, evtchn_port_t port);
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/evtchn/libxenevtchn.map b/tools/libs/evtchn/libxenevtchn.map
new file mode 100644
index 0000000..1437940
--- /dev/null
+++ b/tools/libs/evtchn/libxenevtchn.map
@@ -0,0 +1,14 @@
+VERS_1.0 {
+ global:
+ xenevtchn_fd;
+ xenevtchn_bind_unbound_port;
+ xenevtchn_unbind;
+ xenevtchn_unmask;
+ xenevtchn_notify;
+ xenevtchn_bind_interdomain;
+ xenevtchn_bind_virq;
+ xenevtchn_open;
+ xenevtchn_close;
+ xenevtchn_pending;
+ local: *; /* Do not expose anything by default */
+};
diff --git a/tools/libs/evtchn/linux.c b/tools/libs/evtchn/linux.c
new file mode 100644
index 0000000..27fd6e9
--- /dev/null
+++ b/tools/libs/evtchn/linux.c
@@ -0,0 +1,136 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_linus_osdep.c:
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ */
+
+#include <errno.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <stdlib.h>
+
+#include <sys/ioctl.h>
+
+#include <xen/sys/evtchn.h>
+
+#include "private.h"
+
+int osdep_evtchn_open(xenevtchn_handle *xce)
+{
+ int fd = open("/dev/xen/evtchn", O_RDWR);
+ if ( fd == -1 )
+ return -1;
+ xce->fd = fd;
+ return 0;
+}
+
+int osdep_evtchn_close(xenevtchn_handle *xce)
+{
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
+}
+
+int xenevtchn_fd(xenevtchn_handle *xce)
+{
+ return xce->fd;
+}
+
+int xenevtchn_notify(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_notify notify;
+
+ notify.port = port;
+
+ return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_unbound_port(xenevtchn_handle *xce,
+ int domid)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_unbound_port bind;
+
+ bind.remote_domain = domid;
+
+ return ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_interdomain(xenevtchn_handle *xce,
+ int domid,
+ evtchn_port_t remote_port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_interdomain bind;
+
+ bind.remote_domain = domid;
+ bind.remote_port = remote_port;
+
+ return ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_virq(xenevtchn_handle *xce,
+ unsigned int virq)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_virq bind;
+
+ bind.virq = virq;
+
+ return ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
+}
+
+int xenevtchn_unbind(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_unbind unbind;
+
+ unbind.port = port;
+
+ return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
+}
+
+evtchn_port_or_error_t xenevtchn_pending(xenevtchn_handle *xce)
+{
+ int fd = xce->fd;
+ evtchn_port_t port;
+
+ if ( read(fd, &port, sizeof(port)) != sizeof(port) )
+ return -1;
+
+ return port;
+}
+
+int xenevtchn_unmask(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+
+ if ( write(fd, &port, sizeof(port)) != sizeof(port) )
+ return -1;
+ return 0;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/evtchn/minios.c b/tools/libs/evtchn/minios.c
new file mode 100644
index 0000000..fb913a2
--- /dev/null
+++ b/tools/libs/evtchn/minios.c
@@ -0,0 +1,269 @@
+/******************************************************************************
+ *
+ * Copyright 2007-2008 Samuel Thibault <samuel.thibault@eu.citrix.com>.
+ * All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split off from xc_minios.c
+ */
+
+#include "xen-external/bsd-sys-queue.h"
+#include <mini-os/types.h>
+#include <mini-os/os.h>
+#include <mini-os/lib.h>
+#include <mini-os/events.h>
+#include <mini-os/wait.h>
+
+#include <sys/socket.h>
+
+#include <assert.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <inttypes.h>
+#include <malloc.h>
+
+#include "private.h"
+
+extern void minios_evtchn_close_fd(int fd);
+
+extern struct wait_queue_head event_queue;
+
+//void minios_evtchn_close_fd(int fd);
+
+/* XXX Note: This is not threadsafe */
+static struct evtchn_port_info* port_alloc(int fd) {
+ struct evtchn_port_info *port_info;
+ port_info = malloc(sizeof(struct evtchn_port_info));
+ if (port_info == NULL)
+ return NULL;
+ port_info->pending = 0;
+ port_info->port = -1;
+ port_info->bound = 0;
+
+ LIST_INSERT_HEAD(&files[fd].evtchn.ports, port_info, list);
+ return port_info;
+}
+
+static void port_dealloc(struct evtchn_port_info *port_info) {
+ if (port_info->bound)
+ unbind_evtchn(port_info->port);
+ LIST_REMOVE(port_info, list);
+ free(port_info);
+}
+
+int osdep_evtchn_open(xenevtchn_handle *xce)
+{
+ int fd = alloc_fd(FTYPE_EVTCHN);
+ if ( fd == -1 )
+ return -1;
+ LIST_INIT(&files[fd].evtchn.ports);
+ xce->fd = fd;
+ printf("evtchn_open() -> %d\n", fd);
+ return 0;
+}
+
+int osdep_evtchn_close(xenevtchn_handle *xce)
+{
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
+}
+
+void minios_evtchn_close_fd(int fd)
+{
+ struct evtchn_port_info *port_info, *tmp;
+ LIST_FOREACH_SAFE(port_info, &files[fd].evtchn.ports, list, tmp)
+ port_dealloc(port_info);
+
+ files[fd].type = FTYPE_NONE;
+}
+
+int xenevtchn_fd(xenevtchn_handle *xce)
+{
+ return xce->fd;
+}
+
+int xenevtchn_notify(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int ret;
+
+ ret = notify_remote_via_evtchn(port);
+
+ if (ret < 0) {
+ errno = -ret;
+ ret = -1;
+ }
+ return ret;
+}
+
+static void evtchn_handler(evtchn_port_t port, struct pt_regs *regs, void *data)
+{
+ int fd = (int)(intptr_t)data;
+ struct evtchn_port_info *port_info;
+ assert(files[fd].type == FTYPE_EVTCHN);
+ mask_evtchn(port);
+ LIST_FOREACH(port_info, &files[fd].evtchn.ports, list) {
+ if (port_info->port == port)
+ goto found;
+ }
+ printk("Unknown port for handle %d\n", fd);
+ return;
+
+ found:
+ port_info->pending = 1;
+ files[fd].read = 1;
+ wake_up(&event_queue);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_unbound_port(xenevtchn_handle *xce, int domid)
+{
+ int fd = xce->fd;
+ struct evtchn_port_info *port_info;
+ int ret;
+ evtchn_port_t port;
+
+ assert(get_current() == main_thread);
+ port_info = port_alloc(fd);
+ if (port_info == NULL)
+ return -1;
+
+ printf("xenevtchn_bind_unbound_port(%d)", domid);
+ ret = evtchn_alloc_unbound(domid, evtchn_handler, (void*)(intptr_t)fd, &port);
+ printf(" = %d\n", ret);
+
+ if (ret < 0) {
+ port_dealloc(port_info);
+ errno = -ret;
+ return -1;
+ }
+ port_info->bound = 1;
+ port_info->port = port;
+ unmask_evtchn(port);
+ return port;
+}
+
+evtchn_port_or_error_t xenevtchn_bind_interdomain(xenevtchn_handle *xce, int domid,
+ evtchn_port_t remote_port)
+{
+ int fd = xce->fd;
+ struct evtchn_port_info *port_info;
+ evtchn_port_t local_port;
+ int ret;
+
+ assert(get_current() == main_thread);
+ port_info = port_alloc(fd);
+ if (port_info == NULL)
+ return -1;
+
+ printf("xenevtchn_bind_interdomain(%d, %"PRId32")", domid, remote_port);
+ ret = evtchn_bind_interdomain(domid, remote_port, evtchn_handler, (void*)(intptr_t)fd, &local_port);
+ printf(" = %d\n", ret);
+
+ if (ret < 0) {
+ port_dealloc(port_info);
+ errno = -ret;
+ return -1;
+ }
+ port_info->bound = 1;
+ port_info->port = local_port;
+ unmask_evtchn(local_port);
+ return local_port;
+}
+
+int xenevtchn_unbind(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct evtchn_port_info *port_info;
+
+ LIST_FOREACH(port_info, &files[fd].evtchn.ports, list) {
+ if (port_info->port == port) {
+ port_dealloc(port_info);
+ return 0;
+ }
+ }
+ printf("Warning: couldn't find port %"PRId32" for xc handle %x\n", port, fd);
+ errno = EINVAL;
+ return -1;
+}
+
+evtchn_port_or_error_t xenevtchn_bind_virq(xenevtchn_handle *xce, unsigned int virq)
+{
+ int fd = xce->fd;
+ struct evtchn_port_info *port_info;
+ evtchn_port_t port;
+
+ assert(get_current() == main_thread);
+ port_info = port_alloc(fd);
+ if (port_info == NULL)
+ return -1;
+
+ printf("xenevtchn_bind_virq(%d)", virq);
+ port = bind_virq(virq, evtchn_handler, (void*)(intptr_t)fd);
+
+ if (port < 0) {
+ port_dealloc(port_info);
+ errno = -port;
+ return -1;
+ }
+ port_info->bound = 1;
+ port_info->port = port;
+ unmask_evtchn(port);
+ return port;
+}
+
+evtchn_port_or_error_t xenevtchn_pending(xenevtchn_handle *xce)
+{
+ int fd = xce->fd;
+ struct evtchn_port_info *port_info;
+ unsigned long flags;
+ evtchn_port_t ret = -1;
+
+ local_irq_save(flags);
+ files[fd].read = 0;
+
+ LIST_FOREACH(port_info, &files[fd].evtchn.ports, list) {
+ if (port_info->port != -1 && port_info->pending) {
+ if (ret == -1) {
+ ret = port_info->port;
+ port_info->pending = 0;
+ } else {
+ files[fd].read = 1;
+ break;
+ }
+ }
+ }
+ local_irq_restore(flags);
+ return ret;
+}
+
+int xenevtchn_unmask(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ unmask_evtchn(port);
+ return 0;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/evtchn/netbsd.c b/tools/libs/evtchn/netbsd.c
new file mode 100644
index 0000000..c4123fe
--- /dev/null
+++ b/tools/libs/evtchn/netbsd.c
@@ -0,0 +1,147 @@
+/******************************************************************************
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_netbsd.c
+ */
+
+#include <fcntl.h>
+#include <unistd.h>
+
+#include <sys/ioctl.h>
+
+#include <xen/sys/evtchn.h>
+
+#include "private.h"
+
+#define EVTCHN_DEV_NAME "/dev/xenevt"
+
+int osdep_evtchn_open(xenevtchn_handle *xce)
+{
+ int fd = open(EVTCHN_DEV_NAME, O_NONBLOCK|O_RDWR);
+ if ( fd == -1 )
+ return -1;
+ xce->fd = fd;
+ return 0;
+}
+
+int osdep_evtchn_close(xenevtchn_handle *xce)
+{
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
+}
+
+int xenevtchn_fd(xenevtchn_handle *xce)
+{
+ return xce->fd;
+}
+
+int xenevtchn_notify(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_notify notify;
+
+ notify.port = port;
+
+ return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_unbound_port(xenevtchn_handle * xce, int domid)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_unbound_port bind;
+ int ret;
+
+ bind.remote_domain = domid;
+
+ ret = ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
+ if (ret == 0)
+ return bind.port;
+ else
+ return -1;
+}
+
+evtchn_port_or_error_t xenevtchn_bind_interdomain(xenevtchn_handle *xce, int domid,
+ evtchn_port_t remote_port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_interdomain bind;
+ int ret;
+
+ bind.remote_domain = domid;
+ bind.remote_port = remote_port;
+
+ ret = ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
+ if (ret == 0)
+ return bind.port;
+ else
+ return -1;
+}
+
+int xenevtchn_unbind(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_unbind unbind;
+
+ unbind.port = port;
+
+ return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_virq(xenevtchn_handle *xce, unsigned int virq)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_virq bind;
+ int err;
+
+ bind.virq = virq;
+
+ err = ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
+ if (err)
+ return -1;
+ else
+ return bind.port;
+}
+
+evtchn_port_or_error_t xenevtchn_pending(xenevtchn_handle *xce)
+{
+ int fd = xce->fd;
+ evtchn_port_t port;
+
+ if ( read_exact(fd, (char *)&port, sizeof(port)) == -1 )
+ return -1;
+
+ return port;
+}
+
+int xenevtchn_unmask(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ return write_exact(fd, (char *)&port, sizeof(port));
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/evtchn/private.h b/tools/libs/evtchn/private.h
new file mode 100644
index 0000000..fcd0e96
--- /dev/null
+++ b/tools/libs/evtchn/private.h
@@ -0,0 +1,25 @@
+#ifndef XENEVTCHN_PRIVATE_H
+#define XENEVTCHN_PRIVATE_H
+
+#include <xentoollog.h>
+#include <xenevtchn.h>
+
+struct xenevtchn_handle {
+ xentoollog_logger *logger, *logger_tofree;
+ int fd;
+};
+
+int osdep_evtchn_open(xenevtchn_handle *xce);
+int osdep_evtchn_close(xenevtchn_handle *xce);
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/evtchn/solaris.c b/tools/libs/evtchn/solaris.c
new file mode 100644
index 0000000..114cefb
--- /dev/null
+++ b/tools/libs/evtchn/solaris.c
@@ -0,0 +1,135 @@
+/******************************************************************************
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_solaris.c
+ */
+
+#include <fcntl.h>
+#include <unistd.h>
+
+#include <sys/ioctl.h>
+
+#include <xen/sys/evtchn.h>
+
+#include "private.h"
+
+int osdep_evtchn_open(xenevtchn_handle *xce)
+{
+ int fd;
+
+ if ( (fd = open("/dev/xen/evtchn", O_RDWR)) == -1 )
+ {
+ PERROR("Could not open event channel interface");
+ return -1;
+ }
+
+ xce->fd = fd;
+ return 0;
+}
+
+int osdep_evtchn_close(xenevtchn_handle *xce)
+{
+ if ( xce->fd == -1 )
+ return 0;
+
+ return close(xce->fd);
+}
+
+int xenevtchn_fd(xenevtchn_handle *xce)
+{
+ return xce->fd;
+}
+
+int xenevtchn_notify(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_notify notify;
+
+ notify.port = port;
+
+ return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_unbound_port(xenevtchn_handle *xce, int domid)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_unbound_port bind;
+
+ bind.remote_domain = domid;
+
+ return ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_interdomain(xenevtchn_handle *xce, int domid,
+ evtchn_port_t remote_port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_interdomain bind;
+
+ bind.remote_domain = domid;
+ bind.remote_port = remote_port;
+
+ return ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
+}
+
+evtchn_port_or_error_t xenevtchn_bind_virq(xenevtchn_handle *xce, unsigned int virq)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_bind_virq bind;
+
+ bind.virq = virq;
+
+ return ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
+}
+
+int xenevtchn_unbind(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ struct ioctl_evtchn_unbind unbind;
+
+ unbind.port = port;
+
+ return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
+}
+
+evtchn_port_or_error_t xenevtchn_pending(xenevtchn_handle *xce)
+{
+ int fd = xce->fd;
+ evtchn_port_t port;
+
+ if ( read_exact(fd, (char *)&port, sizeof(port)) == -1 )
+ return -1;
+
+ return port;
+}
+
+int xenevtchn_unmask(xenevtchn_handle *xce, evtchn_port_t port)
+{
+ int fd = xce->fd;
+ return write_exact(fd, (char *)&port, sizeof(port));
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/toollog/include/xentoollog.h b/tools/libs/toollog/include/xentoollog.h
index 853e9c7..76f17fe 100644
--- a/tools/libs/toollog/include/xentoollog.h
+++ b/tools/libs/toollog/include/xentoollog.h
@@ -134,3 +134,13 @@ const char *xtl_level_to_string(xentoollog_level); /* never fails */
#endif /* XENTOOLLOG_H */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/toollog/xtl_core.c b/tools/libs/toollog/xtl_core.c
index c4724a0..099d2f3 100644
--- a/tools/libs/toollog/xtl_core.c
+++ b/tools/libs/toollog/xtl_core.c
@@ -81,3 +81,13 @@ void xtl_logger_destroy(struct xentoollog_logger *logger) {
if (!logger) return;
logger->destroy(logger);
}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/toollog/xtl_logger_stdio.c b/tools/libs/toollog/xtl_logger_stdio.c
index 0cd9206..f9c5bd8 100644
--- a/tools/libs/toollog/xtl_logger_stdio.c
+++ b/tools/libs/toollog/xtl_logger_stdio.c
@@ -190,3 +190,13 @@ xentoollog_logger_stdiostream *xtl_createlogger_stdiostream
return XTL_NEW_LOGGER(stdiostream, newlogger);
}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libvchan/Makefile b/tools/libvchan/Makefile
index 3c50fe6..84128a3 100644
--- a/tools/libvchan/Makefile
+++ b/tools/libvchan/Makefile
@@ -10,9 +10,9 @@ NODE_OBJS = node.o
NODE2_OBJS = node-select.o
LIBVCHAN_PIC_OBJS = $(patsubst %.o,%.opic,$(LIBVCHAN_OBJS))
-LIBVCHAN_LIBS = $(LDLIBS_libxenstore) $(LDLIBS_libxenctrl)
-$(LIBVCHAN_OBJS) $(LIBVCHAN_PIC_OBJS): CFLAGS += $(CFLAGS_libxenstore) $(CFLAGS_libxenctrl)
-$(NODE_OBJS) $(NODE2_OBJS): CFLAGS += $(CFLAGS_libxenctrl)
+LIBVCHAN_LIBS = $(LDLIBS_libxenstore) $(LDLIBS_libxenctrl) $(LDLIBS_libxenevtchn)
+$(LIBVCHAN_OBJS) $(LIBVCHAN_PIC_OBJS): CFLAGS += $(CFLAGS_libxenstore) $(CFLAGS_libxenctrl) $(CFLAGS_libxenevtchn)
+$(NODE_OBJS) $(NODE2_OBJS): CFLAGS += $(CFLAGS_libxenctrl) $(CFLAGS_libxenevtchn)
MAJOR = 1.0
MINOR = 0
diff --git a/tools/libvchan/init.c b/tools/libvchan/init.c
index 77be4e7..66cb103 100644
--- a/tools/libvchan/init.c
+++ b/tools/libvchan/init.c
@@ -216,25 +216,25 @@ static int init_evt_srv(struct libxenvchan *ctrl, int domain, xentoollog_logger
{
evtchn_port_or_error_t port;
- ctrl->event = xc_evtchn_open(logger, 0);
+ ctrl->event = xenevtchn_open(logger, 0);
if (!ctrl->event)
return -1;
- port = xc_evtchn_bind_unbound_port(ctrl->event, domain);
+ port = xenevtchn_bind_unbound_port(ctrl->event, domain);
if (port < 0)
goto fail;
ctrl->event_port = port;
- if (xc_evtchn_unmask(ctrl->event, ctrl->event_port))
+ if (xenevtchn_unmask(ctrl->event, ctrl->event_port))
goto fail;
return 0;
fail:
if (port >= 0)
- xc_evtchn_unbind(ctrl->event, port);
+ xenevtchn_unbind(ctrl->event, port);
- xc_evtchn_close(ctrl->event);
+ xenevtchn_close(ctrl->event);
ctrl->event = NULL;
return -1;
@@ -346,26 +346,26 @@ static int init_evt_cli(struct libxenvchan *ctrl, int domain, xentoollog_logger
{
evtchn_port_or_error_t port;
- ctrl->event = xc_evtchn_open(logger, 0);
+ ctrl->event = xenevtchn_open(logger, 0);
if (!ctrl->event)
return -1;
- port = xc_evtchn_bind_interdomain(ctrl->event,
+ port = xenevtchn_bind_interdomain(ctrl->event,
domain, ctrl->event_port);
if (port < 0)
goto fail;
ctrl->event_port = port;
- if (xc_evtchn_unmask(ctrl->event, ctrl->event_port))
+ if (xenevtchn_unmask(ctrl->event, ctrl->event_port))
goto fail;
return 0;
fail:
if (port >= 0)
- xc_evtchn_unbind(ctrl->event, port);
+ xenevtchn_unbind(ctrl->event, port);
- xc_evtchn_close(ctrl->event);
+ xenevtchn_close(ctrl->event);
ctrl->event = NULL;
return -1;
diff --git a/tools/libvchan/io.c b/tools/libvchan/io.c
index 8a9629b..4461490 100644
--- a/tools/libvchan/io.c
+++ b/tools/libvchan/io.c
@@ -105,7 +105,7 @@ static inline int send_notify(struct libxenvchan *ctrl, uint8_t bit)
notify = ctrl->is_server ? &ctrl->ring->srv_notify : &ctrl->ring->cli_notify;
prev = __sync_fetch_and_and(notify, ~bit);
if (prev & bit)
- return xc_evtchn_notify(ctrl->event, ctrl->event_port);
+ return xenevtchn_notify(ctrl->event, ctrl->event_port);
else
return 0;
}
@@ -194,10 +194,10 @@ int libxenvchan_buffer_space(struct libxenvchan *ctrl)
int libxenvchan_wait(struct libxenvchan *ctrl)
{
- int ret = xc_evtchn_pending(ctrl->event);
+ int ret = xenevtchn_pending(ctrl->event);
if (ret < 0)
return -1;
- xc_evtchn_unmask(ctrl->event, ret);
+ xenevtchn_unmask(ctrl->event, ret);
return 0;
}
@@ -350,7 +350,7 @@ int libxenvchan_is_open(struct libxenvchan* ctrl)
int libxenvchan_fd_for_select(struct libxenvchan *ctrl)
{
- return xc_evtchn_fd(ctrl->event);
+ return xenevtchn_fd(ctrl->event);
}
void libxenvchan_close(struct libxenvchan *ctrl)
@@ -372,8 +372,8 @@ void libxenvchan_close(struct libxenvchan *ctrl)
}
if (ctrl->event) {
if (ctrl->ring)
- xc_evtchn_notify(ctrl->event, ctrl->event_port);
- xc_evtchn_close(ctrl->event);
+ xenevtchn_notify(ctrl->event, ctrl->event_port);
+ xenevtchn_close(ctrl->event);
}
if (ctrl->is_server) {
if (ctrl->gntshr)
diff --git a/tools/libvchan/libxenvchan.h b/tools/libvchan/libxenvchan.h
index 0944a0e..1544378 100644
--- a/tools/libvchan/libxenvchan.h
+++ b/tools/libvchan/libxenvchan.h
@@ -44,6 +44,7 @@
#include <xen/io/libxenvchan.h>
#include <xen/sys/evtchn.h>
+#include <xenevtchn.h>
#include <xenctrl.h>
struct libxenvchan_ring {
@@ -71,7 +72,7 @@ struct libxenvchan {
/* Pointer to shared ring page */
struct vchan_interface *ring;
/* event channel interface */
- xc_evtchn *event;
+ xenevtchn_handle *event;
uint32_t event_port;
/* informative flags: are we acting as server? */
int is_server:1;
diff --git a/tools/libxc/Makefile b/tools/libxc/Makefile
index 940708f..b8fc6a5 100644
--- a/tools/libxc/Makefile
+++ b/tools/libxc/Makefile
@@ -48,6 +48,7 @@ CTRL_SRCS-$(CONFIG_SunOS) += xc_solaris.c
CTRL_SRCS-$(CONFIG_NetBSD) += xc_netbsd.c
CTRL_SRCS-$(CONFIG_NetBSDRump) += xc_netbsd.c
CTRL_SRCS-$(CONFIG_MiniOS) += xc_minios.c
+CTRL_SRCS-y += xc_evtchn_compat.c
GUEST_SRCS-y :=
GUEST_SRCS-y += xg_private.c xc_suspend.c
@@ -111,6 +112,7 @@ CFLAGS-$(CONFIG_Linux) += -D_GNU_SOURCE
CFLAGS += $(PTHREAD_CFLAGS)
CFLAGS += $(CFLAGS_libxentoollog)
+CFLAGS += $(CFLAGS_libxenevtchn)
CTRL_LIB_OBJS := $(patsubst %.c,%.o,$(CTRL_SRCS-y))
CTRL_PIC_OBJS := $(patsubst %.c,%.opic,$(CTRL_SRCS-y))
@@ -164,7 +166,7 @@ install: build
$(INSTALL_DATA) libxenctrl.a $(DESTDIR)$(libdir)
$(SYMLINK_SHLIB) libxenctrl.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxenctrl.so.$(MAJOR)
$(SYMLINK_SHLIB) libxenctrl.so.$(MAJOR) $(DESTDIR)$(libdir)/libxenctrl.so
- $(INSTALL_DATA) include/xenctrl.h include/xenctrlosdep.h $(DESTDIR)$(includedir)
+ $(INSTALL_DATA) include/xenctrl.h include/xenctrl_compat.h include/xenctrlosdep.h $(DESTDIR)$(includedir)
$(INSTALL_SHLIB) libxenguest.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)
$(INSTALL_DATA) libxenguest.a $(DESTDIR)$(libdir)
$(SYMLINK_SHLIB) libxenguest.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxenguest.so.$(MAJOR)
@@ -207,7 +209,7 @@ libxenctrl.so.$(MAJOR): libxenctrl.so.$(MAJOR).$(MINOR)
$(SYMLINK_SHLIB) $< $@
libxenctrl.so.$(MAJOR).$(MINOR): $(CTRL_PIC_OBJS)
- $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(DLOPEN_LIBS) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(DLOPEN_LIBS) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
# libxenguest
@@ -230,7 +232,7 @@ xc_dom_bzimageloader.opic: CFLAGS += $(call zlib-options,D)
libxenguest.so.$(MAJOR).$(MINOR): COMPRESSION_LIBS = $(call zlib-options,l)
libxenguest.so.$(MAJOR).$(MINOR): $(GUEST_PIC_OBJS) libxenctrl.so
- $(CC) $(LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenguest.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $(GUEST_PIC_OBJS) $(COMPRESSION_LIBS) -lz $(LDLIBS_libxenctrl) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenguest.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $(GUEST_PIC_OBJS) $(COMPRESSION_LIBS) -lz $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
xenctrl_osdep_ENOSYS.so: $(OSDEP_PIC_OBJS) libxenctrl.so
$(CC) $(LDFLAGS) $(SHLIB_LDFLAGS) -o $@ $(OSDEP_PIC_OBJS) $(LDLIBS_libxenctrl) $(APPEND_LDFLAGS)
diff --git a/tools/libxc/include/xenctrl.h b/tools/libxc/include/xenctrl.h
index 29d55ad..ca3f776 100644
--- a/tools/libxc/include/xenctrl.h
+++ b/tools/libxc/include/xenctrl.h
@@ -117,7 +117,6 @@
*/
typedef struct xc_interface_core xc_interface;
-typedef struct xenevtchn_handle xc_evtchn;
typedef struct xc_interface_core xc_gnttab;
typedef struct xc_interface_core xc_gntshr;
@@ -1093,7 +1092,6 @@ int xc_cpupool_movedomain(xc_interface *xch,
*/
xc_cpumap_t xc_cpupool_freeinfo(xc_interface *xch);
-
/*
* EVENT CHANNEL FUNCTIONS
*
@@ -1128,101 +1126,7 @@ int xc_evtchn_reset(xc_interface *xch,
typedef struct evtchn_status xc_evtchn_status_t;
int xc_evtchn_status(xc_interface *xch, xc_evtchn_status_t *status);
-/*
- * Return a handle to the event channel driver, or NULL on failure, in
- * which case errno will be set appropriately.
- *
- * Note:
- * After fork a child process must not use any opened xc evtchn
- * handle inherited from their parent. They must open a new handle if
- * they want to interact with xc.
- *
- * Before Xen pre-4.1 this function would sometimes report errors with perror.
- */
-xc_evtchn *xc_evtchn_open(xentoollog_logger *logger,
- unsigned open_flags);
-/*
- * Close a handle previously allocated with xc_evtchn_open().
- */
-int xc_evtchn_close(xc_evtchn *xce);
-
-/*
- * Return an fd that can be select()ed on.
- *
- * Note that due to bugs, setting this fd to non blocking may not
- * work: you would hope that it would result in xc_evtchn_pending
- * failing with EWOULDBLOCK if there are no events signaled, but in
- * fact it may block. (Bug is present in at least Linux 3.12, and
- * perhaps on other platforms or later version.)
- *
- * To be safe, you must use poll() or select() before each call to
- * xc_evtchn_pending. If you have multiple threads (or processes)
- * sharing a single xce handle this will not work, and there is no
- * straightforward workaround. Please design your program some other
- * way.
- */
-int xc_evtchn_fd(xc_evtchn *xce);
-
-/*
- * Notify the given event channel. Returns -1 on failure, in which case
- * errno will be set appropriately.
- */
-int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port);
-
-/*
- * Returns a new event port awaiting interdomain connection from the given
- * domain ID, or -1 on failure, in which case errno will be set appropriately.
- */
-evtchn_port_or_error_t
-xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid);
-
-/*
- * Returns a new event port bound to the remote port for the given domain ID,
- * or -1 on failure, in which case errno will be set appropriately.
- */
-evtchn_port_or_error_t
-xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
- evtchn_port_t remote_port);
-
-/*
- * Bind an event channel to the given VIRQ. Returns the event channel bound to
- * the VIRQ, or -1 on failure, in which case errno will be set appropriately.
- */
-evtchn_port_or_error_t
-xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq);
-
-/*
- * Unbind the given event channel. Returns -1 on failure, in which case errno
- * will be set appropriately.
- */
-int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port);
-
-/*
- * Return the next event channel to become pending, or -1 on failure, in which
- * case errno will be set appropriately.
- *
- * At the hypervisor level the event channel will have been masked,
- * and then cleared, by the underlying machinery (evtchn kernel
- * driver, or equivalent). So if the event channel is signaled again
- * after it is returned here, it will be queued up, and delivered
- * again after you unmask it. (See the documentation in the Xen
- * public header event_channel.h.)
- *
- * On receiving the notification from xc_evtchn_pending, you should
- * normally: check (by other means) what work needs doing; do the
- * necessary work (if any); unmask the event channel with
- * xc_evtchn_unmask (if you want to receive any further
- * notifications).
- */
-evtchn_port_or_error_t
-xc_evtchn_pending(xc_evtchn *xce);
-
-/*
- * Unmask the given event channel. Returns -1 on failure, in which case errno
- * will be set appropriately.
- */
-int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port);
int xc_physdev_pci_access_modify(xc_interface *xch,
uint32_t domid,
@@ -2859,6 +2763,9 @@ int xc_psr_cat_get_l3_info(xc_interface *xch, uint32_t socket,
bool *cdp_enabled);
#endif
+/* Compat shims */
+#include "xenctrl_compat.h"
+
#endif /* XENCTRL_H */
/*
diff --git a/tools/libxc/include/xenctrl_compat.h b/tools/libxc/include/xenctrl_compat.h
new file mode 100644
index 0000000..48daeb2
--- /dev/null
+++ b/tools/libxc/include/xenctrl_compat.h
@@ -0,0 +1,48 @@
+/*
+ * Compat shims for use of 3rd party consumers of libxenctrl
+ * functionality which has been split into separate libraries.
+ *
+ * New code should use the separate libraries.
+ *
+ * Each interface must be opted-into separately by defining:
+ *
+ * XC_WANT_COMPAT_EVTCHN_API
+ * - Functions relating to /dev/xen/evtchn
+ */
+#ifndef XENCTRL_COMPAT_H
+#define XENCTRL_COMPAT_H
+
+#ifdef XC_WANT_COMPAT_EVTCHN_API
+
+typedef struct xenevtchn_handle xc_evtchn;
+
+xc_evtchn *xc_evtchn_open(xentoollog_logger *logger,
+ unsigned open_flags);
+int xc_evtchn_close(xc_evtchn *xce);
+int xc_evtchn_fd(xc_evtchn *xce);
+int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port);
+evtchn_port_or_error_t
+xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid);
+evtchn_port_or_error_t
+xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
+ evtchn_port_t remote_port);
+evtchn_port_or_error_t
+xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq);
+int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port);
+evtchn_port_or_error_t
+xc_evtchn_pending(xc_evtchn *xce);
+int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port);
+
+#endif /* XC_WANT_COMPAT_EVTCHN_API */
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libxc/include/xenguest.h b/tools/libxc/include/xenguest.h
index ec67fbd..c7bffe1 100644
--- a/tools/libxc/include/xenguest.h
+++ b/tools/libxc/include/xenguest.h
@@ -34,6 +34,12 @@
#define X86_64_B_SIZE 64
#define X86_32_B_SIZE 32
+/*
+ * User not using xc_suspend_* / xc_await_suspent may not want to
+ * include the full libxenevtchn API here.
+ */
+typedef struct xenevtchn_handle xenevtchn_handle;
+
/* callbacks provided by xc_domain_save */
struct save_callbacks {
/* Called after expiration of checkpoint interval,
@@ -209,18 +215,18 @@ struct xc_hvm_firmware_module {
* Sets *lockfd to -1.
* Has deallocated everything even on error.
*/
-int xc_suspend_evtchn_release(xc_interface *xch, xc_evtchn *xce, int domid, int suspend_evtchn, int *lockfd);
+int xc_suspend_evtchn_release(xc_interface *xch, xenevtchn_handle *xce, int domid, int suspend_evtchn, int *lockfd);
/**
* This function eats the initial notification.
* xce must not be used for anything else
* See xc_suspend_evtchn_init_sane re lockfd.
*/
-int xc_suspend_evtchn_init_exclusive(xc_interface *xch, xc_evtchn *xce,
+int xc_suspend_evtchn_init_exclusive(xc_interface *xch, xenevtchn_handle *xce,
int domid, int port, int *lockfd);
/* xce must not be used for anything else */
-int xc_await_suspend(xc_interface *xch, xc_evtchn *xce, int suspend_evtchn);
+int xc_await_suspend(xc_interface *xch, xenevtchn_handle *xce, int suspend_evtchn);
/**
* The port will be signaled immediately after this call
@@ -229,7 +235,7 @@ int xc_await_suspend(xc_interface *xch, xc_evtchn *xce, int suspend_evtchn);
* and fed to xc_suspend_evtchn_release. (On error *lockfd is
* undefined and xc_suspend_evtchn_release is not allowed.)
*/
-int xc_suspend_evtchn_init_sane(xc_interface *xch, xc_evtchn *xce,
+int xc_suspend_evtchn_init_sane(xc_interface *xch, xenevtchn_handle *xce,
int domid, int port, int *lockfd);
int xc_get_bit_size(xc_interface *xch,
diff --git a/tools/libxc/xc_evtchn_compat.c b/tools/libxc/xc_evtchn_compat.c
new file mode 100644
index 0000000..5d3e4ba
--- /dev/null
+++ b/tools/libxc/xc_evtchn_compat.c
@@ -0,0 +1,75 @@
+/*
+ * Compat shims for use of 3rd party consumers of libxenctrl xc_evtchn
+ * functionality which has been split into separate libraries.
+ */
+
+#include <xenevtchn.h>
+
+#define XC_WANT_COMPAT_EVTCHN_API
+#include "xenctrl.h"
+
+xc_evtchn *xc_evtchn_open(xentoollog_logger *logger,
+ unsigned open_flags)
+{
+ return xenevtchn_open(logger, open_flags);
+}
+
+int xc_evtchn_close(xc_evtchn *xce)
+{
+ return xenevtchn_close(xce);
+}
+
+int xc_evtchn_fd(xc_evtchn *xce)
+{
+ return xenevtchn_fd(xce);
+}
+
+int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
+{
+ return xenevtchn_notify(xce, port);
+}
+
+evtchn_port_or_error_t
+xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
+{
+ return xenevtchn_bind_unbound_port(xce, domid);
+}
+
+evtchn_port_or_error_t
+xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
+ evtchn_port_t remote_port)
+{
+ return xenevtchn_bind_interdomain(xce, domid, remote_port);
+}
+
+evtchn_port_or_error_t
+xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
+{
+ return xenevtchn_bind_virq(xce, virq);
+}
+
+int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
+{
+ return xenevtchn_unbind(xce, port);
+}
+
+evtchn_port_or_error_t
+xc_evtchn_pending(xc_evtchn *xce)
+{
+ return xenevtchn_pending(xce);
+}
+
+int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
+{
+ return xenevtchn_unmask(xce, port);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libxc/xc_freebsd_osdep.c b/tools/libxc/xc_freebsd_osdep.c
index 4323e16..339997c 100644
--- a/tools/libxc/xc_freebsd_osdep.c
+++ b/tools/libxc/xc_freebsd_osdep.c
@@ -31,13 +31,11 @@
#include <sys/ioctl.h>
#include <xen/memory.h>
-#include <xen/sys/evtchn.h>
#include "xenctrl.h"
#include "xenctrlosdep.h"
#define PRIVCMD_DEV "/dev/xen/privcmd"
-#define EVTCHN_DEV "/dev/xen/evtchn"
#define PERROR(_m, _a...) xc_osdep_log(xch,XTL_ERROR,XC_INTERNAL_ERROR,_m \
" (%d = %s)", ## _a , errno, xc_strerror(xch, errno))
@@ -251,105 +249,6 @@ static struct xc_osdep_ops freebsd_privcmd_ops = {
},
};
-/*-------------------------- Evtchn device interface -------------------------*/
-int osdep_evtchn_open(xc_evtchn *xce)
-{
- int fd = open(EVTCHN_DEV, O_RDWR);
- if ( fd == -1 )
- return -1;
- xce->fd = fd;
- return 0;
-}
-
-int osdep_evtchn_close(xc_evtchn *xce)
-{
- if ( xce->fd == -1 )
- return 0;
-
- return close(xce->fd);
-}
-
-int xc_evtchn_fd(xc_evtchn *xce)
-{
- return xce->fd;
-}
-
-/*------------------------------ Evtchn interface ----------------------------*/
-int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_notify notify;
-
- notify.port = port;
-
- return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
-{
- int ret, fd = xce->fd;
- struct ioctl_evtchn_bind_unbound_port bind;
-
- bind.remote_domain = domid;
-
- ret = ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
- return ( ret == 0 ) ? bind.port : ret;
-}
-
-evtchn_port_or_error_t
-xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid, evtchn_port_t remote_port)
-{
- int ret, fd = xce->fd;
- struct ioctl_evtchn_bind_interdomain bind;
-
- bind.remote_domain = domid;
- bind.remote_port = remote_port;
-
- ret = ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
- return ( ret == 0 ) ? bind.port : ret;
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
-{
- int ret, fd = xce->fd;
- struct ioctl_evtchn_bind_virq bind;
-
- bind.virq = virq;
-
- ret = ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
- return ( ret == 0 ) ? bind.port : ret;
-}
-
-int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_unbind unbind;
-
- unbind.port = port;
-
- return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
-}
-
-evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
-{
- int fd = xce->fd;
- evtchn_port_t port;
-
- if ( read(fd, &port, sizeof(port)) != sizeof(port) )
- return -1;
-
- return port;
-}
-
-int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
-
- if ( write(fd, &port, sizeof(port)) != sizeof(port) )
- return -1;
- return 0;
-}
-
/*---------------------------- FreeBSD interface -----------------------------*/
static struct xc_osdep_ops *
freebsd_osdep_init(xc_interface *xch, enum xc_osdep_type type)
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index c1f34de..1883aba 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -31,7 +31,6 @@
#include <sys/ioctl.h>
#include <xen/memory.h>
-#include <xen/sys/evtchn.h>
#include <xen/sys/gntdev.h>
#include <xen/sys/gntalloc.h>
@@ -455,100 +454,6 @@ static struct xc_osdep_ops linux_privcmd_ops = {
#define DEVXEN "/dev/xen/"
-int osdep_evtchn_open(xc_evtchn *xce)
-{
- int fd = open(DEVXEN "evtchn", O_RDWR);
- if ( fd == -1 )
- return -1;
- xce->fd = fd;
- return 0;
-}
-
-int osdep_evtchn_close(xc_evtchn *xce)
-{
- if ( xce->fd == -1 )
- return 0;
-
- return close(xce->fd);
-}
-
-int xc_evtchn_fd(xc_evtchn *xce)
-{
- return xce->fd;
-}
-
-int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_notify notify;
-
- notify.port = port;
-
- return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_unbound_port bind;
-
- bind.remote_domain = domid;
-
- return ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
- evtchn_port_t remote_port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_interdomain bind;
-
- bind.remote_domain = domid;
- bind.remote_port = remote_port;
-
- return ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_virq bind;
-
- bind.virq = virq;
-
- return ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
-}
-
-int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_unbind unbind;
-
- unbind.port = port;
-
- return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
-}
-
-evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
-{
- int fd = xce->fd;
- evtchn_port_t port;
-
- if ( read(fd, &port, sizeof(port)) != sizeof(port) )
- return -1;
-
- return port;
-}
-
-int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
-
- if ( write(fd, &port, sizeof(port)) != sizeof(port) )
- return -1;
- return 0;
-}
-
static xc_osdep_handle linux_gnttab_open(xc_gnttab *xcg)
{
int fd = open(DEVXEN "gntdev", O_RDWR);
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index 4ace1b5..db7b344 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -19,14 +19,11 @@
*/
#undef NDEBUG
-#include "xen-external/bsd-sys-queue.h"
#include <mini-os/types.h>
#include <mini-os/os.h>
#include <mini-os/mm.h>
#include <mini-os/lib.h>
#include <mini-os/gntmap.h>
-#include <mini-os/events.h>
-#include <mini-os/wait.h>
#include <sys/mman.h>
#include <xen/memory.h>
@@ -41,13 +38,9 @@
#include "xc_private.h"
void minios_interface_close_fd(int fd);
-void minios_evtchn_close_fd(int fd);
void minios_gnttab_close_fd(int fd);
extern void minios_interface_close_fd(int fd);
-extern void minios_evtchn_close_fd(int fd);
-
-extern struct wait_queue_head event_queue;
static xc_osdep_handle minios_privcmd_open(xc_interface *xch)
{
@@ -197,220 +190,6 @@ static struct xc_osdep_ops minios_privcmd_ops = {
},
};
-
-/* XXX Note: This is not threadsafe */
-static struct evtchn_port_info* port_alloc(int fd) {
- struct evtchn_port_info *port_info;
- port_info = malloc(sizeof(struct evtchn_port_info));
- if (port_info == NULL)
- return NULL;
- port_info->pending = 0;
- port_info->port = -1;
- port_info->bound = 0;
-
- LIST_INSERT_HEAD(&files[fd].evtchn.ports, port_info, list);
- return port_info;
-}
-
-static void port_dealloc(struct evtchn_port_info *port_info) {
- if (port_info->bound)
- unbind_evtchn(port_info->port);
- LIST_REMOVE(port_info, list);
- free(port_info);
-}
-
-int osdep_evtchn_open(xc_evtchn *xce)
-{
- int fd = alloc_fd(FTYPE_EVTCHN);
- if ( fd == -1 )
- return -1;
- LIST_INIT(&files[fd].evtchn.ports);
- xce->fd = fd;
- printf("evtchn_open() -> %d\n", fd);
- return 0;
-}
-
-int osdep_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
-{
- if ( xce->fd == -1 )
- return 0;
-
- return close(xce->fd);
-}
-
-void minios_evtchn_close_fd(int fd)
-{
- struct evtchn_port_info *port_info, *tmp;
- LIST_FOREACH_SAFE(port_info, &files[fd].evtchn.ports, list, tmp)
- port_dealloc(port_info);
-
- files[fd].type = FTYPE_NONE;
-}
-
-int xc_evtchn_fd(xc_evtchn *xce)
-{
- return xce->fd;
-}
-
-int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
-{
- int ret;
-
- ret = notify_remote_via_evtchn(port);
-
- if (ret < 0) {
- errno = -ret;
- ret = -1;
- }
- return ret;
-}
-
-static void evtchn_handler(evtchn_port_t port, struct pt_regs *regs, void *data)
-{
- int fd = (int)(intptr_t)data;
- struct evtchn_port_info *port_info;
- assert(files[fd].type == FTYPE_EVTCHN);
- mask_evtchn(port);
- LIST_FOREACH(port_info, &files[fd].evtchn.ports, list) {
- if (port_info->port == port)
- goto found;
- }
- printk("Unknown port for handle %d\n", fd);
- return;
-
- found:
- port_info->pending = 1;
- files[fd].read = 1;
- wake_up(&event_queue);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
-{
- int fd = xce->fd;
- struct evtchn_port_info *port_info;
- int ret;
- evtchn_port_t port;
-
- assert(get_current() == main_thread);
- port_info = port_alloc(fd);
- if (port_info == NULL)
- return -1;
-
- printf("xc_evtchn_bind_unbound_port(%d)", domid);
- ret = evtchn_alloc_unbound(domid, evtchn_handler, (void*)(intptr_t)fd, &port);
- printf(" = %d\n", ret);
-
- if (ret < 0) {
- port_dealloc(port_info);
- errno = -ret;
- return -1;
- }
- port_info->bound = 1;
- port_info->port = port;
- unmask_evtchn(port);
- return port;
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
- evtchn_port_t remote_port)
-{
- int fd = xce->fd;
- struct evtchn_port_info *port_info;
- evtchn_port_t local_port;
- int ret;
-
- assert(get_current() == main_thread);
- port_info = port_alloc(fd);
- if (port_info == NULL)
- return -1;
-
- printf("xc_evtchn_bind_interdomain(%d, %"PRId32")", domid, remote_port);
- ret = evtchn_bind_interdomain(domid, remote_port, evtchn_handler, (void*)(intptr_t)fd, &local_port);
- printf(" = %d\n", ret);
-
- if (ret < 0) {
- port_dealloc(port_info);
- errno = -ret;
- return -1;
- }
- port_info->bound = 1;
- port_info->port = local_port;
- unmask_evtchn(local_port);
- return local_port;
-}
-
-int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct evtchn_port_info *port_info;
-
- LIST_FOREACH(port_info, &files[fd].evtchn.ports, list) {
- if (port_info->port == port) {
- port_dealloc(port_info);
- return 0;
- }
- }
- printf("Warning: couldn't find port %"PRId32" for xc handle %x\n", port, fd);
- errno = EINVAL;
- return -1;
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
-{
- int fd = xce->fd;
- struct evtchn_port_info *port_info;
- evtchn_port_t port;
-
- assert(get_current() == main_thread);
- port_info = port_alloc(fd);
- if (port_info == NULL)
- return -1;
-
- printf("xc_evtchn_bind_virq(%d)", virq);
- port = bind_virq(virq, evtchn_handler, (void*)(intptr_t)fd);
-
- if (port < 0) {
- port_dealloc(port_info);
- errno = -port;
- return -1;
- }
- port_info->bound = 1;
- port_info->port = port;
- unmask_evtchn(port);
- return port;
-}
-
-evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
-{
- int fd = xce->fd;
- struct evtchn_port_info *port_info;
- unsigned long flags;
- evtchn_port_t ret = -1;
-
- local_irq_save(flags);
- files[fd].read = 0;
-
- LIST_FOREACH(port_info, &files[fd].evtchn.ports, list) {
- if (port_info->port != -1 && port_info->pending) {
- if (ret == -1) {
- ret = port_info->port;
- port_info->pending = 0;
- } else {
- files[fd].read = 1;
- break;
- }
- }
- }
- local_irq_restore(flags);
- return ret;
-}
-
-int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
-{
- unmask_evtchn(port);
- return 0;
-}
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
diff --git a/tools/libxc/xc_netbsd.c b/tools/libxc/xc_netbsd.c
index b1b828f..fe4f0a1 100644
--- a/tools/libxc/xc_netbsd.c
+++ b/tools/libxc/xc_netbsd.c
@@ -19,7 +19,6 @@
#include "xc_private.h"
-#include <xen/sys/evtchn.h>
#include <unistd.h>
#include <fcntl.h>
#include <malloc.h>
@@ -223,114 +222,6 @@ static struct xc_osdep_ops netbsd_privcmd_ops = {
},
};
-#define EVTCHN_DEV_NAME "/dev/xenevt"
-
-int osdep_evtchn_open(xc_evtchn *xce)
-{
- int fd = open(EVTCHN_DEV_NAME, O_NONBLOCK|O_RDWR);
- if ( fd == -1 )
- return -1;
- xce->fd = fd;
- return 0;
-}
-
-int osdep_evtchn_close(xc_evtchn *xce, xc_osdep_handle h)
-{
- if ( xce->fd == -1 )
- return 0;
-
- return close(xce->fd);
-}
-
-int xc_evtchn_fd(xc_evtchn *xce)
-{
- return xce->fd;
-}
-
-int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_notify notify;
-
- notify.port = port;
-
- return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn * xce, int domid)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_unbound_port bind;
- int ret;
-
- bind.remote_domain = domid;
-
- ret = ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
- if (ret == 0)
- return bind.port;
- else
- return -1;
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
- evtchn_port_t remote_port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_interdomain bind;
- int ret;
-
- bind.remote_domain = domid;
- bind.remote_port = remote_port;
-
- ret = ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
- if (ret == 0)
- return bind.port;
- else
- return -1;
-}
-
-int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_unbind unbind;
-
- unbind.port = port;
-
- return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_virq bind;
- int err;
-
- bind.virq = virq;
-
- err = ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
- if (err)
- return -1;
- else
- return bind.port;
-}
-
-evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
-{
- int fd = xce->fd;
- evtchn_port_t port;
-
- if ( read_exact(fd, (char *)&port, sizeof(port)) == -1 )
- return -1;
-
- return port;
-}
-
-int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- return write_exact(fd, (char *)&port, sizeof(port));
-}
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
diff --git a/tools/libxc/xc_private.c b/tools/libxc/xc_private.c
index 154e553..c6876c3 100644
--- a/tools/libxc/xc_private.c
+++ b/tools/libxc/xc_private.c
@@ -251,46 +251,6 @@ int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
return xch->ops->u.privcmd.hypercall(xch, xch->ops_handle, hypercall);
}
-xc_evtchn *xc_evtchn_open(xentoollog_logger *logger, unsigned open_flags)
-{
- xc_evtchn *xce = malloc(sizeof(*xce));
- int rc;
-
- if (!xce) return NULL;
-
- xce->fd = -1;
- xce->logger = logger;
- xce->logger_tofree = NULL;
-
- if (!xce->logger) {
- xce->logger = xce->logger_tofree =
- (xentoollog_logger*)
- xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
- if (!xce->logger) goto err;
- }
-
- rc = osdep_evtchn_open(xce);
- if ( rc < 0 ) goto err;
-
- return xce;
-
-err:
- osdep_evtchn_close(xce);
- xtl_logger_destroy(xce->logger_tofree);
- free(xce);
- return NULL;
-}
-
-int xc_evtchn_close(xc_evtchn *xce)
-{
- int rc;
-
- rc = osdep_evtchn_close(xce);
- xtl_logger_destroy(xce->logger_tofree);
- free(xce);
- return rc;
-}
-
xc_gnttab *xc_gnttab_open(xentoollog_logger *logger,
unsigned open_flags)
{
diff --git a/tools/libxc/xc_private.h b/tools/libxc/xc_private.h
index a5bf79a..46e0e75 100644
--- a/tools/libxc/xc_private.h
+++ b/tools/libxc/xc_private.h
@@ -123,13 +123,6 @@ struct xc_interface_core {
xc_osdep_handle ops_handle; /* opaque data for xc_osdep_ops */
};
-struct xenevtchn_handle {
- xentoollog_logger *logger, *logger_tofree;
- int fd;
-};
-int osdep_evtchn_open(xc_evtchn *xce);
-int osdep_evtchn_close(xc_evtchn *xce);
-
void xc_report_error(xc_interface *xch, int code, const char *fmt, ...)
__attribute__((format(printf,3,4)));
void xc_reportv(xc_interface *xch, xentoollog_logger *lg, xentoollog_level,
diff --git a/tools/libxc/xc_solaris.c b/tools/libxc/xc_solaris.c
index 088228a..ed7987c 100644
--- a/tools/libxc/xc_solaris.c
+++ b/tools/libxc/xc_solaris.c
@@ -20,7 +20,6 @@
#include "xc_private.h"
#include <xen/memory.h>
-#include <xen/sys/evtchn.h>
#include <unistd.h>
#include <fcntl.h>
#include <malloc.h>
@@ -194,102 +193,6 @@ static struct xc_osdep_ops solaris_privcmd_ops = {
},
};
-int osdep_evtchn_open(xc_evtchn *xce)
-{
- int fd;
-
- if ( (fd = open("/dev/xen/evtchn", O_RDWR)) == -1 )
- {
- PERROR("Could not open event channel interface");
- return -1;
- }
-
- xce->fd = fd;
- return 0;
-}
-
-int osdep_evtchn_close(xc_evtchn *xce)
-{
- if ( xce->fd == -1 )
- return 0;
-
- return close(xce->fd);
-}
-
-int xc_evtchn_fd(xc_evtchn *xce)
-{
- return xce->fd;
-}
-
-int xc_evtchn_notify(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_notify notify;
-
- notify.port = port;
-
- return ioctl(fd, IOCTL_EVTCHN_NOTIFY, ¬ify);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_unbound_port(xc_evtchn *xce, int domid)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_unbound_port bind;
-
- bind.remote_domain = domid;
-
- return ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_interdomain(xc_evtchn *xce, int domid,
- evtchn_port_t remote_port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_interdomain bind;
-
- bind.remote_domain = domid;
- bind.remote_port = remote_port;
-
- return ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
-}
-
-evtchn_port_or_error_t xc_evtchn_bind_virq(xc_evtchn *xce, unsigned int virq)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_bind_virq bind;
-
- bind.virq = virq;
-
- return ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
-}
-
-int xc_evtchn_unbind(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- struct ioctl_evtchn_unbind unbind;
-
- unbind.port = port;
-
- return ioctl(fd, IOCTL_EVTCHN_UNBIND, &unbind);
-}
-
-evtchn_port_or_error_t xc_evtchn_pending(xc_evtchn *xce)
-{
- int fd = xce->fd;
- evtchn_port_t port;
-
- if ( read_exact(fd, (char *)&port, sizeof(port)) == -1 )
- return -1;
-
- return port;
-}
-
-int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port)
-{
- int fd = xce->fd;
- return write_exact(fd, (char *)&port, sizeof(port));
-}
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
diff --git a/tools/libxc/xc_suspend.c b/tools/libxc/xc_suspend.c
index bba36e7..8361c7b 100644
--- a/tools/libxc/xc_suspend.c
+++ b/tools/libxc/xc_suspend.c
@@ -16,6 +16,8 @@
#include <unistd.h>
#include <fcntl.h>
+#include <xenevtchn.h>
+
#include "xc_private.h"
#include "xenguest.h"
@@ -124,12 +126,12 @@ static int unlock_suspend_event(xc_interface *xch, int domid, int *lockfd)
return -1;
}
-int xc_await_suspend(xc_interface *xch, xc_evtchn *xce, int suspend_evtchn)
+int xc_await_suspend(xc_interface *xch, xenevtchn_handle *xce, int suspend_evtchn)
{
int rc;
do {
- rc = xc_evtchn_pending(xce);
+ rc = xenevtchn_pending(xce);
if (rc < 0) {
ERROR("error polling suspend notification channel: %d", rc);
return -1;
@@ -137,7 +139,7 @@ int xc_await_suspend(xc_interface *xch, xc_evtchn *xce, int suspend_evtchn)
} while (rc != suspend_evtchn);
/* harmless for one-off suspend */
- if (xc_evtchn_unmask(xce, suspend_evtchn) < 0)
+ if (xenevtchn_unmask(xce, suspend_evtchn) < 0)
ERROR("failed to unmask suspend notification channel: %d", rc);
return 0;
@@ -145,16 +147,16 @@ int xc_await_suspend(xc_interface *xch, xc_evtchn *xce, int suspend_evtchn)
/* Internal callers are allowed to call this with suspend_evtchn<0
* but *lockfd>0. */
-int xc_suspend_evtchn_release(xc_interface *xch, xc_evtchn *xce,
+int xc_suspend_evtchn_release(xc_interface *xch, xenevtchn_handle *xce,
int domid, int suspend_evtchn, int *lockfd)
{
if (suspend_evtchn >= 0)
- xc_evtchn_unbind(xce, suspend_evtchn);
+ xenevtchn_unbind(xce, suspend_evtchn);
return unlock_suspend_event(xch, domid, lockfd);
}
-int xc_suspend_evtchn_init_sane(xc_interface *xch, xc_evtchn *xce,
+int xc_suspend_evtchn_init_sane(xc_interface *xch, xenevtchn_handle *xce,
int domid, int port, int *lockfd)
{
int rc, suspend_evtchn = -1;
@@ -164,7 +166,7 @@ int xc_suspend_evtchn_init_sane(xc_interface *xch, xc_evtchn *xce,
goto cleanup;
}
- suspend_evtchn = xc_evtchn_bind_interdomain(xce, domid, port);
+ suspend_evtchn = xenevtchn_bind_interdomain(xce, domid, port);
if (suspend_evtchn < 0) {
ERROR("failed to bind suspend event channel: %d", suspend_evtchn);
goto cleanup;
@@ -184,7 +186,7 @@ cleanup:
return -1;
}
-int xc_suspend_evtchn_init_exclusive(xc_interface *xch, xc_evtchn *xce,
+int xc_suspend_evtchn_init_exclusive(xc_interface *xch, xenevtchn_handle *xce,
int domid, int port, int *lockfd)
{
int suspend_evtchn;
diff --git a/tools/libxl/Makefile b/tools/libxl/Makefile
index 2abae0c..10b1741 100644
--- a/tools/libxl/Makefile
+++ b/tools/libxl/Makefile
@@ -20,12 +20,13 @@ LIBUUID_LIBS += -luuid
endif
LIBXL_LIBS =
-LIBXL_LIBS = $(LDLIBS_libxentoollog) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) $(LDLIBS_libblktapctl) $(PTYFUNCS_LIBS) $(LIBUUID_LIBS)
+LIBXL_LIBS = $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) $(LDLIBS_libblktapctl) $(PTYFUNCS_LIBS) $(LIBUUID_LIBS)
ifeq ($(CONFIG_REMUS_NETBUF),y)
LIBXL_LIBS += $(LIBNL3_LIBS)
endif
CFLAGS_LIBXL += $(CFLAGS_libxentoollog)
+CFLAGS_LIBXL += $(CFLAGS_libxenevtchn)
CFLAGS_LIBXL += $(CFLAGS_libxenctrl)
CFLAGS_LIBXL += $(CFLAGS_libxenguest)
CFLAGS_LIBXL += $(CFLAGS_libxenstore)
@@ -159,7 +160,7 @@ $(XEN_INIT_DOM0_OBJS): CFLAGS += $(CFLAGS_libxenctrl)
$(XEN_INIT_DOM0_OBJS): CFLAGS += $(CFLAGS_libxenstore)
SAVE_HELPER_OBJS = libxl_save_helper.o _libxl_save_msgs_helper.o
-$(SAVE_HELPER_OBJS): CFLAGS += $(CFLAGS_libxenctrl)
+$(SAVE_HELPER_OBJS): CFLAGS += $(CFLAGS_libxenctrl) $(CFLAGS_libxenevtchn)
PKG_CONFIG = xenlight.pc xlutil.pc
diff --git a/tools/libxl/libxl.c b/tools/libxl/libxl.c
index 854e957..a3b12c2 100644
--- a/tools/libxl/libxl.c
+++ b/tools/libxl/libxl.c
@@ -186,7 +186,7 @@ int libxl_ctx_free(libxl_ctx *ctx)
if (ctx->xch) xc_interface_close(ctx->xch);
libxl_version_info_dispose(&ctx->version_info);
if (ctx->xsh) xs_daemon_close(ctx->xsh);
- if (ctx->xce) xc_evtchn_close(ctx->xce);
+ if (ctx->xce) xenevtchn_close(ctx->xce);
libxl__poller_put(ctx, ctx->poller_app);
ctx->poller_app = NULL;
diff --git a/tools/libxl/libxl_dom_suspend.c b/tools/libxl/libxl_dom_suspend.c
index 4cc01ad..a2e8d86 100644
--- a/tools/libxl/libxl_dom_suspend.c
+++ b/tools/libxl/libxl_dom_suspend.c
@@ -107,9 +107,9 @@ static void domain_suspend_callback_common(libxl__egc *egc,
if ((hvm_s_state == 0) && (dss->guest_evtchn.port >= 0)) {
LOG(DEBUG, "issuing %s suspend request via event channel",
dss->hvm ? "PVHVM" : "PV");
- ret = xc_evtchn_notify(CTX->xce, dss->guest_evtchn.port);
+ ret = xenevtchn_notify(CTX->xce, dss->guest_evtchn.port);
if (ret < 0) {
- LOG(ERROR, "xc_evtchn_notify failed ret=%d", ret);
+ LOG(ERROR, "xenevtchn_notify failed ret=%d", ret);
rc = ERROR_FAIL;
goto err;
}
diff --git a/tools/libxl/libxl_event.c b/tools/libxl/libxl_event.c
index 0df6d6c..b323409 100644
--- a/tools/libxl/libxl_event.c
+++ b/tools/libxl/libxl_event.c
@@ -739,7 +739,7 @@ static void evtchn_fd_callback(libxl__egc *egc, libxl__ev_fd *ev,
/* OK, that's that workaround done. We can actually check for
* work for us to do: */
- port = xc_evtchn_pending(CTX->xce);
+ port = xenevtchn_pending(CTX->xce);
if (port < 0) {
if (errno == EAGAIN)
break;
@@ -765,20 +765,20 @@ static void evtchn_fd_callback(libxl__egc *egc, libxl__ev_fd *ev,
}
int libxl__ctx_evtchn_init(libxl__gc *gc) {
- xc_evtchn *xce;
+ xenevtchn_handle *xce;
int rc, fd;
if (CTX->xce)
return 0;
- xce = xc_evtchn_open(CTX->lg, 0);
+ xce = xenevtchn_open(CTX->lg, 0);
if (!xce) {
LOGE(ERROR,"cannot open libxc evtchn handle");
rc = ERROR_FAIL;
goto out;
}
- fd = xc_evtchn_fd(xce);
+ fd = xenevtchn_fd(xce);
assert(fd >= 0);
rc = libxl_fd_set_nonblock(CTX, fd, 1);
@@ -788,7 +788,7 @@ int libxl__ctx_evtchn_init(libxl__gc *gc) {
return 0;
out:
- xc_evtchn_close(xce);
+ xenevtchn_close(xce);
return rc;
}
@@ -810,14 +810,14 @@ int libxl__ev_evtchn_wait(libxl__gc *gc, libxl__ev_evtchn *evev)
if (!libxl__ev_fd_isregistered(&CTX->evtchn_efd)) {
rc = libxl__ev_fd_register(gc, &CTX->evtchn_efd, evtchn_fd_callback,
- xc_evtchn_fd(CTX->xce), POLLIN);
+ xenevtchn_fd(CTX->xce), POLLIN);
if (rc) goto out;
}
if (evev->waiting)
return 0;
- r = xc_evtchn_unmask(CTX->xce, evev->port);
+ r = xenevtchn_unmask(CTX->xce, evev->port);
if (r) {
LOGE(ERROR,"cannot unmask event channel %d",evev->port);
rc = ERROR_FAIL;
diff --git a/tools/libxl/libxl_internal.h b/tools/libxl/libxl_internal.h
index b00add0..3caa40d 100644
--- a/tools/libxl/libxl_internal.h
+++ b/tools/libxl/libxl_internal.h
@@ -48,6 +48,7 @@
#include <sys/file.h>
#include <sys/ioctl.h>
+#include <xenevtchn.h>
#include <xenstore.h>
#include <xenctrl.h>
#include <xenguest.h>
@@ -440,7 +441,7 @@ struct libxl__ctx {
uint32_t watch_counter; /* helps disambiguate slot reuse */
libxl__ev_fd watch_efd;
- xc_evtchn *xce; /* waiting must be done only with libxl__ev_evtchn* */
+ xenevtchn_handle *xce; /* waiting must be done only with libxl__ev_evtchn* */
LIBXL_LIST_HEAD(, libxl__ev_evtchn) evtchns_waiting;
libxl__ev_fd evtchn_efd;
@@ -914,7 +915,7 @@ static inline int libxl__ev_xswatch_isregistered(const libxl__ev_xswatch *xw)
* When the event is signaled then the callback will be made, once.
* Then you must call libxl__ev_evtchn_wait again, if desired.
*
- * You must NOT call xc_evtchn_unmask. wait will do that for you.
+ * You must NOT call xenevtchn_unmask. wait will do that for you.
*
* Calling libxl__ev_evtchn_cancel will arrange for libxl to disregard
* future occurrences of event. Both libxl__ev_evtchn_wait and
diff --git a/tools/misc/Makefile b/tools/misc/Makefile
index c4490f3..cf6a475 100644
--- a/tools/misc/Makefile
+++ b/tools/misc/Makefile
@@ -4,6 +4,7 @@ include $(XEN_ROOT)/tools/Rules.mk
CFLAGS += -Werror
# Include configure output (config.h)
CFLAGS += -include $(XEN_ROOT)/tools/config.h
+CFLAGS += $(CFLAGS_libxenevtchn)
CFLAGS += $(CFLAGS_libxenctrl)
CFLAGS += $(CFLAGS_xeninclude)
CFLAGS += $(CFLAGS_libxenstore)
@@ -88,18 +89,18 @@ xenlockprof: xenlockprof.o
# xen-hptool incorrectly uses libxc internals
xen-hptool.o: CFLAGS += -I$(XEN_ROOT)/tools/libxc
xen-hptool: xen-hptool.o
- $(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) $(APPEND_LDFLAGS)
# xen-mfndump incorrectly uses libxc internals
xen-mfndump.o: CFLAGS += -I$(XEN_ROOT)/tools/libxc
xen-mfndump: xen-mfndump.o
- $(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(APPEND_LDFLAGS)
xenwatchdogd: xenwatchdogd.o
$(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenctrl) $(APPEND_LDFLAGS)
xen-lowmemd: xen-lowmemd.o
- $(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenctrl) $(LDLIBS_libxenstore) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenstore) $(APPEND_LDFLAGS)
gtraceview: gtraceview.o
$(CC) $(LDFLAGS) -o $@ $< $(CURSES_LIBS) $(TINFO_LIBS) $(APPEND_LDFLAGS)
diff --git a/tools/misc/xen-hptool.c b/tools/misc/xen-hptool.c
index c7561a9..ebcc9e8 100644
--- a/tools/misc/xen-hptool.c
+++ b/tools/misc/xen-hptool.c
@@ -1,3 +1,4 @@
+#include <xenevtchn.h>
#include <xenctrl.h>
#include <xc_private.h>
#include <xc_core.h>
@@ -98,7 +99,7 @@ static int hp_mem_query_func(int argc, char *argv[])
return ret;
}
-static int suspend_guest(xc_interface *xch, xc_evtchn *xce, int domid,
+static int suspend_guest(xc_interface *xch, xenevtchn_handle *xce, int domid,
int *evtchn, int *lockfd)
{
int port, rc, suspend_evtchn = -1;
@@ -123,7 +124,7 @@ static int suspend_guest(xc_interface *xch, xc_evtchn *xce, int domid,
}
*evtchn = suspend_evtchn;
- rc = xc_evtchn_notify(xce, suspend_evtchn);
+ rc = xenevtchn_notify(xce, suspend_evtchn);
if (rc < 0)
{
fprintf(stderr, "Failed to notify suspend channel: errno %d\n", rc);
@@ -198,8 +199,8 @@ static int hp_mem_offline_func(int argc, char *argv[])
else if (status & PG_OFFLINE_OWNED)
{
int result, suspend_evtchn = -1, suspend_lockfd = -1;
- xc_evtchn *xce;
- xce = xc_evtchn_open(NULL, 0);
+ xenevtchn_handle *xce;
+ xce = xenevtchn_open(NULL, 0);
if (xce == NULL)
{
@@ -214,7 +215,7 @@ static int hp_mem_offline_func(int argc, char *argv[])
{
fprintf(stderr, "Failed to suspend guest %d for"
" mfn %lx\n", domid, mfn);
- xc_evtchn_close(xce);
+ xenevtchn_close(xce);
return -1;
}
@@ -238,7 +239,7 @@ static int hp_mem_offline_func(int argc, char *argv[])
xc_domain_resume(xch, domid, 1);
xc_suspend_evtchn_release(xch, xce, domid,
suspend_evtchn, &suspend_lockfd);
- xc_evtchn_close(xce);
+ xenevtchn_close(xce);
}
break;
}
diff --git a/tools/misc/xen-lowmemd.c b/tools/misc/xen-lowmemd.c
index 82ffd75..3200404 100644
--- a/tools/misc/xen-lowmemd.c
+++ b/tools/misc/xen-lowmemd.c
@@ -4,22 +4,23 @@
*/
#include <stdio.h>
+#include <xenevtchn.h>
#include <xenctrl.h>
#include <xenstore.h>
#include <stdlib.h>
#include <string.h>
static evtchn_port_t virq_port = -1;
-static xc_evtchn *xce_handle = NULL;
+static xenevtchn_handle *xce_handle = NULL;
static xc_interface *xch = NULL;
static struct xs_handle *xs_handle = NULL;
void cleanup(void)
{
if (virq_port > -1)
- xc_evtchn_unbind(xce_handle, virq_port);
+ xenevtchn_unbind(xce_handle, virq_port);
if (xce_handle)
- xc_evtchn_close(xce_handle);
+ xenevtchn_close(xce_handle);
if (xch)
xc_interface_close(xch);
if (xs_handle)
@@ -94,7 +95,7 @@ int main(int argc, char *argv[])
return 1;
}
- xce_handle = xc_evtchn_open(NULL, 0);
+ xce_handle = xenevtchn_open(NULL, 0);
if (xce_handle == NULL)
{
perror("Failed to open evtchn device");
@@ -108,7 +109,7 @@ int main(int argc, char *argv[])
return 3;
}
- if ((rc = xc_evtchn_bind_virq(xce_handle, VIRQ_ENOMEM)) == -1)
+ if ((rc = xenevtchn_bind_virq(xce_handle, VIRQ_ENOMEM)) == -1)
{
perror("Failed to bind to domain exception virq port");
return 4;
@@ -120,7 +121,7 @@ int main(int argc, char *argv[])
{
evtchn_port_t port;
- if ((port = xc_evtchn_pending(xce_handle)) == -1)
+ if ((port = xenevtchn_pending(xce_handle)) == -1)
{
perror("Failed to listen for pending event channel");
return 5;
@@ -134,7 +135,7 @@ int main(int argc, char *argv[])
return 6;
}
- if (xc_evtchn_unmask(xce_handle, port) == -1)
+ if (xenevtchn_unmask(xce_handle, port) == -1)
{
perror("Failed to unmask port");
return 7;
diff --git a/tools/ocaml/libs/eventchn/Makefile b/tools/ocaml/libs/eventchn/Makefile
index 2d8d618..154efd4 100644
--- a/tools/ocaml/libs/eventchn/Makefile
+++ b/tools/ocaml/libs/eventchn/Makefile
@@ -2,13 +2,13 @@ TOPLEVEL=$(CURDIR)/../..
XEN_ROOT=$(TOPLEVEL)/../..
include $(TOPLEVEL)/common.make
-CFLAGS += $(CFLAGS_libxenctrl) $(CFLAGS_xeninclude)
+CFLAGS += $(CFLAGS_libxenevtchn) $(CFLAGS_xeninclude)
OBJS = xeneventchn
INTF = $(foreach obj, $(OBJS),$(obj).cmi)
LIBS = xeneventchn.cma xeneventchn.cmxa
-LIBS_xeneventchn = $(LDLIBS_libxenctrl)
+LIBS_xeneventchn = $(LDLIBS_libxenevtchn)
all: $(INTF) $(LIBS) $(PROGRAMS)
diff --git a/tools/ocaml/libs/eventchn/xeneventchn_stubs.c b/tools/ocaml/libs/eventchn/xeneventchn_stubs.c
index 5939e7c..c2d4737 100644
--- a/tools/ocaml/libs/eventchn/xeneventchn_stubs.c
+++ b/tools/ocaml/libs/eventchn/xeneventchn_stubs.c
@@ -24,7 +24,7 @@
#include <xen/sysctl.h>
#include <xen/xen.h>
#include <xen/sys/evtchn.h>
-#include <xenctrl.h>
+#include <xenevtchn.h>
#define CAML_NAME_SPACE
#include <caml/mlvalues.h>
@@ -34,14 +34,14 @@
#include <caml/callback.h>
#include <caml/fail.h>
-#define _H(__h) ((xc_evtchn *)(__h))
+#define _H(__h) ((xenevtchn_handle *)(__h))
CAMLprim value stub_eventchn_init(void)
{
CAMLparam0();
CAMLlocal1(result);
- xc_evtchn *xce = xc_evtchn_open(NULL, XC_OPENFLAG_NON_REENTRANT);
+ xenevtchn_handle *xce = xenevtchn_open(NULL, 0);
if (xce == NULL)
caml_failwith("open failed");
@@ -55,7 +55,7 @@ CAMLprim value stub_eventchn_fd(value xce)
CAMLlocal1(result);
int fd;
- fd = xc_evtchn_fd(_H(xce));
+ fd = xenevtchn_fd(_H(xce));
if (fd == -1)
caml_failwith("evtchn fd failed");
@@ -69,7 +69,7 @@ CAMLprim value stub_eventchn_notify(value xce, value port)
CAMLparam2(xce, port);
int rc;
- rc = xc_evtchn_notify(_H(xce), Int_val(port));
+ rc = xenevtchn_notify(_H(xce), Int_val(port));
if (rc == -1)
caml_failwith("evtchn notify failed");
@@ -83,7 +83,7 @@ CAMLprim value stub_eventchn_bind_interdomain(value xce, value domid,
CAMLlocal1(port);
evtchn_port_or_error_t rc;
- rc = xc_evtchn_bind_interdomain(_H(xce), Int_val(domid), Int_val(remote_port));
+ rc = xenevtchn_bind_interdomain(_H(xce), Int_val(domid), Int_val(remote_port));
if (rc == -1)
caml_failwith("evtchn bind_interdomain failed");
port = Val_int(rc);
@@ -97,7 +97,7 @@ CAMLprim value stub_eventchn_bind_dom_exc_virq(value xce)
CAMLlocal1(port);
evtchn_port_or_error_t rc;
- rc = xc_evtchn_bind_virq(_H(xce), VIRQ_DOM_EXC);
+ rc = xenevtchn_bind_virq(_H(xce), VIRQ_DOM_EXC);
if (rc == -1)
caml_failwith("evtchn bind_dom_exc_virq failed");
port = Val_int(rc);
@@ -110,7 +110,7 @@ CAMLprim value stub_eventchn_unbind(value xce, value port)
CAMLparam2(xce, port);
int rc;
- rc = xc_evtchn_unbind(_H(xce), Int_val(port));
+ rc = xenevtchn_unbind(_H(xce), Int_val(port));
if (rc == -1)
caml_failwith("evtchn unbind failed");
@@ -123,7 +123,7 @@ CAMLprim value stub_eventchn_pending(value xce)
CAMLlocal1(result);
evtchn_port_or_error_t port;
- port = xc_evtchn_pending(_H(xce));
+ port = xenevtchn_pending(_H(xce));
if (port == -1)
caml_failwith("evtchn pending failed");
result = Val_int(port);
@@ -137,7 +137,7 @@ CAMLprim value stub_eventchn_unmask(value xce, value _port)
evtchn_port_t port;
port = Int_val(_port);
- if (xc_evtchn_unmask(_H(xce), port))
+ if (xenevtchn_unmask(_H(xce), port))
caml_failwith("evtchn unmask failed");
CAMLreturn(Val_unit);
}
diff --git a/tools/python/setup.py b/tools/python/setup.py
index 9771cc4..ce317b4 100644
--- a/tools/python/setup.py
+++ b/tools/python/setup.py
@@ -8,13 +8,18 @@ extra_compile_args = [ "-fno-strict-aliasing", "-Werror" ]
PATH_XEN = XEN_ROOT + "/tools/include"
PATH_LIBXENTOOLLOG = XEN_ROOT + "/tools/libs/toollog"
+PATH_LIBXENEVTCHN = XEN_ROOT + "/tools/libs/evtchn"
PATH_LIBXC = XEN_ROOT + "/tools/libxc"
PATH_LIBXL = XEN_ROOT + "/tools/libxl"
PATH_XENSTORE = XEN_ROOT + "/tools/xenstore"
xc = Extension("xc",
extra_compile_args = extra_compile_args,
- include_dirs = [ PATH_XEN, PATH_LIBXENTOOLLOG + "/include", PATH_LIBXC + "/include", "xen/lowlevel/xc" ],
+ include_dirs = [ PATH_XEN,
+ PATH_LIBXENTOOLLOG + "/include",
+ PATH_LIBXENEVTCHN + "/include",
+ PATH_LIBXC + "/include",
+ "xen/lowlevel/xc" ],
library_dirs = [ PATH_LIBXC ],
libraries = [ "xenctrl", "xenguest" ],
depends = [ PATH_LIBXC + "/libxenctrl.so", PATH_LIBXC + "/libxenguest.so", "-Wl,-rpath-link="+PATH_LIBXENTOOLLOG ],
diff --git a/tools/tests/xen-access/xen-access.c b/tools/tests/xen-access/xen-access.c
index a52ca6e..f639860 100644
--- a/tools/tests/xen-access/xen-access.c
+++ b/tools/tests/xen-access/xen-access.c
@@ -79,7 +79,7 @@ static void close_handler(int sig)
int xc_wait_for_event_or_timeout(xc_interface *xch, xc_evtchn *xce, unsigned long ms)
{
- struct pollfd fd = { .fd = xc_evtchn_fd(xce), .events = POLLIN | POLLERR };
+ struct pollfd fd = { .fd = xenevtchn_fd(xce), .events = POLLIN | POLLERR };
int port;
int rc;
diff --git a/tools/xcutils/Makefile b/tools/xcutils/Makefile
index fff519d..2d1f112 100644
--- a/tools/xcutils/Makefile
+++ b/tools/xcutils/Makefile
@@ -16,8 +16,8 @@ PROGRAMS = readnotes lsevtchn
CFLAGS += -Werror
# incorrectly uses libxc internals
-CFLAGS_readnotes.o := $(CFLAGS_libxenctrl) $(CFLAGS_libxenguest) -I$(XEN_ROOT)/tools/libxc
-CFLAGS_lsevtchn.o := $(CFLAGS_libxenctrl)
+CFLAGS_readnotes.o := $(CFLAGS_libxenevtchn) $(CFLAGS_libxenctrl) $(CFLAGS_libxenguest) -I$(XEN_ROOT)/tools/libxc
+CFLAGS_lsevtchn.o := $(CFLAGS_libxenevtchn) $(CFLAGS_libxenctrl)
.PHONY: all
all: build
diff --git a/tools/xenmon/Makefile b/tools/xenmon/Makefile
index 20ea100..98056f2 100644
--- a/tools/xenmon/Makefile
+++ b/tools/xenmon/Makefile
@@ -14,8 +14,10 @@ XEN_ROOT=$(CURDIR)/../..
include $(XEN_ROOT)/tools/Rules.mk
CFLAGS += -Werror
+CFLAGS += $(CFLAGS_libxenevtchn)
CFLAGS += $(CFLAGS_libxenctrl)
LDLIBS += $(LDLIBS_libxenctrl)
+LDLIBS += $(LDLIBS_libxenevtchn)
SCRIPTS = xenmon.py
diff --git a/tools/xenmon/xenbaked.c b/tools/xenmon/xenbaked.c
index 1ddb29b..e4602ef 100644
--- a/tools/xenmon/xenbaked.c
+++ b/tools/xenmon/xenbaked.c
@@ -37,6 +37,7 @@
#include <unistd.h>
#include <errno.h>
#include <signal.h>
+#include <xenevtchn.h>
#include <xenctrl.h>
#include <xen/xen.h>
#include <string.h>
@@ -267,7 +268,7 @@ static void log_event(int event_id)
}
int virq_port;
-xc_evtchn *xce_handle = NULL;
+xenevtchn_handle *xce_handle = NULL;
/* Returns the event channel handle. */
/* Stolen from xenstore code */
@@ -279,12 +280,12 @@ static int eventchn_init(void)
if (0)
return -1;
- xce_handle = xc_evtchn_open(NULL, 0);
+ xce_handle = xenevtchn_open(NULL, 0);
if (xce_handle == NULL)
perror("Failed to open evtchn device");
- if ((rc = xc_evtchn_bind_virq(xce_handle, VIRQ_TBUF)) == -1)
+ if ((rc = xenevtchn_bind_virq(xce_handle, VIRQ_TBUF)) == -1)
perror("Failed to bind to domain exception virq port");
virq_port = rc;
@@ -304,7 +305,7 @@ static void wait_for_event(void)
return;
}
- evtchn_fd = xc_evtchn_fd(xce_handle);
+ evtchn_fd = xenevtchn_fd(xce_handle);
FD_ZERO(&inset);
FD_SET(evtchn_fd, &inset);
@@ -314,13 +315,13 @@ static void wait_for_event(void)
ret = select(evtchn_fd+1, &inset, NULL, NULL, &tv);
if ( (ret == 1) && FD_ISSET(evtchn_fd, &inset)) {
- if ((port = xc_evtchn_pending(xce_handle)) == -1)
+ if ((port = xenevtchn_pending(xce_handle)) == -1)
perror("Failed to read from event fd");
// if (port == virq_port)
// printf("got the event I was looking for\r\n");
- if (xc_evtchn_unmask(xce_handle, port) == -1)
+ if (xenevtchn_unmask(xce_handle, port) == -1)
perror("Failed to write to event fd");
}
}
diff --git a/tools/xenpaging/Makefile b/tools/xenpaging/Makefile
index e63d894..d491867 100644
--- a/tools/xenpaging/Makefile
+++ b/tools/xenpaging/Makefile
@@ -2,8 +2,8 @@ XEN_ROOT=$(CURDIR)/../..
include $(XEN_ROOT)/tools/Rules.mk
# xenpaging.c and file_ops.c incorrectly use libxc internals
-CFLAGS += $(CFLAGS_libxenctrl) $(CFLAGS_libxenstore) $(PTHREAD_CFLAGS) -I$(XEN_ROOT)/tools/libxc
-LDLIBS += $(LDLIBS_libxentoollog) $(LDLIBS_libxenctrl) $(LDLIBS_libxenstore) $(PTHREAD_LIBS)
+CFLAGS += $(CFLAGS_libxentoollog) $(CFLAGS_libxenevtchn) $(CFLAGS_libxenctrl) $(CFLAGS_libxenstore) $(PTHREAD_CFLAGS) -I$(XEN_ROOT)/tools/libxc
+LDLIBS += $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenstore) $(PTHREAD_LIBS)
LDFLAGS += $(PTHREAD_LDFLAGS)
POLICY = default
diff --git a/tools/xenpaging/xenpaging.c b/tools/xenpaging/xenpaging.c
index b5ffee6..df99c6a 100644
--- a/tools/xenpaging/xenpaging.c
+++ b/tools/xenpaging/xenpaging.c
@@ -73,7 +73,7 @@ static void xenpaging_mem_paging_flush_ioemu_cache(struct xenpaging *paging)
static int xenpaging_wait_for_event_or_timeout(struct xenpaging *paging)
{
xc_interface *xch = paging->xc_handle;
- xc_evtchn *xce = paging->vm_event.xce_handle;
+ xenevtchn_handle *xce = paging->vm_event.xce_handle;
char **vec, *val;
unsigned int num;
struct pollfd fd[2];
@@ -82,7 +82,7 @@ static int xenpaging_wait_for_event_or_timeout(struct xenpaging *paging)
int timeout;
/* Wait for event channel and xenstore */
- fd[0].fd = xc_evtchn_fd(xce);
+ fd[0].fd = xenevtchn_fd(xce);
fd[0].events = POLLIN | POLLERR;
fd[1].fd = xs_fileno(paging->xs_handle);
fd[1].events = POLLIN | POLLERR;
@@ -146,7 +146,7 @@ static int xenpaging_wait_for_event_or_timeout(struct xenpaging *paging)
if ( rc && fd[0].revents & POLLIN )
{
DPRINTF("Got event from evtchn\n");
- port = xc_evtchn_pending(xce);
+ port = xenevtchn_pending(xce);
if ( port == -1 )
{
PERROR("Failed to read port from event channel");
@@ -154,7 +154,7 @@ static int xenpaging_wait_for_event_or_timeout(struct xenpaging *paging)
goto err;
}
- rc = xc_evtchn_unmask(xce, port);
+ rc = xenevtchn_unmask(xce, port);
if ( rc < 0 )
{
PERROR("Failed to unmask event channel port");
@@ -393,7 +393,7 @@ static struct xenpaging *xenpaging_init(int argc, char *argv[])
}
/* Open event channel */
- paging->vm_event.xce_handle = xc_evtchn_open(NULL, 0);
+ paging->vm_event.xce_handle = xenevtchn_open(NULL, 0);
if ( paging->vm_event.xce_handle == NULL )
{
PERROR("Failed to open event channel");
@@ -401,7 +401,7 @@ static struct xenpaging *xenpaging_init(int argc, char *argv[])
}
/* Bind event notification */
- rc = xc_evtchn_bind_interdomain(paging->vm_event.xce_handle,
+ rc = xenevtchn_bind_interdomain(paging->vm_event.xce_handle,
paging->vm_event.domain_id,
paging->vm_event.evtchn_port);
if ( rc < 0 )
@@ -531,7 +531,7 @@ static void xenpaging_teardown(struct xenpaging *paging)
}
/* Unbind VIRQ */
- rc = xc_evtchn_unbind(paging->vm_event.xce_handle, paging->vm_event.port);
+ rc = xenevtchn_unbind(paging->vm_event.xce_handle, paging->vm_event.port);
if ( rc != 0 )
{
PERROR("Error unbinding event port");
@@ -539,7 +539,7 @@ static void xenpaging_teardown(struct xenpaging *paging)
paging->vm_event.port = -1;
/* Close event channel */
- rc = xc_evtchn_close(paging->vm_event.xce_handle);
+ rc = xenevtchn_close(paging->vm_event.xce_handle);
if ( rc != 0 )
{
PERROR("Error closing event channel");
@@ -692,7 +692,7 @@ static int xenpaging_resume_page(struct xenpaging *paging, vm_event_response_t *
}
/* Tell Xen page is ready */
- return xc_evtchn_notify(paging->vm_event.xce_handle, paging->vm_event.port);
+ return xenevtchn_notify(paging->vm_event.xce_handle, paging->vm_event.port);
}
static int xenpaging_populate_page(struct xenpaging *paging, unsigned long gfn, int i)
diff --git a/tools/xenpaging/xenpaging.h b/tools/xenpaging/xenpaging.h
index c6ab77c..d0f8d20 100644
--- a/tools/xenpaging/xenpaging.h
+++ b/tools/xenpaging/xenpaging.h
@@ -24,6 +24,7 @@
#define __XEN_PAGING2_H__
+#include <xenevtchn.h>
#include <xc_private.h>
#include <xen/event_channel.h>
#include <xen/vm_event.h>
@@ -32,7 +33,7 @@
struct vm_event {
domid_t domain_id;
- xc_evtchn *xce_handle;
+ xenevtchn_handle *xce_handle;
int port;
vm_event_back_ring_t back_ring;
uint32_t evtchn_port;
diff --git a/tools/xenstore/Makefile b/tools/xenstore/Makefile
index 1b4a494..c161046 100644
--- a/tools/xenstore/Makefile
+++ b/tools/xenstore/Makefile
@@ -9,6 +9,7 @@ CFLAGS += -I.
# Include configure output (config.h)
CFLAGS += -include $(XEN_ROOT)/tools/config.h
CFLAGS += -I./include
+CFLAGS += $(CFLAGS_libxenevtchn)
CFLAGS += $(CFLAGS_libxenctrl)
CFLAGS += -DXEN_LIB_STORED="\"$(XEN_LIB_STORED)\""
@@ -75,10 +76,10 @@ endif
init-xenstore-domain.o: CFLAGS += $(CFLAGS_libxenguest)
init-xenstore-domain: init-xenstore-domain.o $(LIBXENSTORE)
- $(CC) $^ $(LDFLAGS) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) -o $@ $(APPEND_LDFLAGS)
+ $(CC) $^ $(LDFLAGS) $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) -o $@ $(APPEND_LDFLAGS)
xenstored: $(XENSTORED_OBJS)
- $(CC) $^ $(LDFLAGS) $(LDLIBS_libxenctrl) $(SOCKET_LIBS) -o $@ $(APPEND_LDFLAGS)
+ $(CC) $^ $(LDFLAGS) $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(SOCKET_LIBS) -o $@ $(APPEND_LDFLAGS)
xenstored.a: $(XENSTORED_OBJS)
$(AR) cr $@ $^
diff --git a/tools/xenstore/xenstored_core.c b/tools/xenstore/xenstored_core.c
index 3c0307e..f62c192 100644
--- a/tools/xenstore/xenstored_core.c
+++ b/tools/xenstore/xenstored_core.c
@@ -40,6 +40,8 @@
#include <assert.h>
#include <setjmp.h>
+#include <xenevtchn.h>
+
#include "utils.h"
#include "list.h"
#include "talloc.h"
@@ -63,7 +65,7 @@
#include <systemd/sd-daemon.h>
#endif
-extern xc_evtchn *xce_handle; /* in xenstored_domain.c */
+extern xenevtchn_handle *xce_handle; /* in xenstored_domain.c */
static int xce_pollfd_idx = -1;
static struct pollfd *fds;
static unsigned int current_array_size;
@@ -372,7 +374,7 @@ static void initialize_fds(int sock, int *p_sock_pollfd_idx,
set_fd(reopen_log_pipe[0], POLLIN|POLLPRI);
if (xce_handle != NULL)
- xce_pollfd_idx = set_fd(xc_evtchn_fd(xce_handle),
+ xce_pollfd_idx = set_fd(xenevtchn_fd(xce_handle),
POLLIN|POLLPRI);
list_for_each_entry(conn, &connections, list) {
diff --git a/tools/xenstore/xenstored_domain.c b/tools/xenstore/xenstored_domain.c
index dcd6581..6ceec29 100644
--- a/tools/xenstore/xenstored_domain.c
+++ b/tools/xenstore/xenstored_domain.c
@@ -29,6 +29,7 @@
#include "xenstored_transaction.h"
#include "xenstored_watch.h"
+#include <xenevtchn.h>
#include <xenctrl.h>
#include <xen/grant_table.h>
@@ -36,7 +37,7 @@ static xc_interface **xc_handle;
xc_gnttab **xcg_handle;
static evtchn_port_t virq_port;
-xc_evtchn *xce_handle = NULL;
+xenevtchn_handle *xce_handle = NULL;
struct domain
{
@@ -128,7 +129,7 @@ static int writechn(struct connection *conn,
xen_mb();
intf->rsp_prod += len;
- xc_evtchn_notify(xce_handle, conn->domain->port);
+ xenevtchn_notify(xce_handle, conn->domain->port);
return len;
}
@@ -158,7 +159,7 @@ static int readchn(struct connection *conn, void *data, unsigned int len)
xen_mb();
intf->req_cons += len;
- xc_evtchn_notify(xce_handle, conn->domain->port);
+ xenevtchn_notify(xce_handle, conn->domain->port);
return len;
}
@@ -190,7 +191,7 @@ static int destroy_domain(void *_domain)
list_del(&domain->list);
if (domain->port) {
- if (xc_evtchn_unbind(xce_handle, domain->port) == -1)
+ if (xenevtchn_unbind(xce_handle, domain->port) == -1)
eprintf("> Unbinding port %i failed!\n", domain->port);
}
@@ -239,13 +240,13 @@ void handle_event(void)
{
evtchn_port_t port;
- if ((port = xc_evtchn_pending(xce_handle)) == -1)
+ if ((port = xenevtchn_pending(xce_handle)) == -1)
barf_perror("Failed to read from event fd");
if (port == virq_port)
domain_cleanup();
- if (xc_evtchn_unmask(xce_handle, port) == -1)
+ if (xenevtchn_unmask(xce_handle, port) == -1)
barf_perror("Failed to write to event fd");
}
@@ -287,7 +288,7 @@ static struct domain *new_domain(void *context, unsigned int domid,
talloc_set_destructor(domain, destroy_domain);
/* Tell kernel we're interested in this event. */
- rc = xc_evtchn_bind_interdomain(xce_handle, domid, port);
+ rc = xenevtchn_bind_interdomain(xce_handle, domid, port);
if (rc == -1)
return NULL;
domain->port = rc;
@@ -392,8 +393,8 @@ void do_introduce(struct connection *conn, struct buffered_data *in)
} else if ((domain->mfn == mfn) && (domain->conn != conn)) {
/* Use XS_INTRODUCE for recreating the xenbus event-channel. */
if (domain->port)
- xc_evtchn_unbind(xce_handle, domain->port);
- rc = xc_evtchn_bind_interdomain(xce_handle, domid, port);
+ xenevtchn_unbind(xce_handle, domain->port);
+ rc = xenevtchn_bind_interdomain(xce_handle, domid, port);
domain->port = (rc == -1) ? 0 : rc;
domain->remote_port = port;
} else {
@@ -614,7 +615,7 @@ static int dom0_init(void)
talloc_steal(dom0->conn, dom0);
- xc_evtchn_notify(xce_handle, dom0->port);
+ xenevtchn_notify(xce_handle, dom0->port);
return 0;
}
@@ -643,7 +644,7 @@ void domain_init(void)
else
talloc_set_destructor(xcg_handle, close_xcg_handle);
- xce_handle = xc_evtchn_open(NULL, 0);
+ xce_handle = xenevtchn_open(NULL, 0);
if (xce_handle == NULL)
barf_perror("Failed to open evtchn device");
@@ -651,7 +652,7 @@ void domain_init(void)
if (dom0_init() != 0)
barf_perror("Failed to initialize dom0 state");
- if ((rc = xc_evtchn_bind_virq(xce_handle, VIRQ_DOM_EXC)) == -1)
+ if ((rc = xenevtchn_bind_virq(xce_handle, VIRQ_DOM_EXC)) == -1)
barf_perror("Failed to bind to domain exception virq port");
virq_port = rc;
}
diff --git a/tools/xentrace/Makefile b/tools/xentrace/Makefile
index 6c13cd1..0157be2 100644
--- a/tools/xentrace/Makefile
+++ b/tools/xentrace/Makefile
@@ -3,8 +3,11 @@ include $(XEN_ROOT)/tools/Rules.mk
CFLAGS += -Werror
+CFLAGS += $(CFLAGS_libxenevtchn)
CFLAGS += $(CFLAGS_libxenctrl)
-LDLIBS += $(LDLIBS_libxenctrl) $(ARGP_LDFLAGS)
+LDLIBS += $(LDLIBS_libxenevtchn)
+LDLIBS += $(LDLIBS_libxenctrl)
+LDLIBS += $(ARGP_LDFLAGS)
BIN-$(CONFIG_X86) = xenalyze
BIN = $(BIN-y)
diff --git a/tools/xentrace/xentrace.c b/tools/xentrace/xentrace.c
index 4ee1458..c970d42 100644
--- a/tools/xentrace/xentrace.c
+++ b/tools/xentrace/xentrace.c
@@ -30,6 +30,7 @@
#include <xen/xen.h>
#include <xen/trace.h>
+#include <xenevtchn.h>
#include <xenctrl.h>
#define PERROR(_m, _a...) \
@@ -74,7 +75,7 @@ settings_t opts;
int interrupted = 0; /* gets set if we get a SIGHUP */
static xc_interface *xc_handle;
-static xc_evtchn *xce_handle = NULL;
+static xenevtchn_handle *xce_handle = NULL;
static int virq_port = -1;
static int outfd = 1;
@@ -602,13 +603,13 @@ static void event_init(void)
{
int rc;
- xce_handle = xc_evtchn_open(NULL, 0);
+ xce_handle = xenevtchn_open(NULL, 0);
if (xce_handle == NULL) {
perror("event channel open");
exit(EXIT_FAILURE);
}
- rc = xc_evtchn_bind_virq(xce_handle, VIRQ_TBUF);
+ rc = xenevtchn_bind_virq(xce_handle, VIRQ_TBUF);
if (rc == -1) {
PERROR("failed to bind to VIRQ port");
exit(EXIT_FAILURE);
@@ -623,7 +624,7 @@ static void event_init(void)
static void wait_for_event_or_timeout(unsigned long milliseconds)
{
int rc;
- struct pollfd fd = { .fd = xc_evtchn_fd(xce_handle),
+ struct pollfd fd = { .fd = xenevtchn_fd(xce_handle),
.events = POLLIN | POLLERR };
int port;
@@ -636,7 +637,7 @@ static void wait_for_event_or_timeout(unsigned long milliseconds)
}
if (rc == 1) {
- port = xc_evtchn_pending(xce_handle);
+ port = xenevtchn_pending(xce_handle);
if (port == -1) {
PERROR("failed to read port from evtchn");
exit(EXIT_FAILURE);
@@ -647,7 +648,7 @@ static void wait_for_event_or_timeout(unsigned long milliseconds)
port, virq_port);
exit(EXIT_FAILURE);
}
- rc = xc_evtchn_unmask(xce_handle, port);
+ rc = xenevtchn_unmask(xce_handle, port);
if (rc == -1) {
PERROR("failed to write port to evtchn");
exit(EXIT_FAILURE);
--
2.1.4
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 05/23] tools: Arrange to check public headers for ANSI compatiblity
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (3 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 04/23] tools: Refactor /dev/xen/evtchn wrappers into libxenevtchn Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 06/23] tools/libxc: Remove osdep indirection for xc_gnt{shr, tab} Ian Campbell
` (18 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
Using the same rune as we use for the Xen public headers, except we do
not need stdint.h here and we use -pedantic too.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
.gitignore | 2 ++
tools/Rules.mk | 8 ++++++++
tools/libs/evtchn/Makefile | 4 +++-
tools/libs/toollog/Makefile | 5 ++++-
4 files changed, 17 insertions(+), 2 deletions(-)
diff --git a/.gitignore b/.gitignore
index bf382e5..b34dc3c 100644
--- a/.gitignore
+++ b/.gitignore
@@ -86,6 +86,8 @@ tools/config.cache
config/Tools.mk
config/Stubdom.mk
config/Docs.mk
+tools/libs/toollog/headers.chk
+tools/libs/evtchn/headers.chk
tools/blktap2/daemon/blktapctrl
tools/blktap2/drivers/img2qcow
tools/blktap2/drivers/lock-util
diff --git a/tools/Rules.mk b/tools/Rules.mk
index 75d02c4..8ed1507 100644
--- a/tools/Rules.mk
+++ b/tools/Rules.mk
@@ -176,6 +176,14 @@ INSTALL_PYTHON_PROG = \
%.opic: %.S
$(CC) $(CPPFLAGS) -DPIC $(CFLAGS) $(CFLAGS.opic) -fPIC -c -o $@ $< $(APPEND_CFLAGS)
+headers.chk:
+ for i in $(filter %.h,$^); do \
+ $(CC) -x c -ansi -Wall -Werror -pedantic $(CFLAGS_xeninclude) \
+ -S -o /dev/null $$i || exit 1; \
+ echo $$i; \
+ done >$@.new
+ mv $@.new $@
+
subdirs-all subdirs-clean subdirs-install subdirs-distclean: .phony
@set -e; for subdir in $(SUBDIRS) $(SUBDIRS-y); do \
$(MAKE) subdir-$(patsubst subdirs-%,%,$@)-$$subdir; \
diff --git a/tools/libs/evtchn/Makefile b/tools/libs/evtchn/Makefile
index 85ed6dc..892892c 100644
--- a/tools/libs/evtchn/Makefile
+++ b/tools/libs/evtchn/Makefile
@@ -32,8 +32,9 @@ build:
$(MAKE) libs
.PHONY: libs
-libs: $(LIB)
+libs: headers.chk $(LIB)
+headers.chk: $(wildcard include/*.h)
libxenevtchn.a: $(LIB_OBJS)
$(AR) rc $@ $^
@@ -63,3 +64,4 @@ TAGS:
.PHONY: clean
clean:
rm -rf *.rpm $(LIB) *~ $(DEPS) $(LIB_OBJS) $(PIC_OBJS)
+ rm -f headers.chk
diff --git a/tools/libs/toollog/Makefile b/tools/libs/toollog/Makefile
index bd12403..72c70c9 100644
--- a/tools/libs/toollog/Makefile
+++ b/tools/libs/toollog/Makefile
@@ -27,7 +27,9 @@ build:
$(MAKE) libs
.PHONY: libs
-libs: $(LIB)
+libs: headers.chk $(LIB)
+
+headers.chk: $(wildcard include/*.h)
libxentoollog.a: $(LIB_OBJS)
$(AR) rc $@ $^
@@ -57,3 +59,4 @@ TAGS:
.PHONY: clean
clean:
rm -rf *.rpm $(LIB) *~ $(DEPS) $(LIB_OBJS) $(PIC_OBJS)
+ rm -f headers.chk
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 06/23] tools/libxc: Remove osdep indirection for xc_gnt{shr, tab}
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (4 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 05/23] tools: Arrange to check public headers for ANSI compatiblity Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab Ian Campbell
` (17 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
The alternative backend (a xen-api/xapi shim) is no longer around and
so this stuff is now just baggage which is getting in the way of
refactoring libxenctrl.
Nested virt probably suffices for this use case now.
It is now necessary to provide explicit versions of things for
platforms which do not implement this functionality, since the osdep
dispatcher cannot fulfil this need any more. These are provided by
appropriate xc_nognt???.c files which are compiled and linked on the
appropriate platforms. In them open and close return failure and
everything else aborts, since if open fails they should never be
called.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libxc/Makefile | 10 ++--
tools/libxc/include/xenctrl.h | 4 +-
tools/libxc/include/xenctrlosdep.h | 23 -------
tools/libxc/xc_gnttab.c | 57 ++++--------------
tools/libxc/xc_linux_osdep.c | 119 +++++++++++++++----------------------
tools/libxc/xc_minios.c | 51 ++++++----------
tools/libxc/xc_nogntshr.c | 46 ++++++++++++++
tools/libxc/xc_nognttab.c | 50 ++++++++++++++++
tools/libxc/xc_private.c | 83 +++++++++++++++++++++-----
tools/libxc/xc_private.h | 24 ++++++++
10 files changed, 274 insertions(+), 193 deletions(-)
create mode 100644 tools/libxc/xc_nogntshr.c
create mode 100644 tools/libxc/xc_nognttab.c
diff --git a/tools/libxc/Makefile b/tools/libxc/Makefile
index b8fc6a5..184cbb7 100644
--- a/tools/libxc/Makefile
+++ b/tools/libxc/Makefile
@@ -43,11 +43,11 @@ CTRL_SRCS-y += xc_resource.c
CTRL_SRCS-$(CONFIG_X86) += xc_psr.c
CTRL_SRCS-$(CONFIG_X86) += xc_pagetab.c
CTRL_SRCS-$(CONFIG_Linux) += xc_linux.c xc_linux_osdep.c
-CTRL_SRCS-$(CONFIG_FreeBSD) += xc_freebsd.c xc_freebsd_osdep.c
-CTRL_SRCS-$(CONFIG_SunOS) += xc_solaris.c
-CTRL_SRCS-$(CONFIG_NetBSD) += xc_netbsd.c
-CTRL_SRCS-$(CONFIG_NetBSDRump) += xc_netbsd.c
-CTRL_SRCS-$(CONFIG_MiniOS) += xc_minios.c
+CTRL_SRCS-$(CONFIG_FreeBSD) += xc_freebsd.c xc_freebsd_osdep.c xc_nognttab.c xc_nogntshr.c
+CTRL_SRCS-$(CONFIG_SunOS) += xc_solaris.c xc_nognttab.c xc_nogntshr.c
+CTRL_SRCS-$(CONFIG_NetBSD) += xc_netbsd.c xc_nognttab.c xc_nogntshr.c
+CTRL_SRCS-$(CONFIG_NetBSDRump) += xc_netbsd.c xc_nognttab.c xc_nogntshr.c
+CTRL_SRCS-$(CONFIG_MiniOS) += xc_minios.c xc_nogntshr.c
CTRL_SRCS-y += xc_evtchn_compat.c
GUEST_SRCS-y :=
diff --git a/tools/libxc/include/xenctrl.h b/tools/libxc/include/xenctrl.h
index ca3f776..acc4af1 100644
--- a/tools/libxc/include/xenctrl.h
+++ b/tools/libxc/include/xenctrl.h
@@ -117,8 +117,8 @@
*/
typedef struct xc_interface_core xc_interface;
-typedef struct xc_interface_core xc_gnttab;
-typedef struct xc_interface_core xc_gntshr;
+typedef struct xengntdev_handle xc_gnttab;
+typedef struct xengntdev_handle xc_gntshr;
enum xc_error_code {
XC_ERROR_NONE = 0,
diff --git a/tools/libxc/include/xenctrlosdep.h b/tools/libxc/include/xenctrlosdep.h
index 89564e1..423660d 100644
--- a/tools/libxc/include/xenctrlosdep.h
+++ b/tools/libxc/include/xenctrlosdep.h
@@ -51,8 +51,6 @@
enum xc_osdep_type {
XC_OSDEP_PRIVCMD,
- XC_OSDEP_GNTTAB,
- XC_OSDEP_GNTSHR,
};
/* Opaque handle internal to the backend */
@@ -88,27 +86,6 @@ struct xc_osdep_ops
size_t chunksize, privcmd_mmap_entry_t entries[],
int nentries);
} privcmd;
- struct {
-#define XC_GRANT_MAP_SINGLE_DOMAIN 0x1
- void *(*grant_map)(xc_gnttab *xcg, xc_osdep_handle h,
- uint32_t count, int flags, int prot,
- uint32_t *domids, uint32_t *refs,
- uint32_t notify_offset,
- evtchn_port_t notify_port);
- int (*munmap)(xc_gnttab *xcg, xc_osdep_handle h,
- void *start_address,
- uint32_t count);
- int (*set_max_grants)(xc_gnttab *xcg, xc_osdep_handle h, uint32_t count);
- } gnttab;
- struct {
- void *(*share_pages)(xc_gntshr *xcg, xc_osdep_handle h,
- uint32_t domid, int count,
- uint32_t *refs, int writable,
- uint32_t notify_offset,
- evtchn_port_t notify_port);
- int (*munmap)(xc_gntshr *xcg, xc_osdep_handle h,
- void *start_address, uint32_t count);
- } gntshr;
} u;
};
typedef struct xc_osdep_ops xc_osdep_ops;
diff --git a/tools/libxc/xc_gnttab.c b/tools/libxc/xc_gnttab.c
index 60335d8..a51f405 100644
--- a/tools/libxc/xc_gnttab.c
+++ b/tools/libxc/xc_gnttab.c
@@ -143,68 +143,48 @@ grant_entry_v2_t *xc_gnttab_map_table_v2(xc_interface *xch, int domid,
return _gnttab_map_table(xch, domid, gnt_num);
}
-void *xc_gnttab_map_grant_ref(xc_gnttab *xcg,
+void *xc_gnttab_map_grant_ref(xc_gnttab *xgt,
uint32_t domid,
uint32_t ref,
int prot)
{
- return xcg->ops->u.gnttab.grant_map(xcg, xcg->ops_handle, 1, 0, prot,
- &domid, &ref, -1, -1);
+ return osdep_gnttab_grant_map(xgt, 1, 0, prot, &domid, &ref, -1, -1);
}
-void *xc_gnttab_map_grant_refs(xc_gnttab *xcg,
+void *xc_gnttab_map_grant_refs(xc_gnttab *xgt,
uint32_t count,
uint32_t *domids,
uint32_t *refs,
int prot)
{
- return xcg->ops->u.gnttab.grant_map(xcg, xcg->ops_handle, count, 0,
- prot, domids, refs, -1, -1);
+ return osdep_gnttab_grant_map(xgt, count, 0, prot, domids, refs, -1, -1);
}
-void *xc_gnttab_map_domain_grant_refs(xc_gnttab *xcg,
+void *xc_gnttab_map_domain_grant_refs(xc_gnttab *xgt,
uint32_t count,
uint32_t domid,
uint32_t *refs,
int prot)
{
- return xcg->ops->u.gnttab.grant_map(xcg, xcg->ops_handle, count,
- XC_GRANT_MAP_SINGLE_DOMAIN,
- prot, &domid, refs, -1, -1);
+ return osdep_gnttab_grant_map(xgt, count, XC_GRANT_MAP_SINGLE_DOMAIN,
+ prot, &domid, refs, -1, -1);
}
-void *xc_gnttab_map_grant_ref_notify(xc_gnttab *xcg,
+void *xc_gnttab_map_grant_ref_notify(xc_gnttab *xgt,
uint32_t domid,
uint32_t ref,
int prot,
uint32_t notify_offset,
evtchn_port_t notify_port)
{
- return xcg->ops->u.gnttab.grant_map(xcg, xcg->ops_handle, 1, 0, prot,
- &domid, &ref, notify_offset, notify_port);
-}
-
-
-int xc_gnttab_munmap(xc_gnttab *xcg,
- void *start_address,
- uint32_t count)
-{
- return xcg->ops->u.gnttab.munmap(xcg, xcg->ops_handle,
- start_address, count);
-}
-
-int xc_gnttab_set_max_grants(xc_gnttab *xcg, uint32_t count)
-{
- if (!xcg->ops->u.gnttab.set_max_grants)
- return 0;
- return xcg->ops->u.gnttab.set_max_grants(xcg, xcg->ops_handle, count);
+ return osdep_gnttab_grant_map(xgt, 1, 0, prot, &domid, &ref,
+ notify_offset, notify_port);
}
void *xc_gntshr_share_pages(xc_gntshr *xcg, uint32_t domid,
int count, uint32_t *refs, int writable)
{
- return xcg->ops->u.gntshr.share_pages(xcg, xcg->ops_handle, domid,
- count, refs, writable, -1, -1);
+ return osdep_gntshr_share_pages(xcg, domid, count, refs, writable, -1, -1);
}
void *xc_gntshr_share_page_notify(xc_gntshr *xcg, uint32_t domid,
@@ -212,22 +192,11 @@ void *xc_gntshr_share_page_notify(xc_gntshr *xcg, uint32_t domid,
uint32_t notify_offset,
evtchn_port_t notify_port)
{
- return xcg->ops->u.gntshr.share_pages(xcg, xcg->ops_handle,
- domid, 1, ref, writable, notify_offset, notify_port);
+ return osdep_gntshr_share_pages(xcg, domid, 1, ref, writable,
+ notify_offset, notify_port);
}
/*
- * Unmaps the @count pages starting at @start_address, which were mapped by a
- * call to xc_gntshr_share_*. Never logs.
- */
-int xc_gntshr_munmap(xc_gntshr *xcg, void *start_address, uint32_t count)
-{
- return xcg->ops->u.gntshr.munmap(xcg, xcg->ops_handle,
- start_address, count);
-}
-
-
-/*
* Local variables:
* mode: C
* c-file-style: "BSD"
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index 1883aba..685852f 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -41,6 +41,9 @@
#define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & ~((1UL<<(_w))-1))
+#define GTERROR(_l, _f...) xtl_log(_l, XTL_ERROR, errno, "gnttab", _f)
+#define GSERROR(_l, _f...) xtl_log(_l, XTL_ERROR, errno, "gntshr", _f)
+
static xc_osdep_handle linux_privcmd_open(xc_interface *xch)
{
int flags, saved_errno;
@@ -454,26 +457,26 @@ static struct xc_osdep_ops linux_privcmd_ops = {
#define DEVXEN "/dev/xen/"
-static xc_osdep_handle linux_gnttab_open(xc_gnttab *xcg)
+int osdep_gnttab_open(xc_gnttab *xgt)
{
int fd = open(DEVXEN "gntdev", O_RDWR);
-
if ( fd == -1 )
- return XC_OSDEP_OPEN_ERROR;
-
- return (xc_osdep_handle)fd;
+ return -1;
+ xgt->fd = fd;
+ return 0;
}
-static int linux_gnttab_close(xc_gnttab *xcg, xc_osdep_handle h)
+int osdep_gnttab_close(xc_gnttab *xgt)
{
- int fd = (int)h;
- return close(fd);
+ if ( xgt->fd == -1 )
+ return 0;
+
+ return close(xgt->fd);
}
-static int linux_gnttab_set_max_grants(xc_gnttab *xch, xc_osdep_handle h,
- uint32_t count)
+int xc_gnttab_set_max_grants(xc_gnttab *xgt, uint32_t count)
{
- int fd = (int)h, rc;
+ int fd = xgt->fd, rc;
struct ioctl_gntdev_set_max_grants max_grants = { .count = count };
rc = ioctl(fd, IOCTL_GNTDEV_SET_MAX_GRANTS, &max_grants);
@@ -485,19 +488,19 @@ static int linux_gnttab_set_max_grants(xc_gnttab *xch, xc_osdep_handle h,
if (errno == ENOTTY)
rc = 0;
else
- PERROR("linux_gnttab_set_max_grants: ioctl SET_MAX_GRANTS failed");
+ GTERROR(xgt->logger, "ioctl SET_MAX_GRANTS failed");
}
return rc;
}
-static void *linux_gnttab_grant_map(xc_gnttab *xch, xc_osdep_handle h,
- uint32_t count, int flags, int prot,
- uint32_t *domids, uint32_t *refs,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
+void *osdep_gnttab_grant_map(xc_gnttab *xgt,
+ uint32_t count, int flags, int prot,
+ uint32_t *domids, uint32_t *refs,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
{
- int fd = (int)h;
+ int fd = xgt->fd;
struct ioctl_gntdev_map_grant_ref *map;
unsigned int map_size = ROUNDUP((sizeof(*map) + (count - 1) *
sizeof(struct ioctl_gntdev_map_grant_ref)),
@@ -518,7 +521,7 @@ static void *linux_gnttab_grant_map(xc_gnttab *xch, xc_osdep_handle h,
MAP_PRIVATE | MAP_ANON | MAP_POPULATE, -1, 0);
if ( map == MAP_FAILED )
{
- PERROR("linux_gnttab_grant_map: mmap of map failed");
+ GTERROR(xgt->logger, "mmap of map failed");
return NULL;
}
}
@@ -532,7 +535,7 @@ static void *linux_gnttab_grant_map(xc_gnttab *xch, xc_osdep_handle h,
map->count = count;
if ( ioctl(fd, IOCTL_GNTDEV_MAP_GRANT_REF, map) ) {
- PERROR("linux_gnttab_grant_map: ioctl MAP_GRANT_REF failed");
+ GTERROR(xgt->logger, "ioctl MAP_GRANT_REF failed");
goto out;
}
@@ -571,7 +574,7 @@ static void *linux_gnttab_grant_map(xc_gnttab *xch, xc_osdep_handle h,
if (notify.action)
rv = ioctl(fd, IOCTL_GNTDEV_SET_UNMAP_NOTIFY, ¬ify);
if (rv) {
- PERROR("linux_gnttab_grant_map: ioctl SET_UNMAP_NOTIFY failed");
+ GTERROR(xgt->logger, "ioctl SET_UNMAP_NOTIFY failed");
munmap(addr, count * XC_PAGE_SIZE);
addr = MAP_FAILED;
}
@@ -583,7 +586,7 @@ static void *linux_gnttab_grant_map(xc_gnttab *xch, xc_osdep_handle h,
struct ioctl_gntdev_unmap_grant_ref unmap_grant;
/* Unmap the driver slots used to store the grant information. */
- PERROR("xc_gnttab_map_grant_refs: mmap failed");
+ GTERROR(xgt->logger, "mmap failed");
unmap_grant.index = map->index;
unmap_grant.count = count;
ioctl(fd, IOCTL_GNTDEV_UNMAP_GRANT_REF, &unmap_grant);
@@ -598,12 +601,9 @@ static void *linux_gnttab_grant_map(xc_gnttab *xch, xc_osdep_handle h,
return addr;
}
-
-
-static int linux_gnttab_munmap(xc_gnttab *xcg, xc_osdep_handle h,
- void *start_address, uint32_t count)
+int xc_gnttab_munmap(xc_gnttab *xgt, void *start_address, uint32_t count)
{
- int fd = (int)h;
+ int fd = xgt->fd;
struct ioctl_gntdev_get_offset_for_vaddr get_offset;
struct ioctl_gntdev_unmap_grant_ref unmap_grant;
int rc;
@@ -641,43 +641,33 @@ static int linux_gnttab_munmap(xc_gnttab *xcg, xc_osdep_handle h,
return 0;
}
-static struct xc_osdep_ops linux_gnttab_ops = {
- .open = &linux_gnttab_open,
- .close = &linux_gnttab_close,
-
- .u.gnttab = {
- .set_max_grants = linux_gnttab_set_max_grants,
- .grant_map = &linux_gnttab_grant_map,
- .munmap = &linux_gnttab_munmap,
- },
-};
-
-static xc_osdep_handle linux_gntshr_open(xc_gntshr *xcg)
+int osdep_gntshr_open(xc_gntshr *xgs)
{
int fd = open(DEVXEN "gntalloc", O_RDWR);
-
if ( fd == -1 )
- return XC_OSDEP_OPEN_ERROR;
-
- return (xc_osdep_handle)fd;
+ return -1;
+ xgs->fd = fd;
+ return 0;
}
-static int linux_gntshr_close(xc_gntshr *xcg, xc_osdep_handle h)
+int osdep_gntshr_close(xc_gntshr *xgs)
{
- int fd = (int)h;
- return close(fd);
+ if ( xgs->fd == -1 )
+ return 0;
+
+ return close(xgs->fd);
}
-static void *linux_gntshr_share_pages(xc_gntshr *xch, xc_osdep_handle h,
- uint32_t domid, int count,
- uint32_t *refs, int writable,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
+void *osdep_gntshr_share_pages(xc_gntshr *xgs,
+ uint32_t domid, int count,
+ uint32_t *refs, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
{
struct ioctl_gntalloc_alloc_gref *gref_info = NULL;
struct ioctl_gntalloc_unmap_notify notify;
struct ioctl_gntalloc_dealloc_gref gref_drop;
- int fd = (int)h;
+ int fd = xgs->fd;
int err;
void *area = NULL;
gref_info = malloc(sizeof(*gref_info) + count * sizeof(uint32_t));
@@ -689,7 +679,7 @@ static void *linux_gntshr_share_pages(xc_gntshr *xch, xc_osdep_handle h,
err = ioctl(fd, IOCTL_GNTALLOC_ALLOC_GREF, gref_info);
if (err) {
- PERROR("linux_gntshr_share_pages: ioctl failed");
+ GSERROR(xgs->logger, "ioctl failed");
goto out;
}
@@ -698,7 +688,7 @@ static void *linux_gntshr_share_pages(xc_gntshr *xch, xc_osdep_handle h,
if (area == MAP_FAILED) {
area = NULL;
- PERROR("linux_gntshr_share_pages: mmap failed");
+ GSERROR(xgs->logger, "mmap failed");
goto out_remove_fdmap;
}
@@ -715,7 +705,7 @@ static void *linux_gntshr_share_pages(xc_gntshr *xch, xc_osdep_handle h,
if (notify.action)
err = ioctl(fd, IOCTL_GNTALLOC_SET_UNMAP_NOTIFY, ¬ify);
if (err) {
- PERROR("linux_gntshr_share_page_notify: ioctl SET_UNMAP_NOTIFY failed");
+ GSERROR(xgs->logger, "ioctl SET_UNMAP_NOTIFY failed");
munmap(area, count * XC_PAGE_SIZE);
area = NULL;
}
@@ -734,33 +724,18 @@ static void *linux_gntshr_share_pages(xc_gntshr *xch, xc_osdep_handle h,
return area;
}
-static int linux_gntshr_munmap(xc_gntshr *xcg, xc_osdep_handle h,
- void *start_address, uint32_t count)
+int xc_gntshr_munmap(xc_gntshr *xgs,
+ void *start_address, uint32_t count)
{
return munmap(start_address, count * XC_PAGE_SIZE);
}
-static struct xc_osdep_ops linux_gntshr_ops = {
- .open = &linux_gntshr_open,
- .close = &linux_gntshr_close,
-
- .u.gntshr = {
- .share_pages = &linux_gntshr_share_pages,
- .munmap = &linux_gntshr_munmap,
- },
-};
-
-
static struct xc_osdep_ops *linux_osdep_init(xc_interface *xch, enum xc_osdep_type type)
{
switch ( type )
{
case XC_OSDEP_PRIVCMD:
return &linux_privcmd_ops;
- case XC_OSDEP_GNTTAB:
- return &linux_gnttab_ops;
- case XC_OSDEP_GNTSHR:
- return &linux_gntshr_ops;
default:
return NULL;
}
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index db7b344..fd7def6 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -202,19 +202,22 @@ void *xc_memalign(xc_interface *xch, size_t alignment, size_t size)
return memalign(alignment, size);
}
-static xc_osdep_handle minios_gnttab_open(xc_gnttab *xcg)
+int osdep_gnttab_open(xc_gnttab *xgt)
{
int fd = alloc_fd(FTYPE_GNTMAP);
if ( fd == -1 )
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
gntmap_init(&files[fd].gntmap);
- return (xc_osdep_handle)fd;
+ xgt->fd = fd;
+ return 0;
}
-static int minios_gnttab_close(xc_gnttab *xcg, xc_osdep_handle h)
+int osdep_gnttab_close(xc_gnttab *xgt)
{
- int fd = (int)h;
- return close(fd);
+ if ( xgt->fd == -1 )
+ return 0;
+
+ return close(xgt->fd);
}
void minios_gnttab_close_fd(int fd)
@@ -223,13 +226,13 @@ void minios_gnttab_close_fd(int fd)
files[fd].type = FTYPE_NONE;
}
-static void *minios_gnttab_grant_map(xc_gnttab *xcg, xc_osdep_handle h,
- uint32_t count, int flags, int prot,
- uint32_t *domids, uint32_t *refs,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
+void *osdep_gnttab_grant_map(xc_gnttab *xgt,
+ uint32_t count, int flags, int prot,
+ uint32_t *domids, uint32_t *refs,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
{
- int fd = (int)h;
+ int fd = xgt->fd;
int stride = 1;
if (flags & XC_GRANT_MAP_SINGLE_DOMAIN)
stride = 0;
@@ -242,11 +245,9 @@ static void *minios_gnttab_grant_map(xc_gnttab *xcg, xc_osdep_handle h,
refs, prot & PROT_WRITE);
}
-static int minios_gnttab_munmap(xc_gnttab *xcg, xc_osdep_handle h,
- void *start_address,
- uint32_t count)
+int xc_gnttab_munmap(xc_gnttab *xgt, void *start_address, uint32_t count)
{
- int fd = (int)h;
+ int fd = xgt->fd;
int ret;
ret = gntmap_munmap(&files[fd].gntmap,
(unsigned long) start_address,
@@ -258,10 +259,9 @@ static int minios_gnttab_munmap(xc_gnttab *xcg, xc_osdep_handle h,
return ret;
}
-static int minios_gnttab_set_max_grants(xc_gnttab *xcg, xc_osdep_handle h,
- uint32_t count)
+int xc_gnttab_set_max_grants(xc_gnttab *xgt, uint32_t count)
{
- int fd = (int)h;
+ int fd = xgt->fd;
int ret;
ret = gntmap_set_max_grants(&files[fd].gntmap,
count);
@@ -272,25 +272,12 @@ static int minios_gnttab_set_max_grants(xc_gnttab *xcg, xc_osdep_handle h,
return ret;
}
-static struct xc_osdep_ops minios_gnttab_ops = {
- .open = &minios_gnttab_open,
- .close = &minios_gnttab_close,
-
- .u.gnttab = {
- .grant_map = &minios_gnttab_grant_map,
- .munmap = &minios_gnttab_munmap,
- .set_max_grants = &minios_gnttab_set_max_grants,
- },
-};
-
static struct xc_osdep_ops *minios_osdep_init(xc_interface *xch, enum xc_osdep_type type)
{
switch ( type )
{
case XC_OSDEP_PRIVCMD:
return &minios_privcmd_ops;
- case XC_OSDEP_GNTTAB:
- return &minios_gnttab_ops;
default:
return NULL;
}
diff --git a/tools/libxc/xc_nogntshr.c b/tools/libxc/xc_nogntshr.c
new file mode 100644
index 0000000..9aa6064
--- /dev/null
+++ b/tools/libxc/xc_nogntshr.c
@@ -0,0 +1,46 @@
+/******************************************************************************
+ *
+ * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <stdlib.h>
+
+#include "xc_private.h"
+
+int osdep_gntshr_open(xc_gnttab *xgt)
+{
+ return -1;
+}
+
+int osdep_gntshr_close(xc_gnttab *xgt)
+{
+ return 0;
+}
+
+void *osdep_gntshr_share_pages(xc_gntshr *xgs,
+ uint32_t domid, int count,
+ uint32_t *refs, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ abort()
+}
+
+int xc_gntshr_munmap(xc_gntshr *xgs,
+ void *start_address, uint32_t count)
+{
+ abort();
+}
diff --git a/tools/libxc/xc_nognttab.c b/tools/libxc/xc_nognttab.c
new file mode 100644
index 0000000..e8a0fcb
--- /dev/null
+++ b/tools/libxc/xc_nognttab.c
@@ -0,0 +1,50 @@
+/******************************************************************************
+ *
+ * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <stdlib.h>
+
+#include "xc_private.h"
+
+int osdep_gnttab_open(xc_gnttab *xgt)
+{
+ return -1;
+}
+
+int osdep_gnttab_close(xc_gnttab *xgt)
+{
+ return 0;
+}
+
+int xc_gnttab_set_max_grants(xc_gnttab *xgt, uint32_t count)
+{
+ abort();
+}
+
+void *osdep_gnttab_grant_map(xc_gnttab *xgt,
+ uint32_t count, int flags, int prot,
+ uint32_t *domids, uint32_t *refs,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ abort();
+}
+
+int xc_gnttab_munmap(xc_gnttab *xgt, void *start_address, uint32_t count)
+{
+ abort();
+}
diff --git a/tools/libxc/xc_private.c b/tools/libxc/xc_private.c
index c6876c3..5285614 100644
--- a/tools/libxc/xc_private.c
+++ b/tools/libxc/xc_private.c
@@ -119,8 +119,6 @@ static const char *xc_osdep_type_name(enum xc_osdep_type type)
switch ( type )
{
case XC_OSDEP_PRIVCMD: return "privcmd";
- case XC_OSDEP_GNTTAB: return "gnttab";
- case XC_OSDEP_GNTSHR: return "gntshr";
}
return "unknown";
}
@@ -251,30 +249,85 @@ int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
return xch->ops->u.privcmd.hypercall(xch, xch->ops_handle, hypercall);
}
-xc_gnttab *xc_gnttab_open(xentoollog_logger *logger,
- unsigned open_flags)
+xc_gnttab *xc_gnttab_open(xentoollog_logger *logger, unsigned open_flags)
{
- return xc_interface_open_common(logger, NULL, open_flags,
- XC_OSDEP_GNTTAB);
+ xc_gnttab *xgt = malloc(sizeof(*xgt));
+ int rc;
+
+ if (!xgt) return NULL;
+
+ xgt->fd = -1;
+ xgt->logger = logger;
+ xgt->logger_tofree = NULL;
+
+ if (!xgt->logger) {
+ xgt->logger = xgt->logger_tofree =
+ (xentoollog_logger*)
+ xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
+ if (!xgt->logger) goto err;
+ }
+
+ rc = osdep_gnttab_open(xgt);
+ if ( rc < 0 ) goto err;
+
+ return xgt;
+
+err:
+ osdep_gnttab_close(xgt);
+ xtl_logger_destroy(xgt->logger_tofree);
+ free(xgt);
+ return NULL;
}
-int xc_gnttab_close(xc_gnttab *xcg)
+int xc_gnttab_close(xc_gnttab *xgt)
{
- return xc_interface_close_common(xcg);
+ int rc;
+
+ rc = osdep_gnttab_close(xgt);
+ xtl_logger_destroy(xgt->logger_tofree);
+ free(xgt);
+ return rc;
}
-xc_gntshr *xc_gntshr_open(xentoollog_logger *logger,
- unsigned open_flags)
+xc_gntshr *xc_gntshr_open(xentoollog_logger *logger, unsigned open_flags)
{
- return xc_interface_open_common(logger, NULL, open_flags,
- XC_OSDEP_GNTSHR);
+ xc_gntshr *xgs = malloc(sizeof(*xgs));
+ int rc;
+
+ if (!xgs) return NULL;
+
+ xgs->fd = -1;
+ xgs->logger = logger;
+ xgs->logger_tofree = NULL;
+
+ if (!xgs->logger) {
+ xgs->logger = xgs->logger_tofree =
+ (xentoollog_logger*)
+ xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
+ if (!xgs->logger) goto err;
+ }
+
+ rc = osdep_gntshr_open(xgs);
+ if ( rc < 0 ) goto err;
+
+ return xgs;
+
+err:
+ osdep_gntshr_close(xgs);
+ xtl_logger_destroy(xgs->logger_tofree);
+ free(xgs);
+ return NULL;
}
-int xc_gntshr_close(xc_gntshr *xcg)
+int xc_gntshr_close(xc_gntshr *xgs)
{
- return xc_interface_close_common(xcg);
-}
+ int rc;
+ rc = osdep_gntshr_close(xgs);
+ xtl_logger_destroy(xgs->logger_tofree);
+ free(xgs);
+ return rc;
+}
static pthread_key_t errbuf_pkey;
static pthread_once_t errbuf_pkey_once = PTHREAD_ONCE_INIT;
diff --git a/tools/libxc/xc_private.h b/tools/libxc/xc_private.h
index 46e0e75..1e9dc93 100644
--- a/tools/libxc/xc_private.h
+++ b/tools/libxc/xc_private.h
@@ -123,6 +123,30 @@ struct xc_interface_core {
xc_osdep_handle ops_handle; /* opaque data for xc_osdep_ops */
};
+struct xengntdev_handle {
+ xentoollog_logger *logger, *logger_tofree;
+ int fd;
+};
+
+int osdep_gnttab_open(xc_gnttab *xgt);
+int osdep_gnttab_close(xc_gnttab *xgt);
+
+#define XC_GRANT_MAP_SINGLE_DOMAIN 0x1
+void *osdep_gnttab_grant_map(xc_gnttab *xgt,
+ uint32_t count, int flags, int prot,
+ uint32_t *domids, uint32_t *refs,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port);
+
+int osdep_gntshr_open(xc_gntshr *xgs);
+int osdep_gntshr_close(xc_gntshr *xgs);
+
+void *osdep_gntshr_share_pages(xc_gntshr *xgs,
+ uint32_t domid, int count,
+ uint32_t *refs, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port);
+
void xc_report_error(xc_interface *xch, int code, const char *fmt, ...)
__attribute__((format(printf,3,4)));
void xc_reportv(xc_interface *xch, xentoollog_logger *lg, xentoollog_level,
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (5 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 06/23] tools/libxc: Remove osdep indirection for xc_gnt{shr, tab} Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 14:34 ` Wei Liu
2015-11-11 15:03 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 08/23] tools/libxc: Remove osdep indirection for privcmd Ian Campbell
` (16 subsequent siblings)
23 siblings, 2 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
libxengnttab will provide a stable API and ABI for accessing the
grant table devices.
The functions are moved into the xengnt{tab,shr} namespace to make a
clean break from libxc and avoid ambiguity regarding which interfaces
are stable.
XXX consider combining into a single namespace (i.e. with
xengnttab_handle having two open fd's in it on Linux)
All in-tree users are updated to use the new names.
Upon request (via #define XC_WANT_COMPAT_GNTTAB_API) libxenctrl will
provide a compat API for the old names. This is used by qemu-xen for
the time being. qemu-xen-traditional is updated in lockstep.
This leaves a few grant table related functions which go via privcmd
(EVTCHNOP) rather than ioctls on the /dev/xen/gnt* devices in
libxenctrl. Specifically:
- xc_gnttab_get_version
- xc_gnttab_map_table_v1
- xc_gnttab_map_table_v2
- xc_gnttab_op
These functions do not appear to be needed by qemu-dm, qemu-pv
(provision of device model to HVM guests and PV backends respectively)
or by libvchan suggesting they are not needed by non-toolstack uses of
event channels.
The new library uses a version script to ensure that only expected
symbols are exported and to version them such that ABI guarantees can
be kept in the future.
After this change libxenvchan no longer needs to link against
libxenctrl. It still needs xenctrl.h in one file for xen_mb and
friends.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
Must be applied with:
- "qemu-xen-traditional: Use libxengnttab" and a corresponding
QEMU_TAG update folded here.
- "mini-os: Include libxengnttab with libxc"" and a corresponding
bump to MINIOS_UPSTREAM_REVISION folded in here.
v3:
- Remove SHLIB_libxenctrl from SHDEPS_libxenvchan, and replace with
SHLIB_libxentoollog.
- Move to tools/libs/gnttab
- Adjust for rebase over 31cf2ca75181 "tools/libxc: linux: Don't use
getpagesize() when unmapping the grants"
v5: Allow _close(NULL).
---
.gitignore | 2 +
stubdom/Makefile | 19 +-
tools/Makefile | 3 +
tools/Rules.mk | 14 +-
tools/console/Makefile | 5 +-
tools/console/daemon/io.c | 21 +-
tools/libs/Makefile | 1 +
tools/libs/evtchn/minios.c | 5 +-
tools/libs/gnttab/Makefile | 69 +++++
tools/libs/gnttab/gntshr_core.c | 95 ++++++
.../xc_nognttab.c => libs/gnttab/gntshr_unimp.c} | 34 ++-
tools/libs/gnttab/gnttab_core.c | 124 ++++++++
tools/libs/gnttab/gnttab_unimp.c | 89 ++++++
tools/libs/gnttab/include/xengnttab.h | 218 ++++++++++++++
tools/libs/gnttab/libxengnttab.map | 23 ++
tools/libs/gnttab/linux.c | 329 +++++++++++++++++++++
tools/libs/gnttab/minios.c | 117 ++++++++
tools/libs/gnttab/private.h | 47 +++
tools/libvchan/Makefile | 8 +-
tools/libvchan/init.c | 24 +-
tools/libvchan/io.c | 8 +-
tools/libvchan/libxenvchan.h | 6 +-
tools/libxc/Makefile | 15 +-
tools/libxc/include/xenctrl.h | 168 -----------
tools/libxc/include/xenctrl_compat.h | 48 +++
tools/libxc/xc_gnttab.c | 53 ----
tools/libxc/xc_gnttab_compat.c | 111 +++++++
tools/libxc/xc_linux_osdep.c | 280 ------------------
tools/libxc/xc_minios.c | 73 -----
tools/libxc/xc_nogntshr.c | 46 ---
tools/libxc/xc_private.c | 80 -----
tools/libxc/xc_private.h | 24 --
tools/xenstore/Makefile | 4 +-
tools/xenstore/xenstored_core.h | 4 +-
tools/xenstore/xenstored_domain.c | 24 +-
tools/xenstore/xenstored_minios.c | 5 +-
36 files changed, 1396 insertions(+), 800 deletions(-)
create mode 100644 tools/libs/gnttab/Makefile
create mode 100644 tools/libs/gnttab/gntshr_core.c
rename tools/{libxc/xc_nognttab.c => libs/gnttab/gntshr_unimp.c} (52%)
create mode 100644 tools/libs/gnttab/gnttab_core.c
create mode 100644 tools/libs/gnttab/gnttab_unimp.c
create mode 100644 tools/libs/gnttab/include/xengnttab.h
create mode 100644 tools/libs/gnttab/libxengnttab.map
create mode 100644 tools/libs/gnttab/linux.c
create mode 100644 tools/libs/gnttab/minios.c
create mode 100644 tools/libs/gnttab/private.h
create mode 100644 tools/libxc/xc_gnttab_compat.c
delete mode 100644 tools/libxc/xc_nogntshr.c
diff --git a/.gitignore b/.gitignore
index b34dc3c..9241c54 100644
--- a/.gitignore
+++ b/.gitignore
@@ -60,6 +60,7 @@ stubdom/ioemu
stubdom/xenstore
stubdom/libxentoollog-*
stubdom/libxenevtchn-*
+stubdom/libxengnttab-*
stubdom/libxc-*
stubdom/lwip-*
stubdom/mini-os-*
@@ -88,6 +89,7 @@ config/Stubdom.mk
config/Docs.mk
tools/libs/toollog/headers.chk
tools/libs/evtchn/headers.chk
+tools/libs/gnttab/headers.chk
tools/blktap2/daemon/blktapctrl
tools/blktap2/drivers/img2qcow
tools/blktap2/drivers/lock-util
diff --git a/stubdom/Makefile b/stubdom/Makefile
index 3b3bfc9..d4576eb 100644
--- a/stubdom/Makefile
+++ b/stubdom/Makefile
@@ -324,6 +324,12 @@ mk-headers-$(XEN_TARGET_ARCH): $(IOEMU_LINKFARM_TARGET)
ln -sf $(XEN_ROOT)/tools/libs/evtchn/include/*.h include/ && \
ln -sf $(XEN_ROOT)/tools/libs/evtchn/*.c . && \
ln -sf $(XEN_ROOT)/tools/libs/evtchn/Makefile . )
+ mkdir -p libs-$(XEN_TARGET_ARCH)/gnttab/include
+ [ -h libs-$(XEN_TARGET_ARCH)/gnttab/Makefile ] || ( cd libs-$(XEN_TARGET_ARCH)/gnttab && \
+ ln -sf $(XEN_ROOT)/tools/libs/gnttab/*.h . && \
+ ln -sf $(XEN_ROOT)/tools/libs/gnttab/include/*.h include/ && \
+ ln -sf $(XEN_ROOT)/tools/libs/gnttab/*.c . && \
+ ln -sf $(XEN_ROOT)/tools/libs/gnttab/Makefile . )
mkdir -p libxc-$(XEN_TARGET_ARCH)
[ -h libxc-$(XEN_TARGET_ARCH)/Makefile ] || ( cd libxc-$(XEN_TARGET_ARCH) && \
ln -sf $(XEN_ROOT)/tools/libxc/*.h . && \
@@ -369,12 +375,23 @@ libs-$(XEN_TARGET_ARCH)/evtchn/libxenevtchn.a: $(NEWLIB_STAMPFILE)
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/evtchn
#######
+# libxengnttab
+#######
+
+.PHONY: libxengnttab
+libxengnttab: libs-$(XEN_TARGET_ARCH)/gnttab/libxengnttab.a
+libs-$(XEN_TARGET_ARCH)/gnttab/libxengnttab.a: $(NEWLIB_STAMPFILE)
+ $(MAKE) -C $(XEN_ROOT)/tools/include
+ $(MAKE) DESTDIR= -C $(MINI_OS) links
+ CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/gnttab
+
+#######
# libxc
#######
.PHONY: libxc
libxc: libxc-$(XEN_TARGET_ARCH)/libxenctrl.a libxc-$(XEN_TARGET_ARCH)/libxenguest.a
-libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog libxenevtchn cross-zlib
+libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog libxenevtchn libxengnttab cross-zlib
$(MAKE) -C $(XEN_ROOT)/tools/include
$(MAKE) DESTDIR= -C $(MINI_OS) links
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= CONFIG_LIBXC_MINIOS=y -C libxc-$(XEN_TARGET_ARCH)
diff --git a/tools/Makefile b/tools/Makefile
index 99e016a..f373e71 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -249,9 +249,11 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
--includedir=$(LIBEXEC_INC) \
--source-path=$$source \
--extra-cflags="-DXC_WANT_COMPAT_EVTCHN_API=1 \
+ -DXC_WANT_COMPAT_GNTTAB_API=1 \
-I$(XEN_ROOT)/tools/include \
-I$(XEN_ROOT)/tools/libs/toollog/include \
-I$(XEN_ROOT)/tools/libs/evtchn/include \
+ -I$(XEN_ROOT)/tools/libs/gnttab/include \
-I$(XEN_ROOT)/tools/libxc/include \
-I$(XEN_ROOT)/tools/xenstore/include \
-I$(XEN_ROOT)/tools/xenstore/compat/include \
@@ -260,6 +262,7 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
-L$(XEN_ROOT)/tools/xenstore \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/evtchn \
+ -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/gnttab \
$(QEMU_UPSTREAM_RPATH)" \
--bindir=$(LIBEXEC_BIN) \
--datadir=$(SHAREDIR)/qemu-xen \
diff --git a/tools/Rules.mk b/tools/Rules.mk
index 8ed1507..5824ede 100644
--- a/tools/Rules.mk
+++ b/tools/Rules.mk
@@ -12,6 +12,7 @@ INSTALL = $(XEN_ROOT)/tools/cross-install
XEN_INCLUDE = $(XEN_ROOT)/tools/include
XEN_LIBXENTOOLLOG = $(XEN_ROOT)/tools/libs/toollog
XEN_LIBXENEVTCHN = $(XEN_ROOT)/tools/libs/evtchn
+XEN_LIBXENGNTTAB = $(XEN_ROOT)/tools/libs/gnttab
XEN_LIBXC = $(XEN_ROOT)/tools/libxc
XEN_XENLIGHT = $(XEN_ROOT)/tools/libxl
XEN_XENSTORE = $(XEN_ROOT)/tools/xenstore
@@ -88,8 +89,17 @@ SHDEPS_libxenevtchn =
LDLIBS_libxenevtchn = $(XEN_LIBXENEVTCHN)/libxenevtchn$(libextension)
SHLIB_libxenevtchn = -Wl,-rpath-link=$(XEN_LIBXENEVTCHN)
+CFLAGS_libxengnttab = -I$(XEN_LIBXENGNTTAB)/include $(CFLAGS_xeninclude)
+LDLIBS_libxengnttab = $(XEN_LIBXENGNTTAB)/libxengnttab$(libextension)
+SHLIB_libxengnttab = -Wl,-rpath-link=$(XEN_LIBXENGNTTAB)
+
+# xengntshr_* interfaces are actually part of libxengnttab.so
+CFLAGS_libxengntshr = -I$(XEN_LIBXENGNTTAB)/include $(CFLAGS_xeninclude)
+LDLIBS_libxengntshr = $(XEN_LIBXENGNTTAB)/libxengnttab$(libextension)
+SHLIB_libxengntshr = -Wl,-rpath-link=$(XEN_LIBXENGNTTAB)
+
CFLAGS_libxenctrl = -I$(XEN_LIBXC)/include $(CFLAGS_libxentoollog) $(CFLAGS_xeninclude)
-SHDEPS_libxenctrl = $(SHLIB_libxentoollog) $(SHLIB_libxenevtchn)
+SHDEPS_libxenctrl = $(SHLIB_libxentoollog) $(SHLIB_libxenevtchn) $(SHLIB_libxengnttab) $(SHLIB_libxengntshr)
LDLIBS_libxenctrl = $(SHDEPS_libxenctrl) $(XEN_LIBXC)/libxenctrl$(libextension)
SHLIB_libxenctrl = $(SHDEPS_libxenctrl) -Wl,-rpath-link=$(XEN_LIBXC)
@@ -109,7 +119,7 @@ LDLIBS_libxenstat = $(SHDEPS_libxenstat) $(XEN_LIBXENSTAT)/libxenstat$(libexten
SHLIB_libxenstat = $(SHDEPS_libxenstat) -Wl,-rpath-link=$(XEN_LIBXENSTAT)
CFLAGS_libxenvchan = -I$(XEN_LIBVCHAN)
-SHDEPS_libxenvchan = $(SHLIB_libxenctrl) $(SHLIB_libxenstore) $(SHLIB_libxenevtchn)
+SHDEPS_libxenvchan = $(SHLIB_libxentoollog) $(SHLIB_libxenstore) $(SHLIB_libxenevtchn) $(SHLIB_libxengnttab) $(SHLIB_libxengntshr)
LDLIBS_libxenvchan = $(SHDEPS_libxenvchan) $(XEN_LIBVCHAN)/libxenvchan$(libextension)
SHLIB_libxenvchan = $(SHDEPS_libxenvchan) -Wl,-rpath-link=$(XEN_LIBVCHAN)
diff --git a/tools/console/Makefile b/tools/console/Makefile
index 4b3a492..6eeac8f 100644
--- a/tools/console/Makefile
+++ b/tools/console/Makefile
@@ -3,10 +3,8 @@ include $(XEN_ROOT)/tools/Rules.mk
CFLAGS += -Werror
-CFLAGS += $(CFLAGS_libxenevtchn)
CFLAGS += $(CFLAGS_libxenctrl)
CFLAGS += $(CFLAGS_libxenstore)
-LDLIBS += $(LDLIBS_libxenevtchn)
LDLIBS += $(LDLIBS_libxenctrl)
LDLIBS += $(LDLIBS_libxenstore)
LDLIBS += $(SOCKET_LIBS)
@@ -28,8 +26,9 @@ clean:
.PHONY: distclean
distclean: clean
+daemon/io.o: CFLAGS += $(CFLAGS_libxenevtchn) $(CFLAGS_libxengnttab)
xenconsoled: $(patsubst %.c,%.o,$(wildcard daemon/*.c))
- $(CC) $(LDFLAGS) $^ -o $@ $(LDLIBS) $(LDLIBS_xenconsoled) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) $^ -o $@ $(LDLIBS) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_xenconsoled) $(APPEND_LDFLAGS)
xenconsole: client/_paths.h $(patsubst %.c,%.o,$(wildcard client/*.c))
$(CC) $(LDFLAGS) $^ -o $@ $(LDLIBS) $(LDLIBS_xenconsole) $(APPEND_LDFLAGS)
diff --git a/tools/console/daemon/io.c b/tools/console/daemon/io.c
index 2f2e9c5..f265f4d 100644
--- a/tools/console/daemon/io.c
+++ b/tools/console/daemon/io.c
@@ -22,6 +22,7 @@
#include "utils.h"
#include "io.h"
#include <xenevtchn.h>
+#include <xengnttab.h>
#include <xenstore.h>
#include <xen/io/console.h>
#include <xen/grant_table.h>
@@ -72,7 +73,7 @@ static int log_time_hv_needts = 1;
static int log_time_guest_needts = 1;
static int log_hv_fd = -1;
-static xc_gnttab *xcg_handle = NULL;
+static xengnttab_handle *xgt_handle = NULL;
static struct pollfd *fds;
static unsigned int current_array_size;
@@ -520,8 +521,8 @@ static void domain_unmap_interface(struct domain *dom)
{
if (dom->interface == NULL)
return;
- if (xcg_handle && dom->ring_ref == -1)
- xc_gnttab_munmap(xcg_handle, dom->interface, 1);
+ if (xgt_handle && dom->ring_ref == -1)
+ xengnttab_munmap(xgt_handle, dom->interface, 1);
else
munmap(dom->interface, XC_PAGE_SIZE);
dom->interface = NULL;
@@ -552,9 +553,9 @@ static int domain_create_ring(struct domain *dom)
if (ring_ref != dom->ring_ref && dom->ring_ref != -1)
domain_unmap_interface(dom);
- if (!dom->interface && xcg_handle) {
+ if (!dom->interface && xgt_handle) {
/* Prefer using grant table */
- dom->interface = xc_gnttab_map_grant_ref(xcg_handle,
+ dom->interface = xengnttab_map_grant_ref(xgt_handle,
dom->domid, GNTTAB_RESERVED_CONSOLE,
PROT_READ|PROT_WRITE);
dom->ring_ref = -1;
@@ -1029,8 +1030,8 @@ void handle_io(void)
handle_hv_logs(xce_handle, true);
}
- xcg_handle = xc_gnttab_open(NULL, 0);
- if (xcg_handle == NULL) {
+ xgt_handle = xengnttab_open(NULL, 0);
+ if (xgt_handle == NULL) {
dolog(LOG_DEBUG, "Failed to open xcg handle: %d (%s)",
errno, strerror(errno));
}
@@ -1206,9 +1207,9 @@ void handle_io(void)
xenevtchn_close(xce_handle);
xce_handle = NULL;
}
- if (xcg_handle != NULL) {
- xc_gnttab_close(xcg_handle);
- xcg_handle = NULL;
+ if (xgt_handle != NULL) {
+ xengnttab_close(xgt_handle);
+ xgt_handle = NULL;
}
log_hv_evtchn = -1;
}
diff --git a/tools/libs/Makefile b/tools/libs/Makefile
index 0e3f523..00156ae 100644
--- a/tools/libs/Makefile
+++ b/tools/libs/Makefile
@@ -4,5 +4,6 @@ include $(XEN_ROOT)/tools/Rules.mk
SUBDIRS-y :=
SUBDIRS-y += toollog
SUBDIRS-y += evtchn
+SUBDIRS-y += gnttab
all clean install distclean: %: subdirs-%
diff --git a/tools/libs/evtchn/minios.c b/tools/libs/evtchn/minios.c
index fb913a2..b839cd0 100644
--- a/tools/libs/evtchn/minios.c
+++ b/tools/libs/evtchn/minios.c
@@ -27,13 +27,12 @@
#include <mini-os/events.h>
#include <mini-os/wait.h>
-#include <sys/socket.h>
-
#include <assert.h>
#include <errno.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdint.h>
+#include <unistd.h>
#include <inttypes.h>
#include <malloc.h>
@@ -43,8 +42,6 @@ extern void minios_evtchn_close_fd(int fd);
extern struct wait_queue_head event_queue;
-//void minios_evtchn_close_fd(int fd);
-
/* XXX Note: This is not threadsafe */
static struct evtchn_port_info* port_alloc(int fd) {
struct evtchn_port_info *port_info;
diff --git a/tools/libs/gnttab/Makefile b/tools/libs/gnttab/Makefile
new file mode 100644
index 0000000..53ba960
--- /dev/null
+++ b/tools/libs/gnttab/Makefile
@@ -0,0 +1,69 @@
+XEN_ROOT = $(CURDIR)/../../..
+include $(XEN_ROOT)/tools/Rules.mk
+
+MAJOR = 1
+MINOR = 0
+SHLIB_LDFLAGS += -Wl,--version-script=libxengnttab.map
+
+CFLAGS += -Werror -Wmissing-prototypes
+CFLAGS += -I./include $(CFLAGS_xeninclude)
+CFLAGS += $(CFLAGS_libxentoollog)
+
+SRCS-GNTTAB += gnttab_core.c
+SRCS-GNTSHR += gntshr_core.c
+
+SRCS-$(CONFIG_Linux) += $(SRCS-GNTTAB) $(SRCS-GNTSHR) linux.c
+SRCS-$(CONFIG_MiniOS) += $(SRCS-GNTTAB) gntshr_unimp.c minios.c
+SRCS-$(CONFIG_FreeBSD) += gnttab_unimp.c gntshr_unimp.c
+SRCS-$(CONFIG_SunOS) += gnttab_unimp.c gntshr_unimp.c
+SRCS-$(CONFIG_NetBSD) += gnttab_unimp.c gntshr_unimp.c
+
+LIB_OBJS := $(patsubst %.c,%.o,$(SRCS-y))
+PIC_OBJS := $(patsubst %.c,%.opic,$(SRCS-y))
+
+LIB := libxengnttab.a
+ifneq ($(nosharedlibs),y)
+LIB += libxengnttab.so
+endif
+
+.PHONY: all
+all: build
+
+.PHONY: build
+build:
+ $(MAKE) libs
+
+.PHONY: libs
+libs: headers.chk $(LIB)
+
+headers.chk: $(wildcard include/*.h)
+
+libxengnttab.a: $(LIB_OBJS)
+ $(AR) rc $@ $^
+
+libxengnttab.so: libxengnttab.so.$(MAJOR)
+ $(SYMLINK_SHLIB) $< $@
+libxengnttab.so.$(MAJOR): libxengnttab.so.$(MAJOR).$(MINOR)
+ $(SYMLINK_SHLIB) $< $@
+
+libxengnttab.so.$(MAJOR).$(MINOR): $(PIC_OBJS) libxengnttab.map
+ $(CC) $(LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxengnttab.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $(PIC_OBJS) $(LDLIBS_libxentoollog) $(APPEND_LDFLAGS)
+
+.PHONY: install
+install: build
+ $(INSTALL_DIR) $(DESTDIR)$(libdir)
+ $(INSTALL_DIR) $(DESTDIR)$(includedir)
+ $(INSTALL_SHLIB) libxengnttab.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)
+ $(INSTALL_DATA) libxengnttab.a $(DESTDIR)$(libdir)
+ $(SYMLINK_SHLIB) libxengnttab.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxengnttab.so.$(MAJOR)
+ $(SYMLINK_SHLIB) libxengnttab.so.$(MAJOR) $(DESTDIR)$(libdir)/libxengnttab.so
+ $(INSTALL_DATA) include/xengnttab.h $(DESTDIR)$(includedir)
+
+.PHONY: TAGS
+TAGS:
+ etags -t *.c *.h
+
+.PHONY: clean
+clean:
+ rm -rf *.rpm $(LIB) *~ $(DEPS) $(LIB_OBJS) $(PIC_OBJS)
+ rm -f headers.chk
diff --git a/tools/libs/gnttab/gntshr_core.c b/tools/libs/gnttab/gntshr_core.c
new file mode 100644
index 0000000..72dd86a
--- /dev/null
+++ b/tools/libs/gnttab/gntshr_core.c
@@ -0,0 +1,95 @@
+/******************************************************************************
+ *
+ * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_gnttab.c
+ */
+
+#include <stdlib.h>
+
+#include "private.h"
+
+xengntshr_handle *xengntshr_open(xentoollog_logger *logger, unsigned open_flags)
+{
+ xengntshr_handle *xgs = malloc(sizeof(*xgs));
+ int rc;
+
+ if (!xgs) return NULL;
+
+ xgs->fd = -1;
+ xgs->logger = logger;
+ xgs->logger_tofree = NULL;
+
+ if (!xgs->logger) {
+ xgs->logger = xgs->logger_tofree =
+ (xentoollog_logger*)
+ xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
+ if (!xgs->logger) goto err;
+ }
+
+ rc = osdep_gntshr_open(xgs);
+ if ( rc < 0 ) goto err;
+
+ return xgs;
+
+err:
+ osdep_gntshr_close(xgs);
+ xtl_logger_destroy(xgs->logger_tofree);
+ free(xgs);
+ return NULL;
+}
+
+int xengntshr_close(xengntshr_handle *xgs)
+{
+ int rc;
+
+ if ( !xgs )
+ return 0;
+
+ rc = osdep_gntshr_close(xgs);
+ xtl_logger_destroy(xgs->logger_tofree);
+ free(xgs);
+ return rc;
+}
+void *xengntshr_share_pages(xengntshr_handle *xcg, uint32_t domid,
+ int count, uint32_t *refs, int writable)
+{
+ return osdep_gntshr_share_pages(xcg, domid, count, refs, writable, -1, -1);
+}
+
+void *xengntshr_share_page_notify(xengntshr_handle *xcg, uint32_t domid,
+ uint32_t *ref, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ return osdep_gntshr_share_pages(xcg, domid, 1, ref, writable,
+ notify_offset, notify_port);
+}
+
+int xengntshr_munmap(xengntshr_handle *xgs, void *start_address, uint32_t count)
+{
+ return osdep_gntshr_munmap(xgs, start_address, count);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libxc/xc_nognttab.c b/tools/libs/gnttab/gntshr_unimp.c
similarity index 52%
rename from tools/libxc/xc_nognttab.c
rename to tools/libs/gnttab/gntshr_unimp.c
index e8a0fcb..71dd558 100644
--- a/tools/libxc/xc_nognttab.c
+++ b/tools/libs/gnttab/gntshr_unimp.c
@@ -14,37 +14,49 @@
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_gnttab.c
*/
#include <stdlib.h>
-#include "xc_private.h"
+#include "private.h"
-int osdep_gnttab_open(xc_gnttab *xgt)
+xengntshr_handle *xengntshr_open(xentoollog_logger *logger, unsigned open_flags)
{
- return -1;
+ return NULL;
}
-int osdep_gnttab_close(xc_gnttab *xgt)
+int xengntshr_close(xengntshr_handle *xgs)
{
return 0;
}
-int xc_gnttab_set_max_grants(xc_gnttab *xgt, uint32_t count)
+void *xengntshr_share_pages(xengntshr_handle *xcg, uint32_t domid,
+ int count, uint32_t *refs, int writable)
{
abort();
}
-void *osdep_gnttab_grant_map(xc_gnttab *xgt,
- uint32_t count, int flags, int prot,
- uint32_t *domids, uint32_t *refs,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
+void *xengntshr_share_page_notify(xengntshr_handle *xcg, uint32_t domid,
+ uint32_t *ref, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
{
abort();
}
-int xc_gnttab_munmap(xc_gnttab *xgt, void *start_address, uint32_t count)
+int xengntshr_munmap(xengntshr_handle *xgs, void *start_address, uint32_t count)
{
abort();
}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/gnttab/gnttab_core.c b/tools/libs/gnttab/gnttab_core.c
new file mode 100644
index 0000000..1a752a9
--- /dev/null
+++ b/tools/libs/gnttab/gnttab_core.c
@@ -0,0 +1,124 @@
+/******************************************************************************
+ *
+ * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_gnttab.c
+ */
+
+#include <stdlib.h>
+
+#include "private.h"
+
+xengnttab_handle *xengnttab_open(xentoollog_logger *logger, unsigned open_flags)
+{
+ xengnttab_handle *xgt = malloc(sizeof(*xgt));
+ int rc;
+
+ if (!xgt) return NULL;
+
+ xgt->fd = -1;
+ xgt->logger = logger;
+ xgt->logger_tofree = NULL;
+
+ if (!xgt->logger) {
+ xgt->logger = xgt->logger_tofree =
+ (xentoollog_logger*)
+ xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
+ if (!xgt->logger) goto err;
+ }
+
+ rc = osdep_gnttab_open(xgt);
+ if ( rc < 0 ) goto err;
+
+ return xgt;
+
+err:
+ osdep_gnttab_close(xgt);
+ xtl_logger_destroy(xgt->logger_tofree);
+ free(xgt);
+ return NULL;
+}
+
+int xengnttab_close(xengnttab_handle *xgt)
+{
+ int rc;
+
+ if ( !xgt )
+ return 0;
+
+ rc = osdep_gnttab_close(xgt);
+ xtl_logger_destroy(xgt->logger_tofree);
+ free(xgt);
+ return rc;
+}
+
+int xengnttab_set_max_grants(xengnttab_handle *xgt, uint32_t count)
+{
+ return osdep_gnttab_set_max_grants(xgt, count);
+}
+
+void *xengnttab_map_grant_ref(xengnttab_handle *xgt,
+ uint32_t domid,
+ uint32_t ref,
+ int prot)
+{
+ return osdep_gnttab_grant_map(xgt, 1, 0, prot, &domid, &ref, -1, -1);
+}
+
+void *xengnttab_map_grant_refs(xengnttab_handle *xgt,
+ uint32_t count,
+ uint32_t *domids,
+ uint32_t *refs,
+ int prot)
+{
+ return osdep_gnttab_grant_map(xgt, count, 0, prot, domids, refs, -1, -1);
+}
+
+void *xengnttab_map_domain_grant_refs(xengnttab_handle *xgt,
+ uint32_t count,
+ uint32_t domid,
+ uint32_t *refs,
+ int prot)
+{
+ return osdep_gnttab_grant_map(xgt, count, XENGNTTAB_GRANT_MAP_SINGLE_DOMAIN,
+ prot, &domid, refs, -1, -1);
+}
+
+void *xengnttab_map_grant_ref_notify(xengnttab_handle *xgt,
+ uint32_t domid,
+ uint32_t ref,
+ int prot,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ return osdep_gnttab_grant_map(xgt, 1, 0, prot, &domid, &ref,
+ notify_offset, notify_port);
+}
+
+int xengnttab_munmap(xengnttab_handle *xgt, void *start_address, uint32_t count)
+{
+ return osdep_gnttab_munmap(xgt, start_address, count);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/gnttab/gnttab_unimp.c b/tools/libs/gnttab/gnttab_unimp.c
new file mode 100644
index 0000000..3f930be
--- /dev/null
+++ b/tools/libs/gnttab/gnttab_unimp.c
@@ -0,0 +1,89 @@
+/******************************************************************************
+ *
+ * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_gnttab.c
+ */
+
+#include <stdlib.h>
+
+#include "private.h"
+
+xengnttab_handle *xengnttab_open(xentoollog_logger *logger, unsigned open_flags)
+{
+ return NULL;
+}
+
+int xengnttab_close(xengnttab_handle *xgt)
+{
+ return 0;
+}
+
+int xengnttab_set_max_grants(xengnttab_handle *xgt, uint32_t count)
+{
+ abort();
+}
+
+void *xengnttab_map_grant_ref(xengnttab_handle *xgt,
+ uint32_t domid,
+ uint32_t ref,
+ int prot)
+{
+ abort();
+}
+
+void *xengnttab_map_grant_refs(xengnttab_handle *xgt,
+ uint32_t count,
+ uint32_t *domids,
+ uint32_t *refs,
+ int prot)
+{
+ abort();
+}
+
+void *xengnttab_map_domain_grant_refs(xengnttab_handle *xgt,
+ uint32_t count,
+ uint32_t domid,
+ uint32_t *refs,
+ int prot)
+{
+ abort();
+}
+
+void *xengnttab_map_grant_ref_notify(xengnttab_handle *xgt,
+ uint32_t domid,
+ uint32_t ref,
+ int prot,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ abort();
+}
+
+int xengnttab_munmap(xengnttab_handle *xgt, void *start_address, uint32_t count)
+{
+ abort();
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/gnttab/include/xengnttab.h b/tools/libs/gnttab/include/xengnttab.h
new file mode 100644
index 0000000..007e6db
--- /dev/null
+++ b/tools/libs/gnttab/include/xengnttab.h
@@ -0,0 +1,218 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split off from:
+ * xenctrl.h
+ *
+ * A library for low-level access to the Xen control interfaces.
+ *
+ * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
+ */
+#ifndef XENGNTTAB_H
+#define XENGNTTAB_H
+
+#include <stdint.h>
+
+#include <xen/grant_table.h>
+#include <xen/event_channel.h>
+
+/* Callers who don't care don't need to #include <xentoollog.h> */
+typedef struct xentoollog_logger xentoollog_logger;
+
+/*
+ * Grant Table Interface (making use of grants from other domains)
+ */
+
+typedef struct xengntdev_handle xengnttab_handle;
+
+/*
+ * Note:
+ * After fork a child process must not use any opened xc gnttab
+ * handle inherited from their parent. They must open a new handle if
+ * they want to interact with xc.
+ *
+ * Return an fd onto the grant table driver. Logs errors.
+ */
+xengnttab_handle *xengnttab_open(xentoollog_logger *logger, unsigned open_flags);
+
+/*
+ * Close a handle previously allocated with xengnttab_open().
+ * Never logs errors.
+ */
+int xengnttab_close(xengnttab_handle *xgt);
+
+/*
+ * Memory maps a grant reference from one domain to a local address range.
+ * Mappings should be unmapped with xengnttab_munmap. Logs errors.
+ *
+ * @parm xgt a handle on an open grant table interface
+ * @parm domid the domain to map memory from
+ * @parm ref the grant reference ID to map
+ * @parm prot same flag as in mmap()
+ */
+void *xengnttab_map_grant_ref(xengnttab_handle *xgt,
+ uint32_t domid,
+ uint32_t ref,
+ int prot);
+
+/**
+ * Memory maps one or more grant references from one or more domains to a
+ * contiguous local address range. Mappings should be unmapped with
+ * xengnttab_munmap. Logs errors.
+ *
+ * @parm xgt a handle on an open grant table interface
+ * @parm count the number of grant references to be mapped
+ * @parm domids an array of @count domain IDs by which the corresponding @refs
+ * were granted
+ * @parm refs an array of @count grant references to be mapped
+ * @parm prot same flag as in mmap()
+ */
+void *xengnttab_map_grant_refs(xengnttab_handle *xgt,
+ uint32_t count,
+ uint32_t *domids,
+ uint32_t *refs,
+ int prot);
+
+/**
+ * Memory maps one or more grant references from one domain to a
+ * contiguous local address range. Mappings should be unmapped with
+ * xengnttab_munmap. Logs errors.
+ *
+ * @parm xgt a handle on an open grant table interface
+ * @parm count the number of grant references to be mapped
+ * @parm domid the domain to map memory from
+ * @parm refs an array of @count grant references to be mapped
+ * @parm prot same flag as in mmap()
+ */
+void *xengnttab_map_domain_grant_refs(xengnttab_handle *xgt,
+ uint32_t count,
+ uint32_t domid,
+ uint32_t *refs,
+ int prot);
+
+/**
+ * Memory maps a grant reference from one domain to a local address range.
+ * Mappings should be unmapped with xengnttab_munmap. If notify_offset or
+ * notify_port are not -1, this version will attempt to set up an unmap
+ * notification at the given offset and event channel. When the page is
+ * unmapped, the byte at the given offset will be zeroed and a wakeup will be
+ * sent to the given event channel. Logs errors.
+ *
+ * @parm xgt a handle on an open grant table interface
+ * @parm domid the domain to map memory from
+ * @parm ref the grant reference ID to map
+ * @parm prot same flag as in mmap()
+ * @parm notify_offset The byte offset in the page to use for unmap
+ * notification; -1 for none.
+ * @parm notify_port The event channel port to use for unmap notify, or -1
+ */
+void *xengnttab_map_grant_ref_notify(xengnttab_handle *xgt,
+ uint32_t domid,
+ uint32_t ref,
+ int prot,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port);
+
+/*
+ * Unmaps the @count pages starting at @start_address, which were mapped by a
+ * call to xengnttab_map_grant_ref or xengnttab_map_grant_refs. Never logs.
+ */
+int xengnttab_munmap(xengnttab_handle *xgt,
+ void *start_address,
+ uint32_t count);
+
+/*
+ * Sets the maximum number of grants that may be mapped by the given instance
+ * to @count. Never logs.
+ *
+ * N.B. This function must be called after opening the handle, and before any
+ * other functions are invoked on it.
+ *
+ * N.B. When variable-length grants are mapped, fragmentation may be observed,
+ * and it may not be possible to satisfy requests up to the maximum number
+ * of grants.
+ */
+int xengnttab_set_max_grants(xengnttab_handle *xgt,
+ uint32_t count);
+
+/*
+ * Grant Sharing Interface (allocating and granting pages)
+ */
+
+typedef struct xengntdev_handle xengntshr_handle;
+
+/*
+ * Return an fd onto the grant sharing driver. Logs errors.
+ *
+ * Note:
+ * After fork a child process must not use any opened xc gntshr
+ * handle inherited from their parent. They must open a new handle if
+ * they want to interact with xc.
+ *
+ */
+xengntshr_handle *xengntshr_open(xentoollog_logger *logger,
+ unsigned open_flags);
+
+/*
+ * Close a handle previously allocated with xengntshr_open().
+ * Never logs errors.
+ */
+int xengntshr_close(xengntshr_handle *xgs);
+
+/*
+ * Creates and shares pages with another domain.
+ *
+ * @parm xgs a handle to an open grant sharing instance
+ * @parm domid the domain to share memory with
+ * @parm count the number of pages to share
+ * @parm refs the grant references of the pages (output)
+ * @parm writable true if the other domain can write to the pages
+ * @return local mapping of the pages
+ */
+void *xengntshr_share_pages(xengntshr_handle *xgs, uint32_t domid,
+ int count, uint32_t *refs, int writable);
+
+/*
+ * Creates and shares a page with another domain, with unmap notification.
+ *
+ * @parm xgs a handle to an open grant sharing instance
+ * @parm domid the domain to share memory with
+ * @parm refs the grant reference of the pages (output)
+ * @parm writable true if the other domain can write to the page
+ * @parm notify_offset The byte offset in the page to use for unmap
+ * notification; -1 for none.
+ * @parm notify_port The event channel port to use for unmap notify, or -1
+ * @return local mapping of the page
+ */
+void *xengntshr_share_page_notify(xengntshr_handle *xgs, uint32_t domid,
+ uint32_t *ref, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port);
+/*
+ * Unmaps the @count pages starting at @start_address, which were mapped by a
+ * call to xengntshr_share_*. Never logs.
+ */
+int xengntshr_munmap(xengntshr_handle *xgs, void *start_address, uint32_t count);
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/gnttab/libxengnttab.map b/tools/libs/gnttab/libxengnttab.map
new file mode 100644
index 0000000..b46f9aa
--- /dev/null
+++ b/tools/libs/gnttab/libxengnttab.map
@@ -0,0 +1,23 @@
+VERS_1.0 {
+ global:
+ xengnttab_open;
+ xengnttab_close;
+
+ xengnttab_set_max_grants;
+
+ xengnttab_map_domain_grant_refs;
+ xengnttab_map_grant_ref;
+ xengnttab_map_grant_ref_notify;
+ xengnttab_map_grant_refs;
+
+ xengnttab_munmap;
+
+ xengntshr_open;
+ xengntshr_close;
+
+ xengntshr_share_page_notify;
+ xengntshr_share_pages;
+
+ xengntshr_munmap;
+ local: *; /* Do not expose anything by default */
+};
diff --git a/tools/libs/gnttab/linux.c b/tools/libs/gnttab/linux.c
new file mode 100644
index 0000000..635765b
--- /dev/null
+++ b/tools/libs/gnttab/linux.c
@@ -0,0 +1,329 @@
+/*
+ * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_linux_osdep.c
+ */
+
+#include <fcntl.h>
+#include <errno.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <string.h>
+
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+
+#include <xen/sys/gntdev.h>
+#include <xen/sys/gntalloc.h>
+
+#include "private.h"
+
+#define DEVXEN "/dev/xen/"
+
+#define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & ~((1UL<<(_w))-1))
+
+#define GTERROR(_l, _f...) xtl_log(_l, XTL_ERROR, errno, "gnttab", _f)
+#define GSERROR(_l, _f...) xtl_log(_l, XTL_ERROR, errno, "gntshr", _f)
+
+#define PAGE_SHIFT 12
+#define PAGE_SIZE (1UL << PAGE_SHIFT)
+#define PAGE_MASK (~(PAGE_SIZE-1))
+
+int osdep_gnttab_open(xengnttab_handle *xgt)
+{
+ int fd = open(DEVXEN "gntdev", O_RDWR);
+ if ( fd == -1 )
+ return -1;
+ xgt->fd = fd;
+ return 0;
+}
+
+int osdep_gnttab_close(xengnttab_handle *xgt)
+{
+ if ( xgt->fd == -1 )
+ return 0;
+
+ return close(xgt->fd);
+}
+
+int osdep_gnttab_set_max_grants(xengnttab_handle *xgt, uint32_t count)
+{
+ int fd = xgt->fd, rc;
+ struct ioctl_gntdev_set_max_grants max_grants = { .count = count };
+
+ rc = ioctl(fd, IOCTL_GNTDEV_SET_MAX_GRANTS, &max_grants);
+ if (rc) {
+ /*
+ * Newer (e.g. pv-ops) kernels don't implement this IOCTL,
+ * so ignore the resulting specific failure.
+ */
+ if (errno == ENOTTY)
+ rc = 0;
+ else
+ GTERROR(xgt->logger, "ioctl SET_MAX_GRANTS failed");
+ }
+
+ return rc;
+}
+
+void *osdep_gnttab_grant_map(xengnttab_handle *xgt,
+ uint32_t count, int flags, int prot,
+ uint32_t *domids, uint32_t *refs,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ int fd = xgt->fd;
+ struct ioctl_gntdev_map_grant_ref *map;
+ unsigned int map_size = ROUNDUP((sizeof(*map) + (count - 1) *
+ sizeof(struct ioctl_gntdev_map_grant_ref)),
+ PAGE_SHIFT);
+ void *addr = NULL;
+ int domids_stride = 1;
+ int i;
+
+ if (flags & XENGNTTAB_GRANT_MAP_SINGLE_DOMAIN)
+ domids_stride = 0;
+
+ if ( map_size <= PAGE_SIZE )
+ map = alloca(sizeof(*map) +
+ (count - 1) * sizeof(struct ioctl_gntdev_map_grant_ref));
+ else
+ {
+ map = mmap(NULL, map_size, PROT_READ | PROT_WRITE,
+ MAP_PRIVATE | MAP_ANON | MAP_POPULATE, -1, 0);
+ if ( map == MAP_FAILED )
+ {
+ GTERROR(xgt->logger, "mmap of map failed");
+ return NULL;
+ }
+ }
+
+ for ( i = 0; i < count; i++ )
+ {
+ map->refs[i].domid = domids[i * domids_stride];
+ map->refs[i].ref = refs[i];
+ }
+
+ map->count = count;
+
+ if ( ioctl(fd, IOCTL_GNTDEV_MAP_GRANT_REF, map) ) {
+ GTERROR(xgt->logger, "ioctl MAP_GRANT_REF failed");
+ goto out;
+ }
+
+ retry:
+ addr = mmap(NULL, PAGE_SIZE * count, prot, MAP_SHARED, fd,
+ map->index);
+
+ if (addr == MAP_FAILED && errno == EAGAIN)
+ {
+ /*
+ * The grant hypercall can return EAGAIN if the granted page is
+ * swapped out. Since the paging daemon may be in the same domain, the
+ * hypercall cannot block without causing a deadlock.
+ *
+ * Because there are no notificaitons when the page is swapped in, wait
+ * a bit before retrying, and hope that the page will arrive eventually.
+ */
+ usleep(1000);
+ goto retry;
+ }
+
+ if (addr != MAP_FAILED)
+ {
+ int rv = 0;
+ struct ioctl_gntdev_unmap_notify notify;
+ notify.index = map->index;
+ notify.action = 0;
+ if (notify_offset < PAGE_SIZE * count) {
+ notify.index += notify_offset;
+ notify.action |= UNMAP_NOTIFY_CLEAR_BYTE;
+ }
+ if (notify_port != -1) {
+ notify.event_channel_port = notify_port;
+ notify.action |= UNMAP_NOTIFY_SEND_EVENT;
+ }
+ if (notify.action)
+ rv = ioctl(fd, IOCTL_GNTDEV_SET_UNMAP_NOTIFY, ¬ify);
+ if (rv) {
+ GTERROR(xgt->logger, "ioctl SET_UNMAP_NOTIFY failed");
+ munmap(addr, count * PAGE_SIZE);
+ addr = MAP_FAILED;
+ }
+ }
+
+ if (addr == MAP_FAILED)
+ {
+ int saved_errno = errno;
+ struct ioctl_gntdev_unmap_grant_ref unmap_grant;
+
+ /* Unmap the driver slots used to store the grant information. */
+ GTERROR(xgt->logger, "mmap failed");
+ unmap_grant.index = map->index;
+ unmap_grant.count = count;
+ ioctl(fd, IOCTL_GNTDEV_UNMAP_GRANT_REF, &unmap_grant);
+ errno = saved_errno;
+ addr = NULL;
+ }
+
+ out:
+ if ( map_size > PAGE_SIZE )
+ munmap(map, map_size);
+
+ return addr;
+}
+
+int osdep_gnttab_munmap(xengnttab_handle *xgt,
+ void *start_address,
+ uint32_t count)
+{
+ int fd = xgt->fd;
+ struct ioctl_gntdev_get_offset_for_vaddr get_offset;
+ struct ioctl_gntdev_unmap_grant_ref unmap_grant;
+ int rc;
+
+ if ( start_address == NULL )
+ {
+ errno = EINVAL;
+ return -1;
+ }
+
+ /* First, it is necessary to get the offset which was initially used to
+ * mmap() the pages.
+ */
+ get_offset.vaddr = (unsigned long)start_address;
+ if ( (rc = ioctl(fd, IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR,
+ &get_offset)) )
+ return rc;
+
+ if ( get_offset.count != count )
+ {
+ errno = EINVAL;
+ return -1;
+ }
+
+ /* Next, unmap the memory. */
+ if ( (rc = munmap(start_address, count * PAGE_SIZE)) )
+ return rc;
+
+ /* Finally, unmap the driver slots used to store the grant information. */
+ unmap_grant.index = get_offset.offset;
+ unmap_grant.count = count;
+ if ( (rc = ioctl(fd, IOCTL_GNTDEV_UNMAP_GRANT_REF, &unmap_grant)) )
+ return rc;
+
+ return 0;
+}
+
+int osdep_gntshr_open(xengntshr_handle *xgs)
+{
+ int fd = open(DEVXEN "gntalloc", O_RDWR);
+ if ( fd == -1 )
+ return -1;
+ xgs->fd = fd;
+ return 0;
+}
+
+int osdep_gntshr_close(xengntshr_handle *xgs)
+{
+ if ( xgs->fd == -1 )
+ return 0;
+
+ return close(xgs->fd);
+}
+
+void *osdep_gntshr_share_pages(xengntshr_handle *xgs,
+ uint32_t domid, int count,
+ uint32_t *refs, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ struct ioctl_gntalloc_alloc_gref *gref_info = NULL;
+ struct ioctl_gntalloc_unmap_notify notify;
+ struct ioctl_gntalloc_dealloc_gref gref_drop;
+ int fd = xgs->fd;
+ int err;
+ void *area = NULL;
+ gref_info = malloc(sizeof(*gref_info) + count * sizeof(uint32_t));
+ if (!gref_info)
+ return NULL;
+ gref_info->domid = domid;
+ gref_info->flags = writable ? GNTALLOC_FLAG_WRITABLE : 0;
+ gref_info->count = count;
+
+ err = ioctl(fd, IOCTL_GNTALLOC_ALLOC_GREF, gref_info);
+ if (err) {
+ GSERROR(xgs->logger, "ioctl failed");
+ goto out;
+ }
+
+ area = mmap(NULL, count * PAGE_SIZE, PROT_READ | PROT_WRITE,
+ MAP_SHARED, fd, gref_info->index);
+
+ if (area == MAP_FAILED) {
+ area = NULL;
+ GSERROR(xgs->logger, "mmap failed");
+ goto out_remove_fdmap;
+ }
+
+ notify.index = gref_info->index;
+ notify.action = 0;
+ if (notify_offset < PAGE_SIZE * count) {
+ notify.index += notify_offset;
+ notify.action |= UNMAP_NOTIFY_CLEAR_BYTE;
+ }
+ if (notify_port != -1) {
+ notify.event_channel_port = notify_port;
+ notify.action |= UNMAP_NOTIFY_SEND_EVENT;
+ }
+ if (notify.action)
+ err = ioctl(fd, IOCTL_GNTALLOC_SET_UNMAP_NOTIFY, ¬ify);
+ if (err) {
+ GSERROR(xgs->logger, "ioctl SET_UNMAP_NOTIFY failed");
+ munmap(area, count * PAGE_SIZE);
+ area = NULL;
+ }
+
+ memcpy(refs, gref_info->gref_ids, count * sizeof(uint32_t));
+
+ out_remove_fdmap:
+ /* Removing the mapping from the file descriptor does not cause the pages to
+ * be deallocated until the mapping is removed.
+ */
+ gref_drop.index = gref_info->index;
+ gref_drop.count = count;
+ ioctl(fd, IOCTL_GNTALLOC_DEALLOC_GREF, &gref_drop);
+ out:
+ free(gref_info);
+ return area;
+}
+
+int osdep_gntshr_munmap(xengntshr_handle *xgs,
+ void *start_address, uint32_t count)
+{
+ return munmap(start_address, count * PAGE_SIZE);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/gnttab/minios.c b/tools/libs/gnttab/minios.c
new file mode 100644
index 0000000..4b70d03
--- /dev/null
+++ b/tools/libs/gnttab/minios.c
@@ -0,0 +1,117 @@
+/*
+ *
+ * Copyright 2007-2008 Samuel Thibault <samuel.thibault@eu.citrix.com>.
+ * All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Splitfrom xc_minios.c
+ */
+
+#include <mini-os/types.h>
+#include <mini-os/os.h>
+#include <mini-os/lib.h>
+
+#include <mini-os/gntmap.h>
+#include <sys/mman.h>
+
+#include <errno.h>
+#include <unistd.h>
+
+#include "private.h"
+
+void minios_gnttab_close_fd(int fd);
+
+int osdep_gnttab_open(xengnttab_handle *xgt)
+{
+ int fd = alloc_fd(FTYPE_GNTMAP);
+ if ( fd == -1 )
+ return -1;
+ gntmap_init(&files[fd].gntmap);
+ xgt->fd = fd;
+ return 0;
+}
+
+int osdep_gnttab_close(xengnttab_handle *xgt)
+{
+ if ( xgt->fd == -1 )
+ return 0;
+
+ return close(xgt->fd);
+}
+
+void minios_gnttab_close_fd(int fd)
+{
+ gntmap_fini(&files[fd].gntmap);
+ files[fd].type = FTYPE_NONE;
+}
+
+void *osdep_gnttab_grant_map(xengnttab_handle *xgt,
+ uint32_t count, int flags, int prot,
+ uint32_t *domids, uint32_t *refs,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ int fd = xgt->fd;
+ int stride = 1;
+ if (flags & XENGNTTAB_GRANT_MAP_SINGLE_DOMAIN)
+ stride = 0;
+ if (notify_offset != -1 || notify_port != -1) {
+ errno = ENOSYS;
+ return NULL;
+ }
+ return gntmap_map_grant_refs(&files[fd].gntmap,
+ count, domids, stride,
+ refs, prot & PROT_WRITE);
+}
+
+int osdep_gnttab_munmap(xengnttab_handle *xgt,
+ void *start_address,
+ uint32_t count)
+{
+ int fd = xgt->fd;
+ int ret;
+ ret = gntmap_munmap(&files[fd].gntmap,
+ (unsigned long) start_address,
+ count);
+ if (ret < 0) {
+ errno = -ret;
+ return -1;
+ }
+ return ret;
+}
+
+int osdep_gnttab_set_max_grants(xengnttab_handle *xgt, uint32_t count)
+{
+ int fd = xgt->fd;
+ int ret;
+ ret = gntmap_set_max_grants(&files[fd].gntmap,
+ count);
+ if (ret < 0) {
+ errno = -ret;
+ return -1;
+ }
+ return ret;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/gnttab/private.h b/tools/libs/gnttab/private.h
new file mode 100644
index 0000000..cb7e2cc
--- /dev/null
+++ b/tools/libs/gnttab/private.h
@@ -0,0 +1,47 @@
+#ifndef XENGNTTAB_PRIVATE_H
+#define XENGNTTAB_PRIVATE_H
+
+#include <xentoollog.h>
+#include <xengnttab.h>
+
+struct xengntdev_handle {
+ xentoollog_logger *logger, *logger_tofree;
+ int fd;
+};
+
+int osdep_gnttab_open(xengnttab_handle *xgt);
+int osdep_gnttab_close(xengnttab_handle *xgt);
+
+int osdep_gnttab_set_max_grants(xengnttab_handle *xgt, uint32_t count);
+
+#define XENGNTTAB_GRANT_MAP_SINGLE_DOMAIN 0x1
+void *osdep_gnttab_grant_map(xengnttab_handle *xgt,
+ uint32_t count, int flags, int prot,
+ uint32_t *domids, uint32_t *refs,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port);
+int osdep_gnttab_munmap(xengnttab_handle *xgt,
+ void *start_address,
+ uint32_t count);
+int osdep_gntshr_open(xengntshr_handle *xgs);
+int osdep_gntshr_close(xengntshr_handle *xgs);
+
+void *osdep_gntshr_share_pages(xengntshr_handle *xgs,
+ uint32_t domid, int count,
+ uint32_t *refs, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port);
+int osdep_gntshr_munmap(xengntshr_handle *xgs,
+ void *start_address, uint32_t count);
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libvchan/Makefile b/tools/libvchan/Makefile
index 84128a3..0573d2f 100644
--- a/tools/libvchan/Makefile
+++ b/tools/libvchan/Makefile
@@ -10,15 +10,17 @@ NODE_OBJS = node.o
NODE2_OBJS = node-select.o
LIBVCHAN_PIC_OBJS = $(patsubst %.o,%.opic,$(LIBVCHAN_OBJS))
-LIBVCHAN_LIBS = $(LDLIBS_libxenstore) $(LDLIBS_libxenctrl) $(LDLIBS_libxenevtchn)
-$(LIBVCHAN_OBJS) $(LIBVCHAN_PIC_OBJS): CFLAGS += $(CFLAGS_libxenstore) $(CFLAGS_libxenctrl) $(CFLAGS_libxenevtchn)
-$(NODE_OBJS) $(NODE2_OBJS): CFLAGS += $(CFLAGS_libxenctrl) $(CFLAGS_libxenevtchn)
+LIBVCHAN_LIBS = $(LDLIBS_libxenstore) $(LDLIBS_libxengnttab) $(LDLIBS_libxengntshr) $(LDLIBS_libxenevtchn)
+$(LIBVCHAN_OBJS) $(LIBVCHAN_PIC_OBJS): CFLAGS += $(CFLAGS_libxenstore) $(CFLAGS_libxengnttab) $(CFLAGS_libxengntshr) $(CFLAGS_libxenevtchn)
+$(NODE_OBJS) $(NODE2_OBJS): CFLAGS += $(CFLAGS_libxengnttab) $(CFLAGS_libxengntshr) $(CFLAGS_libxenevtchn)
MAJOR = 1.0
MINOR = 0
CFLAGS += -I../include -I.
+io.o io.opic: CFLAGS += $(CFLAGS_libxenctrl) # for xen_mb et al
+
.PHONY: all
all: libxenvchan.so vchan-node1 vchan-node2 libxenvchan.a
diff --git a/tools/libvchan/init.c b/tools/libvchan/init.c
index 66cb103..18cbbf4 100644
--- a/tools/libvchan/init.c
+++ b/tools/libvchan/init.c
@@ -78,7 +78,7 @@ static int init_gnt_srv(struct libxenvchan *ctrl, int domain)
uint32_t ring_ref = -1;
void *ring;
- ring = xc_gntshr_share_page_notify(ctrl->gntshr, domain,
+ ring = xengntshr_share_page_notify(ctrl->gntshr, domain,
&ring_ref, 1, offsetof(struct vchan_interface, srv_live),
ctrl->event_port);
@@ -104,7 +104,7 @@ static int init_gnt_srv(struct libxenvchan *ctrl, int domain)
ctrl->read.buffer = ((void*)ctrl->ring) + LARGE_RING_OFFSET;
break;
default:
- ctrl->read.buffer = xc_gntshr_share_pages(ctrl->gntshr, domain,
+ ctrl->read.buffer = xengntshr_share_pages(ctrl->gntshr, domain,
pages_left, ctrl->ring->grants, 1);
if (!ctrl->read.buffer)
goto out_ring;
@@ -118,7 +118,7 @@ static int init_gnt_srv(struct libxenvchan *ctrl, int domain)
ctrl->write.buffer = ((void*)ctrl->ring) + LARGE_RING_OFFSET;
break;
default:
- ctrl->write.buffer = xc_gntshr_share_pages(ctrl->gntshr, domain,
+ ctrl->write.buffer = xengntshr_share_pages(ctrl->gntshr, domain,
pages_right, ctrl->ring->grants + pages_left, 1);
if (!ctrl->write.buffer)
goto out_unmap_left;
@@ -128,9 +128,9 @@ out:
return ring_ref;
out_unmap_left:
if (pages_left)
- xc_gntshr_munmap(ctrl->gntshr, ctrl->read.buffer, pages_left);
+ xengntshr_munmap(ctrl->gntshr, ctrl->read.buffer, pages_left);
out_ring:
- xc_gntshr_munmap(ctrl->gntshr, ring, 1);
+ xengntshr_munmap(ctrl->gntshr, ring, 1);
ring_ref = -1;
ctrl->ring = NULL;
ctrl->write.order = ctrl->read.order = 0;
@@ -142,7 +142,7 @@ static int init_gnt_cli(struct libxenvchan *ctrl, int domain, uint32_t ring_ref)
int rv = -1;
uint32_t *grants;
- ctrl->ring = xc_gnttab_map_grant_ref_notify(ctrl->gnttab,
+ ctrl->ring = xengnttab_map_grant_ref_notify(ctrl->gnttab,
domain, ring_ref, PROT_READ|PROT_WRITE,
offsetof(struct vchan_interface, cli_live), ctrl->event_port);
@@ -172,7 +172,7 @@ static int init_gnt_cli(struct libxenvchan *ctrl, int domain, uint32_t ring_ref)
default:
{
int pages_left = 1 << (ctrl->write.order - PAGE_SHIFT);
- ctrl->write.buffer = xc_gnttab_map_domain_grant_refs(ctrl->gnttab,
+ ctrl->write.buffer = xengnttab_map_domain_grant_refs(ctrl->gnttab,
pages_left, domain, grants, PROT_READ|PROT_WRITE);
if (!ctrl->write.buffer)
goto out_unmap_ring;
@@ -190,7 +190,7 @@ static int init_gnt_cli(struct libxenvchan *ctrl, int domain, uint32_t ring_ref)
default:
{
int pages_right = 1 << (ctrl->read.order - PAGE_SHIFT);
- ctrl->read.buffer = xc_gnttab_map_domain_grant_refs(ctrl->gnttab,
+ ctrl->read.buffer = xengnttab_map_domain_grant_refs(ctrl->gnttab,
pages_right, domain, grants, PROT_READ);
if (!ctrl->read.buffer)
goto out_unmap_left;
@@ -202,10 +202,10 @@ static int init_gnt_cli(struct libxenvchan *ctrl, int domain, uint32_t ring_ref)
return rv;
out_unmap_left:
if (ctrl->write.order >= PAGE_SHIFT)
- xc_gnttab_munmap(ctrl->gnttab, ctrl->write.buffer,
+ xengnttab_munmap(ctrl->gnttab, ctrl->write.buffer,
1 << (ctrl->write.order - PAGE_SHIFT));
out_unmap_ring:
- xc_gnttab_munmap(ctrl->gnttab, ctrl->ring, 1);
+ xengnttab_munmap(ctrl->gnttab, ctrl->ring, 1);
ctrl->ring = 0;
ctrl->write.order = ctrl->read.order = 0;
rv = -1;
@@ -325,7 +325,7 @@ struct libxenvchan *libxenvchan_server_init(xentoollog_logger *logger, int domai
ctrl->write.order = LARGE_RING_SHIFT;
}
- ctrl->gntshr = xc_gntshr_open(logger, 0);
+ ctrl->gntshr = xengntshr_open(logger, 0);
if (!ctrl->gntshr)
goto out;
@@ -413,7 +413,7 @@ struct libxenvchan *libxenvchan_client_init(xentoollog_logger *logger, int domai
if (!ctrl->event_port)
goto fail;
- ctrl->gnttab = xc_gnttab_open(logger, 0);
+ ctrl->gnttab = xengnttab_open(logger, 0);
if (!ctrl->gnttab)
goto fail;
diff --git a/tools/libvchan/io.c b/tools/libvchan/io.c
index 4461490..bfffd73 100644
--- a/tools/libvchan/io.c
+++ b/tools/libvchan/io.c
@@ -364,10 +364,10 @@ void libxenvchan_close(struct libxenvchan *ctrl)
if (ctrl->ring) {
if (ctrl->is_server) {
ctrl->ring->srv_live = 0;
- xc_gntshr_munmap(ctrl->gntshr, ctrl->ring, 1);
+ xengntshr_munmap(ctrl->gntshr, ctrl->ring, 1);
} else {
ctrl->ring->cli_live = 0;
- xc_gnttab_munmap(ctrl->gnttab, ctrl->ring, 1);
+ xengnttab_munmap(ctrl->gnttab, ctrl->ring, 1);
}
}
if (ctrl->event) {
@@ -377,10 +377,10 @@ void libxenvchan_close(struct libxenvchan *ctrl)
}
if (ctrl->is_server) {
if (ctrl->gntshr)
- xc_gntshr_close(ctrl->gntshr);
+ xengntshr_close(ctrl->gntshr);
} else {
if (ctrl->gnttab)
- xc_gnttab_close(ctrl->gnttab);
+ xengnttab_close(ctrl->gnttab);
}
free(ctrl);
}
diff --git a/tools/libvchan/libxenvchan.h b/tools/libvchan/libxenvchan.h
index 1544378..341c375 100644
--- a/tools/libvchan/libxenvchan.h
+++ b/tools/libvchan/libxenvchan.h
@@ -45,7 +45,7 @@
#include <xen/io/libxenvchan.h>
#include <xen/sys/evtchn.h>
#include <xenevtchn.h>
-#include <xenctrl.h>
+#include <xengnttab.h>
struct libxenvchan_ring {
/* Pointer into the shared page. Offsets into buffer. */
@@ -66,8 +66,8 @@ struct libxenvchan_ring {
struct libxenvchan {
/* Mapping handle for shared ring page */
union {
- xc_gntshr *gntshr; /* for server */
- xc_gnttab *gnttab; /* for client */
+ xengntshr_handle *gntshr; /* for server */
+ xengnttab_handle *gnttab; /* for client */
};
/* Pointer to shared ring page */
struct vchan_interface *ring;
diff --git a/tools/libxc/Makefile b/tools/libxc/Makefile
index 184cbb7..33d18db 100644
--- a/tools/libxc/Makefile
+++ b/tools/libxc/Makefile
@@ -43,12 +43,13 @@ CTRL_SRCS-y += xc_resource.c
CTRL_SRCS-$(CONFIG_X86) += xc_psr.c
CTRL_SRCS-$(CONFIG_X86) += xc_pagetab.c
CTRL_SRCS-$(CONFIG_Linux) += xc_linux.c xc_linux_osdep.c
-CTRL_SRCS-$(CONFIG_FreeBSD) += xc_freebsd.c xc_freebsd_osdep.c xc_nognttab.c xc_nogntshr.c
-CTRL_SRCS-$(CONFIG_SunOS) += xc_solaris.c xc_nognttab.c xc_nogntshr.c
-CTRL_SRCS-$(CONFIG_NetBSD) += xc_netbsd.c xc_nognttab.c xc_nogntshr.c
-CTRL_SRCS-$(CONFIG_NetBSDRump) += xc_netbsd.c xc_nognttab.c xc_nogntshr.c
-CTRL_SRCS-$(CONFIG_MiniOS) += xc_minios.c xc_nogntshr.c
+CTRL_SRCS-$(CONFIG_FreeBSD) += xc_freebsd.c xc_freebsd_osdep.c
+CTRL_SRCS-$(CONFIG_SunOS) += xc_solaris.c
+CTRL_SRCS-$(CONFIG_NetBSD) += xc_netbsd.c
+CTRL_SRCS-$(CONFIG_NetBSDRump) += xc_netbsd.c
+CTRL_SRCS-$(CONFIG_MiniOS) += xc_minios.c
CTRL_SRCS-y += xc_evtchn_compat.c
+CTRL_SRCS-y += xc_gnttab_compat.c
GUEST_SRCS-y :=
GUEST_SRCS-y += xg_private.c xc_suspend.c
@@ -126,6 +127,8 @@ OSDEP_PIC_OBJS := $(patsubst %.c,%.opic,$(OSDEP_SRCS-y))
$(CTRL_LIB_OBJS) $(GUEST_LIB_OBJS) $(OSDEP_LIB_OBJS) \
$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS) $(OSDEP_PIC_OBJS) : CFLAGS += -include $(XEN_ROOT)/tools/config.h
+$(CTRL_LIB_OBJS) $(CTRL_PIC_OBJS): CFLAGS += $(CFLAGS_libxengnttab) $(CFLAGS_libxengntshr)
+
LIB := libxenctrl.a
ifneq ($(nosharedlibs),y)
LIB += libxenctrl.so libxenctrl.so.$(MAJOR) libxenctrl.so.$(MAJOR).$(MINOR)
@@ -209,7 +212,7 @@ libxenctrl.so.$(MAJOR): libxenctrl.so.$(MAJOR).$(MINOR)
$(SYMLINK_SHLIB) $< $@
libxenctrl.so.$(MAJOR).$(MINOR): $(CTRL_PIC_OBJS)
- $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(DLOPEN_LIBS) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_libxengntshr) $(DLOPEN_LIBS) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
# libxenguest
diff --git a/tools/libxc/include/xenctrl.h b/tools/libxc/include/xenctrl.h
index acc4af1..bfc8592 100644
--- a/tools/libxc/include/xenctrl.h
+++ b/tools/libxc/include/xenctrl.h
@@ -5,9 +5,6 @@
*
* Copyright (c) 2003-2004, K A Fraser.
*
- * xc_gnttab functions:
- * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
- *
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation;
@@ -117,8 +114,6 @@
*/
typedef struct xc_interface_core xc_interface;
-typedef struct xengntdev_handle xc_gnttab;
-typedef struct xengntdev_handle xc_gntshr;
enum xc_error_code {
XC_ERROR_NONE = 0,
@@ -1556,116 +1551,6 @@ int xc_domain_subscribe_for_suspend(
* These functions sometimes log messages as above, but not always.
*/
-/*
- * Note:
- * After fork a child process must not use any opened xc gnttab
- * handle inherited from their parent. They must open a new handle if
- * they want to interact with xc.
- *
- * Return an fd onto the grant table driver. Logs errors.
- */
-xc_gnttab *xc_gnttab_open(xentoollog_logger *logger,
- unsigned open_flags);
-
-/*
- * Close a handle previously allocated with xc_gnttab_open().
- * Never logs errors.
- */
-int xc_gnttab_close(xc_gnttab *xcg);
-
-/*
- * Memory maps a grant reference from one domain to a local address range.
- * Mappings should be unmapped with xc_gnttab_munmap. Logs errors.
- *
- * @parm xcg a handle on an open grant table interface
- * @parm domid the domain to map memory from
- * @parm ref the grant reference ID to map
- * @parm prot same flag as in mmap()
- */
-void *xc_gnttab_map_grant_ref(xc_gnttab *xcg,
- uint32_t domid,
- uint32_t ref,
- int prot);
-
-/**
- * Memory maps one or more grant references from one or more domains to a
- * contiguous local address range. Mappings should be unmapped with
- * xc_gnttab_munmap. Logs errors.
- *
- * @parm xcg a handle on an open grant table interface
- * @parm count the number of grant references to be mapped
- * @parm domids an array of @count domain IDs by which the corresponding @refs
- * were granted
- * @parm refs an array of @count grant references to be mapped
- * @parm prot same flag as in mmap()
- */
-void *xc_gnttab_map_grant_refs(xc_gnttab *xcg,
- uint32_t count,
- uint32_t *domids,
- uint32_t *refs,
- int prot);
-
-/**
- * Memory maps one or more grant references from one domain to a
- * contiguous local address range. Mappings should be unmapped with
- * xc_gnttab_munmap. Logs errors.
- *
- * @parm xcg a handle on an open grant table interface
- * @parm count the number of grant references to be mapped
- * @parm domid the domain to map memory from
- * @parm refs an array of @count grant references to be mapped
- * @parm prot same flag as in mmap()
- */
-void *xc_gnttab_map_domain_grant_refs(xc_gnttab *xcg,
- uint32_t count,
- uint32_t domid,
- uint32_t *refs,
- int prot);
-
-/**
- * Memory maps a grant reference from one domain to a local address range.
- * Mappings should be unmapped with xc_gnttab_munmap. If notify_offset or
- * notify_port are not -1, this version will attempt to set up an unmap
- * notification at the given offset and event channel. When the page is
- * unmapped, the byte at the given offset will be zeroed and a wakeup will be
- * sent to the given event channel. Logs errors.
- *
- * @parm xcg a handle on an open grant table interface
- * @parm domid the domain to map memory from
- * @parm ref the grant reference ID to map
- * @parm prot same flag as in mmap()
- * @parm notify_offset The byte offset in the page to use for unmap
- * notification; -1 for none.
- * @parm notify_port The event channel port to use for unmap notify, or -1
- */
-void *xc_gnttab_map_grant_ref_notify(xc_gnttab *xcg,
- uint32_t domid,
- uint32_t ref,
- int prot,
- uint32_t notify_offset,
- evtchn_port_t notify_port);
-
-/*
- * Unmaps the @count pages starting at @start_address, which were mapped by a
- * call to xc_gnttab_map_grant_ref or xc_gnttab_map_grant_refs. Never logs.
- */
-int xc_gnttab_munmap(xc_gnttab *xcg,
- void *start_address,
- uint32_t count);
-
-/*
- * Sets the maximum number of grants that may be mapped by the given instance
- * to @count. Never logs.
- *
- * N.B. This function must be called after opening the handle, and before any
- * other functions are invoked on it.
- *
- * N.B. When variable-length grants are mapped, fragmentation may be observed,
- * and it may not be possible to satisfy requests up to the maximum number
- * of grants.
- */
-int xc_gnttab_set_max_grants(xc_gnttab *xcg,
- uint32_t count);
int xc_gnttab_op(xc_interface *xch, int cmd,
void * op, int op_size, int count);
@@ -1676,59 +1561,6 @@ grant_entry_v1_t *xc_gnttab_map_table_v1(xc_interface *xch, int domid, int *gnt_
grant_entry_v2_t *xc_gnttab_map_table_v2(xc_interface *xch, int domid, int *gnt_num);
/* Sometimes these don't set errno [fixme], and sometimes they don't log. */
-/*
- * Return an fd onto the grant sharing driver. Logs errors.
- *
- * Note:
- * After fork a child process must not use any opened xc gntshr
- * handle inherited from their parent. They must open a new handle if
- * they want to interact with xc.
- *
- */
-xc_gntshr *xc_gntshr_open(xentoollog_logger *logger,
- unsigned open_flags);
-
-/*
- * Close a handle previously allocated with xc_gntshr_open().
- * Never logs errors.
- */
-int xc_gntshr_close(xc_gntshr *xcg);
-
-/*
- * Creates and shares pages with another domain.
- *
- * @parm xcg a handle to an open grant sharing instance
- * @parm domid the domain to share memory with
- * @parm count the number of pages to share
- * @parm refs the grant references of the pages (output)
- * @parm writable true if the other domain can write to the pages
- * @return local mapping of the pages
- */
-void *xc_gntshr_share_pages(xc_gntshr *xcg, uint32_t domid,
- int count, uint32_t *refs, int writable);
-
-/*
- * Creates and shares a page with another domain, with unmap notification.
- *
- * @parm xcg a handle to an open grant sharing instance
- * @parm domid the domain to share memory with
- * @parm refs the grant reference of the pages (output)
- * @parm writable true if the other domain can write to the page
- * @parm notify_offset The byte offset in the page to use for unmap
- * notification; -1 for none.
- * @parm notify_port The event channel port to use for unmap notify, or -1
- * @return local mapping of the page
- */
-void *xc_gntshr_share_page_notify(xc_gntshr *xcg, uint32_t domid,
- uint32_t *ref, int writable,
- uint32_t notify_offset,
- evtchn_port_t notify_port);
-/*
- * Unmaps the @count pages starting at @start_address, which were mapped by a
- * call to xc_gntshr_share_*. Never logs.
- */
-int xc_gntshr_munmap(xc_gntshr *xcg, void *start_address, uint32_t count);
-
int xc_physdev_map_pirq(xc_interface *xch,
int domid,
int index,
diff --git a/tools/libxc/include/xenctrl_compat.h b/tools/libxc/include/xenctrl_compat.h
index 48daeb2..d99fa11 100644
--- a/tools/libxc/include/xenctrl_compat.h
+++ b/tools/libxc/include/xenctrl_compat.h
@@ -35,6 +35,54 @@ int xc_evtchn_unmask(xc_evtchn *xce, evtchn_port_t port);
#endif /* XC_WANT_COMPAT_EVTCHN_API */
+#ifdef XC_WANT_COMPAT_GNTTAB_API
+
+typedef struct xengntdev_handle xc_gnttab;
+
+xc_gnttab *xc_gnttab_open(xentoollog_logger *logger,
+ unsigned open_flags);
+int xc_gnttab_close(xc_gnttab *xcg);
+void *xc_gnttab_map_grant_ref(xc_gnttab *xcg,
+ uint32_t domid,
+ uint32_t ref,
+ int prot);
+void *xc_gnttab_map_grant_refs(xc_gnttab *xcg,
+ uint32_t count,
+ uint32_t *domids,
+ uint32_t *refs,
+ int prot);
+void *xc_gnttab_map_domain_grant_refs(xc_gnttab *xcg,
+ uint32_t count,
+ uint32_t domid,
+ uint32_t *refs,
+ int prot);
+void *xc_gnttab_map_grant_ref_notify(xc_gnttab *xcg,
+ uint32_t domid,
+ uint32_t ref,
+ int prot,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port);
+int xc_gnttab_munmap(xc_gnttab *xcg,
+ void *start_address,
+ uint32_t count);
+int xc_gnttab_set_max_grants(xc_gnttab *xcg,
+ uint32_t count);
+
+typedef struct xengntdev_handle xc_gntshr;
+
+xc_gntshr *xc_gntshr_open(xentoollog_logger *logger,
+ unsigned open_flags);
+int xc_gntshr_close(xc_gntshr *xcg);
+void *xc_gntshr_share_pages(xc_gntshr *xcg, uint32_t domid,
+ int count, uint32_t *refs, int writable);
+void *xc_gntshr_share_page_notify(xc_gntshr *xcg, uint32_t domid,
+ uint32_t *ref, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port);
+int xc_gntshr_munmap(xc_gntshr *xcg, void *start_address, uint32_t count);
+
+#endif /* XC_WANT_COMPAT_GNTTAB_API */
+
#endif
/*
diff --git a/tools/libxc/xc_gnttab.c b/tools/libxc/xc_gnttab.c
index a51f405..dd32aa2 100644
--- a/tools/libxc/xc_gnttab.c
+++ b/tools/libxc/xc_gnttab.c
@@ -143,59 +143,6 @@ grant_entry_v2_t *xc_gnttab_map_table_v2(xc_interface *xch, int domid,
return _gnttab_map_table(xch, domid, gnt_num);
}
-void *xc_gnttab_map_grant_ref(xc_gnttab *xgt,
- uint32_t domid,
- uint32_t ref,
- int prot)
-{
- return osdep_gnttab_grant_map(xgt, 1, 0, prot, &domid, &ref, -1, -1);
-}
-
-void *xc_gnttab_map_grant_refs(xc_gnttab *xgt,
- uint32_t count,
- uint32_t *domids,
- uint32_t *refs,
- int prot)
-{
- return osdep_gnttab_grant_map(xgt, count, 0, prot, domids, refs, -1, -1);
-}
-
-void *xc_gnttab_map_domain_grant_refs(xc_gnttab *xgt,
- uint32_t count,
- uint32_t domid,
- uint32_t *refs,
- int prot)
-{
- return osdep_gnttab_grant_map(xgt, count, XC_GRANT_MAP_SINGLE_DOMAIN,
- prot, &domid, refs, -1, -1);
-}
-
-void *xc_gnttab_map_grant_ref_notify(xc_gnttab *xgt,
- uint32_t domid,
- uint32_t ref,
- int prot,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
-{
- return osdep_gnttab_grant_map(xgt, 1, 0, prot, &domid, &ref,
- notify_offset, notify_port);
-}
-
-void *xc_gntshr_share_pages(xc_gntshr *xcg, uint32_t domid,
- int count, uint32_t *refs, int writable)
-{
- return osdep_gntshr_share_pages(xcg, domid, count, refs, writable, -1, -1);
-}
-
-void *xc_gntshr_share_page_notify(xc_gntshr *xcg, uint32_t domid,
- uint32_t *ref, int writable,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
-{
- return osdep_gntshr_share_pages(xcg, domid, 1, ref, writable,
- notify_offset, notify_port);
-}
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xc_gnttab_compat.c b/tools/libxc/xc_gnttab_compat.c
new file mode 100644
index 0000000..1f496a1
--- /dev/null
+++ b/tools/libxc/xc_gnttab_compat.c
@@ -0,0 +1,111 @@
+/*
+ * Compat shims for use of 3rd party consumers of libxenctrl xc_gnt{tab,shr}
+ * functionality which has been split into separate libraries.
+ */
+
+#include <xengnttab.h>
+
+#define XC_WANT_COMPAT_GNTTAB_API
+#include "xenctrl.h"
+
+xc_gnttab *xc_gnttab_open(xentoollog_logger *logger,
+ unsigned open_flags)
+{
+ return xengnttab_open(logger, open_flags);
+}
+
+int xc_gnttab_close(xc_gnttab *xcg)
+{
+ return xengnttab_close(xcg);
+}
+
+void *xc_gnttab_map_grant_ref(xc_gnttab *xcg,
+ uint32_t domid,
+ uint32_t ref,
+ int prot)
+{
+ return xengnttab_map_grant_ref(xcg, domid, ref, prot);
+}
+
+void *xc_gnttab_map_grant_refs(xc_gnttab *xcg,
+ uint32_t count,
+ uint32_t *domids,
+ uint32_t *refs,
+ int prot)
+{
+ return xengnttab_map_grant_refs(xcg, count, domids, refs, prot);
+}
+
+void *xc_gnttab_map_domain_grant_refs(xc_gnttab *xcg,
+ uint32_t count,
+ uint32_t domid,
+ uint32_t *refs,
+ int prot)
+{
+ return xengnttab_map_domain_grant_refs(xcg, count, domid, refs, prot);
+}
+
+void *xc_gnttab_map_grant_ref_notify(xc_gnttab *xcg,
+ uint32_t domid,
+ uint32_t ref,
+ int prot,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ return xengnttab_map_grant_ref_notify(xcg, domid, ref, prot,
+ notify_offset, notify_port);
+}
+
+int xc_gnttab_munmap(xc_gnttab *xcg,
+ void *start_address,
+ uint32_t count)
+{
+ return xengnttab_munmap(xcg, start_address, count);
+}
+
+int xc_gnttab_set_max_grants(xc_gnttab *xcg,
+ uint32_t count)
+{
+ return xengnttab_set_max_grants(xcg, count);
+}
+
+xc_gntshr *xc_gntshr_open(xentoollog_logger *logger,
+ unsigned open_flags)
+{
+ return xengntshr_open(logger, open_flags);
+}
+
+int xc_gntshr_close(xc_gntshr *xcg)
+{
+ return xengntshr_close(xcg);
+}
+
+void *xc_gntshr_share_pages(xc_gntshr *xcg, uint32_t domid,
+ int count, uint32_t *refs, int writable)
+{
+ return xengntshr_share_pages(xcg, domid, count, refs, writable);
+}
+
+void *xc_gntshr_share_page_notify(xc_gntshr *xcg, uint32_t domid,
+ uint32_t *ref, int writable,
+ uint32_t notify_offset,
+ evtchn_port_t notify_port)
+{
+ return xengntshr_share_page_notify(xcg, domid, ref, writable,
+ notify_offset, notify_port);
+}
+
+int xc_gntshr_munmap(xc_gntshr *xcg, void *start_address, uint32_t count)
+{
+ return xengntshr_munmap(xcg, start_address, count);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index 685852f..96903f7 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -31,8 +31,6 @@
#include <sys/ioctl.h>
#include <xen/memory.h>
-#include <xen/sys/gntdev.h>
-#include <xen/sys/gntalloc.h>
#include "xenctrl.h"
#include "xenctrlosdep.h"
@@ -41,9 +39,6 @@
#define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & ~((1UL<<(_w))-1))
-#define GTERROR(_l, _f...) xtl_log(_l, XTL_ERROR, errno, "gnttab", _f)
-#define GSERROR(_l, _f...) xtl_log(_l, XTL_ERROR, errno, "gntshr", _f)
-
static xc_osdep_handle linux_privcmd_open(xc_interface *xch)
{
int flags, saved_errno;
@@ -455,281 +450,6 @@ static struct xc_osdep_ops linux_privcmd_ops = {
},
};
-#define DEVXEN "/dev/xen/"
-
-int osdep_gnttab_open(xc_gnttab *xgt)
-{
- int fd = open(DEVXEN "gntdev", O_RDWR);
- if ( fd == -1 )
- return -1;
- xgt->fd = fd;
- return 0;
-}
-
-int osdep_gnttab_close(xc_gnttab *xgt)
-{
- if ( xgt->fd == -1 )
- return 0;
-
- return close(xgt->fd);
-}
-
-int xc_gnttab_set_max_grants(xc_gnttab *xgt, uint32_t count)
-{
- int fd = xgt->fd, rc;
- struct ioctl_gntdev_set_max_grants max_grants = { .count = count };
-
- rc = ioctl(fd, IOCTL_GNTDEV_SET_MAX_GRANTS, &max_grants);
- if (rc) {
- /*
- * Newer (e.g. pv-ops) kernels don't implement this IOCTL,
- * so ignore the resulting specific failure.
- */
- if (errno == ENOTTY)
- rc = 0;
- else
- GTERROR(xgt->logger, "ioctl SET_MAX_GRANTS failed");
- }
-
- return rc;
-}
-
-void *osdep_gnttab_grant_map(xc_gnttab *xgt,
- uint32_t count, int flags, int prot,
- uint32_t *domids, uint32_t *refs,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
-{
- int fd = xgt->fd;
- struct ioctl_gntdev_map_grant_ref *map;
- unsigned int map_size = ROUNDUP((sizeof(*map) + (count - 1) *
- sizeof(struct ioctl_gntdev_map_grant_ref)),
- XC_PAGE_SHIFT);
- void *addr = NULL;
- int domids_stride = 1;
- int i;
-
- if (flags & XC_GRANT_MAP_SINGLE_DOMAIN)
- domids_stride = 0;
-
- if ( map_size <= XC_PAGE_SIZE )
- map = alloca(sizeof(*map) +
- (count - 1) * sizeof(struct ioctl_gntdev_map_grant_ref));
- else
- {
- map = mmap(NULL, map_size, PROT_READ | PROT_WRITE,
- MAP_PRIVATE | MAP_ANON | MAP_POPULATE, -1, 0);
- if ( map == MAP_FAILED )
- {
- GTERROR(xgt->logger, "mmap of map failed");
- return NULL;
- }
- }
-
- for ( i = 0; i < count; i++ )
- {
- map->refs[i].domid = domids[i * domids_stride];
- map->refs[i].ref = refs[i];
- }
-
- map->count = count;
-
- if ( ioctl(fd, IOCTL_GNTDEV_MAP_GRANT_REF, map) ) {
- GTERROR(xgt->logger, "ioctl MAP_GRANT_REF failed");
- goto out;
- }
-
- retry:
- addr = mmap(NULL, XC_PAGE_SIZE * count, prot, MAP_SHARED, fd,
- map->index);
-
- if (addr == MAP_FAILED && errno == EAGAIN)
- {
- /*
- * The grant hypercall can return EAGAIN if the granted page is
- * swapped out. Since the paging daemon may be in the same domain, the
- * hypercall cannot block without causing a deadlock.
- *
- * Because there are no notificaitons when the page is swapped in, wait
- * a bit before retrying, and hope that the page will arrive eventually.
- */
- usleep(1000);
- goto retry;
- }
-
- if (addr != MAP_FAILED)
- {
- int rv = 0;
- struct ioctl_gntdev_unmap_notify notify;
- notify.index = map->index;
- notify.action = 0;
- if (notify_offset < XC_PAGE_SIZE * count) {
- notify.index += notify_offset;
- notify.action |= UNMAP_NOTIFY_CLEAR_BYTE;
- }
- if (notify_port != -1) {
- notify.event_channel_port = notify_port;
- notify.action |= UNMAP_NOTIFY_SEND_EVENT;
- }
- if (notify.action)
- rv = ioctl(fd, IOCTL_GNTDEV_SET_UNMAP_NOTIFY, ¬ify);
- if (rv) {
- GTERROR(xgt->logger, "ioctl SET_UNMAP_NOTIFY failed");
- munmap(addr, count * XC_PAGE_SIZE);
- addr = MAP_FAILED;
- }
- }
-
- if (addr == MAP_FAILED)
- {
- int saved_errno = errno;
- struct ioctl_gntdev_unmap_grant_ref unmap_grant;
-
- /* Unmap the driver slots used to store the grant information. */
- GTERROR(xgt->logger, "mmap failed");
- unmap_grant.index = map->index;
- unmap_grant.count = count;
- ioctl(fd, IOCTL_GNTDEV_UNMAP_GRANT_REF, &unmap_grant);
- errno = saved_errno;
- addr = NULL;
- }
-
- out:
- if ( map_size > XC_PAGE_SIZE )
- munmap(map, map_size);
-
- return addr;
-}
-
-int xc_gnttab_munmap(xc_gnttab *xgt, void *start_address, uint32_t count)
-{
- int fd = xgt->fd;
- struct ioctl_gntdev_get_offset_for_vaddr get_offset;
- struct ioctl_gntdev_unmap_grant_ref unmap_grant;
- int rc;
-
- if ( start_address == NULL )
- {
- errno = EINVAL;
- return -1;
- }
-
- /* First, it is necessary to get the offset which was initially used to
- * mmap() the pages.
- */
- get_offset.vaddr = (unsigned long)start_address;
- if ( (rc = ioctl(fd, IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR,
- &get_offset)) )
- return rc;
-
- if ( get_offset.count != count )
- {
- errno = EINVAL;
- return -1;
- }
-
- /* Next, unmap the memory. */
- if ( (rc = munmap(start_address, count * XC_PAGE_SIZE)) )
- return rc;
-
- /* Finally, unmap the driver slots used to store the grant information. */
- unmap_grant.index = get_offset.offset;
- unmap_grant.count = count;
- if ( (rc = ioctl(fd, IOCTL_GNTDEV_UNMAP_GRANT_REF, &unmap_grant)) )
- return rc;
-
- return 0;
-}
-
-int osdep_gntshr_open(xc_gntshr *xgs)
-{
- int fd = open(DEVXEN "gntalloc", O_RDWR);
- if ( fd == -1 )
- return -1;
- xgs->fd = fd;
- return 0;
-}
-
-int osdep_gntshr_close(xc_gntshr *xgs)
-{
- if ( xgs->fd == -1 )
- return 0;
-
- return close(xgs->fd);
-}
-
-void *osdep_gntshr_share_pages(xc_gntshr *xgs,
- uint32_t domid, int count,
- uint32_t *refs, int writable,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
-{
- struct ioctl_gntalloc_alloc_gref *gref_info = NULL;
- struct ioctl_gntalloc_unmap_notify notify;
- struct ioctl_gntalloc_dealloc_gref gref_drop;
- int fd = xgs->fd;
- int err;
- void *area = NULL;
- gref_info = malloc(sizeof(*gref_info) + count * sizeof(uint32_t));
- if (!gref_info)
- return NULL;
- gref_info->domid = domid;
- gref_info->flags = writable ? GNTALLOC_FLAG_WRITABLE : 0;
- gref_info->count = count;
-
- err = ioctl(fd, IOCTL_GNTALLOC_ALLOC_GREF, gref_info);
- if (err) {
- GSERROR(xgs->logger, "ioctl failed");
- goto out;
- }
-
- area = mmap(NULL, count * XC_PAGE_SIZE, PROT_READ | PROT_WRITE,
- MAP_SHARED, fd, gref_info->index);
-
- if (area == MAP_FAILED) {
- area = NULL;
- GSERROR(xgs->logger, "mmap failed");
- goto out_remove_fdmap;
- }
-
- notify.index = gref_info->index;
- notify.action = 0;
- if (notify_offset < XC_PAGE_SIZE * count) {
- notify.index += notify_offset;
- notify.action |= UNMAP_NOTIFY_CLEAR_BYTE;
- }
- if (notify_port != -1) {
- notify.event_channel_port = notify_port;
- notify.action |= UNMAP_NOTIFY_SEND_EVENT;
- }
- if (notify.action)
- err = ioctl(fd, IOCTL_GNTALLOC_SET_UNMAP_NOTIFY, ¬ify);
- if (err) {
- GSERROR(xgs->logger, "ioctl SET_UNMAP_NOTIFY failed");
- munmap(area, count * XC_PAGE_SIZE);
- area = NULL;
- }
-
- memcpy(refs, gref_info->gref_ids, count * sizeof(uint32_t));
-
- out_remove_fdmap:
- /* Removing the mapping from the file descriptor does not cause the pages to
- * be deallocated until the mapping is removed.
- */
- gref_drop.index = gref_info->index;
- gref_drop.count = count;
- ioctl(fd, IOCTL_GNTALLOC_DEALLOC_GREF, &gref_drop);
- out:
- free(gref_info);
- return area;
-}
-
-int xc_gntshr_munmap(xc_gntshr *xgs,
- void *start_address, uint32_t count)
-{
- return munmap(start_address, count * XC_PAGE_SIZE);
-}
-
static struct xc_osdep_ops *linux_osdep_init(xc_interface *xch, enum xc_osdep_type type)
{
switch ( type )
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index fd7def6..22d985c 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -23,8 +23,6 @@
#include <mini-os/os.h>
#include <mini-os/mm.h>
#include <mini-os/lib.h>
-#include <mini-os/gntmap.h>
-#include <sys/mman.h>
#include <xen/memory.h>
#include <unistd.h>
@@ -38,7 +36,6 @@
#include "xc_private.h"
void minios_interface_close_fd(int fd);
-void minios_gnttab_close_fd(int fd);
extern void minios_interface_close_fd(int fd);
@@ -202,76 +199,6 @@ void *xc_memalign(xc_interface *xch, size_t alignment, size_t size)
return memalign(alignment, size);
}
-int osdep_gnttab_open(xc_gnttab *xgt)
-{
- int fd = alloc_fd(FTYPE_GNTMAP);
- if ( fd == -1 )
- return -1;
- gntmap_init(&files[fd].gntmap);
- xgt->fd = fd;
- return 0;
-}
-
-int osdep_gnttab_close(xc_gnttab *xgt)
-{
- if ( xgt->fd == -1 )
- return 0;
-
- return close(xgt->fd);
-}
-
-void minios_gnttab_close_fd(int fd)
-{
- gntmap_fini(&files[fd].gntmap);
- files[fd].type = FTYPE_NONE;
-}
-
-void *osdep_gnttab_grant_map(xc_gnttab *xgt,
- uint32_t count, int flags, int prot,
- uint32_t *domids, uint32_t *refs,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
-{
- int fd = xgt->fd;
- int stride = 1;
- if (flags & XC_GRANT_MAP_SINGLE_DOMAIN)
- stride = 0;
- if (notify_offset != -1 || notify_port != -1) {
- errno = ENOSYS;
- return NULL;
- }
- return gntmap_map_grant_refs(&files[fd].gntmap,
- count, domids, stride,
- refs, prot & PROT_WRITE);
-}
-
-int xc_gnttab_munmap(xc_gnttab *xgt, void *start_address, uint32_t count)
-{
- int fd = xgt->fd;
- int ret;
- ret = gntmap_munmap(&files[fd].gntmap,
- (unsigned long) start_address,
- count);
- if (ret < 0) {
- errno = -ret;
- return -1;
- }
- return ret;
-}
-
-int xc_gnttab_set_max_grants(xc_gnttab *xgt, uint32_t count)
-{
- int fd = xgt->fd;
- int ret;
- ret = gntmap_set_max_grants(&files[fd].gntmap,
- count);
- if (ret < 0) {
- errno = -ret;
- return -1;
- }
- return ret;
-}
-
static struct xc_osdep_ops *minios_osdep_init(xc_interface *xch, enum xc_osdep_type type)
{
switch ( type )
diff --git a/tools/libxc/xc_nogntshr.c b/tools/libxc/xc_nogntshr.c
deleted file mode 100644
index 9aa6064..0000000
--- a/tools/libxc/xc_nogntshr.c
+++ /dev/null
@@ -1,46 +0,0 @@
-/******************************************************************************
- *
- * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
- *
- * This library is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License as published by the Free Software Foundation;
- * version 2.1 of the License.
- *
- * This library is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this library; If not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <stdlib.h>
-
-#include "xc_private.h"
-
-int osdep_gntshr_open(xc_gnttab *xgt)
-{
- return -1;
-}
-
-int osdep_gntshr_close(xc_gnttab *xgt)
-{
- return 0;
-}
-
-void *osdep_gntshr_share_pages(xc_gntshr *xgs,
- uint32_t domid, int count,
- uint32_t *refs, int writable,
- uint32_t notify_offset,
- evtchn_port_t notify_port)
-{
- abort()
-}
-
-int xc_gntshr_munmap(xc_gntshr *xgs,
- void *start_address, uint32_t count)
-{
- abort();
-}
diff --git a/tools/libxc/xc_private.c b/tools/libxc/xc_private.c
index 5285614..10b2367 100644
--- a/tools/libxc/xc_private.c
+++ b/tools/libxc/xc_private.c
@@ -249,86 +249,6 @@ int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
return xch->ops->u.privcmd.hypercall(xch, xch->ops_handle, hypercall);
}
-xc_gnttab *xc_gnttab_open(xentoollog_logger *logger, unsigned open_flags)
-{
- xc_gnttab *xgt = malloc(sizeof(*xgt));
- int rc;
-
- if (!xgt) return NULL;
-
- xgt->fd = -1;
- xgt->logger = logger;
- xgt->logger_tofree = NULL;
-
- if (!xgt->logger) {
- xgt->logger = xgt->logger_tofree =
- (xentoollog_logger*)
- xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
- if (!xgt->logger) goto err;
- }
-
- rc = osdep_gnttab_open(xgt);
- if ( rc < 0 ) goto err;
-
- return xgt;
-
-err:
- osdep_gnttab_close(xgt);
- xtl_logger_destroy(xgt->logger_tofree);
- free(xgt);
- return NULL;
-}
-
-int xc_gnttab_close(xc_gnttab *xgt)
-{
- int rc;
-
- rc = osdep_gnttab_close(xgt);
- xtl_logger_destroy(xgt->logger_tofree);
- free(xgt);
- return rc;
-}
-
-xc_gntshr *xc_gntshr_open(xentoollog_logger *logger, unsigned open_flags)
-{
- xc_gntshr *xgs = malloc(sizeof(*xgs));
- int rc;
-
- if (!xgs) return NULL;
-
- xgs->fd = -1;
- xgs->logger = logger;
- xgs->logger_tofree = NULL;
-
- if (!xgs->logger) {
- xgs->logger = xgs->logger_tofree =
- (xentoollog_logger*)
- xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
- if (!xgs->logger) goto err;
- }
-
- rc = osdep_gntshr_open(xgs);
- if ( rc < 0 ) goto err;
-
- return xgs;
-
-err:
- osdep_gntshr_close(xgs);
- xtl_logger_destroy(xgs->logger_tofree);
- free(xgs);
- return NULL;
-}
-
-int xc_gntshr_close(xc_gntshr *xgs)
-{
- int rc;
-
- rc = osdep_gntshr_close(xgs);
- xtl_logger_destroy(xgs->logger_tofree);
- free(xgs);
- return rc;
-}
-
static pthread_key_t errbuf_pkey;
static pthread_once_t errbuf_pkey_once = PTHREAD_ONCE_INIT;
diff --git a/tools/libxc/xc_private.h b/tools/libxc/xc_private.h
index 1e9dc93..46e0e75 100644
--- a/tools/libxc/xc_private.h
+++ b/tools/libxc/xc_private.h
@@ -123,30 +123,6 @@ struct xc_interface_core {
xc_osdep_handle ops_handle; /* opaque data for xc_osdep_ops */
};
-struct xengntdev_handle {
- xentoollog_logger *logger, *logger_tofree;
- int fd;
-};
-
-int osdep_gnttab_open(xc_gnttab *xgt);
-int osdep_gnttab_close(xc_gnttab *xgt);
-
-#define XC_GRANT_MAP_SINGLE_DOMAIN 0x1
-void *osdep_gnttab_grant_map(xc_gnttab *xgt,
- uint32_t count, int flags, int prot,
- uint32_t *domids, uint32_t *refs,
- uint32_t notify_offset,
- evtchn_port_t notify_port);
-
-int osdep_gntshr_open(xc_gntshr *xgs);
-int osdep_gntshr_close(xc_gntshr *xgs);
-
-void *osdep_gntshr_share_pages(xc_gntshr *xgs,
- uint32_t domid, int count,
- uint32_t *refs, int writable,
- uint32_t notify_offset,
- evtchn_port_t notify_port);
-
void xc_report_error(xc_interface *xch, int code, const char *fmt, ...)
__attribute__((format(printf,3,4)));
void xc_reportv(xc_interface *xch, xentoollog_logger *lg, xentoollog_level,
diff --git a/tools/xenstore/Makefile b/tools/xenstore/Makefile
index c161046..178771f 100644
--- a/tools/xenstore/Makefile
+++ b/tools/xenstore/Makefile
@@ -78,8 +78,10 @@ init-xenstore-domain.o: CFLAGS += $(CFLAGS_libxenguest)
init-xenstore-domain: init-xenstore-domain.o $(LIBXENSTORE)
$(CC) $^ $(LDFLAGS) $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) -o $@ $(APPEND_LDFLAGS)
+$(XENSTORED_OBJS): CFLAGS += $(CFLAGS_libxengnttab)
+
xenstored: $(XENSTORED_OBJS)
- $(CC) $^ $(LDFLAGS) $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(SOCKET_LIBS) -o $@ $(APPEND_LDFLAGS)
+ $(CC) $^ $(LDFLAGS) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_libxenctrl) $(SOCKET_LIBS) -o $@ $(APPEND_LDFLAGS)
xenstored.a: $(XENSTORED_OBJS)
$(AR) cr $@ $^
diff --git a/tools/xenstore/xenstored_core.h b/tools/xenstore/xenstored_core.h
index 8c853c9..624737d 100644
--- a/tools/xenstore/xenstored_core.h
+++ b/tools/xenstore/xenstored_core.h
@@ -20,12 +20,14 @@
#define _XENSTORED_CORE_H
#include <xenctrl.h>
+#include <xengnttab.h>
#include <sys/types.h>
#include <dirent.h>
#include <stdbool.h>
#include <stdint.h>
#include <errno.h>
+
#include "xenstore_lib.h"
#include "list.h"
#include "tdb.h"
@@ -196,7 +198,7 @@ void finish_daemonize(void);
/* Open a pipe for signal handling */
void init_pipe(int reopen_log_pipe[2]);
-xc_gnttab **xcg_handle;
+xengnttab_handle **xgt_handle;
#endif /* _XENSTORED_CORE_H */
diff --git a/tools/xenstore/xenstored_domain.c b/tools/xenstore/xenstored_domain.c
index 6ceec29..25b8607 100644
--- a/tools/xenstore/xenstored_domain.c
+++ b/tools/xenstore/xenstored_domain.c
@@ -34,7 +34,7 @@
#include <xen/grant_table.h>
static xc_interface **xc_handle;
-xc_gnttab **xcg_handle;
+xengnttab_handle **xgt_handle;
static evtchn_port_t virq_port;
xenevtchn_handle *xce_handle = NULL;
@@ -166,9 +166,9 @@ static int readchn(struct connection *conn, void *data, unsigned int len)
static void *map_interface(domid_t domid, unsigned long mfn)
{
- if (*xcg_handle != NULL) {
+ if (*xgt_handle != NULL) {
/* this is the preferred method */
- return xc_gnttab_map_grant_ref(*xcg_handle, domid,
+ return xengnttab_map_grant_ref(*xgt_handle, domid,
GNTTAB_RESERVED_XENSTORE, PROT_READ|PROT_WRITE);
} else {
return xc_map_foreign_range(*xc_handle, domid,
@@ -178,8 +178,8 @@ static void *map_interface(domid_t domid, unsigned long mfn)
static void unmap_interface(void *interface)
{
- if (*xcg_handle != NULL)
- xc_gnttab_munmap(*xcg_handle, interface, 1);
+ if (*xgt_handle != NULL)
+ xengnttab_munmap(*xgt_handle, interface, 1);
else
munmap(interface, XC_PAGE_SIZE);
}
@@ -577,9 +577,9 @@ static int close_xc_handle(void *_handle)
return 0;
}
-static int close_xcg_handle(void *_handle)
+static int close_xgt_handle(void *_handle)
{
- xc_gnttab_close(*(xc_gnttab **)_handle);
+ xengnttab_close(*(xengnttab_handle **)_handle);
return 0;
}
@@ -634,15 +634,15 @@ void domain_init(void)
talloc_set_destructor(xc_handle, close_xc_handle);
- xcg_handle = talloc(talloc_autofree_context(), xc_gnttab*);
- if (!xcg_handle)
+ xgt_handle = talloc(talloc_autofree_context(), xengnttab_handle*);
+ if (!xgt_handle)
barf_perror("Failed to allocate domain gnttab handle");
- *xcg_handle = xc_gnttab_open(NULL, 0);
- if (*xcg_handle == NULL)
+ *xgt_handle = xengnttab_open(NULL, 0);
+ if (*xgt_handle == NULL)
xprintf("WARNING: Failed to open connection to gnttab\n");
else
- talloc_set_destructor(xcg_handle, close_xcg_handle);
+ talloc_set_destructor(xgt_handle, close_xgt_handle);
xce_handle = xenevtchn_open(NULL, 0);
diff --git a/tools/xenstore/xenstored_minios.c b/tools/xenstore/xenstored_minios.c
index b686e1c..6ef1c46 100644
--- a/tools/xenstore/xenstored_minios.c
+++ b/tools/xenstore/xenstored_minios.c
@@ -17,7 +17,6 @@
*/
#include <sys/types.h>
#include <sys/mman.h>
-#include <xenctrl.h>
#include "xenstored_core.h"
#include <xen/grant_table.h>
@@ -50,12 +49,12 @@ evtchn_port_t xenbus_evtchn(void)
void *xenbus_map(void)
{
- return xc_gnttab_map_grant_ref(*xcg_handle, xenbus_master_domid(),
+ return xengnttab_map_grant_ref(*xgt_handle, xenbus_master_domid(),
GNTTAB_RESERVED_XENSTORE, PROT_READ|PROT_WRITE);
}
void unmap_xenbus(void *interface)
{
- xc_gnttab_munmap(*xcg_handle, interface, 1);
+ xengnttab_munmap(*xgt_handle, interface, 1);
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 08/23] tools/libxc: Remove osdep indirection for privcmd
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (6 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall Ian Campbell
` (15 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: David Scott, Ian Campbell
The alternative backend (a xen-api/xapi shim) is no longer around and
so this stuff is now just baggage which is getting in the way of
refactoring libxenctrl.
Nested virt probably suffices for this use case now.
This was the last component of the osdep infrastructure, so all the
dynamic loading etc stuff all falls away too.
As part of this I was forced to investigate the twisty
xc_map_foreign_* maze, which I have added to the
toolstack-library-apis doc in the hopes of doing something sensible.
NetBSD and Solaris now call xc_map_foreign_bulk_compat directly from
their xc_map_foreign_bulk, which could have been achieved by using
some ifdefs around a renamed function. This will fall out in the wash
when these functions move to their own library.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: David Scott <dave.scott@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Cc: David Scott <dave.scott@citrix.com>
---
config/FreeBSD.mk | 2 -
config/NetBSD.mk | 3 -
config/NetBSDRump.mk | 1 -
config/StdGNU.mk | 1 -
config/SunOS.mk | 1 -
tools/libxc/Makefile | 27 ++----
tools/libxc/include/xenctrl.h | 9 --
tools/libxc/include/xenctrlosdep.h | 133 ---------------------------
tools/libxc/xc_foreign_memory.c | 31 +------
tools/libxc/xc_freebsd_osdep.c | 100 ++++++---------------
tools/libxc/xc_hcall_buf.c | 6 +-
tools/libxc/xc_linux_osdep.c | 88 ++++++------------
tools/libxc/xc_minios.c | 82 +++++------------
tools/libxc/xc_netbsd.c | 90 +++++++------------
tools/libxc/xc_private.c | 174 ++----------------------------------
tools/libxc/xc_private.h | 19 ++--
tools/libxc/xc_solaris.c | 90 +++++++------------
tools/libxc/xenctrl_osdep_ENOSYS.c | 123 -------------------------
tools/ocaml/libs/xc/xenctrl.ml | 2 -
tools/ocaml/libs/xc/xenctrl.mli | 1 -
tools/ocaml/libs/xc/xenctrl_stubs.c | 7 --
tools/ocaml/xenstored/domains.ml | 6 +-
tools/ocaml/xenstored/xenstored.ml | 5 +-
23 files changed, 178 insertions(+), 823 deletions(-)
delete mode 100644 tools/libxc/include/xenctrlosdep.h
delete mode 100644 tools/libxc/xenctrl_osdep_ENOSYS.c
diff --git a/config/FreeBSD.mk b/config/FreeBSD.mk
index 5a13d607..bb3a5d0 100644
--- a/config/FreeBSD.mk
+++ b/config/FreeBSD.mk
@@ -1,6 +1,4 @@
include $(XEN_ROOT)/config/StdGNU.mk
-DLOPEN_LIBS =
-
# No wget on FreeBSD base system
WGET = ftp
diff --git a/config/NetBSD.mk b/config/NetBSD.mk
index 21318d6..cf766e5 100644
--- a/config/NetBSD.mk
+++ b/config/NetBSD.mk
@@ -1,6 +1,3 @@
include $(XEN_ROOT)/config/StdGNU.mk
-# Override settings for this OS
-DLOPEN_LIBS =
-
WGET = ftp
diff --git a/config/NetBSDRump.mk b/config/NetBSDRump.mk
index 2a87218..74755a1 100644
--- a/config/NetBSDRump.mk
+++ b/config/NetBSDRump.mk
@@ -1,6 +1,5 @@
include $(XEN_ROOT)/config/StdGNU.mk
-DLOPEN_LIBS =
PTHREAD_LIBS =
WGET = ftp
diff --git a/config/StdGNU.mk b/config/StdGNU.mk
index 129d5c8..39d36b2 100644
--- a/config/StdGNU.mk
+++ b/config/StdGNU.mk
@@ -31,7 +31,6 @@ DEBUG_DIR ?= /usr/lib/debug
SOCKET_LIBS =
UTIL_LIBS = -lutil
-DLOPEN_LIBS = -ldl
SONAME_LDFLAG = -soname
SHLIB_LDFLAGS = -shared
diff --git a/config/SunOS.mk b/config/SunOS.mk
index db5e898..86a384d 100644
--- a/config/SunOS.mk
+++ b/config/SunOS.mk
@@ -27,7 +27,6 @@ SunOS_LIBDIR_x86_64 = /usr/sfw/lib/amd64
SOCKET_LIBS = -lsocket
PTHREAD_LIBS = -lpthread
UTIL_LIBS =
-DLOPEN_LIBS = -ldl
SONAME_LDFLAG = -h
SHLIB_LDFLAGS = -R $(SunOS_LIBDIR) -shared
diff --git a/tools/libxc/Makefile b/tools/libxc/Makefile
index 33d18db..3305fdd 100644
--- a/tools/libxc/Makefile
+++ b/tools/libxc/Makefile
@@ -101,8 +101,6 @@ GUEST_SRCS-y += xc_dom_decompress_unsafe_lzo1x.c
GUEST_SRCS-y += xc_dom_decompress_unsafe_xz.c
endif
-OSDEP_SRCS-y += xenctrl_osdep_ENOSYS.c
-
-include $(XEN_TARGET_ARCH)/Makefile
CFLAGS += -Werror -Wmissing-prototypes
@@ -121,11 +119,8 @@ CTRL_PIC_OBJS := $(patsubst %.c,%.opic,$(CTRL_SRCS-y))
GUEST_LIB_OBJS := $(patsubst %.c,%.o,$(GUEST_SRCS-y))
GUEST_PIC_OBJS := $(patsubst %.c,%.opic,$(GUEST_SRCS-y))
-OSDEP_LIB_OBJS := $(patsubst %.c,%.o,$(OSDEP_SRCS-y))
-OSDEP_PIC_OBJS := $(patsubst %.c,%.opic,$(OSDEP_SRCS-y))
-
-$(CTRL_LIB_OBJS) $(GUEST_LIB_OBJS) $(OSDEP_LIB_OBJS) \
-$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS) $(OSDEP_PIC_OBJS) : CFLAGS += -include $(XEN_ROOT)/tools/config.h
+$(CTRL_LIB_OBJS) $(GUEST_LIB_OBJS) \
+$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS): CFLAGS += -include $(XEN_ROOT)/tools/config.h
$(CTRL_LIB_OBJS) $(CTRL_PIC_OBJS): CFLAGS += $(CFLAGS_libxengnttab) $(CFLAGS_libxengntshr)
@@ -139,17 +134,13 @@ ifneq ($(nosharedlibs),y)
LIB += libxenguest.so libxenguest.so.$(MAJOR) libxenguest.so.$(MAJOR).$(MINOR)
endif
-ifneq ($(nosharedlibs),y)
-LIB += xenctrl_osdep_ENOSYS.so
-endif
-
genpath-target = $(call buildmakevars2header,_paths.h)
$(eval $(genpath-target))
xc_private.h: _paths.h
-$(CTRL_LIB_OBJS) $(GUEST_LIB_OBJS) $(OSDEP_LIB_OBJS) \
-$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS) $(OSDEP_PIC_OBJS): xc_private.h
+$(CTRL_LIB_OBJS) $(GUEST_LIB_OBJS) \
+$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS): xc_private.h
.PHONY: all
all: build
@@ -169,7 +160,7 @@ install: build
$(INSTALL_DATA) libxenctrl.a $(DESTDIR)$(libdir)
$(SYMLINK_SHLIB) libxenctrl.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxenctrl.so.$(MAJOR)
$(SYMLINK_SHLIB) libxenctrl.so.$(MAJOR) $(DESTDIR)$(libdir)/libxenctrl.so
- $(INSTALL_DATA) include/xenctrl.h include/xenctrl_compat.h include/xenctrlosdep.h $(DESTDIR)$(includedir)
+ $(INSTALL_DATA) include/xenctrl.h include/xenctrl_compat.h $(DESTDIR)$(includedir)
$(INSTALL_SHLIB) libxenguest.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)
$(INSTALL_DATA) libxenguest.a $(DESTDIR)$(libdir)
$(SYMLINK_SHLIB) libxenguest.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxenguest.so.$(MAJOR)
@@ -185,8 +176,7 @@ clean:
rm -rf *.rpm $(LIB) *~ $(DEPS) \
_paths.h \
$(CTRL_LIB_OBJS) $(CTRL_PIC_OBJS) \
- $(GUEST_LIB_OBJS) $(GUEST_PIC_OBJS) \
- $(OSDEP_LIB_OBJS) $(OSDEP_PIC_OBJS)
+ $(GUEST_LIB_OBJS) $(GUEST_PIC_OBJS)
.PHONY: distclean
distclean: clean
@@ -212,7 +202,7 @@ libxenctrl.so.$(MAJOR): libxenctrl.so.$(MAJOR).$(MINOR)
$(SYMLINK_SHLIB) $< $@
libxenctrl.so.$(MAJOR).$(MINOR): $(CTRL_PIC_OBJS)
- $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_libxengntshr) $(DLOPEN_LIBS) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_libxengntshr) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
# libxenguest
@@ -237,8 +227,5 @@ libxenguest.so.$(MAJOR).$(MINOR): COMPRESSION_LIBS = $(call zlib-options,l)
libxenguest.so.$(MAJOR).$(MINOR): $(GUEST_PIC_OBJS) libxenctrl.so
$(CC) $(LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenguest.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $(GUEST_PIC_OBJS) $(COMPRESSION_LIBS) -lz $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
-xenctrl_osdep_ENOSYS.so: $(OSDEP_PIC_OBJS) libxenctrl.so
- $(CC) $(LDFLAGS) $(SHLIB_LDFLAGS) -o $@ $(OSDEP_PIC_OBJS) $(LDLIBS_libxenctrl) $(APPEND_LDFLAGS)
-
-include $(DEPS)
diff --git a/tools/libxc/include/xenctrl.h b/tools/libxc/include/xenctrl.h
index bfc8592..bdf4f2f 100644
--- a/tools/libxc/include/xenctrl.h
+++ b/tools/libxc/include/xenctrl.h
@@ -181,15 +181,6 @@ enum xc_open_flags {
*/
int xc_interface_close(xc_interface *xch);
-/**
- * Query the active OS interface (i.e. that which would be returned by
- * xc_interface_open) to find out if it is fake (i.e. backends onto
- * something other than an actual Xen hypervisor).
- *
- * @return 0 is "real", >0 if fake, -1 on error.
- */
-int xc_interface_is_fake(void);
-
/*
* HYPERCALL SAFE MEMORY BUFFER
*
diff --git a/tools/libxc/include/xenctrlosdep.h b/tools/libxc/include/xenctrlosdep.h
deleted file mode 100644
index 423660d..0000000
--- a/tools/libxc/include/xenctrlosdep.h
+++ /dev/null
@@ -1,133 +0,0 @@
-/******************************************************************************
- *
- * Interface to OS specific low-level operations
- *
- * Copyright (c) 2010, Citrix Systems Inc.
- *
- * This library is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License as published by the Free Software Foundation;
- * version 2.1 of the License.
- *
- * This library is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this library; If not, see <http://www.gnu.org/licenses/>.
- */
-
-/*
- * This interface defines the interactions between the Xen control
- * libraries and the OS facilities used to communicate with the
- * hypervisor.
- *
- * It is possible to override the default (native) implementation by
- * setting the XENCTRL_OSDEP environment variable to point to a
- * plugin library. Userspace can use this facility to intercept
- * hypervisor operations. This can be used e.g. to implement a
- * userspace simulator for Xen hypercalls.
- *
- * The plugin must contain a data structure:
- * xc_osdep_info_t xc_osdep_info;
- *
- * xc_osdep_init:
- * Must return a suitable struct xc_osdep_ops pointer or NULL on failure.
- */
-
-#ifndef XC_OSDEP_H
-#define XC_OSDEP_H
-
-/* Tell the Xen public headers we are a user-space tools build. */
-#ifndef __XEN_TOOLS__
-#define __XEN_TOOLS__ 1
-#endif
-
-#include <sys/mman.h>
-#include <sys/types.h>
-
-#include <xen/sys/privcmd.h>
-
-enum xc_osdep_type {
- XC_OSDEP_PRIVCMD,
-};
-
-/* Opaque handle internal to the backend */
-typedef unsigned long xc_osdep_handle;
-
-#define XC_OSDEP_OPEN_ERROR ((xc_osdep_handle)-1)
-
-struct xc_osdep_ops
-{
- /* Opens an interface.
- *
- * Must return an opaque handle on success or
- * XC_OSDEP_OPEN_ERROR on failure
- */
- xc_osdep_handle (*open)(xc_interface *xch);
-
- int (*close)(xc_interface *xch, xc_osdep_handle h);
-
- union {
- struct {
- void *(*alloc_hypercall_buffer)(xc_interface *xch, xc_osdep_handle h, int npages);
- void (*free_hypercall_buffer)(xc_interface *xch, xc_osdep_handle h, void *ptr, int npages);
-
- int (*hypercall)(xc_interface *xch, xc_osdep_handle h, privcmd_hypercall_t *hypercall);
-
- void *(*map_foreign_batch)(xc_interface *xch, xc_osdep_handle h, uint32_t dom, int prot,
- xen_pfn_t *arr, int num);
- void *(*map_foreign_bulk)(xc_interface *xch, xc_osdep_handle h, uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num);
- void *(*map_foreign_range)(xc_interface *xch, xc_osdep_handle h, uint32_t dom, int size, int prot,
- unsigned long mfn);
- void *(*map_foreign_ranges)(xc_interface *xch, xc_osdep_handle h, uint32_t dom, size_t size, int prot,
- size_t chunksize, privcmd_mmap_entry_t entries[],
- int nentries);
- } privcmd;
- } u;
-};
-typedef struct xc_osdep_ops xc_osdep_ops;
-
-typedef xc_osdep_ops *(*xc_osdep_init_fn)(xc_interface *xch, enum xc_osdep_type);
-
-struct xc_osdep_info
-{
- /* Describes this backend. */
- const char *name;
-
- /* Returns ops function. */
- xc_osdep_init_fn init;
-
- /* True if this interface backs onto a fake Xen. */
- int fake;
-
- /* For internal use by loader. */
- void *dl_handle;
-};
-typedef struct xc_osdep_info xc_osdep_info_t;
-
-/* All backends, including the builtin backend, must supply this structure. */
-extern xc_osdep_info_t xc_osdep_info;
-
-/* Stub for not yet converted OSes */
-void *xc_map_foreign_bulk_compat(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num);
-
-/* Report errors through xc_interface */
-void xc_osdep_log(xc_interface *xch, xentoollog_level level, int code,
- const char *fmt, ...) __attribute__((format(printf, 4, 5)));
-
-#endif
-
-/*
- * Local variables:
- * mode: C
- * c-file-style: "BSD"
- * c-basic-offset: 4
- * tab-width: 4
- * indent-tabs-mode: nil
- * End:
- */
diff --git a/tools/libxc/xc_foreign_memory.c b/tools/libxc/xc_foreign_memory.c
index f42d140..9c705b6 100644
--- a/tools/libxc/xc_foreign_memory.c
+++ b/tools/libxc/xc_foreign_memory.c
@@ -50,37 +50,8 @@ void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
return res;
}
-void *xc_map_foreign_range(xc_interface *xch, uint32_t dom,
- int size, int prot, unsigned long mfn)
-{
- return xch->ops->u.privcmd.map_foreign_range(xch, xch->ops_handle,
- dom, size, prot, mfn);
-}
-
-void *xc_map_foreign_ranges(xc_interface *xch, uint32_t dom,
- size_t size, int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[], int nentries)
-{
- return xch->ops->u.privcmd.map_foreign_ranges(xch, xch->ops_handle,
- dom, size, prot, chunksize, entries, nentries);
-}
-
-void *xc_map_foreign_batch(xc_interface *xch, uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
-{
- return xch->ops->u.privcmd.map_foreign_batch(xch, xch->ops_handle,
- dom, prot, arr, num);
-}
-
-void *xc_map_foreign_bulk(xc_interface *xch, uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
-{
- return xch->ops->u.privcmd.map_foreign_bulk(xch, xch->ops_handle,
- dom, prot, arr, err, num);
-}
-
/* stub for all not yet converted OSes */
-void *xc_map_foreign_bulk_compat(xc_interface *xch, xc_osdep_handle h,
+void *xc_map_foreign_bulk_compat(xc_interface *xch,
uint32_t dom, int prot,
const xen_pfn_t *arr, int *err, unsigned int num)
{
diff --git a/tools/libxc/xc_freebsd_osdep.c b/tools/libxc/xc_freebsd_osdep.c
index 339997c..f6a2ccd 100644
--- a/tools/libxc/xc_freebsd_osdep.c
+++ b/tools/libxc/xc_freebsd_osdep.c
@@ -32,16 +32,12 @@
#include <xen/memory.h>
-#include "xenctrl.h"
-#include "xenctrlosdep.h"
+#include "xc_private.h"
#define PRIVCMD_DEV "/dev/xen/privcmd"
-#define PERROR(_m, _a...) xc_osdep_log(xch,XTL_ERROR,XC_INTERNAL_ERROR,_m \
- " (%d = %s)", ## _a , errno, xc_strerror(xch, errno))
-
/*------------------------- Privcmd device interface -------------------------*/
-static xc_osdep_handle freebsd_privcmd_open(xc_interface *xch)
+int osdep_privcmd_open(xc_interface *xch)
{
int flags, saved_errno;
int fd = open(PRIVCMD_DEV, O_RDWR);
@@ -50,7 +46,7 @@ static xc_osdep_handle freebsd_privcmd_open(xc_interface *xch)
{
PERROR("Could not obtain handle on privileged command interface "
PRIVCMD_DEV);
- return XC_OSDEP_OPEN_ERROR;
+ return -1
}
/*
@@ -73,27 +69,27 @@ static xc_osdep_handle freebsd_privcmd_open(xc_interface *xch)
goto error;
}
- return (xc_osdep_handle)fd;
+ xch->privcmdfd = fd;
+ return 0;
error:
saved_errno = errno;
close(fd);
errno = saved_errno;
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
}
-static int freebsd_privcmd_close(xc_interface *xch, xc_osdep_handle h)
+int osdep_privcmd_close(xc_interface *xch)
{
- int fd = (int)h;
-
+ int fd = xch->privcmdfd;
+ if ( fd == -1 )
+ return 0;
return close(fd);
}
/*------------------------ Privcmd hypercall interface -----------------------*/
-static void *freebsd_privcmd_alloc_hypercall_buffer(xc_interface *xch,
- xc_osdep_handle h,
- int npages)
+void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
{
size_t size = npages * XC_PAGE_SIZE;
void *p;
@@ -117,9 +113,7 @@ static void *freebsd_privcmd_alloc_hypercall_buffer(xc_interface *xch,
return p;
}
-static void freebsd_privcmd_free_hypercall_buffer(xc_interface *xch,
- xc_osdep_handle h, void *ptr,
- int npages)
+void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
{
int saved_errno = errno;
@@ -131,10 +125,9 @@ static void freebsd_privcmd_free_hypercall_buffer(xc_interface *xch,
errno = saved_errno;
}
-static int freebsd_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h,
- privcmd_hypercall_t *hypercall)
+int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
int ret;
ret = ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
@@ -143,13 +136,12 @@ static int freebsd_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h,
}
/*----------------------- Privcmd foreign map interface ----------------------*/
-static void *freebsd_privcmd_map_foreign_bulk(xc_interface *xch,
- xc_osdep_handle h,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err,
- unsigned int num)
+void *xc_map_foreign_bulk(xc_interface *xch,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err,
+ unsigned int num)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
privcmd_mmapbatch_t ioctlx;
void *addr;
int rc;
@@ -180,10 +172,9 @@ static void *freebsd_privcmd_map_foreign_bulk(xc_interface *xch,
return addr;
}
-static void *freebsd_privcmd_map_foreign_range(xc_interface *xch,
- xc_osdep_handle h,
- uint32_t dom, int size, int prot,
- unsigned long mfn)
+void *xc_map_foreign_range(xc_interface *xch,
+ uint32_t dom, int size, int prot,
+ unsigned long mfn)
{
xen_pfn_t *arr;
int num;
@@ -203,12 +194,11 @@ static void *freebsd_privcmd_map_foreign_range(xc_interface *xch,
return ret;
}
-static void *freebsd_privcmd_map_foreign_ranges(xc_interface *xch,
- xc_osdep_handle h,
- uint32_t dom, size_t size,
- int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[],
- int nentries)
+void *xc_map_foreign_ranges(xc_interface *xch,
+ uint32_t dom, size_t size,
+ int prot, size_t chunksize,
+ privcmd_mmap_entry_t entries[],
+ int nentries)
{
xen_pfn_t *arr;
int num_per_entry;
@@ -232,42 +222,6 @@ static void *freebsd_privcmd_map_foreign_ranges(xc_interface *xch,
return ret;
}
-/*----------------------------- Privcmd handlers -----------------------------*/
-static struct xc_osdep_ops freebsd_privcmd_ops = {
- .open = &freebsd_privcmd_open,
- .close = &freebsd_privcmd_close,
-
- .u.privcmd = {
- .alloc_hypercall_buffer = &freebsd_privcmd_alloc_hypercall_buffer,
- .free_hypercall_buffer = &freebsd_privcmd_free_hypercall_buffer,
-
- .hypercall = &freebsd_privcmd_hypercall,
-
- .map_foreign_bulk = &freebsd_privcmd_map_foreign_bulk,
- .map_foreign_range = &freebsd_privcmd_map_foreign_range,
- .map_foreign_ranges = &freebsd_privcmd_map_foreign_ranges,
- },
-};
-
-/*---------------------------- FreeBSD interface -----------------------------*/
-static struct xc_osdep_ops *
-freebsd_osdep_init(xc_interface *xch, enum xc_osdep_type type)
-{
- switch ( type )
- {
- case XC_OSDEP_PRIVCMD:
- return &freebsd_privcmd_ops;
- default:
- return NULL;
- }
-}
-
-xc_osdep_info_t xc_osdep_info = {
- .name = "FreeBSD Native OS interface",
- .init = &freebsd_osdep_init,
- .fake = 0,
-};
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xc_hcall_buf.c b/tools/libxc/xc_hcall_buf.c
index 6e3c958..a0e66cf 100644
--- a/tools/libxc/xc_hcall_buf.c
+++ b/tools/libxc/xc_hcall_buf.c
@@ -122,7 +122,7 @@ void xc__hypercall_buffer_cache_release(xc_interface *xch)
while ( xch->hypercall_buffer_cache_nr > 0 )
{
p = xch->hypercall_buffer_cache[--xch->hypercall_buffer_cache_nr];
- xch->ops->u.privcmd.free_hypercall_buffer(xch, xch->ops_handle, p, 1);
+ osdep_free_hypercall_buffer(xch, p, 1);
}
hypercall_buffer_cache_unlock(xch);
@@ -133,7 +133,7 @@ void *xc__hypercall_buffer_alloc_pages(xc_interface *xch, xc_hypercall_buffer_t
void *p = hypercall_buffer_cache_alloc(xch, nr_pages);
if ( !p )
- p = xch->ops->u.privcmd.alloc_hypercall_buffer(xch, xch->ops_handle, nr_pages);
+ p = osdep_alloc_hypercall_buffer(xch, nr_pages);
if (!p)
return NULL;
@@ -151,7 +151,7 @@ void xc__hypercall_buffer_free_pages(xc_interface *xch, xc_hypercall_buffer_t *b
return;
if ( !hypercall_buffer_cache_free(xch, b->hbuf, nr_pages) )
- xch->ops->u.privcmd.free_hypercall_buffer(xch, xch->ops_handle, b->hbuf, nr_pages);
+ osdep_free_hypercall_buffer(xch, b->hbuf, nr_pages);
}
struct allocation_header {
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index 96903f7..e858ad9 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -33,13 +33,12 @@
#include <xen/memory.h>
#include "xenctrl.h"
-#include "xenctrlosdep.h"
#include "xc_private.h"
#define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & ~((1UL<<(_w))-1))
-static xc_osdep_handle linux_privcmd_open(xc_interface *xch)
+int osdep_privcmd_open(xc_interface *xch)
{
int flags, saved_errno;
int fd = open("/proc/xen/privcmd", O_RDWR);
@@ -47,7 +46,7 @@ static xc_osdep_handle linux_privcmd_open(xc_interface *xch)
if ( fd == -1 )
{
PERROR("Could not obtain handle on privileged command interface");
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
}
/* Although we return the file handle as the 'xc handle' the API
@@ -68,22 +67,25 @@ static xc_osdep_handle linux_privcmd_open(xc_interface *xch)
goto error;
}
- return (xc_osdep_handle)fd;
+ xch->privcmdfd = fd;
+ return 0;
error:
saved_errno = errno;
close(fd);
errno = saved_errno;
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
}
-static int linux_privcmd_close(xc_interface *xch, xc_osdep_handle h)
+int osdep_privcmd_close(xc_interface *xch)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
+ if (fd == -1)
+ return 0;
return close(fd);
}
-static void *linux_privcmd_alloc_hypercall_buffer(xc_interface *xch, xc_osdep_handle h, int npages)
+void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
{
size_t size = npages * XC_PAGE_SIZE;
void *p;
@@ -115,7 +117,7 @@ out:
return NULL;
}
-static void linux_privcmd_free_hypercall_buffer(xc_interface *xch, xc_osdep_handle h, void *ptr, int npages)
+void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
{
int saved_errno = errno;
/* Recover the VMA flags. Maybe it's not necessary */
@@ -126,9 +128,9 @@ static void linux_privcmd_free_hypercall_buffer(xc_interface *xch, xc_osdep_hand
errno = saved_errno;
}
-static int linux_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h, privcmd_hypercall_t *hypercall)
+int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
return ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
}
@@ -154,11 +156,11 @@ static int xc_map_foreign_batch_single(int fd, uint32_t dom,
return rc;
}
-static void *linux_privcmd_map_foreign_batch(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
+void *xc_map_foreign_batch(xc_interface *xch,
+ uint32_t dom, int prot,
+ xen_pfn_t *arr, int num)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
privcmd_mmapbatch_t ioctlx;
void *addr;
int rc;
@@ -260,11 +262,11 @@ out:
return rc;
}
-static void *linux_privcmd_map_foreign_bulk(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
+void *xc_map_foreign_bulk(xc_interface *xch,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
privcmd_mmapbatch_v2_t ioctlx;
void *addr;
unsigned int i;
@@ -384,9 +386,9 @@ static void *linux_privcmd_map_foreign_bulk(xc_interface *xch, xc_osdep_handle h
return addr;
}
-static void *linux_privcmd_map_foreign_range(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, int size, int prot,
- unsigned long mfn)
+void *xc_map_foreign_range(xc_interface *xch,
+ uint32_t dom, int size, int prot,
+ unsigned long mfn)
{
xen_pfn_t *arr;
int num;
@@ -406,10 +408,10 @@ static void *linux_privcmd_map_foreign_range(xc_interface *xch, xc_osdep_handle
return ret;
}
-static void *linux_privcmd_map_foreign_ranges(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, size_t size, int prot,
- size_t chunksize, privcmd_mmap_entry_t entries[],
- int nentries)
+void *xc_map_foreign_ranges(xc_interface *xch,
+ uint32_t dom, size_t size, int prot,
+ size_t chunksize, privcmd_mmap_entry_t entries[],
+ int nentries)
{
xen_pfn_t *arr;
int num_per_entry;
@@ -433,40 +435,6 @@ static void *linux_privcmd_map_foreign_ranges(xc_interface *xch, xc_osdep_handle
return ret;
}
-static struct xc_osdep_ops linux_privcmd_ops = {
- .open = &linux_privcmd_open,
- .close = &linux_privcmd_close,
-
- .u.privcmd = {
- .alloc_hypercall_buffer = &linux_privcmd_alloc_hypercall_buffer,
- .free_hypercall_buffer = &linux_privcmd_free_hypercall_buffer,
-
- .hypercall = &linux_privcmd_hypercall,
-
- .map_foreign_batch = &linux_privcmd_map_foreign_batch,
- .map_foreign_bulk = &linux_privcmd_map_foreign_bulk,
- .map_foreign_range = &linux_privcmd_map_foreign_range,
- .map_foreign_ranges = &linux_privcmd_map_foreign_ranges,
- },
-};
-
-static struct xc_osdep_ops *linux_osdep_init(xc_interface *xch, enum xc_osdep_type type)
-{
- switch ( type )
- {
- case XC_OSDEP_PRIVCMD:
- return &linux_privcmd_ops;
- default:
- return NULL;
- }
-}
-
-xc_osdep_info_t xc_osdep_info = {
- .name = "Linux Native OS interface",
- .init = &linux_osdep_init,
- .fake = 0,
-};
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index 22d985c..047e13b 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -39,19 +39,20 @@ void minios_interface_close_fd(int fd);
extern void minios_interface_close_fd(int fd);
-static xc_osdep_handle minios_privcmd_open(xc_interface *xch)
+int osdep_privcmd_open(xc_interface *xch)
{
int fd = alloc_fd(FTYPE_XC);
if ( fd == -1)
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
- return (xc_osdep_handle)fd;
+ xch->privcmdfd = fd;
+ return 0;
}
-static int minios_privcmd_close(xc_interface *xch, xc_osdep_handle h)
+int osdep_privcmd_close(xc_interface *xch)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
return close(fd);
}
@@ -60,17 +61,17 @@ void minios_interface_close_fd(int fd)
files[fd].type = FTYPE_NONE;
}
-static void *minios_privcmd_alloc_hypercall_buffer(xc_interface *xch, xc_osdep_handle h, int npages)
+void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
{
return xc_memalign(xch, PAGE_SIZE, npages * PAGE_SIZE);
}
-static void minios_privcmd_free_hypercall_buffer(xc_interface *xch, xc_osdep_handle h, void *ptr, int npages)
+void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
{
free(ptr);
}
-static int minios_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h, privcmd_hypercall_t *hypercall)
+int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
{
multicall_entry_t call;
int i, ret;
@@ -92,21 +93,21 @@ static int minios_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h, privcm
return call.result;
}
-static void *minios_privcmd_map_foreign_bulk(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
+void *xc_map_foreign_bulk(xc_interface *xch,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num)
{
unsigned long pt_prot = 0;
if (prot & PROT_READ)
pt_prot = L1_PROT_RO;
if (prot & PROT_WRITE)
pt_prot = L1_PROT;
- return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
+ return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
}
-static void *minios_privcmd_map_foreign_batch(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
+void *xc_map_foreign_batch(xc_interface *xch,
+ uint32_t dom, int prot,
+ xen_pfn_t *arr, int num)
{
unsigned long pt_prot = 0;
int err[num];
@@ -126,10 +127,10 @@ static void *minios_privcmd_map_foreign_batch(xc_interface *xch, xc_osdep_handl
return (void *) addr;
}
-static void *minios_privcmd_map_foreign_range(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom,
- int size, int prot,
- unsigned long mfn)
+void *xc_map_foreign_range(xc_interface *xch,
+ uint32_t dom,
+ int size, int prot,
+ unsigned long mfn)
{
unsigned long pt_prot = 0;
@@ -142,10 +143,10 @@ static void *minios_privcmd_map_foreign_range(xc_interface *xch, xc_osdep_handle
return map_frames_ex(&mfn, size / getpagesize(), 0, 1, 1, dom, NULL, pt_prot);
}
-static void *minios_privcmd_map_foreign_ranges(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom,
- size_t size, int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[], int nentries)
+void *xc_map_foreign_ranges(xc_interface *xch,
+ uint32_t dom,
+ size_t size, int prot, size_t chunksize,
+ privcmd_mmap_entry_t entries[], int nentries)
{
unsigned long *mfns;
int i, j, n;
@@ -169,24 +170,6 @@ static void *minios_privcmd_map_foreign_ranges(xc_interface *xch, xc_osdep_handl
return ret;
}
-
-static struct xc_osdep_ops minios_privcmd_ops = {
- .open = &minios_privcmd_open,
- .close = &minios_privcmd_close,
-
- .u.privcmd = {
- .alloc_hypercall_buffer = &minios_privcmd_alloc_hypercall_buffer,
- .free_hypercall_buffer = &minios_privcmd_free_hypercall_buffer,
-
- .hypercall = &minios_privcmd_hypercall,
-
- .map_foreign_batch = &minios_privcmd_map_foreign_batch,
- .map_foreign_bulk = &minios_privcmd_map_foreign_bulk,
- .map_foreign_range = &minios_privcmd_map_foreign_range,
- .map_foreign_ranges = &minios_privcmd_map_foreign_ranges,
- },
-};
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
@@ -199,23 +182,6 @@ void *xc_memalign(xc_interface *xch, size_t alignment, size_t size)
return memalign(alignment, size);
}
-static struct xc_osdep_ops *minios_osdep_init(xc_interface *xch, enum xc_osdep_type type)
-{
- switch ( type )
- {
- case XC_OSDEP_PRIVCMD:
- return &minios_privcmd_ops;
- default:
- return NULL;
- }
-}
-
-xc_osdep_info_t xc_osdep_info = {
- .name = "Minios Native OS interface",
- .init = &minios_osdep_init,
- .fake = 0,
-};
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xc_netbsd.c b/tools/libxc/xc_netbsd.c
index fe4f0a1..9abb3b6 100644
--- a/tools/libxc/xc_netbsd.c
+++ b/tools/libxc/xc_netbsd.c
@@ -24,7 +24,7 @@
#include <malloc.h>
#include <sys/mman.h>
-static xc_osdep_handle netbsd_privcmd_open(xc_interface *xch)
+int osdep_privcmd_open(xc_interface *xch)
{
int flags, saved_errno;
int fd = open("/kern/xen/privcmd", O_RDWR);
@@ -32,7 +32,7 @@ static xc_osdep_handle netbsd_privcmd_open(xc_interface *xch)
if ( fd == -1 )
{
PERROR("Could not obtain handle on privileged command interface");
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
}
/* Although we return the file handle as the 'xc handle' the API
@@ -51,22 +51,23 @@ static xc_osdep_handle netbsd_privcmd_open(xc_interface *xch)
goto error;
}
- return (xc_osdep_handle)fd;
+ xch->privcmdfd = fd;
+ return 0;
error:
saved_errno = errno;
close(fd);
errno = saved_errno;
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
}
-static int netbsd_privcmd_close(xc_interface *xch, xc_osdep_handle h)
+int osdep_privcmd_close(xc_interface *xch)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
return close(fd);
}
-static void *netbsd_privcmd_alloc_hypercall_buffer(xc_interface *xch, xc_osdep_handle h, int npages)
+void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
{
size_t size = npages * XC_PAGE_SIZE;
void *p;
@@ -83,15 +84,15 @@ static void *netbsd_privcmd_alloc_hypercall_buffer(xc_interface *xch, xc_osdep_h
return p;
}
-static void netbsd_privcmd_free_hypercall_buffer(xc_interface *xch, xc_osdep_handle h, void *ptr, int npages)
+void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
{
(void) munlock(ptr, npages * XC_PAGE_SIZE);
free(ptr);
}
-static int netbsd_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h, privcmd_hypercall_t *hypercall)
+int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
int error = ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
/*
@@ -106,11 +107,18 @@ static int netbsd_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h, privcm
return hypercall->retval;
}
-static void *netbsd_privcmd_map_foreign_batch(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
+void *xc_map_foreign_bulk(xc_interface *xch,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num)
{
- int fd = (int)h;
+ return xc_map_foreign_bulk_compat(xch, dom, prot, arr, err, num);
+}
+
+void *xc_map_foreign_batch(xc_interface *xch,
+ uint32_t dom, int prot,
+ xen_pfn_t *arr, int num)
+{
+ int fd = xch->privcmdfd;
privcmd_mmapbatch_t ioctlx;
void *addr;
addr = mmap(NULL, num*XC_PAGE_SIZE, prot, MAP_ANON | MAP_SHARED, -1, 0);
@@ -135,12 +143,12 @@ static void *netbsd_privcmd_map_foreign_batch(xc_interface *xch, xc_osdep_handle
}
-static void *netbsd_privcmd_map_foreign_range(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom,
- int size, int prot,
- unsigned long mfn)
+void *xc_map_foreign_range(xc_interface *xch,
+ uint32_t dom,
+ int size, int prot,
+ unsigned long mfn)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
privcmd_mmap_t ioctlx;
privcmd_mmap_entry_t entry;
void *addr;
@@ -167,12 +175,12 @@ static void *netbsd_privcmd_map_foreign_range(xc_interface *xch, xc_osdep_handle
return addr;
}
-static void *netbsd_privcmd_map_foreign_ranges(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom,
- size_t size, int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[], int nentries)
+void *xc_map_foreign_ranges(xc_interface *xch,
+ uint32_t dom,
+ size_t size, int prot, size_t chunksize,
+ privcmd_mmap_entry_t entries[], int nentries)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
privcmd_mmap_t ioctlx;
int i, rc;
void *addr;
@@ -205,23 +213,6 @@ mmap_failed:
return NULL;
}
-static struct xc_osdep_ops netbsd_privcmd_ops = {
- .open = &netbsd_privcmd_open,
- .close = &netbsd_privcmd_close,
-
- .u.privcmd = {
- .alloc_hypercall_buffer = &netbsd_privcmd_alloc_hypercall_buffer,
- .free_hypercall_buffer = &netbsd_privcmd_free_hypercall_buffer,
-
- .hypercall = &netbsd_privcmd_hypercall,
-
- .map_foreign_batch = &netbsd_privcmd_map_foreign_batch,
- .map_foreign_bulk = &xc_map_foreign_bulk_compat,
- .map_foreign_range = &netbsd_privcmd_map_foreign_range,
- .map_foreign_ranges = &netbsd_privcmd_map_foreign_ranges,
- },
-};
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
@@ -262,23 +253,6 @@ void *xc_memalign(xc_interface *xch, size_t alignment, size_t size)
return valloc(size);
}
-static struct xc_osdep_ops *netbsd_osdep_init(xc_interface *xch, enum xc_osdep_type type)
-{
- switch ( type )
- {
- case XC_OSDEP_PRIVCMD:
- return &netbsd_privcmd_ops;
- default:
- return NULL;
- }
-}
-
-xc_osdep_info_t xc_osdep_info = {
- .name = "Netbsd Native OS interface",
- .init = &netbsd_osdep_init,
- .fake = 0,
-};
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xc_private.c b/tools/libxc/xc_private.c
index 10b2367..309eaad 100644
--- a/tools/libxc/xc_private.c
+++ b/tools/libxc/xc_private.c
@@ -26,111 +26,12 @@
#include <pthread.h>
#include <assert.h>
-#ifndef __MINIOS__
-#include <dlfcn.h>
-#endif
-
-#define XENCTRL_OSDEP "XENCTRL_OSDEP"
-
-#if !defined (__MINIOS__) && !defined(__RUMPUSER_XEN__) && !defined(__RUMPRUN__)
-#define DO_DYNAMIC_OSDEP
-#endif
-
-/*
- * Returns a (shallow) copy of the xc_osdep_info_t for the
- * active OS interface.
- *
- * On success a handle to the relevant library is opened. The user
- * must subsequently call xc_osdep_put_info() when it is
- * finished with the library.
- *
- * Logs IFF xch != NULL.
- *
- * Returns:
- * 0 - on success
- * -1 - on error
- */
-static int xc_osdep_get_info(xc_interface *xch, xc_osdep_info_t *info)
-{
- int rc = -1;
-#ifdef DO_DYNAMIC_OSDEP
- const char *lib = getenv(XENCTRL_OSDEP);
- xc_osdep_info_t *pinfo;
- void *dl_handle = NULL;
-
- if ( lib != NULL )
- {
- if ( getuid() != geteuid() )
- {
- if ( xch ) ERROR("cannot use %s=%s with setuid application", XENCTRL_OSDEP, lib);
- abort();
- }
- if ( getgid() != getegid() )
- {
- if ( xch ) ERROR("cannot use %s=%s with setgid application", XENCTRL_OSDEP, lib);
- abort();
- }
-
- dl_handle = dlopen(lib, RTLD_LAZY|RTLD_LOCAL);
- if ( !dl_handle )
- {
- if ( xch ) ERROR("unable to open osdep library %s: %s", lib, dlerror());
- goto out;
- }
-
- pinfo = dlsym(dl_handle, "xc_osdep_info");
- if ( !pinfo )
- {
- if ( xch ) ERROR("unable to find xc_osinteface_info in %s: %s", lib, dlerror());
- goto out;
- }
-
- *info = *pinfo;
- info->dl_handle = dl_handle;
- }
- else
-#endif /*DO_DYNAMIC_OSDEP*/
- {
- *info = xc_osdep_info;
- info->dl_handle = NULL;
- }
-
- rc = 0;
-
-#ifdef DO_DYNAMIC_OSDEP
-out:
- if ( dl_handle && rc == -1 )
- dlclose(dl_handle);
-#endif /*DO_DYNAMIC_OSDEP*/
-
- return rc;
-}
-
-static void xc_osdep_put(xc_osdep_info_t *info)
-{
-#ifdef DO_DYNAMIC_OSDEP
- if ( info->dl_handle )
- dlclose(info->dl_handle);
-#endif /*DO_DYNAMIC_OSDEP*/
-}
-
-static const char *xc_osdep_type_name(enum xc_osdep_type type)
-{
- switch ( type )
- {
- case XC_OSDEP_PRIVCMD: return "privcmd";
- }
- return "unknown";
-}
-
-static struct xc_interface_core *xc_interface_open_common(xentoollog_logger *logger,
- xentoollog_logger *dombuild_logger,
- unsigned open_flags,
- enum xc_osdep_type type)
+struct xc_interface_core *xc_interface_open(xentoollog_logger *logger,
+ xentoollog_logger *dombuild_logger,
+ unsigned open_flags)
{
struct xc_interface_core xch_buf, *xch = &xch_buf;
- xch->type = type;
xch->flags = open_flags;
xch->dombuild_logger_file = 0;
xc_clear_last_error(xch);
@@ -148,9 +49,6 @@ static struct xc_interface_core *xc_interface_open_common(xentoollog_logger *log
xch->hypercall_buffer_cache_misses = 0;
xch->hypercall_buffer_cache_toobig = 0;
- xch->ops_handle = XC_OSDEP_OPEN_ERROR;
- xch->ops = NULL;
-
if (!xch->error_handler) {
xch->error_handler = xch->error_handler_tofree =
(xentoollog_logger*)
@@ -168,40 +66,26 @@ static struct xc_interface_core *xc_interface_open_common(xentoollog_logger *log
*xch = xch_buf;
if (!(open_flags & XC_OPENFLAG_DUMMY)) {
- if ( xc_osdep_get_info(xch, &xch->osdep) < 0 )
+ if ( osdep_privcmd_open(xch) < 0 )
goto err;
-
- xch->ops = xch->osdep.init(xch, type);
- if ( xch->ops == NULL )
- {
- DPRINTF("OSDEP: interface %d (%s) not supported on this platform",
- type, xc_osdep_type_name(type));
- goto err_put_iface;
- }
-
- xch->ops_handle = xch->ops->open(xch);
- if (xch->ops_handle == XC_OSDEP_OPEN_ERROR)
- goto err_put_iface;
}
return xch;
-err_put_iface:
- xc_osdep_put(&xch->osdep);
err:
xtl_logger_destroy(xch->error_handler_tofree);
if (xch != &xch_buf) free(xch);
return NULL;
}
-static int xc_interface_close_common(xc_interface *xch)
+int xc_interface_close(xc_interface *xch)
{
int rc = 0;
if (!xch)
- return 0;
+ return 0;
- rc = xch->ops->close(xch, xch->ops_handle);
+ rc = osdep_privcmd_close(xch);
if (rc) PERROR("Could not close hypervisor interface");
xc__hypercall_buffer_cache_release(xch);
@@ -213,42 +97,6 @@ static int xc_interface_close_common(xc_interface *xch)
return rc;
}
-int xc_interface_is_fake(void)
-{
- xc_osdep_info_t info;
-
- if ( xc_osdep_get_info(NULL, &info) < 0 )
- return -1;
-
- /* Have a copy of info so can release the interface now. */
- xc_osdep_put(&info);
-
- return info.fake;
-}
-
-xc_interface *xc_interface_open(xentoollog_logger *logger,
- xentoollog_logger *dombuild_logger,
- unsigned open_flags)
-{
- xc_interface *xch;
-
- xch = xc_interface_open_common(logger, dombuild_logger, open_flags,
- XC_OSDEP_PRIVCMD);
-
- return xch;
-}
-
-int xc_interface_close(xc_interface *xch)
-{
- return xc_interface_close_common(xch);
-}
-
-
-int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
-{
- return xch->ops->u.privcmd.hypercall(xch, xch->ops_handle, hypercall);
-}
-
static pthread_key_t errbuf_pkey;
static pthread_once_t errbuf_pkey_once = PTHREAD_ONCE_INIT;
@@ -335,14 +183,6 @@ void xc_report_error(xc_interface *xch, int code, const char *fmt, ...)
va_end(args);
}
-void xc_osdep_log(xc_interface *xch, xentoollog_level level, int code, const char *fmt, ...)
-{
- va_list args;
- va_start(args, fmt);
- xc_reportv(xch, xch->error_handler, level, code, fmt, args);
- va_end(args);
-}
-
const char *xc_set_progress_prefix(xc_interface *xch, const char *doing)
{
const char *old = xch->currently_progress_reporting;
diff --git a/tools/libxc/xc_private.h b/tools/libxc/xc_private.h
index 46e0e75..9f846b8 100644
--- a/tools/libxc/xc_private.h
+++ b/tools/libxc/xc_private.h
@@ -30,7 +30,6 @@
#include "_paths.h"
#include "xenctrl.h"
-#include "xenctrlosdep.h"
#include <xen/sys/privcmd.h>
@@ -88,7 +87,6 @@ struct iovec {
#define MAX_PAGECACHE_USAGE (4*1024)
struct xc_interface_core {
- enum xc_osdep_type type;
int flags;
xentoollog_logger *error_handler, *error_handler_tofree;
xentoollog_logger *dombuild_logger, *dombuild_logger_tofree;
@@ -117,12 +115,21 @@ struct xc_interface_core {
int hypercall_buffer_cache_misses;
int hypercall_buffer_cache_toobig;
- /* Low lovel OS interface */
- xc_osdep_info_t osdep;
- xc_osdep_ops *ops; /* backend operations */
- xc_osdep_handle ops_handle; /* opaque data for xc_osdep_ops */
+ /* Privcmd interface */
+ int privcmdfd;
};
+int osdep_privcmd_open(xc_interface *xch);
+int osdep_privcmd_close(xc_interface *xch);
+
+void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages);
+void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages);
+
+/* Stub for not yet converted OSes */
+void *xc_map_foreign_bulk_compat(xc_interface *xch,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num);
+
void xc_report_error(xc_interface *xch, int code, const char *fmt, ...)
__attribute__((format(printf,3,4)));
void xc_reportv(xc_interface *xch, xentoollog_logger *lg, xentoollog_level,
diff --git a/tools/libxc/xc_solaris.c b/tools/libxc/xc_solaris.c
index ed7987c..6f84b82 100644
--- a/tools/libxc/xc_solaris.c
+++ b/tools/libxc/xc_solaris.c
@@ -24,7 +24,7 @@
#include <fcntl.h>
#include <malloc.h>
-static xc_osdep_handle solaris_privcmd_open(xc_interface *xch)
+int osdep_privcmd_open(xc_interface *xch)
{
int flags, saved_errno;
int fd = open("/dev/xen/privcmd", O_RDWR);
@@ -32,7 +32,7 @@ static xc_osdep_handle solaris_privcmd_open(xc_interface *xch)
if ( fd == -1 )
{
PERROR("Could not obtain handle on privileged command interface");
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
}
/* Although we return the file handle as the 'xc handle' the API
@@ -51,42 +51,43 @@ static xc_osdep_handle solaris_privcmd_open(xc_interface *xch)
goto error;
}
- return (xc_osdep_handle)fd;
+ xch->privcmdfd = fd;
+ return 0;
error:
saved_errno = errno;
close(fd);
errno = saved_errno;
- return XC_OSDEP_OPEN_ERROR;
+ return -1;
}
-static int solaris_privcmd_close(xc_interface *xch, xc_osdep_handle h)
+int osdep_privcmd_close(xc_interface *xch)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
return close(fd);
}
-static void *solaris_privcmd_alloc_hypercall_buffer(xc_interface *xch, xc_osdep_handle h, int npages)
+void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
{
return xc_memalign(xch, XC_PAGE_SIZE, npages * XC_PAGE_SIZE);
}
-static void solaris_privcmd_free_hypercall_buffer(xc_interface *xch, xc_osdep_handle h, void *ptr, int npages)
+static void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
{
free(ptr);
}
-static int solaris_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h, privcmd_hypercall_t *hypercall)
+int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
return ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
}
-static void *solaris_privcmd_map_foreign_batch(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
+void *xc_map_foreign_batch(xc_interface *xch,
+ uint32_t dom, int prot,
+ xen_pfn_t *arr, int num)
{
- int fd = (int)h;
+ int fd = xch->privcmdfd;
privcmd_mmapbatch_t ioctlx;
void *addr;
addr = mmap(NULL, num*XC_PAGE_SIZE, prot, MAP_SHARED, fd, 0);
@@ -109,12 +110,19 @@ static void *solaris_privcmd_map_foreign_batch(xc_interface *xch, xc_osdep_handl
}
-static void *xc_map_foreign_range(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom,
- int size, int prot,
- unsigned long mfn)
+void *xc_map_foreign_bulk(xc_interface *xch,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num)
{
- int fd = (int)fd;
+ return xc_map_foreign_bulk_compat(xch, dom, prot, arr, err, num);
+}
+
+void *xc_map_foreign_range(xc_interface *xch,
+ uint32_t dom,
+ int size, int prot,
+ unsigned long mfn)
+{
+ int fd = xch->privcmdfd;
privcmd_mmap_t ioctlx;
privcmd_mmap_entry_t entry;
void *addr;
@@ -138,12 +146,12 @@ static void *xc_map_foreign_range(xc_interface *xch, xc_osdep_handle h,
return addr;
}
-static void *solaric_privcmd_map_foreign_ranges(xc_interface *xch, xc_osdep_handle h,
- uint32_t dom,
- size_t size, int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[], int nentries)
+void *xc_map_foreign_ranges(xc_interface *xch,
+ uint32_t dom,
+ size_t size, int prot, size_t chunksize,
+ privcmd_mmap_entry_t entries[], int nentries)
{
- int fd = (int)fd;
+ int fd = xch->privcmdfd;
privcmd_mmap_t ioctlx;
int i, rc;
void *addr;
@@ -176,23 +184,6 @@ mmap_failed:
return NULL;
}
-static struct xc_osdep_ops solaris_privcmd_ops = {
- .open = &solaris_privcmd_open,
- .close = &solaris_privcmd_close,
-
- .u.privcmd = {
- .alloc_hypercall_buffer = &solaris_privcmd_alloc_hypercall_buffer,
- .free_hypercall_buffer = &solaris_privcmd_free_hypercall_buffer,
-
- .hypercall = &solaris_privcmd_hypercall;
-
- .map_foreign_batch = &solaris_privcmd_map_foreign_batch,
- .map_foreign_bulk = &xc_map_foreign_bulk_compat,
- .map_foreign_range = &solaris_privcmd_map_foreign_range,
- .map_foreign_ranges = &solaris_privcmd_map_foreign_ranges,
- },
-};
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
@@ -204,23 +195,6 @@ void *xc_memalign(xc_interface *xch, size_t alignment, size_t size)
return memalign(alignment, size);
}
-static struct xc_osdep_ops *solaris_osdep_init(xc_interface *xch, enum xc_osdep_type type)
-{
- switch ( type )
- {
- case XC_OSDEP_PRIVCMD:
- return &solaris_privcmd_ops;
- default:
- return NULL;
- }
-}
-
-xc_osdep_info_t xc_osdep_info = {
- .name = "Solaris Native OS interface",
- .init = &solaris_osdep_init,
- .fake = 0,
-};
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xenctrl_osdep_ENOSYS.c b/tools/libxc/xenctrl_osdep_ENOSYS.c
deleted file mode 100644
index 5182532..0000000
--- a/tools/libxc/xenctrl_osdep_ENOSYS.c
+++ /dev/null
@@ -1,123 +0,0 @@
-/* Dummy backend which just logs and returns ENOSYS. */
-
-#include <errno.h>
-#include <inttypes.h>
-#include <stdlib.h>
-
-#include "xenctrl.h"
-#include "xenctrlosdep.h"
-
-#define IPRINTF(_x, _f, _a...) xc_osdep_log(_x,XTL_INFO,0, _f , ## _a)
-
-#define ERROR(_x, _m, _a...) xc_osdep_log(_x,XTL_ERROR,XC_INTERNAL_ERROR,_m , ## _a )
-#define PERROR(_x, _m, _a...) xc_osdep_log(_x,XTL_ERROR,XC_INTERNAL_ERROR,_m \
- " (%d = %s)", ## _a , errno, xc_strerror(xch, errno))
-
-static xc_osdep_handle ENOSYS_privcmd_open(xc_interface *xch)
-{
- IPRINTF(xch, "ENOSYS_privcmd: opening handle %d\n", 1);
- return (xc_osdep_handle)1; /*dummy*/
-}
-
-static int ENOSYS_privcmd_close(xc_interface *xch, xc_osdep_handle h)
-{
- IPRINTF(xch, "ENOSYS_privcmd: closing handle %lx\n", h);
- return 0;
-}
-
-static int ENOSYS_privcmd_hypercall(xc_interface *xch, xc_osdep_handle h, privcmd_hypercall_t *hypercall)
-{
-#if defined(__FreeBSD__) || defined(__NetBSD__)
- IPRINTF(xch, "ENOSYS_privcmd %lx: hypercall: %02lu(%#lx,%#lx,%#lx,%#lx,%#lx)\n",
-#else
- IPRINTF(xch, "ENOSYS_privcmd %lx: hypercall: %02lld(%#llx,%#llx,%#llx,%#llx,%#llx)\n",
-#endif
- h, hypercall->op,
- hypercall->arg[0], hypercall->arg[1], hypercall->arg[2],
- hypercall->arg[3], hypercall->arg[4]);
- return -ENOSYS;
-}
-
-static void *ENOSYS_privcmd_map_foreign_batch(xc_interface *xch, xc_osdep_handle h, uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
-{
- IPRINTF(xch, "ENOSYS_privcmd %lx: map_foreign_batch: dom%d prot %#x arr %p num %d\n", h, dom, prot, arr, num);
- return MAP_FAILED;
-}
-
-static void *ENOSYS_privcmd_map_foreign_bulk(xc_interface *xch, xc_osdep_handle h, uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
-{
- IPRINTF(xch, "ENOSYS_privcmd %lx map_foreign_buld: dom%d prot %#x arr %p err %p num %d\n", h, dom, prot, arr, err, num);
- return MAP_FAILED;
-}
-
-static void *ENOSYS_privcmd_map_foreign_range(xc_interface *xch, xc_osdep_handle h, uint32_t dom, int size, int prot,
- unsigned long mfn)
-{
- IPRINTF(xch, "ENOSYS_privcmd %lx map_foreign_range: dom%d size %#x prot %#x mfn %ld\n", h, dom, size, prot, mfn);
- return MAP_FAILED;
-}
-
-static void *ENOSYS_privcmd_map_foreign_ranges(xc_interface *xch, xc_osdep_handle h, uint32_t dom, size_t size, int prot,
- size_t chunksize, privcmd_mmap_entry_t entries[],
- int nentries)
-{
- IPRINTF(xch, "ENOSYS_privcmd %lx map_foreign_ranges: dom%d size %zd prot %#x chunksize %zd entries %p num %d\n", h, dom, size, prot, chunksize, entries, nentries);
- return MAP_FAILED;
-}
-
-static struct xc_osdep_ops ENOSYS_privcmd_ops =
-{
- .open = &ENOSYS_privcmd_open,
- .close = &ENOSYS_privcmd_close,
- .u.privcmd = {
- .hypercall = &ENOSYS_privcmd_hypercall,
-
- .map_foreign_batch = &ENOSYS_privcmd_map_foreign_batch,
- .map_foreign_bulk = &ENOSYS_privcmd_map_foreign_bulk,
- .map_foreign_range = &ENOSYS_privcmd_map_foreign_range,
- .map_foreign_ranges = &ENOSYS_privcmd_map_foreign_ranges,
- }
-};
-
-static struct xc_osdep_ops * ENOSYS_osdep_init(xc_interface *xch, enum xc_osdep_type type)
-{
- struct xc_osdep_ops *ops;
-
- if (getenv("ENOSYS") == NULL)
- {
- PERROR(xch, "ENOSYS: not configured\n");
- return NULL;
- }
-
- switch ( type )
- {
- case XC_OSDEP_PRIVCMD:
- ops = &ENOSYS_privcmd_ops;
- break;
- default:
- ops = NULL;
- break;
- }
-
- IPRINTF(xch, "ENOSYS_osdep_init: initialising handle ops at %p\n", ops);
-
- return ops;
-}
-
-xc_osdep_info_t xc_osdep_info = {
- .name = "Pessimistic ENOSYS OS interface",
- .init = &ENOSYS_osdep_init,
- .fake = 1,
-};
-
-/*
- * Local variables:
- * mode: C
- * c-file-style: "BSD"
- * c-basic-offset: 4
- * tab-width: 4
- * indent-tabs-mode: nil
- * End:
- */
diff --git a/tools/ocaml/libs/xc/xenctrl.ml b/tools/ocaml/libs/xc/xenctrl.ml
index b7ba8b7..701b66c 100644
--- a/tools/ocaml/libs/xc/xenctrl.ml
+++ b/tools/ocaml/libs/xc/xenctrl.ml
@@ -108,8 +108,6 @@ external sizeof_xen_pfn: unit -> int = "stub_sizeof_xen_pfn"
external interface_open: unit -> handle = "stub_xc_interface_open"
external interface_close: handle -> unit = "stub_xc_interface_close"
-external is_fake: unit -> bool = "stub_xc_interface_is_fake"
-
let with_intf f =
let xc = interface_open () in
let r = try f xc with exn -> interface_close xc; raise exn in
diff --git a/tools/ocaml/libs/xc/xenctrl.mli b/tools/ocaml/libs/xc/xenctrl.mli
index bc4af56..3f4947c 100644
--- a/tools/ocaml/libs/xc/xenctrl.mli
+++ b/tools/ocaml/libs/xc/xenctrl.mli
@@ -72,7 +72,6 @@ external sizeof_vcpu_guest_context : unit -> int
= "stub_sizeof_vcpu_guest_context"
external sizeof_xen_pfn : unit -> int = "stub_sizeof_xen_pfn"
external interface_open : unit -> handle = "stub_xc_interface_open"
-external is_fake : unit -> bool = "stub_xc_interface_is_fake"
external interface_close : handle -> unit = "stub_xc_interface_close"
val with_intf : (handle -> 'a) -> 'a
val domain_create : handle -> int32 -> domain_create_flag list -> string -> domid
diff --git a/tools/ocaml/libs/xc/xenctrl_stubs.c b/tools/ocaml/libs/xc/xenctrl_stubs.c
index b7de615..29055cc 100644
--- a/tools/ocaml/libs/xc/xenctrl_stubs.c
+++ b/tools/ocaml/libs/xc/xenctrl_stubs.c
@@ -126,13 +126,6 @@ CAMLprim value stub_xc_interface_open(void)
}
-CAMLprim value stub_xc_interface_is_fake(void)
-{
- CAMLparam0();
- int is_fake = xc_interface_is_fake();
- CAMLreturn(Val_int(is_fake));
-}
-
CAMLprim value stub_xc_interface_close(value xch)
{
CAMLparam1(xch);
diff --git a/tools/ocaml/xenstored/domains.ml b/tools/ocaml/xenstored/domains.ml
index 92e438f..395f3a9 100644
--- a/tools/ocaml/xenstored/domains.ml
+++ b/tools/ocaml/xenstored/domains.ml
@@ -65,11 +65,9 @@ let create xc doms domid mfn port =
Domain.bind_interdomain dom;
dom
-let create0 fake doms =
+let create0 doms =
let port, interface =
- if fake then (
- 0, Xenctrl.with_intf (fun xc -> Xenctrl.map_foreign_range xc 0 (Xenmmap.getpagesize()) 0n)
- ) else (
+ (
let port = Utils.read_file_single_integer Define.xenstored_proc_port
and fd = Unix.openfile Define.xenstored_proc_kva
[ Unix.O_RDWR ] 0o600 in
diff --git a/tools/ocaml/xenstored/xenstored.ml b/tools/ocaml/xenstored/xenstored.ml
index 42b8183..25c126a 100644
--- a/tools/ocaml/xenstored/xenstored.ml
+++ b/tools/ocaml/xenstored/xenstored.ml
@@ -176,7 +176,7 @@ let from_channel store cons doms chan =
if domid > 0 then
Domains.create xc doms domid mfn port
else
- Domains.create0 false doms
+ Domains.create0 doms
in
Connections.add_domain cons ndom;
in
@@ -278,8 +278,7 @@ let _ =
Store.mkdir store (Perms.Connection.create 0) localpath;
if cf.domain_init then (
- let usingxiu = Xenctrl.is_fake () in
- Connections.add_domain cons (Domains.create0 usingxiu domains);
+ Connections.add_domain cons (Domains.create0 domains);
Event.bind_dom_exc_virq eventchn
);
);
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (7 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 08/23] tools/libxc: Remove osdep indirection for privcmd Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 15:08 ` Ian Jackson
` (2 more replies)
2015-11-09 12:00 ` [PATCH XEN v5 10/23] tools/libxc: drop xc_map_foreign_bulk_compat wrappers Ian Campbell
` (14 subsequent siblings)
23 siblings, 3 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
libxencall will provide a stable API and ABI for calling hypercalls
(although those hypercalls themselves may not have a stable API). As
well as the hypercall buffer infrastructure needed in order to safely
provide pointer arguments to hypercalls.
libxenctrl encapsulates a instance of this interface, so users of that
library are not currently subjected to any actual changes. However all
hypercalls made internally by libxc now use the correct interface. It
is expected that most users of this library will be other libraries
providing a higher level interface, rather than applications directly.
Only the basic functionality to allocate hypercall safe memory is
moved, the type safe stuff and bounce buffers remain in libxc.
Note that the functionality to map foreign pages using privcmd is not
yet moved, meaning that an xc_interface will now contain two open
privcmd file descriptors. Foreign memory mapping is logically separate
functionality and will be moved into its own library.
The new library uses a version script to ensure that only expected
symbols are exported and to version them such that ABI guarantees can
be kept in the future.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
Must be applied with:
- "qemu-xen-traditional: Add libxencall to rpath-link" and a
corresponding QEMU_TAG update folded here.
- "mini-os: Include libxencall with libxc" and a corresponding bump
to MINIOS_UPSTREAM_REVISION folded in here.
v3: Moved to tools/libs/call
Ported new wrappers (altp2m)
v5: Allow _close(NULL).
---
.gitignore | 2 +
stubdom/Makefile | 20 +++-
tools/Makefile | 1 +
tools/Rules.mk | 7 +-
tools/libs/Makefile | 1 +
tools/libs/call/Makefile | 67 +++++++++++++
tools/libs/call/buffer.c | 192 ++++++++++++++++++++++++++++++++++++++
tools/libs/call/core.c | 147 +++++++++++++++++++++++++++++
tools/libs/call/freebsd.c | 140 +++++++++++++++++++++++++++
tools/libs/call/include/xencall.h | 84 +++++++++++++++++
tools/libs/call/libxencall.map | 19 ++++
tools/libs/call/linux.c | 132 ++++++++++++++++++++++++++
tools/libs/call/minios.c | 81 ++++++++++++++++
tools/libs/call/netbsd.c | 121 ++++++++++++++++++++++++
tools/libs/call/private.h | 68 ++++++++++++++
tools/libs/call/solaris.c | 97 +++++++++++++++++++
tools/libxc/Makefile | 7 +-
tools/libxc/xc_altp2m.c | 64 ++++---------
tools/libxc/xc_domain.c | 105 +++++++--------------
tools/libxc/xc_evtchn.c | 9 +-
tools/libxc/xc_flask.c | 8 +-
tools/libxc/xc_freebsd_osdep.c | 47 ----------
tools/libxc/xc_gnttab.c | 9 +-
tools/libxc/xc_hcall_buf.c | 138 ++-------------------------
tools/libxc/xc_kexec.c | 36 +++----
tools/libxc/xc_linux_osdep.c | 49 ----------
tools/libxc/xc_minios.c | 32 -------
tools/libxc/xc_misc.c | 79 ++++++----------
tools/libxc/xc_netbsd.c | 40 --------
tools/libxc/xc_private.c | 64 +++++--------
tools/libxc/xc_private.h | 76 +++++----------
tools/libxc/xc_solaris.c | 16 ----
tools/libxc/xc_tmem.c | 7 +-
tools/misc/Makefile | 4 +-
tools/xcutils/Makefile | 2 +-
tools/xenpaging/Makefile | 2 +-
36 files changed, 1341 insertions(+), 632 deletions(-)
create mode 100644 tools/libs/call/Makefile
create mode 100644 tools/libs/call/buffer.c
create mode 100644 tools/libs/call/core.c
create mode 100644 tools/libs/call/freebsd.c
create mode 100644 tools/libs/call/include/xencall.h
create mode 100644 tools/libs/call/libxencall.map
create mode 100644 tools/libs/call/linux.c
create mode 100644 tools/libs/call/minios.c
create mode 100644 tools/libs/call/netbsd.c
create mode 100644 tools/libs/call/private.h
create mode 100644 tools/libs/call/solaris.c
diff --git a/.gitignore b/.gitignore
index 9241c54..2899852 100644
--- a/.gitignore
+++ b/.gitignore
@@ -61,6 +61,7 @@ stubdom/xenstore
stubdom/libxentoollog-*
stubdom/libxenevtchn-*
stubdom/libxengnttab-*
+stubdom/libxencall-*
stubdom/libxc-*
stubdom/lwip-*
stubdom/mini-os-*
@@ -90,6 +91,7 @@ config/Docs.mk
tools/libs/toollog/headers.chk
tools/libs/evtchn/headers.chk
tools/libs/gnttab/headers.chk
+tools/libs/call/headers.chk
tools/blktap2/daemon/blktapctrl
tools/blktap2/drivers/img2qcow
tools/blktap2/drivers/lock-util
diff --git a/stubdom/Makefile b/stubdom/Makefile
index d4576eb..24f0e0f 100644
--- a/stubdom/Makefile
+++ b/stubdom/Makefile
@@ -330,6 +330,12 @@ mk-headers-$(XEN_TARGET_ARCH): $(IOEMU_LINKFARM_TARGET)
ln -sf $(XEN_ROOT)/tools/libs/gnttab/include/*.h include/ && \
ln -sf $(XEN_ROOT)/tools/libs/gnttab/*.c . && \
ln -sf $(XEN_ROOT)/tools/libs/gnttab/Makefile . )
+ mkdir -p libs-$(XEN_TARGET_ARCH)/call/include
+ [ -h libs-$(XEN_TARGET_ARCH)/call/Makefile ] || ( cd libs-$(XEN_TARGET_ARCH)/call && \
+ ln -sf $(XEN_ROOT)/tools/libs/call/*.h . && \
+ ln -sf $(XEN_ROOT)/tools/libs/call/include/*.h include/ && \
+ ln -sf $(XEN_ROOT)/tools/libs/call/*.c . && \
+ ln -sf $(XEN_ROOT)/tools/libs/call/Makefile . )
mkdir -p libxc-$(XEN_TARGET_ARCH)
[ -h libxc-$(XEN_TARGET_ARCH)/Makefile ] || ( cd libxc-$(XEN_TARGET_ARCH) && \
ln -sf $(XEN_ROOT)/tools/libxc/*.h . && \
@@ -386,12 +392,24 @@ libs-$(XEN_TARGET_ARCH)/gnttab/libxengnttab.a: $(NEWLIB_STAMPFILE)
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/gnttab
#######
+# libxencall
+#######
+
+.PHONY: libxencall
+libxencall: libs-$(XEN_TARGET_ARCH)/call/libxencall.a
+libs-$(XEN_TARGET_ARCH)/call/libxencall.a: $(NEWLIB_STAMPFILE)
+ $(MAKE) -C $(XEN_ROOT)/tools/include
+ $(MAKE) DESTDIR= -C $(MINI_OS) links
+ CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/call
+APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/call -whole-archive -lxencall -no-whole-archive
+
+#######
# libxc
#######
.PHONY: libxc
libxc: libxc-$(XEN_TARGET_ARCH)/libxenctrl.a libxc-$(XEN_TARGET_ARCH)/libxenguest.a
-libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog libxenevtchn libxengnttab cross-zlib
+libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog libxenevtchn libxengnttab libxencall cross-zlib
$(MAKE) -C $(XEN_ROOT)/tools/include
$(MAKE) DESTDIR= -C $(MINI_OS) links
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= CONFIG_LIBXC_MINIOS=y -C libxc-$(XEN_TARGET_ARCH)
diff --git a/tools/Makefile b/tools/Makefile
index f373e71..e5bbc98 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -263,6 +263,7 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/evtchn \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/gnttab \
+ -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/call \
$(QEMU_UPSTREAM_RPATH)" \
--bindir=$(LIBEXEC_BIN) \
--datadir=$(SHAREDIR)/qemu-xen \
diff --git a/tools/Rules.mk b/tools/Rules.mk
index 5824ede..39426f4 100644
--- a/tools/Rules.mk
+++ b/tools/Rules.mk
@@ -13,6 +13,7 @@ XEN_INCLUDE = $(XEN_ROOT)/tools/include
XEN_LIBXENTOOLLOG = $(XEN_ROOT)/tools/libs/toollog
XEN_LIBXENEVTCHN = $(XEN_ROOT)/tools/libs/evtchn
XEN_LIBXENGNTTAB = $(XEN_ROOT)/tools/libs/gnttab
+XEN_LIBXENCALL = $(XEN_ROOT)/tools/libs/call
XEN_LIBXC = $(XEN_ROOT)/tools/libxc
XEN_XENLIGHT = $(XEN_ROOT)/tools/libxl
XEN_XENSTORE = $(XEN_ROOT)/tools/xenstore
@@ -98,8 +99,12 @@ CFLAGS_libxengntshr = -I$(XEN_LIBXENGNTTAB)/include $(CFLAGS_xeninclude)
LDLIBS_libxengntshr = $(XEN_LIBXENGNTTAB)/libxengnttab$(libextension)
SHLIB_libxengntshr = -Wl,-rpath-link=$(XEN_LIBXENGNTTAB)
+CFLAGS_libxencall = -I$(XEN_LIBXENCALL)/include $(CFLAGS_xeninclude)
+LDLIBS_libxencall = $(XEN_LIBXENCALL)/libxencall$(libextension)
+SHLIB_libxencall = -Wl,-rpath-link=$(XEN_LIBXENCALL)
+
CFLAGS_libxenctrl = -I$(XEN_LIBXC)/include $(CFLAGS_libxentoollog) $(CFLAGS_xeninclude)
-SHDEPS_libxenctrl = $(SHLIB_libxentoollog) $(SHLIB_libxenevtchn) $(SHLIB_libxengnttab) $(SHLIB_libxengntshr)
+SHDEPS_libxenctrl = $(SHLIB_libxentoollog) $(SHLIB_libxenevtchn) $(SHLIB_libxengnttab) $(SHLIB_libxengntshr) $(SHLIB_libxencall)
LDLIBS_libxenctrl = $(SHDEPS_libxenctrl) $(XEN_LIBXC)/libxenctrl$(libextension)
SHLIB_libxenctrl = $(SHDEPS_libxenctrl) -Wl,-rpath-link=$(XEN_LIBXC)
diff --git a/tools/libs/Makefile b/tools/libs/Makefile
index 00156ae..f4f5d57 100644
--- a/tools/libs/Makefile
+++ b/tools/libs/Makefile
@@ -5,5 +5,6 @@ SUBDIRS-y :=
SUBDIRS-y += toollog
SUBDIRS-y += evtchn
SUBDIRS-y += gnttab
+SUBDIRS-y += call
all clean install distclean: %: subdirs-%
diff --git a/tools/libs/call/Makefile b/tools/libs/call/Makefile
new file mode 100644
index 0000000..5cbf62e
--- /dev/null
+++ b/tools/libs/call/Makefile
@@ -0,0 +1,67 @@
+XEN_ROOT = $(CURDIR)/../../..
+include $(XEN_ROOT)/tools/Rules.mk
+
+MAJOR = 1
+MINOR = 0
+SHLIB_LDFLAGS += -Wl,--version-script=libxencall.map
+
+CFLAGS += -Werror -Wmissing-prototypes
+CFLAGS += -I./include $(CFLAGS_xeninclude)
+CFLAGS += $(CFLAGS_libxentoollog)
+
+SRCS-y += core.c buffer.c
+SRCS-$(CONFIG_Linux) += linux.c
+SRCS-$(CONFIG_FreeBSD) += freebsd.c
+SRCS-$(CONFIG_SunOS) += solaris.c
+SRCS-$(CONFIG_NetBSD) += netbsd.c
+SRCS-$(CONFIG_MiniOS) += minios.c
+
+LIB_OBJS := $(patsubst %.c,%.o,$(SRCS-y))
+PIC_OBJS := $(patsubst %.c,%.opic,$(SRCS-y))
+
+LIB := libxencall.a
+ifneq ($(nosharedlibs),y)
+LIB += libxencall.so
+endif
+
+.PHONY: all
+all: build
+
+.PHONY: build
+build:
+ $(MAKE) libs
+
+.PHONY: libs
+libs: headers.chk $(LIB)
+
+headers.chk: $(wildcard include/*.h)
+
+libxencall.a: $(LIB_OBJS)
+ $(AR) rc $@ $^
+
+libxencall.so: libxencall.so.$(MAJOR)
+ $(SYMLINK_SHLIB) $< $@
+libxencall.so.$(MAJOR): libxencall.so.$(MAJOR).$(MINOR)
+ $(SYMLINK_SHLIB) $< $@
+
+libxencall.so.$(MAJOR).$(MINOR): $(PIC_OBJS) libxencall.map
+ $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxencall.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $(PIC_OBJS) $(LDLIBS_libxentoollog) $(APPEND_LDFLAGS)
+
+.PHONY: install
+install: build
+ $(INSTALL_DIR) $(DESTDIR)$(libdir)
+ $(INSTALL_DIR) $(DESTDIR)$(includedir)
+ $(INSTALL_SHLIB) libxencall.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)
+ $(INSTALL_DATA) libxencall.a $(DESTDIR)$(libdir)
+ $(SYMLINK_SHLIB) libxencall.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxencall.so.$(MAJOR)
+ $(SYMLINK_SHLIB) libxencall.so.$(MAJOR) $(DESTDIR)$(libdir)/libxencall.so
+ $(INSTALL_DATA) include/xencall.h $(DESTDIR)$(includedir)
+
+.PHONY: TAGS
+TAGS:
+ etags -t *.c *.h
+
+.PHONY: clean
+clean:
+ rm -rf *.rpm $(LIB) *~ $(DEPS) $(LIB_OBJS) $(PIC_OBJS)
+ rm -f headers.chk
diff --git a/tools/libs/call/buffer.c b/tools/libs/call/buffer.c
new file mode 100644
index 0000000..fbbd7ff
--- /dev/null
+++ b/tools/libs/call/buffer.c
@@ -0,0 +1,192 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <errno.h>
+#include <string.h>
+#include <pthread.h>
+
+#include "private.h"
+
+#define DBGPRINTF(_m...) \
+ xtl_log(xcall->logger, XTL_DEBUG, -1, "xencall:buffer", _m)
+
+#define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & ~((1UL<<(_w))-1))
+
+pthread_mutex_t cache_mutex = PTHREAD_MUTEX_INITIALIZER;
+
+static void cache_lock(xencall_handle *xcall)
+{
+ int saved_errno = errno;
+ if ( xcall->flags & XENCALL_OPENFLAG_NON_REENTRANT )
+ return;
+ pthread_mutex_lock(&cache_mutex);
+ /* Ignore pthread errors. */
+ errno = saved_errno;
+}
+
+static void cache_unlock(xencall_handle *xcall)
+{
+ int saved_errno = errno;
+ if ( xcall->flags & XENCALL_OPENFLAG_NON_REENTRANT )
+ return;
+ pthread_mutex_unlock(&cache_mutex);
+ /* Ignore pthread errors. */
+ errno = saved_errno;
+}
+
+static void *cache_alloc(xencall_handle *xcall, int nr_pages)
+{
+ void *p = NULL;
+
+ cache_lock(xcall);
+
+ xcall->buffer_total_allocations++;
+ xcall->buffer_current_allocations++;
+ if ( xcall->buffer_current_allocations > xcall->buffer_maximum_allocations )
+ xcall->buffer_maximum_allocations = xcall->buffer_current_allocations;
+
+ if ( nr_pages > 1 )
+ {
+ xcall->buffer_cache_toobig++;
+ }
+ else if ( xcall->buffer_cache_nr > 0 )
+ {
+ p = xcall->buffer_cache[--xcall->buffer_cache_nr];
+ xcall->buffer_cache_hits++;
+ }
+ else
+ {
+ xcall->buffer_cache_misses++;
+ }
+
+ cache_unlock(xcall);
+
+ return p;
+}
+
+static int cache_free(xencall_handle *xcall, void *p, int nr_pages)
+{
+ int rc = 0;
+
+ cache_lock(xcall);
+
+ xcall->buffer_total_releases++;
+ xcall->buffer_current_allocations--;
+
+ if ( nr_pages == 1 &&
+ xcall->buffer_cache_nr < BUFFER_CACHE_SIZE )
+ {
+ xcall->buffer_cache[xcall->buffer_cache_nr++] = p;
+ rc = 1;
+ }
+
+ cache_unlock(xcall);
+
+ return rc;
+}
+
+void buffer_release_cache(xencall_handle *xcall)
+{
+ void *p;
+
+ cache_lock(xcall);
+
+ DBGPRINTF("total allocations:%d total releases:%d",
+ xcall->buffer_total_allocations,
+ xcall->buffer_total_releases);
+ DBGPRINTF("current allocations:%d maximum allocations:%d",
+ xcall->buffer_current_allocations,
+ xcall->buffer_maximum_allocations);
+ DBGPRINTF("cache current size:%d",
+ xcall->buffer_cache_nr);
+ DBGPRINTF("cache hits:%d misses:%d toobig:%d",
+ xcall->buffer_cache_hits,
+ xcall->buffer_cache_misses,
+ xcall->buffer_cache_toobig);
+
+ while ( xcall->buffer_cache_nr > 0 )
+ {
+ p = xcall->buffer_cache[--xcall->buffer_cache_nr];
+ osdep_free_pages(xcall, p, 1);
+ }
+
+ cache_unlock(xcall);
+}
+
+void *xencall_alloc_buffer_pages(xencall_handle *xcall, int nr_pages)
+{
+ void *p = cache_alloc(xcall, nr_pages);
+
+ if ( !p )
+ p = osdep_alloc_pages(xcall, nr_pages);
+
+ if (!p)
+ return NULL;
+
+ memset(p, 0, nr_pages * PAGE_SIZE);
+
+ return p;
+}
+
+void xencall_free_buffer_pages(xencall_handle *xcall, void *p, int nr_pages)
+{
+ if ( p == NULL )
+ return;
+
+ if ( !cache_free(xcall, p, nr_pages) )
+ osdep_free_pages(xcall, p, nr_pages);
+}
+
+struct allocation_header {
+ int nr_pages;
+};
+
+void *xencall_alloc_buffer(xencall_handle *xcall, size_t size)
+{
+ size_t actual_size = ROUNDUP(size + sizeof(struct allocation_header), PAGE_SHIFT);
+ int nr_pages = actual_size >> PAGE_SHIFT;
+ struct allocation_header *hdr;
+
+ hdr = xencall_alloc_buffer_pages(xcall, nr_pages);
+ if ( hdr == NULL )
+ return NULL;
+
+ hdr->nr_pages = nr_pages;
+
+ return (void *)(hdr+1);
+}
+
+void xencall_free_buffer(xencall_handle *xcall, void *p)
+{
+ struct allocation_header *hdr;
+
+ if (p == NULL)
+ return;
+
+ hdr = p;
+ --hdr;
+
+ xencall_free_buffer_pages(xcall, hdr, hdr->nr_pages);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/call/core.c b/tools/libs/call/core.c
new file mode 100644
index 0000000..a342871
--- /dev/null
+++ b/tools/libs/call/core.c
@@ -0,0 +1,147 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <stdlib.h>
+
+#include "private.h"
+
+xencall_handle *xencall_open(xentoollog_logger *logger, unsigned open_flags)
+{
+ xencall_handle *xcall = malloc(sizeof(*xcall));
+ int rc;
+
+ if (!xcall) return NULL;
+
+ xcall->flags = open_flags;
+ xcall->buffer_cache_nr = 0;
+
+ xcall->buffer_total_allocations = 0;
+ xcall->buffer_total_releases = 0;
+ xcall->buffer_current_allocations = 0;
+ xcall->buffer_maximum_allocations = 0;
+ xcall->buffer_cache_hits = 0;
+ xcall->buffer_cache_misses = 0;
+ xcall->buffer_cache_toobig = 0;
+ xcall->logger = logger;
+ xcall->logger_tofree = NULL;
+
+ if (!xcall->logger) {
+ xcall->logger = xcall->logger_tofree =
+ (xentoollog_logger*)
+ xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
+ if (!xcall->logger) goto err;
+ }
+
+ rc = osdep_xencall_open(xcall);
+ if ( rc < 0 ) goto err;
+
+ return xcall;
+
+err:
+ osdep_xencall_close(xcall);
+ xtl_logger_destroy(xcall->logger_tofree);
+ free(xcall);
+ return NULL;
+}
+
+int xencall_close(xencall_handle *xcall)
+{
+ int rc;
+
+ if ( !xcall )
+ return 0;
+
+ rc = osdep_xencall_close(xcall);
+ buffer_release_cache(xcall);
+ xtl_logger_destroy(xcall->logger_tofree);
+ free(xcall);
+ return rc;
+}
+
+int xencall0(xencall_handle *xcall, unsigned int op)
+{
+ privcmd_hypercall_t call = {
+ .op = op,
+ };
+
+ return osdep_hypercall(xcall, &call);
+}
+
+int xencall1(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1)
+{
+ privcmd_hypercall_t call = {
+ .op = op,
+ .arg = { arg1 },
+ };
+
+ return osdep_hypercall(xcall, &call);
+}
+
+int xencall2(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1, uint64_t arg2)
+{
+ privcmd_hypercall_t call = {
+ .op = op,
+ .arg = { arg1, arg2 },
+ };
+
+ return osdep_hypercall(xcall, &call);
+}
+
+int xencall3(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1, uint64_t arg2, uint64_t arg3)
+{
+ privcmd_hypercall_t call = {
+ .op = op,
+ .arg = { arg1, arg2, arg3},
+ };
+
+ return osdep_hypercall(xcall, &call);
+}
+
+int xencall4(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1, uint64_t arg2, uint64_t arg3,
+ uint64_t arg4)
+{
+ privcmd_hypercall_t call = {
+ .op = op,
+ .arg = { arg1, arg2, arg3, arg4 },
+ };
+
+ return osdep_hypercall(xcall, &call);
+}
+
+int xencall5(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1, uint64_t arg2, uint64_t arg3,
+ uint64_t arg4, uint64_t arg5)
+{
+ privcmd_hypercall_t call = {
+ .op = op,
+ .arg = { arg1, arg2, arg3, arg4, arg5 },
+ };
+
+ return osdep_hypercall(xcall, &call);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/call/freebsd.c b/tools/libs/call/freebsd.c
new file mode 100644
index 0000000..548be40
--- /dev/null
+++ b/tools/libs/call/freebsd.c
@@ -0,0 +1,140 @@
+ /******************************************************************************
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split from xc_freebsd_osdep.c
+ */
+
+#include <errno.h>
+#include <fcntl.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#include <sys/mman.h>
+#include <sys/ioctl.h>
+
+#include "private.h"
+
+#define PRIVCMD_DEV "/dev/xen/privcmd"
+
+int osdep_xencall_open(xencall_handle *xcall)
+{
+ int flags, saved_errno;
+ int fd = open(PRIVCMD_DEV, O_RDWR);
+
+ if ( fd == -1 )
+ {
+ PERROR("Could not obtain handle on privileged command interface "
+ PRIVCMD_DEV);
+ return -1;
+ }
+
+ /*
+ * Although we return the file handle as the 'xc handle' the API
+ * does not specify / guarentee that this integer is in fact
+ * a file handle. Thus we must take responsiblity to ensure
+ * it doesn't propagate (ie leak) outside the process.
+ */
+ if ( (flags = fcntl(fd, F_GETFD)) < 0 )
+ {
+ PERROR("Could not get file handle flags");
+ goto error;
+ }
+
+ flags |= FD_CLOEXEC;
+
+ if ( fcntl(fd, F_SETFD, flags) < 0 )
+ {
+ PERROR("Could not set file handle flags");
+ goto error;
+ }
+
+ xcall->fd = fd;
+ return 0;
+
+ error:
+ saved_errno = errno;
+ close(fd);
+ errno = saved_errno;
+
+ return -1;
+}
+
+int osdep_xencall_close(xencall_handle *xcall)
+{
+ int fd = xcall->fd;
+ if ( fd == -1 )
+ return 0;
+ return close(fd);
+}
+
+int osdep_hypercall(xencall_handle *xcall, privcmd_hypercall_t *hypercall)
+{
+ int fd = xcall->fd;
+ int ret;
+
+ ret = ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
+
+ return (ret == 0) ? hypercall->retval : ret;
+}
+
+void *osdep_alloc_pages(xencall_handle *xcall, unsigned int npages)
+{
+ size_t size = npages * PAGE_SIZE;
+ void *p;
+
+ /* Address returned by mmap is page aligned. */
+ p = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
+ -1, 0);
+ if (p == NULL)
+ return NULL;
+
+ /*
+ * Since FreeBSD doesn't have the MAP_LOCKED flag,
+ * lock memory using mlock.
+ */
+ if ( mlock(p, size) < 0 )
+ {
+ munmap(p, size);
+ return NULL;
+ }
+
+ return p;
+}
+
+void osdep_free_pages(xencall_handle *xcall, void *ptr, unsigned int npages)
+{
+ int saved_errno = errno;
+ /* Unlock pages */
+ munlock(ptr, npages * PAGE_SIZE);
+
+ munmap(ptr, npages * PAGE_SIZE);
+ /* We MUST propagate the hypercall errno, not unmap call's. */
+ errno = saved_errno;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/call/include/xencall.h b/tools/libs/call/include/xencall.h
new file mode 100644
index 0000000..a0b3591
--- /dev/null
+++ b/tools/libs/call/include/xencall.h
@@ -0,0 +1,84 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+#ifndef XENCALL_H
+#define XENCALL_H
+
+/*
+ * This library allows you to make arbitrary hypercalls (subject to
+ * sufficient permission for the process and the domain itself). Note
+ * that while the library interface is stable the hypercalls are
+ * subject to their own rules.
+ */
+
+#include <stdint.h>
+#include <stddef.h>
+
+/* Callers who don't care don't need to #include <xentoollog.h> */
+typedef struct xentoollog_logger xentoollog_logger;
+
+typedef struct xencall_handle xencall_handle;
+
+/*
+ */
+#define XENCALL_OPENFLAG_NON_REENTRANT (1U<<0)
+
+/*
+ * Return a handle onto the hypercall driver. Logs errors.
+ */
+xencall_handle *xencall_open(xentoollog_logger *logger, unsigned open_flags);
+
+/*
+ * Close a handle previously allocated with xencall_open().
+ */
+int xencall_close(xencall_handle *xcall);
+
+/*
+ * Call hypercalls with varying numbers of arguments.
+ */
+int xencall0(xencall_handle *xcall, unsigned int op);
+int xencall1(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1);
+int xencall2(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1, uint64_t arg2);
+int xencall3(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1, uint64_t arg2, uint64_t arg3);
+int xencall4(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1, uint64_t arg2, uint64_t arg3,
+ uint64_t arg4);
+int xencall5(xencall_handle *xcall, unsigned int op,
+ uint64_t arg1, uint64_t arg2, uint64_t arg3,
+ uint64_t arg4, uint64_t arg5);
+
+/*
+ * Allocate and free memory which is suitable for use as a pointer
+ * argument to a hypercall.
+ */
+void *xencall_alloc_buffer_pages(xencall_handle *xcall, int nr_pages);
+void xencall_free_buffer_pages(xencall_handle *xcall, void *p, int nr_pages);
+
+void *xencall_alloc_buffer(xencall_handle *xcall, size_t size);
+void xencall_free_buffer(xencall_handle *xcall, void *p);
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/call/libxencall.map b/tools/libs/call/libxencall.map
new file mode 100644
index 0000000..2f96144
--- /dev/null
+++ b/tools/libs/call/libxencall.map
@@ -0,0 +1,19 @@
+VERS_1.0 {
+ global:
+ xencall_open;
+ xencall_close;
+
+ xencall0;
+ xencall1;
+ xencall2;
+ xencall3;
+ xencall4;
+ xencall5;
+ xencall6;
+
+ xencall_alloc_buffer;
+ xencall_free_buffer;
+ xencall_alloc_buffer_pages;
+ xencall_free_buffer_pages;
+ local: *; /* Do not expose anything by default */
+};
diff --git a/tools/libs/call/linux.c b/tools/libs/call/linux.c
new file mode 100644
index 0000000..906ca7e
--- /dev/null
+++ b/tools/libs/call/linux.c
@@ -0,0 +1,132 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_linus_osdep.c:
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ */
+
+#include <errno.h>
+#include <fcntl.h>
+#include <unistd.h>
+
+#include <sys/mman.h>
+#include <sys/ioctl.h>
+
+#include "private.h"
+
+int osdep_xencall_open(xencall_handle *xcall)
+{
+ int flags, saved_errno;
+ int fd = open("/proc/xen/privcmd", O_RDWR);
+
+ if ( fd == -1 )
+ {
+ PERROR("Could not obtain handle on privileged command interface");
+ return -1;
+ }
+
+ /* Although we return the file handle as the 'xc handle' the API
+ does not specify / guarentee that this integer is in fact
+ a file handle. Thus we must take responsiblity to ensure
+ it doesn't propagate (ie leak) outside the process */
+ if ( (flags = fcntl(fd, F_GETFD)) < 0 )
+ {
+ PERROR("Could not get file handle flags");
+ goto error;
+ }
+
+ flags |= FD_CLOEXEC;
+
+ if ( fcntl(fd, F_SETFD, flags) < 0 )
+ {
+ PERROR("Could not set file handle flags");
+ goto error;
+ }
+
+ xcall->fd = fd;
+ return 0;
+
+ error:
+ saved_errno = errno;
+ close(fd);
+ errno = saved_errno;
+ return -1;
+}
+
+int osdep_xencall_close(xencall_handle *xcall)
+{
+ int fd = xcall->fd;
+ if (fd == -1)
+ return 0;
+ return close(fd);
+}
+
+int osdep_hypercall(xencall_handle *xcall, privcmd_hypercall_t *hypercall)
+{
+ return ioctl(xcall->fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
+}
+
+void *osdep_alloc_pages(xencall_handle *xcall, unsigned int npages)
+{
+ size_t size = npages * PAGE_SIZE;
+ void *p;
+ int rc, saved_errno;
+
+ /* Address returned by mmap is page aligned. */
+ p = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_LOCKED, -1, 0);
+ if ( p == MAP_FAILED )
+ {
+ PERROR("xc_alloc_hypercall_buffer: mmap failed");
+ return NULL;
+ }
+
+ /* Do not copy the VMA to child process on fork. Avoid the page being COW
+ on hypercall. */
+ rc = madvise(p, npages * PAGE_SIZE, MADV_DONTFORK);
+ if ( rc < 0 )
+ {
+ PERROR("xc_alloc_hypercall_buffer: madvise failed");
+ goto out;
+ }
+
+ return p;
+
+out:
+ saved_errno = errno;
+ (void)munmap(p, size);
+ errno = saved_errno;
+ return NULL;
+}
+
+void osdep_free_pages(xencall_handle *xcall, void *ptr, unsigned int npages)
+{
+ int saved_errno = errno;
+ /* Recover the VMA flags. Maybe it's not necessary */
+ madvise(ptr, npages * PAGE_SIZE, MADV_DOFORK);
+
+ munmap(ptr, npages * PAGE_SIZE);
+ /* We MUST propagate the hypercall errno, not unmap call's. */
+ errno = saved_errno;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/call/minios.c b/tools/libs/call/minios.c
new file mode 100644
index 0000000..1cdc073
--- /dev/null
+++ b/tools/libs/call/minios.c
@@ -0,0 +1,81 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_minios.c
+ *
+ * Copyright 2007-2008 Samuel Thibault <samuel.thibault@eu.citrix.com>.
+ */
+
+#include <mini-os/types.h>
+#include <mini-os/os.h>
+#include <mini-os/mm.h>
+#include <mini-os/lib.h>
+
+#include <errno.h>
+#include <malloc.h>
+
+#include "private.h"
+
+int osdep_xencall_open(xencall_handle *xcall)
+{
+ /* No fd required */
+ return 0;
+}
+
+int osdep_xencall_close(xencall_handle *xcall)
+{
+ return 0;
+}
+
+int osdep_hypercall(xencall_handle *xcall, privcmd_hypercall_t *hypercall)
+{
+ multicall_entry_t call;
+ int i, ret;
+
+ call.op = hypercall->op;
+ for (i = 0; i < 5; i++)
+ call.args[i] = hypercall->arg[i];
+
+ ret = HYPERVISOR_multicall(&call, 1);
+
+ if (ret < 0) {
+ errno = -ret;
+ return -1;
+ }
+ if ((long) call.result < 0) {
+ errno = - (long) call.result;
+ return -1;
+ }
+ return call.result;
+}
+
+void *osdep_alloc_pages(xencall_handle *xcall, unsigned npages)
+{
+ return memalign(PAGE_SIZE, npages * PAGE_SIZE);
+}
+
+void osdep_free_pages(xencall_handle *xcall, void *ptr, unsigned npages)
+{
+ free(ptr);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/call/netbsd.c b/tools/libs/call/netbsd.c
new file mode 100644
index 0000000..12008fb
--- /dev/null
+++ b/tools/libs/call/netbsd.c
@@ -0,0 +1,121 @@
+/******************************************************************************
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split from xc_netbsd.c
+ */
+
+#include "xc_private.h"
+
+#include <unistd.h>
+#include <fcntl.h>
+#include <malloc.h>
+#include <sys/mman.h>
+
+int osdep_xencall_open(xencall_handle *xcall)
+{
+ int flags, saved_errno;
+ int fd = open("/kern/xen/privcmd", O_RDWR);
+
+ if ( fd == -1 )
+ {
+ PERROR("Could not obtain handle on privileged command interface");
+ return -1;
+ }
+
+ /* Although we return the file handle as the 'xc handle' the API
+ does not specify / guarentee that this integer is in fact
+ a file handle. Thus we must take responsiblity to ensure
+ it doesn't propagate (ie leak) outside the process */
+ if ( (flags = fcntl(fd, F_GETFD)) < 0 )
+ {
+ PERROR("Could not get file handle flags");
+ goto error;
+ }
+ flags |= FD_CLOEXEC;
+ if ( fcntl(fd, F_SETFD, flags) < 0 )
+ {
+ PERROR("Could not set file handle flags");
+ goto error;
+ }
+
+ xcall->fd = fd;
+ return 0;
+
+ error:
+ saved_errno = errno;
+ close(fd);
+ errno = saved_errno;
+ return -1;
+}
+
+int osdep_xencall_close(xencall_handle *xcall)
+{
+ int fd = xcall->fd;
+ return close(fd);
+}
+
+void *osdep_alloc_hypercall_buffer(xencall_handle *xcall, unsigned int npages)
+{
+ size_t size = npages * XC_PAGE_SIZE;
+ void *p;
+
+ p = xc_memalign(xcall, XC_PAGE_SIZE, size);
+ if (!p)
+ return NULL;
+
+ if ( mlock(p, size) < 0 )
+ {
+ free(p);
+ return NULL;
+ }
+ return p;
+}
+
+void osdep_free_hypercall_buffer(xencall_handle *xcall, void *ptr,
+ unsigned int npages)
+{
+ (void) munlock(ptr, npages * XC_PAGE_SIZE);
+ free(ptr);
+}
+
+int do_xen_hypercall(xencall_handle *xcall, privcmd_hypercall_t *hypercall)
+{
+ int fd = xcall->fd;
+ int error = ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
+
+ /*
+ * Since NetBSD ioctl can only return 0 on success or < 0 on
+ * error, if we want to return a value from ioctl we should
+ * do so by setting hypercall->retval, to mimic Linux ioctl
+ * implementation.
+ */
+ if (error < 0)
+ return error;
+ else
+ return hypercall->retval;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/call/private.h b/tools/libs/call/private.h
new file mode 100644
index 0000000..af727fd
--- /dev/null
+++ b/tools/libs/call/private.h
@@ -0,0 +1,68 @@
+#ifndef XENCALL_PRIVATE_H
+#define XENCALL_PRIVATE_H
+
+#include <xentoollog.h>
+
+#include <xencall.h>
+
+#include <xen/xen.h>
+#include <xen/sys/privcmd.h>
+
+#ifndef PAGE_SHIFT /* Mini-os, Yukk */
+#define PAGE_SHIFT 12
+#endif
+#ifndef __MINIOS__ /* Yukk */
+#define PAGE_SIZE (1UL << PAGE_SHIFT)
+#define PAGE_MASK (~(PAGE_SIZE-1))
+#endif
+
+struct xencall_handle {
+ xentoollog_logger *logger, *logger_tofree;
+ unsigned flags;
+ int fd;
+
+ /*
+ * A simple cache of unused, single page, hypercall buffers
+ *
+ * Protected by a global lock.
+ */
+#define BUFFER_CACHE_SIZE 4
+ int buffer_cache_nr;
+ void *buffer_cache[BUFFER_CACHE_SIZE];
+
+ /*
+ * Hypercall buffer statistics. All protected by the global
+ * buffer_cache lock.
+ */
+ int buffer_total_allocations;
+ int buffer_total_releases;
+ int buffer_current_allocations;
+ int buffer_maximum_allocations;
+ int buffer_cache_hits;
+ int buffer_cache_misses;
+ int buffer_cache_toobig;
+};
+
+int osdep_xencall_open(xencall_handle *xcall);
+int osdep_xencall_close(xencall_handle *xcall);
+
+int osdep_hypercall(xencall_handle *xcall, privcmd_hypercall_t *hypercall);
+
+void *osdep_alloc_pages(xencall_handle *xcall, unsigned int nr_pages);
+void osdep_free_pages(xencall_handle *xcall, void *p, unsigned int nr_pages);
+
+void buffer_release_cache(xencall_handle *xcall);
+
+#define PERROR(_f...) xtl_log(xcall->logger, XTL_ERROR, errno, "xencall", _f)
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/call/solaris.c b/tools/libs/call/solaris.c
new file mode 100644
index 0000000..972989d
--- /dev/null
+++ b/tools/libs/call/solaris.c
@@ -0,0 +1,97 @@
+/******************************************************************************
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split from xc_solaris.c
+ */
+
+#include "xc_private.h"
+
+#include <xen/memory.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <malloc.h>
+
+int osdep_xencall_open(xencall_handle *xcall)
+{
+ int flags, saved_errno;
+ int fd = open("/dev/xen/privcmd", O_RDWR);
+
+ if ( fd == -1 )
+ {
+ PERROR("Could not obtain handle on privileged command interface");
+ return -1;
+ }
+
+ /* Although we return the file handle as the 'xc handle' the API
+ does not specify / guarentee that this integer is in fact
+ a file handle. Thus we must take responsiblity to ensure
+ it doesn't propagate (ie leak) outside the process */
+ if ( (flags = fcntl(fd, F_GETFD)) < 0 )
+ {
+ PERROR("Could not get file handle flags");
+ goto error;
+ }
+ flags |= FD_CLOEXEC;
+ if ( fcntl(fd, F_SETFD, flags) < 0 )
+ {
+ PERROR("Could not set file handle flags");
+ goto error;
+ }
+
+ xcall->fd = fd;
+ return 0;
+
+ error:
+ saved_errno = errno;
+ close(fd);
+ errno = saved_errno;
+ return -1;
+}
+
+int osdep_xencall_close(xencall_handle *xcall)
+{
+ int fd = xcall->fd;
+ return close(fd);
+}
+
+void *osdep_alloc_hypercall_buffer(xencall_handle *xcall, unsigned int npages)
+{
+ return xc_memalign(xcall, XC_PAGE_SIZE, npages * XC_PAGE_SIZE);
+}
+
+void osdep_free_hypercall_buffer(xencall_handle *xcall, void *ptr,
+ unsigned int npages)
+{
+ free(ptr);
+}
+
+int do_xen_hypercall(xencall_handle *xcall, privcmd_hypercall_t *hypercall)
+{
+ int fd = xcall->fd;
+ return ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libxc/Makefile b/tools/libxc/Makefile
index 3305fdd..a122f73 100644
--- a/tools/libxc/Makefile
+++ b/tools/libxc/Makefile
@@ -122,6 +122,11 @@ GUEST_PIC_OBJS := $(patsubst %.c,%.opic,$(GUEST_SRCS-y))
$(CTRL_LIB_OBJS) $(GUEST_LIB_OBJS) \
$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS): CFLAGS += -include $(XEN_ROOT)/tools/config.h
+# libxenguest includes xc_private.h, so needs this despite not using
+# this functionality directly.
+$(CTRL_LIB_OBJS) $(GUEST_LIB_OBJS) \
+$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS): CFLAGS += $(CFLAGS_libxencall)
+
$(CTRL_LIB_OBJS) $(CTRL_PIC_OBJS): CFLAGS += $(CFLAGS_libxengnttab) $(CFLAGS_libxengntshr)
LIB := libxenctrl.a
@@ -202,7 +207,7 @@ libxenctrl.so.$(MAJOR): libxenctrl.so.$(MAJOR).$(MINOR)
$(SYMLINK_SHLIB) $< $@
libxenctrl.so.$(MAJOR).$(MINOR): $(CTRL_PIC_OBJS)
- $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_libxengntshr) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_libxengntshr) $(LDLIBS_libxencall) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
# libxenguest
diff --git a/tools/libxc/xc_altp2m.c b/tools/libxc/xc_altp2m.c
index 87a0fdf..0639632 100644
--- a/tools/libxc/xc_altp2m.c
+++ b/tools/libxc/xc_altp2m.c
@@ -27,22 +27,18 @@
int xc_altp2m_get_domain_state(xc_interface *handle, domid_t dom, bool *state)
{
int rc;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_altp2m_op_t, arg);
arg = xc_hypercall_buffer_alloc(handle, arg, sizeof(*arg));
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_altp2m;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->version = HVMOP_ALTP2M_INTERFACE_VERSION;
arg->cmd = HVMOP_altp2m_get_domain_state;
arg->domain = dom;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op, HVMOP_altp2m,
+ HYPERCALL_BUFFER_AS_ARG(arg));
if ( !rc )
*state = arg->u.domain_state.state;
@@ -54,23 +50,19 @@ int xc_altp2m_get_domain_state(xc_interface *handle, domid_t dom, bool *state)
int xc_altp2m_set_domain_state(xc_interface *handle, domid_t dom, bool state)
{
int rc;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_altp2m_op_t, arg);
arg = xc_hypercall_buffer_alloc(handle, arg, sizeof(*arg));
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_altp2m;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->version = HVMOP_ALTP2M_INTERFACE_VERSION;
arg->cmd = HVMOP_altp2m_set_domain_state;
arg->domain = dom;
arg->u.domain_state.state = state;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op, HVMOP_altp2m,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(handle, arg);
return rc;
@@ -81,24 +73,20 @@ int xc_altp2m_set_vcpu_enable_notify(xc_interface *handle, domid_t domid,
uint32_t vcpuid, xen_pfn_t gfn)
{
int rc;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_altp2m_op_t, arg);
arg = xc_hypercall_buffer_alloc(handle, arg, sizeof(*arg));
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_altp2m;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->version = HVMOP_ALTP2M_INTERFACE_VERSION;
arg->cmd = HVMOP_altp2m_vcpu_enable_notify;
arg->domain = domid;
arg->u.enable_notify.vcpu_id = vcpuid;
arg->u.enable_notify.gfn = gfn;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op, HVMOP_altp2m,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(handle, arg);
return rc;
@@ -108,24 +96,20 @@ int xc_altp2m_create_view(xc_interface *handle, domid_t domid,
xenmem_access_t default_access, uint16_t *view_id)
{
int rc;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_altp2m_op_t, arg);
arg = xc_hypercall_buffer_alloc(handle, arg, sizeof(*arg));
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_altp2m;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->version = HVMOP_ALTP2M_INTERFACE_VERSION;
arg->cmd = HVMOP_altp2m_create_p2m;
arg->domain = domid;
arg->u.view.view = -1;
arg->u.view.hvmmem_default_access = default_access;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op, HVMOP_altp2m,
+ HYPERCALL_BUFFER_AS_ARG(arg));
if ( !rc )
*view_id = arg->u.view.view;
@@ -138,23 +122,19 @@ int xc_altp2m_destroy_view(xc_interface *handle, domid_t domid,
uint16_t view_id)
{
int rc;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_altp2m_op_t, arg);
arg = xc_hypercall_buffer_alloc(handle, arg, sizeof(*arg));
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_altp2m;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->version = HVMOP_ALTP2M_INTERFACE_VERSION;
arg->cmd = HVMOP_altp2m_destroy_p2m;
arg->domain = domid;
arg->u.view.view = view_id;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op, HVMOP_altp2m,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(handle, arg);
return rc;
@@ -165,23 +145,19 @@ int xc_altp2m_switch_to_view(xc_interface *handle, domid_t domid,
uint16_t view_id)
{
int rc;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_altp2m_op_t, arg);
arg = xc_hypercall_buffer_alloc(handle, arg, sizeof(*arg));
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_altp2m;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->version = HVMOP_ALTP2M_INTERFACE_VERSION;
arg->cmd = HVMOP_altp2m_switch_p2m;
arg->domain = domid;
arg->u.view.view = view_id;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op, HVMOP_altp2m,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(handle, arg);
return rc;
@@ -192,17 +168,12 @@ int xc_altp2m_set_mem_access(xc_interface *handle, domid_t domid,
xenmem_access_t access)
{
int rc;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_altp2m_op_t, arg);
arg = xc_hypercall_buffer_alloc(handle, arg, sizeof(*arg));
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_altp2m;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->version = HVMOP_ALTP2M_INTERFACE_VERSION;
arg->cmd = HVMOP_altp2m_set_mem_access;
arg->domain = domid;
@@ -210,7 +181,8 @@ int xc_altp2m_set_mem_access(xc_interface *handle, domid_t domid,
arg->u.set_mem_access.hvmmem_access = access;
arg->u.set_mem_access.gfn = gfn;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op, HVMOP_altp2m,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(handle, arg);
return rc;
@@ -221,17 +193,12 @@ int xc_altp2m_change_gfn(xc_interface *handle, domid_t domid,
xen_pfn_t new_gfn)
{
int rc;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_altp2m_op_t, arg);
arg = xc_hypercall_buffer_alloc(handle, arg, sizeof(*arg));
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_altp2m;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->version = HVMOP_ALTP2M_INTERFACE_VERSION;
arg->cmd = HVMOP_altp2m_change_gfn;
arg->domain = domid;
@@ -239,7 +206,8 @@ int xc_altp2m_change_gfn(xc_interface *handle, domid_t domid,
arg->u.change_gfn.old_gfn = old_gfn;
arg->u.change_gfn.new_gfn = new_gfn;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op, HVMOP_altp2m,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(handle, arg);
return rc;
diff --git a/tools/libxc/xc_domain.c b/tools/libxc/xc_domain.c
index e7278dd..82039eb 100644
--- a/tools/libxc/xc_domain.c
+++ b/tools/libxc/xc_domain.c
@@ -131,7 +131,6 @@ int xc_domain_shutdown(xc_interface *xch,
int reason)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(sched_remote_shutdown_t, arg);
arg = xc_hypercall_buffer_alloc(xch, arg, sizeof(*arg));
@@ -141,13 +140,11 @@ int xc_domain_shutdown(xc_interface *xch,
goto out1;
}
- hypercall.op = __HYPERVISOR_sched_op;
- hypercall.arg[0] = (unsigned long)SCHEDOP_remote_shutdown;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
arg->domain_id = domid;
arg->reason = reason;
-
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall2(xch->xcall, __HYPERVISOR_sched_op,
+ SCHEDOP_remote_shutdown,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
@@ -570,7 +567,6 @@ int xc_watchdog(xc_interface *xch,
uint32_t timeout)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(sched_watchdog_t, arg);
arg = xc_hypercall_buffer_alloc(xch, arg, sizeof(*arg));
@@ -580,13 +576,12 @@ int xc_watchdog(xc_interface *xch,
goto out1;
}
- hypercall.op = __HYPERVISOR_sched_op;
- hypercall.arg[0] = (unsigned long)SCHEDOP_watchdog;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
arg->id = id;
arg->timeout = timeout;
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall2(xch->xcall, __HYPERVISOR_sched_op,
+ SCHEDOP_watchdog,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
@@ -1365,7 +1360,6 @@ static inline int xc_hvm_param_deprecated_check(uint32_t param)
int xc_hvm_param_set(xc_interface *handle, domid_t dom, uint32_t param, uint64_t value)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_param_t, arg);
int rc = xc_hvm_param_deprecated_check(param);
@@ -1376,20 +1370,18 @@ int xc_hvm_param_set(xc_interface *handle, domid_t dom, uint32_t param, uint64_t
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_set_param;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
arg->domid = dom;
arg->index = param;
arg->value = value;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_set_param,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(handle, arg);
return rc;
}
int xc_hvm_param_get(xc_interface *handle, domid_t dom, uint32_t param, uint64_t *value)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_param_t, arg);
int rc = xc_hvm_param_deprecated_check(param);
@@ -1400,12 +1392,11 @@ int xc_hvm_param_get(xc_interface *handle, domid_t dom, uint32_t param, uint64_t
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_get_param;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
arg->domid = dom;
arg->index = param;
- rc = do_xen_hypercall(handle, &hypercall);
+ rc = xencall2(handle->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_get_param,
+ HYPERCALL_BUFFER_AS_ARG(arg));
*value = arg->value;
xc_hypercall_buffer_free(handle, arg);
return rc;
@@ -1433,7 +1424,6 @@ int xc_hvm_create_ioreq_server(xc_interface *xch,
int handle_bufioreq,
ioservid_t *id)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_create_ioreq_server_t, arg);
int rc;
@@ -1441,14 +1431,12 @@ int xc_hvm_create_ioreq_server(xc_interface *xch,
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_create_ioreq_server;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = domid;
arg->handle_bufioreq = handle_bufioreq;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_create_ioreq_server,
+ HYPERCALL_BUFFER_AS_ARG(arg));
*id = arg->id;
@@ -1463,7 +1451,6 @@ int xc_hvm_get_ioreq_server_info(xc_interface *xch,
xen_pfn_t *bufioreq_pfn,
evtchn_port_t *bufioreq_port)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_get_ioreq_server_info_t, arg);
int rc;
@@ -1471,14 +1458,12 @@ int xc_hvm_get_ioreq_server_info(xc_interface *xch,
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_get_ioreq_server_info;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = domid;
arg->id = id;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_get_ioreq_server_info,
+ HYPERCALL_BUFFER_AS_ARG(arg));
if ( rc != 0 )
goto done;
@@ -1500,7 +1485,6 @@ int xc_hvm_map_io_range_to_ioreq_server(xc_interface *xch, domid_t domid,
ioservid_t id, int is_mmio,
uint64_t start, uint64_t end)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_io_range_t, arg);
int rc;
@@ -1508,17 +1492,15 @@ int xc_hvm_map_io_range_to_ioreq_server(xc_interface *xch, domid_t domid,
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_map_io_range_to_ioreq_server;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = domid;
arg->id = id;
arg->type = is_mmio ? HVMOP_IO_RANGE_MEMORY : HVMOP_IO_RANGE_PORT;
arg->start = start;
arg->end = end;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_map_io_range_to_ioreq_server,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
return rc;
@@ -1528,7 +1510,6 @@ int xc_hvm_unmap_io_range_from_ioreq_server(xc_interface *xch, domid_t domid,
ioservid_t id, int is_mmio,
uint64_t start, uint64_t end)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_io_range_t, arg);
int rc;
@@ -1536,17 +1517,15 @@ int xc_hvm_unmap_io_range_from_ioreq_server(xc_interface *xch, domid_t domid,
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_unmap_io_range_from_ioreq_server;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = domid;
arg->id = id;
arg->type = is_mmio ? HVMOP_IO_RANGE_MEMORY : HVMOP_IO_RANGE_PORT;
arg->start = start;
arg->end = end;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_unmap_io_range_from_ioreq_server,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
return rc;
@@ -1557,7 +1536,6 @@ int xc_hvm_map_pcidev_to_ioreq_server(xc_interface *xch, domid_t domid,
uint8_t bus, uint8_t device,
uint8_t function)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_io_range_t, arg);
int rc;
@@ -1570,10 +1548,6 @@ int xc_hvm_map_pcidev_to_ioreq_server(xc_interface *xch, domid_t domid,
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_map_io_range_to_ioreq_server;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = domid;
arg->id = id;
arg->type = HVMOP_IO_RANGE_PCI;
@@ -1587,7 +1561,9 @@ int xc_hvm_map_pcidev_to_ioreq_server(xc_interface *xch, domid_t domid,
(uint64_t)device,
(uint64_t)function);
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_map_io_range_to_ioreq_server,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
return rc;
@@ -1598,7 +1574,6 @@ int xc_hvm_unmap_pcidev_from_ioreq_server(xc_interface *xch, domid_t domid,
uint8_t bus, uint8_t device,
uint8_t function)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_io_range_t, arg);
int rc;
@@ -1611,10 +1586,6 @@ int xc_hvm_unmap_pcidev_from_ioreq_server(xc_interface *xch, domid_t domid,
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_unmap_io_range_from_ioreq_server;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = domid;
arg->id = id;
arg->type = HVMOP_IO_RANGE_PCI;
@@ -1623,7 +1594,9 @@ int xc_hvm_unmap_pcidev_from_ioreq_server(xc_interface *xch, domid_t domid,
(uint64_t)device,
(uint64_t)function);
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_unmap_io_range_from_ioreq_server,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
return rc;
@@ -1633,7 +1606,6 @@ int xc_hvm_destroy_ioreq_server(xc_interface *xch,
domid_t domid,
ioservid_t id)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_destroy_ioreq_server_t, arg);
int rc;
@@ -1641,14 +1613,12 @@ int xc_hvm_destroy_ioreq_server(xc_interface *xch,
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_destroy_ioreq_server;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = domid;
arg->id = id;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_destroy_ioreq_server,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
return rc;
@@ -1659,7 +1629,6 @@ int xc_hvm_set_ioreq_server_state(xc_interface *xch,
ioservid_t id,
int enabled)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_hvm_set_ioreq_server_state_t, arg);
int rc;
@@ -1667,15 +1636,13 @@ int xc_hvm_set_ioreq_server_state(xc_interface *xch,
if ( arg == NULL )
return -1;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_set_ioreq_server_state;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = domid;
arg->id = id;
arg->enabled = !!enabled;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_set_ioreq_server_state,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
return rc;
diff --git a/tools/libxc/xc_evtchn.c b/tools/libxc/xc_evtchn.c
index ae2fe1a..53f7605 100644
--- a/tools/libxc/xc_evtchn.c
+++ b/tools/libxc/xc_evtchn.c
@@ -25,7 +25,6 @@ static int do_evtchn_op(xc_interface *xch, int cmd, void *arg,
size_t arg_size, int silently_fail)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(arg, arg_size, XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
if ( xc_hypercall_bounce_pre(xch, arg) )
@@ -34,11 +33,9 @@ static int do_evtchn_op(xc_interface *xch, int cmd, void *arg,
goto out;
}
- hypercall.op = __HYPERVISOR_event_channel_op;
- hypercall.arg[0] = cmd;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
- if ((ret = do_xen_hypercall(xch, &hypercall)) < 0 && !silently_fail)
+ ret = xencall2(xch->xcall, __HYPERVISOR_event_channel_op,
+ cmd, HYPERCALL_BUFFER_AS_ARG(arg));
+ if ( ret < 0 && !silently_fail )
ERROR("do_evtchn_op: HYPERVISOR_event_channel_op failed: %d", ret);
xc_hypercall_bounce_post(xch, arg);
diff --git a/tools/libxc/xc_flask.c b/tools/libxc/xc_flask.c
index b533656..ec52b0f 100644
--- a/tools/libxc/xc_flask.c
+++ b/tools/libxc/xc_flask.c
@@ -37,7 +37,6 @@
int xc_flask_op(xc_interface *xch, xen_flask_op_t *op)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(op, sizeof(*op), XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
op->interface_version = XEN_FLASK_INTERFACE_VERSION;
@@ -48,10 +47,9 @@ int xc_flask_op(xc_interface *xch, xen_flask_op_t *op)
goto out;
}
- hypercall.op = __HYPERVISOR_xsm_op;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(op);
-
- if ( (ret = do_xen_hypercall(xch, &hypercall)) < 0 )
+ ret = xencall1(xch->xcall, __HYPERVISOR_xsm_op,
+ HYPERCALL_BUFFER_AS_ARG(op));
+ if ( ret < 0 )
{
if ( errno == EACCES )
fprintf(stderr, "XSM operation failed!\n");
diff --git a/tools/libxc/xc_freebsd_osdep.c b/tools/libxc/xc_freebsd_osdep.c
index f6a2ccd..6b440ee 100644
--- a/tools/libxc/xc_freebsd_osdep.c
+++ b/tools/libxc/xc_freebsd_osdep.c
@@ -88,53 +88,6 @@ int osdep_privcmd_close(xc_interface *xch)
return close(fd);
}
-/*------------------------ Privcmd hypercall interface -----------------------*/
-void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
-{
- size_t size = npages * XC_PAGE_SIZE;
- void *p;
-
- /* Address returned by mmap is page aligned. */
- p = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
- -1, 0);
- if (p == NULL)
- return NULL;
-
- /*
- * Since FreeBSD doesn't have the MAP_LOCKED flag,
- * lock memory using mlock.
- */
- if ( mlock(p, size) < 0 )
- {
- munmap(p, size);
- return NULL;
- }
-
- return p;
-}
-
-void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
-{
-
- int saved_errno = errno;
- /* Unlock pages */
- munlock(ptr, npages * XC_PAGE_SIZE);
-
- munmap(ptr, npages * XC_PAGE_SIZE);
- /* We MUST propagate the hypercall errno, not unmap call's. */
- errno = saved_errno;
-}
-
-int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
-{
- int fd = xch->privcmdfd;
- int ret;
-
- ret = ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
-
- return (ret == 0) ? hypercall->retval : ret;
-}
-
/*----------------------- Privcmd foreign map interface ----------------------*/
void *xc_map_foreign_bulk(xc_interface *xch,
uint32_t dom, int prot,
diff --git a/tools/libxc/xc_gnttab.c b/tools/libxc/xc_gnttab.c
index dd32aa2..af53fac 100644
--- a/tools/libxc/xc_gnttab.c
+++ b/tools/libxc/xc_gnttab.c
@@ -21,7 +21,6 @@
int xc_gnttab_op(xc_interface *xch, int cmd, void * op, int op_size, int count)
{
int ret = 0;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(op, count * op_size, XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
if ( xc_hypercall_bounce_pre(xch, op) )
@@ -30,12 +29,8 @@ int xc_gnttab_op(xc_interface *xch, int cmd, void * op, int op_size, int count)
goto out1;
}
- hypercall.op = __HYPERVISOR_grant_table_op;
- hypercall.arg[0] = cmd;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(op);
- hypercall.arg[2] = count;
-
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall3(xch->xcall, __HYPERVISOR_grant_table_op,
+ cmd, HYPERCALL_BUFFER_AS_ARG(op), count);
xc_hypercall_bounce_post(xch, op);
diff --git a/tools/libxc/xc_hcall_buf.c b/tools/libxc/xc_hcall_buf.c
index a0e66cf..a57895b 100644
--- a/tools/libxc/xc_hcall_buf.c
+++ b/tools/libxc/xc_hcall_buf.c
@@ -17,7 +17,6 @@
#include <stdlib.h>
#include <string.h>
-#include <pthread.h>
#include "xc_private.h"
#include "xg_private.h"
@@ -28,163 +27,38 @@ xc_hypercall_buffer_t XC__HYPERCALL_BUFFER_NAME(HYPERCALL_BUFFER_NULL) = {
HYPERCALL_BUFFER_INIT_NO_BOUNCE
};
-pthread_mutex_t hypercall_buffer_cache_mutex = PTHREAD_MUTEX_INITIALIZER;
-
-static void hypercall_buffer_cache_lock(xc_interface *xch)
-{
- int saved_errno = errno;
- if ( xch->flags & XC_OPENFLAG_NON_REENTRANT )
- return;
- pthread_mutex_lock(&hypercall_buffer_cache_mutex);
- /* Ignore pthread errors. */
- errno = saved_errno;
-}
-
-static void hypercall_buffer_cache_unlock(xc_interface *xch)
-{
- int saved_errno = errno;
- if ( xch->flags & XC_OPENFLAG_NON_REENTRANT )
- return;
- pthread_mutex_unlock(&hypercall_buffer_cache_mutex);
- /* Ignore pthread errors. */
- errno = saved_errno;
-}
-
-static void *hypercall_buffer_cache_alloc(xc_interface *xch, int nr_pages)
-{
- void *p = NULL;
-
- hypercall_buffer_cache_lock(xch);
-
- xch->hypercall_buffer_total_allocations++;
- xch->hypercall_buffer_current_allocations++;
- if ( xch->hypercall_buffer_current_allocations > xch->hypercall_buffer_maximum_allocations )
- xch->hypercall_buffer_maximum_allocations = xch->hypercall_buffer_current_allocations;
-
- if ( nr_pages > 1 )
- {
- xch->hypercall_buffer_cache_toobig++;
- }
- else if ( xch->hypercall_buffer_cache_nr > 0 )
- {
- p = xch->hypercall_buffer_cache[--xch->hypercall_buffer_cache_nr];
- xch->hypercall_buffer_cache_hits++;
- }
- else
- {
- xch->hypercall_buffer_cache_misses++;
- }
-
- hypercall_buffer_cache_unlock(xch);
-
- return p;
-}
-
-static int hypercall_buffer_cache_free(xc_interface *xch, void *p, int nr_pages)
-{
- int rc = 0;
-
- hypercall_buffer_cache_lock(xch);
-
- xch->hypercall_buffer_total_releases++;
- xch->hypercall_buffer_current_allocations--;
-
- if ( nr_pages == 1 && xch->hypercall_buffer_cache_nr < HYPERCALL_BUFFER_CACHE_SIZE )
- {
- xch->hypercall_buffer_cache[xch->hypercall_buffer_cache_nr++] = p;
- rc = 1;
- }
-
- hypercall_buffer_cache_unlock(xch);
-
- return rc;
-}
-
-void xc__hypercall_buffer_cache_release(xc_interface *xch)
-{
- void *p;
-
- hypercall_buffer_cache_lock(xch);
-
- DBGPRINTF("hypercall buffer: total allocations:%d total releases:%d",
- xch->hypercall_buffer_total_allocations,
- xch->hypercall_buffer_total_releases);
- DBGPRINTF("hypercall buffer: current allocations:%d maximum allocations:%d",
- xch->hypercall_buffer_current_allocations,
- xch->hypercall_buffer_maximum_allocations);
- DBGPRINTF("hypercall buffer: cache current size:%d",
- xch->hypercall_buffer_cache_nr);
- DBGPRINTF("hypercall buffer: cache hits:%d misses:%d toobig:%d",
- xch->hypercall_buffer_cache_hits,
- xch->hypercall_buffer_cache_misses,
- xch->hypercall_buffer_cache_toobig);
-
- while ( xch->hypercall_buffer_cache_nr > 0 )
- {
- p = xch->hypercall_buffer_cache[--xch->hypercall_buffer_cache_nr];
- osdep_free_hypercall_buffer(xch, p, 1);
- }
-
- hypercall_buffer_cache_unlock(xch);
-}
-
void *xc__hypercall_buffer_alloc_pages(xc_interface *xch, xc_hypercall_buffer_t *b, int nr_pages)
{
- void *p = hypercall_buffer_cache_alloc(xch, nr_pages);
-
- if ( !p )
- p = osdep_alloc_hypercall_buffer(xch, nr_pages);
+ void *p = xencall_alloc_buffer_pages(xch->xcall, nr_pages);
if (!p)
return NULL;
b->hbuf = p;
- memset(p, 0, nr_pages * PAGE_SIZE);
-
return b->hbuf;
}
void xc__hypercall_buffer_free_pages(xc_interface *xch, xc_hypercall_buffer_t *b, int nr_pages)
{
- if ( b->hbuf == NULL )
- return;
-
- if ( !hypercall_buffer_cache_free(xch, b->hbuf, nr_pages) )
- osdep_free_hypercall_buffer(xch, b->hbuf, nr_pages);
+ xencall_free_buffer_pages(xch->xcall, b->hbuf, nr_pages);
}
-struct allocation_header {
- int nr_pages;
-};
-
void *xc__hypercall_buffer_alloc(xc_interface *xch, xc_hypercall_buffer_t *b, size_t size)
{
- size_t actual_size = ROUNDUP(size + sizeof(struct allocation_header), PAGE_SHIFT);
- int nr_pages = actual_size >> PAGE_SHIFT;
- struct allocation_header *hdr;
+ void *p = xencall_alloc_buffer(xch->xcall, size);
- hdr = xc__hypercall_buffer_alloc_pages(xch, b, nr_pages);
- if ( hdr == NULL )
+ if (!p)
return NULL;
- b->hbuf = (void *)(hdr+1);
+ b->hbuf = p;
- hdr->nr_pages = nr_pages;
return b->hbuf;
}
void xc__hypercall_buffer_free(xc_interface *xch, xc_hypercall_buffer_t *b)
{
- struct allocation_header *hdr;
-
- if (b->hbuf == NULL)
- return;
-
- hdr = b->hbuf;
- b->hbuf = --hdr;
-
- xc__hypercall_buffer_free_pages(xch, b, hdr->nr_pages);
+ xencall_free_buffer(xch->xcall, b->hbuf);
}
int xc__hypercall_bounce_pre(xc_interface *xch, xc_hypercall_buffer_t *b)
diff --git a/tools/libxc/xc_kexec.c b/tools/libxc/xc_kexec.c
index a49cffb..1cceb5d 100644
--- a/tools/libxc/xc_kexec.c
+++ b/tools/libxc/xc_kexec.c
@@ -14,7 +14,6 @@
int xc_kexec_exec(xc_interface *xch, int type)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_kexec_exec_t, exec);
int ret = -1;
@@ -27,11 +26,9 @@ int xc_kexec_exec(xc_interface *xch, int type)
exec->type = type;
- hypercall.op = __HYPERVISOR_kexec_op;
- hypercall.arg[0] = KEXEC_CMD_kexec;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(exec);
-
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall2(xch->xcall, __HYPERVISOR_kexec_op,
+ KEXEC_CMD_kexec,
+ HYPERCALL_BUFFER_AS_ARG(exec));
out:
xc_hypercall_buffer_free(xch, exec);
@@ -42,7 +39,6 @@ out:
int xc_kexec_get_range(xc_interface *xch, int range, int nr,
uint64_t *size, uint64_t *start)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_kexec_range_t, get_range);
int ret = -1;
@@ -56,11 +52,9 @@ int xc_kexec_get_range(xc_interface *xch, int range, int nr,
get_range->range = range;
get_range->nr = nr;
- hypercall.op = __HYPERVISOR_kexec_op;
- hypercall.arg[0] = KEXEC_CMD_kexec_get_range;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(get_range);
-
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall2(xch->xcall, __HYPERVISOR_kexec_op,
+ KEXEC_CMD_kexec_get_range,
+ HYPERCALL_BUFFER_AS_ARG(get_range));
*size = get_range->size;
*start = get_range->start;
@@ -76,7 +70,6 @@ int xc_kexec_load(xc_interface *xch, uint8_t type, uint16_t arch,
uint32_t nr_segments, xen_kexec_segment_t *segments)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(segments, sizeof(*segments) * nr_segments,
XC_HYPERCALL_BUFFER_BOUNCE_IN);
DECLARE_HYPERCALL_BUFFER(xen_kexec_load_t, load);
@@ -99,11 +92,9 @@ int xc_kexec_load(xc_interface *xch, uint8_t type, uint16_t arch,
load->nr_segments = nr_segments;
set_xen_guest_handle(load->segments.h, segments);
- hypercall.op = __HYPERVISOR_kexec_op;
- hypercall.arg[0] = KEXEC_CMD_kexec_load;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(load);
-
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall2(xch->xcall, __HYPERVISOR_kexec_op,
+ KEXEC_CMD_kexec_load,
+ HYPERCALL_BUFFER_AS_ARG(load));
out:
xc_hypercall_buffer_free(xch, load);
@@ -114,7 +105,6 @@ out:
int xc_kexec_unload(xc_interface *xch, int type)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(xen_kexec_unload_t, unload);
int ret = -1;
@@ -127,11 +117,9 @@ int xc_kexec_unload(xc_interface *xch, int type)
unload->type = type;
- hypercall.op = __HYPERVISOR_kexec_op;
- hypercall.arg[0] = KEXEC_CMD_kexec_unload;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(unload);
-
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall2(xch->xcall, __HYPERVISOR_kexec_op,
+ KEXEC_CMD_kexec_unload,
+ HYPERCALL_BUFFER_AS_ARG(unload));
out:
xc_hypercall_buffer_free(xch, unload);
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index e858ad9..7f27966 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -85,55 +85,6 @@ int osdep_privcmd_close(xc_interface *xch)
return close(fd);
}
-void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
-{
- size_t size = npages * XC_PAGE_SIZE;
- void *p;
- int rc, saved_errno;
-
- /* Address returned by mmap is page aligned. */
- p = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_LOCKED, -1, 0);
- if ( p == MAP_FAILED )
- {
- PERROR("xc_alloc_hypercall_buffer: mmap failed");
- return NULL;
- }
-
- /* Do not copy the VMA to child process on fork. Avoid the page being COW
- on hypercall. */
- rc = madvise(p, npages * XC_PAGE_SIZE, MADV_DONTFORK);
- if ( rc < 0 )
- {
- PERROR("xc_alloc_hypercall_buffer: madvise failed");
- goto out;
- }
-
- return p;
-
-out:
- saved_errno = errno;
- (void)munmap(p, size);
- errno = saved_errno;
- return NULL;
-}
-
-void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
-{
- int saved_errno = errno;
- /* Recover the VMA flags. Maybe it's not necessary */
- madvise(ptr, npages * XC_PAGE_SIZE, MADV_DOFORK);
-
- munmap(ptr, npages * XC_PAGE_SIZE);
- /* We MUST propagate the hypercall errno, not unmap call's. */
- errno = saved_errno;
-}
-
-int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
-{
- int fd = xch->privcmdfd;
- return ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
-}
-
static int xc_map_foreign_batch_single(int fd, uint32_t dom,
xen_pfn_t *mfn, unsigned long addr)
{
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index 047e13b..e3c8241 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -61,38 +61,6 @@ void minios_interface_close_fd(int fd)
files[fd].type = FTYPE_NONE;
}
-void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
-{
- return xc_memalign(xch, PAGE_SIZE, npages * PAGE_SIZE);
-}
-
-void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
-{
- free(ptr);
-}
-
-int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
-{
- multicall_entry_t call;
- int i, ret;
-
- call.op = hypercall->op;
- for (i = 0; i < ARRAY_SIZE(hypercall->arg); i++)
- call.args[i] = hypercall->arg[i];
-
- ret = HYPERVISOR_multicall(&call, 1);
-
- if (ret < 0) {
- errno = -ret;
- return -1;
- }
- if ((long) call.result < 0) {
- errno = - (long) call.result;
- return -1;
- }
- return call.result;
-}
-
void *xc_map_foreign_bulk(xc_interface *xch,
uint32_t dom, int prot,
const xen_pfn_t *arr, int *err, unsigned int num)
diff --git a/tools/libxc/xc_misc.c b/tools/libxc/xc_misc.c
index c613545..124537b 100644
--- a/tools/libxc/xc_misc.c
+++ b/tools/libxc/xc_misc.c
@@ -323,7 +323,6 @@ int xc_sched_id(xc_interface *xch,
int xc_mca_op(xc_interface *xch, struct xen_mc *mc)
{
int ret = 0;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(mc, sizeof(*mc), XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
if ( xc_hypercall_bounce_pre(xch, mc) )
@@ -333,9 +332,9 @@ int xc_mca_op(xc_interface *xch, struct xen_mc *mc)
}
mc->interface_version = XEN_MCA_INTERFACE_VERSION;
- hypercall.op = __HYPERVISOR_mca;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(mc);
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall1(xch->xcall, __HYPERVISOR_mca,
+ HYPERCALL_BUFFER_AS_ARG(mc));
+
xc_hypercall_bounce_post(xch, mc);
return ret;
}
@@ -471,7 +470,6 @@ int xc_hvm_set_pci_intx_level(
uint8_t domain, uint8_t bus, uint8_t device, uint8_t intx,
unsigned int level)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(struct xen_hvm_set_pci_intx_level, arg);
int rc;
@@ -482,10 +480,6 @@ int xc_hvm_set_pci_intx_level(
return -1;
}
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_set_pci_intx_level;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = dom;
arg->domain = domain;
arg->bus = bus;
@@ -493,7 +487,9 @@ int xc_hvm_set_pci_intx_level(
arg->intx = intx;
arg->level = level;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_set_pci_intx_level,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
@@ -505,7 +501,6 @@ int xc_hvm_set_isa_irq_level(
uint8_t isa_irq,
unsigned int level)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(struct xen_hvm_set_isa_irq_level, arg);
int rc;
@@ -516,15 +511,13 @@ int xc_hvm_set_isa_irq_level(
return -1;
}
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_set_isa_irq_level;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = dom;
arg->isa_irq = isa_irq;
arg->level = level;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_set_isa_irq_level,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
@@ -534,7 +527,6 @@ int xc_hvm_set_isa_irq_level(
int xc_hvm_set_pci_link_route(
xc_interface *xch, domid_t dom, uint8_t link, uint8_t isa_irq)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(struct xen_hvm_set_pci_link_route, arg);
int rc;
@@ -545,15 +537,13 @@ int xc_hvm_set_pci_link_route(
return -1;
}
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_set_pci_link_route;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = dom;
arg->link = link;
arg->isa_irq = isa_irq;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_set_pci_link_route,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
@@ -563,7 +553,6 @@ int xc_hvm_set_pci_link_route(
int xc_hvm_inject_msi(
xc_interface *xch, domid_t dom, uint64_t addr, uint32_t data)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(struct xen_hvm_inject_msi, arg);
int rc;
@@ -574,15 +563,13 @@ int xc_hvm_inject_msi(
return -1;
}
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_inject_msi;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = dom;
arg->addr = addr;
arg->data = data;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_inject_msi,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
@@ -594,7 +581,6 @@ int xc_hvm_track_dirty_vram(
uint64_t first_pfn, uint64_t nr,
unsigned long *dirty_bitmap)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(dirty_bitmap, (nr+7) / 8, XC_HYPERCALL_BUFFER_BOUNCE_OUT);
DECLARE_HYPERCALL_BUFFER(struct xen_hvm_track_dirty_vram, arg);
int rc;
@@ -607,16 +593,14 @@ int xc_hvm_track_dirty_vram(
goto out;
}
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_track_dirty_vram;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = dom;
arg->first_pfn = first_pfn;
arg->nr = nr;
set_xen_guest_handle(arg->dirty_bitmap, dirty_bitmap);
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_track_dirty_vram,
+ HYPERCALL_BUFFER_AS_ARG(arg));
out:
xc_hypercall_buffer_free(xch, arg);
@@ -627,7 +611,6 @@ out:
int xc_hvm_modified_memory(
xc_interface *xch, domid_t dom, uint64_t first_pfn, uint64_t nr)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(struct xen_hvm_modified_memory, arg);
int rc;
@@ -638,15 +621,13 @@ int xc_hvm_modified_memory(
return -1;
}
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_modified_memory;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
arg->domid = dom;
arg->first_pfn = first_pfn;
arg->nr = nr;
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_modified_memory,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
@@ -656,7 +637,6 @@ int xc_hvm_modified_memory(
int xc_hvm_set_mem_type(
xc_interface *xch, domid_t dom, hvmmem_type_t mem_type, uint64_t first_pfn, uint64_t nr)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(struct xen_hvm_set_mem_type, arg);
int rc;
@@ -672,11 +652,9 @@ int xc_hvm_set_mem_type(
arg->first_pfn = first_pfn;
arg->nr = nr;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_set_mem_type;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_set_mem_type,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
@@ -688,7 +666,6 @@ int xc_hvm_inject_trap(
uint32_t type, uint32_t error_code, uint32_t insn_len,
uint64_t cr2)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER(struct xen_hvm_inject_trap, arg);
int rc;
@@ -707,11 +684,9 @@ int xc_hvm_inject_trap(
arg->insn_len = insn_len;
arg->cr2 = cr2;
- hypercall.op = __HYPERVISOR_hvm_op;
- hypercall.arg[0] = HVMOP_inject_trap;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
- rc = do_xen_hypercall(xch, &hypercall);
+ rc = xencall2(xch->xcall, __HYPERVISOR_hvm_op,
+ HVMOP_inject_trap,
+ HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_buffer_free(xch, arg);
diff --git a/tools/libxc/xc_netbsd.c b/tools/libxc/xc_netbsd.c
index 9abb3b6..5e3b343 100644
--- a/tools/libxc/xc_netbsd.c
+++ b/tools/libxc/xc_netbsd.c
@@ -67,46 +67,6 @@ int osdep_privcmd_close(xc_interface *xch)
return close(fd);
}
-void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
-{
- size_t size = npages * XC_PAGE_SIZE;
- void *p;
-
- p = xc_memalign(xch, XC_PAGE_SIZE, size);
- if (!p)
- return NULL;
-
- if ( mlock(p, size) < 0 )
- {
- free(p);
- return NULL;
- }
- return p;
-}
-
-void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
-{
- (void) munlock(ptr, npages * XC_PAGE_SIZE);
- free(ptr);
-}
-
-int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
-{
- int fd = xch->privcmdfd;
- int error = ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
-
- /*
- * Since NetBSD ioctl can only return 0 on success or < 0 on
- * error, if we want to return a value from ioctl we should
- * do so by setting hypercall->retval, to mimic Linux ioctl
- * implementation.
- */
- if (error < 0)
- return error;
- else
- return hypercall->retval;
-}
-
void *xc_map_foreign_bulk(xc_interface *xch,
uint32_t dom, int prot,
const xen_pfn_t *arr, int *err, unsigned int num)
diff --git a/tools/libxc/xc_private.c b/tools/libxc/xc_private.c
index 309eaad..b3a660e 100644
--- a/tools/libxc/xc_private.c
+++ b/tools/libxc/xc_private.c
@@ -39,16 +39,6 @@ struct xc_interface_core *xc_interface_open(xentoollog_logger *logger,
xch->error_handler = logger; xch->error_handler_tofree = 0;
xch->dombuild_logger = dombuild_logger; xch->dombuild_logger_tofree = 0;
- xch->hypercall_buffer_cache_nr = 0;
-
- xch->hypercall_buffer_total_allocations = 0;
- xch->hypercall_buffer_total_releases = 0;
- xch->hypercall_buffer_current_allocations = 0;
- xch->hypercall_buffer_maximum_allocations = 0;
- xch->hypercall_buffer_cache_hits = 0;
- xch->hypercall_buffer_cache_misses = 0;
- xch->hypercall_buffer_cache_toobig = 0;
-
if (!xch->error_handler) {
xch->error_handler = xch->error_handler_tofree =
(xentoollog_logger*)
@@ -65,14 +55,22 @@ struct xc_interface_core *xc_interface_open(xentoollog_logger *logger,
}
*xch = xch_buf;
- if (!(open_flags & XC_OPENFLAG_DUMMY)) {
- if ( osdep_privcmd_open(xch) < 0 )
- goto err;
- }
+ if (open_flags & XC_OPENFLAG_DUMMY)
+ return xch; /* We are done */
+
+ if ( osdep_privcmd_open(xch) < 0 )
+ goto err;
+
+ xch->xcall = xencall_open(xch->error_handler,
+ open_flags & XC_OPENFLAG_NON_REENTRANT ? XENCALL_OPENFLAG_NON_REENTRANT : 0U);
+
+ if ( xch->xcall == NULL )
+ goto err;
return xch;
err:
+ osdep_privcmd_close(xch);
xtl_logger_destroy(xch->error_handler_tofree);
if (xch != &xch_buf) free(xch);
return NULL;
@@ -85,11 +83,12 @@ int xc_interface_close(xc_interface *xch)
if (!xch)
return 0;
+ rc = xencall_close(xch->xcall);
+ if (rc) PERROR("Could not close xencall interface");
+
rc = osdep_privcmd_close(xch);
if (rc) PERROR("Could not close hypervisor interface");
- xc__hypercall_buffer_cache_release(xch);
-
xtl_logger_destroy(xch->dombuild_logger_tofree);
xtl_logger_destroy(xch->error_handler_tofree);
@@ -228,7 +227,6 @@ int xc_mmuext_op(
unsigned int nr_ops,
domid_t dom)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(op, nr_ops*sizeof(*op), XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
long ret = -1;
@@ -238,13 +236,9 @@ int xc_mmuext_op(
goto out1;
}
- hypercall.op = __HYPERVISOR_mmuext_op;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(op);
- hypercall.arg[1] = (unsigned long)nr_ops;
- hypercall.arg[2] = (unsigned long)0;
- hypercall.arg[3] = (unsigned long)dom;
-
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall4(xch->xcall, __HYPERVISOR_mmuext_op,
+ HYPERCALL_BUFFER_AS_ARG(op),
+ nr_ops, 0, dom);
xc_hypercall_bounce_post(xch, op);
@@ -254,8 +248,7 @@ int xc_mmuext_op(
static int flush_mmu_updates(xc_interface *xch, struct xc_mmu *mmu)
{
- int err = 0;
- DECLARE_HYPERCALL;
+ int rc, err = 0;
DECLARE_NAMED_HYPERCALL_BOUNCE(updates, mmu->updates, mmu->idx*sizeof(*mmu->updates), XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
if ( mmu->idx == 0 )
@@ -268,13 +261,10 @@ static int flush_mmu_updates(xc_interface *xch, struct xc_mmu *mmu)
goto out;
}
- hypercall.op = __HYPERVISOR_mmu_update;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(updates);
- hypercall.arg[1] = (unsigned long)mmu->idx;
- hypercall.arg[2] = 0;
- hypercall.arg[3] = mmu->subject;
-
- if ( do_xen_hypercall(xch, &hypercall) < 0 )
+ rc = xencall4(xch->xcall, __HYPERVISOR_mmu_update,
+ HYPERCALL_BUFFER_AS_ARG(updates),
+ mmu->idx, 0, mmu->subject);
+ if ( rc < 0 )
{
ERROR("Failure when submitting mmu updates");
err = 1;
@@ -317,7 +307,6 @@ int xc_flush_mmu_updates(xc_interface *xch, struct xc_mmu *mmu)
int do_memory_op(xc_interface *xch, int cmd, void *arg, size_t len)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(arg, len, XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
long ret = -1;
@@ -327,11 +316,8 @@ int do_memory_op(xc_interface *xch, int cmd, void *arg, size_t len)
goto out1;
}
- hypercall.op = __HYPERVISOR_memory_op;
- hypercall.arg[0] = (unsigned long) cmd;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(arg);
-
- ret = do_xen_hypercall(xch, &hypercall);
+ ret = xencall2(xch->xcall, __HYPERVISOR_memory_op,
+ cmd, HYPERCALL_BUFFER_AS_ARG(arg));
xc_hypercall_bounce_post(xch, arg);
out1:
diff --git a/tools/libxc/xc_private.h b/tools/libxc/xc_private.h
index 9f846b8..3ca9d29 100644
--- a/tools/libxc/xc_private.h
+++ b/tools/libxc/xc_private.h
@@ -30,6 +30,7 @@
#include "_paths.h"
#include "xenctrl.h"
+#include <xencall.h>
#include <xen/sys/privcmd.h>
@@ -53,7 +54,6 @@ struct iovec {
#include <sys/uio.h>
#endif
-#define DECLARE_HYPERCALL privcmd_hypercall_t hypercall
#define DECLARE_DOMCTL struct xen_domctl domctl
#define DECLARE_SYSCTL struct xen_sysctl sysctl
#define DECLARE_PHYSDEV_OP struct physdev_op physdev_op
@@ -94,29 +94,11 @@ struct xc_interface_core {
FILE *dombuild_logger_file;
const char *currently_progress_reporting;
- /*
- * A simple cache of unused, single page, hypercall buffers
- *
- * Protected by a global lock.
- */
-#define HYPERCALL_BUFFER_CACHE_SIZE 4
- int hypercall_buffer_cache_nr;
- void *hypercall_buffer_cache[HYPERCALL_BUFFER_CACHE_SIZE];
-
- /*
- * Hypercall buffer statistics. All protected by the global
- * hypercall_buffer_cache lock.
- */
- int hypercall_buffer_total_allocations;
- int hypercall_buffer_total_releases;
- int hypercall_buffer_current_allocations;
- int hypercall_buffer_maximum_allocations;
- int hypercall_buffer_cache_hits;
- int hypercall_buffer_cache_misses;
- int hypercall_buffer_cache_toobig;
-
/* Privcmd interface */
int privcmdfd;
+
+ /* Hypercall interface */
+ xencall_handle *xcall;
};
int osdep_privcmd_open(xc_interface *xch);
@@ -232,24 +214,16 @@ void xc__hypercall_buffer_cache_release(xc_interface *xch);
* Hypercall interfaces.
*/
-int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall);
-
static inline int do_xen_version(xc_interface *xch, int cmd, xc_hypercall_buffer_t *dest)
{
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER_ARGUMENT(dest);
-
- hypercall.op = __HYPERVISOR_xen_version;
- hypercall.arg[0] = (unsigned long) cmd;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(dest);
-
- return do_xen_hypercall(xch, &hypercall);
+ return xencall2(xch->xcall, __HYPERVISOR_xen_version,
+ cmd, HYPERCALL_BUFFER_AS_ARG(dest));
}
static inline int do_physdev_op(xc_interface *xch, int cmd, void *op, size_t len)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(op, len, XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
if ( xc_hypercall_bounce_pre(xch, op) )
@@ -258,11 +232,9 @@ static inline int do_physdev_op(xc_interface *xch, int cmd, void *op, size_t len
goto out1;
}
- hypercall.op = __HYPERVISOR_physdev_op;
- hypercall.arg[0] = (unsigned long) cmd;
- hypercall.arg[1] = HYPERCALL_BUFFER_AS_ARG(op);
-
- if ( (ret = do_xen_hypercall(xch, &hypercall)) < 0 )
+ ret = xencall2(xch->xcall, __HYPERVISOR_physdev_op,
+ cmd, HYPERCALL_BUFFER_AS_ARG(op));
+ if ( ret < 0 )
{
if ( errno == EACCES )
DPRINTF("physdev operation failed -- need to"
@@ -277,7 +249,6 @@ out1:
static inline int do_domctl(xc_interface *xch, struct xen_domctl *domctl)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(domctl, sizeof(*domctl), XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
domctl->interface_version = XEN_DOMCTL_INTERFACE_VERSION;
@@ -288,10 +259,9 @@ static inline int do_domctl(xc_interface *xch, struct xen_domctl *domctl)
goto out1;
}
- hypercall.op = __HYPERVISOR_domctl;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(domctl);
-
- if ( (ret = do_xen_hypercall(xch, &hypercall)) < 0 )
+ ret = xencall1(xch->xcall, __HYPERVISOR_domctl,
+ HYPERCALL_BUFFER_AS_ARG(domctl));
+ if ( ret < 0 )
{
if ( errno == EACCES )
DPRINTF("domctl operation failed -- need to"
@@ -306,7 +276,6 @@ static inline int do_domctl(xc_interface *xch, struct xen_domctl *domctl)
static inline int do_sysctl(xc_interface *xch, struct xen_sysctl *sysctl)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(sysctl, sizeof(*sysctl), XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
sysctl->interface_version = XEN_SYSCTL_INTERFACE_VERSION;
@@ -317,9 +286,9 @@ static inline int do_sysctl(xc_interface *xch, struct xen_sysctl *sysctl)
goto out1;
}
- hypercall.op = __HYPERVISOR_sysctl;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(sysctl);
- if ( (ret = do_xen_hypercall(xch, &hypercall)) < 0 )
+ ret = xencall1(xch->xcall, __HYPERVISOR_sysctl,
+ HYPERCALL_BUFFER_AS_ARG(sysctl));
+ if ( ret < 0 )
{
if ( errno == EACCES )
DPRINTF("sysctl operation failed -- need to"
@@ -335,7 +304,6 @@ static inline int do_platform_op(xc_interface *xch,
struct xen_platform_op *platform_op)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(platform_op, sizeof(*platform_op),
XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
@@ -347,9 +315,9 @@ static inline int do_platform_op(xc_interface *xch,
return -1;
}
- hypercall.op = __HYPERVISOR_platform_op;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(platform_op);
- if ( (ret = do_xen_hypercall(xch, &hypercall)) < 0 )
+ ret = xencall1(xch->xcall, __HYPERVISOR_platform_op,
+ HYPERCALL_BUFFER_AS_ARG(platform_op));
+ if ( ret < 0 )
{
if ( errno == EACCES )
DPRINTF("platform operation failed -- need to"
@@ -365,13 +333,11 @@ static inline int do_multicall_op(xc_interface *xch,
uint32_t nr_calls)
{
int ret = -1;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BUFFER_ARGUMENT(call_list);
- hypercall.op = __HYPERVISOR_multicall;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(call_list);
- hypercall.arg[1] = nr_calls;
- if ( (ret = do_xen_hypercall(xch, &hypercall)) < 0 )
+ ret = xencall2(xch->xcall, __HYPERVISOR_multicall,
+ HYPERCALL_BUFFER_AS_ARG(call_list), nr_calls);
+ if ( ret < 0 )
{
if ( errno == EACCES )
DPRINTF("multicall operation failed -- need to"
diff --git a/tools/libxc/xc_solaris.c b/tools/libxc/xc_solaris.c
index 6f84b82..18622fa 100644
--- a/tools/libxc/xc_solaris.c
+++ b/tools/libxc/xc_solaris.c
@@ -67,22 +67,6 @@ int osdep_privcmd_close(xc_interface *xch)
return close(fd);
}
-void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages)
-{
- return xc_memalign(xch, XC_PAGE_SIZE, npages * XC_PAGE_SIZE);
-}
-
-static void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages)
-{
- free(ptr);
-}
-
-int do_xen_hypercall(xc_interface *xch, privcmd_hypercall_t *hypercall)
-{
- int fd = xch->privcmdfd;
- return ioctl(fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
-}
-
void *xc_map_foreign_batch(xc_interface *xch,
uint32_t dom, int prot,
xen_pfn_t *arr, int num)
diff --git a/tools/libxc/xc_tmem.c b/tools/libxc/xc_tmem.c
index 8f4c0cc..4e5c278 100644
--- a/tools/libxc/xc_tmem.c
+++ b/tools/libxc/xc_tmem.c
@@ -23,7 +23,6 @@
static int do_tmem_op(xc_interface *xch, tmem_op_t *op)
{
int ret;
- DECLARE_HYPERCALL;
DECLARE_HYPERCALL_BOUNCE(op, sizeof(*op), XC_HYPERCALL_BUFFER_BOUNCE_BOTH);
if ( xc_hypercall_bounce_pre(xch, op) )
@@ -32,9 +31,9 @@ static int do_tmem_op(xc_interface *xch, tmem_op_t *op)
return -EFAULT;
}
- hypercall.op = __HYPERVISOR_tmem_op;
- hypercall.arg[0] = HYPERCALL_BUFFER_AS_ARG(op);
- if ((ret = do_xen_hypercall(xch, &hypercall)) < 0)
+ ret = xencall1(xch->xcall, __HYPERVISOR_tmem_op,
+ HYPERCALL_BUFFER_AS_ARG(op));
+ if ( ret < 0 )
{
if ( errno == EACCES )
DPRINTF("tmem operation failed -- need to"
diff --git a/tools/misc/Makefile b/tools/misc/Makefile
index cf6a475..a2ef0ec 100644
--- a/tools/misc/Makefile
+++ b/tools/misc/Makefile
@@ -87,12 +87,12 @@ xenlockprof: xenlockprof.o
$(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenctrl) $(APPEND_LDFLAGS)
# xen-hptool incorrectly uses libxc internals
-xen-hptool.o: CFLAGS += -I$(XEN_ROOT)/tools/libxc
+xen-hptool.o: CFLAGS += -I$(XEN_ROOT)/tools/libxc $(CFLAGS_libxencall)
xen-hptool: xen-hptool.o
$(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(LDLIBS_libxenstore) $(APPEND_LDFLAGS)
# xen-mfndump incorrectly uses libxc internals
-xen-mfndump.o: CFLAGS += -I$(XEN_ROOT)/tools/libxc
+xen-mfndump.o: CFLAGS += -I$(XEN_ROOT)/tools/libxc $(CFLAGS_libxencall)
xen-mfndump: xen-mfndump.o
$(CC) $(LDFLAGS) -o $@ $< $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenguest) $(APPEND_LDFLAGS)
diff --git a/tools/xcutils/Makefile b/tools/xcutils/Makefile
index 2d1f112..e127af8 100644
--- a/tools/xcutils/Makefile
+++ b/tools/xcutils/Makefile
@@ -16,7 +16,7 @@ PROGRAMS = readnotes lsevtchn
CFLAGS += -Werror
# incorrectly uses libxc internals
-CFLAGS_readnotes.o := $(CFLAGS_libxenevtchn) $(CFLAGS_libxenctrl) $(CFLAGS_libxenguest) -I$(XEN_ROOT)/tools/libxc
+CFLAGS_readnotes.o := $(CFLAGS_libxenevtchn) $(CFLAGS_libxenctrl) $(CFLAGS_libxenguest) -I$(XEN_ROOT)/tools/libxc $(CFLAGS_libxencall)
CFLAGS_lsevtchn.o := $(CFLAGS_libxenevtchn) $(CFLAGS_libxenctrl)
.PHONY: all
diff --git a/tools/xenpaging/Makefile b/tools/xenpaging/Makefile
index d491867..64876b3 100644
--- a/tools/xenpaging/Makefile
+++ b/tools/xenpaging/Makefile
@@ -2,7 +2,7 @@ XEN_ROOT=$(CURDIR)/../..
include $(XEN_ROOT)/tools/Rules.mk
# xenpaging.c and file_ops.c incorrectly use libxc internals
-CFLAGS += $(CFLAGS_libxentoollog) $(CFLAGS_libxenevtchn) $(CFLAGS_libxenctrl) $(CFLAGS_libxenstore) $(PTHREAD_CFLAGS) -I$(XEN_ROOT)/tools/libxc
+CFLAGS += $(CFLAGS_libxentoollog) $(CFLAGS_libxenevtchn) $(CFLAGS_libxenctrl) $(CFLAGS_libxenstore) $(PTHREAD_CFLAGS) -I$(XEN_ROOT)/tools/libxc $(CFLAGS_libxencall)
LDLIBS += $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxenctrl) $(LDLIBS_libxenstore) $(PTHREAD_LIBS)
LDFLAGS += $(PTHREAD_LDFLAGS)
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 10/23] tools/libxc: drop xc_map_foreign_bulk_compat wrappers
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (8 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 11/23] tools: Remove xc_map_foreign_batch Ian Campbell
` (13 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
On Solaris and NetBSD xc_map_foreign_bulk is implemented by calling
xc_map_foreign_bulk_compat and xc_map_foreign_bulk_compat is exposed
as a symbol by libxenctrl.so.
Remove these wrappers and turn the compat function into the real thing
surrounded by the appropriate ifdef.
As this is a compat function all new ports should instead implement
xc_map_foreign_bulk properly, hence the ifdef should never be
expanded.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libxc/xc_foreign_memory.c | 13 +++++++++----
tools/libxc/xc_netbsd.c | 7 -------
tools/libxc/xc_private.h | 5 -----
tools/libxc/xc_solaris.c | 7 -------
4 files changed, 9 insertions(+), 23 deletions(-)
diff --git a/tools/libxc/xc_foreign_memory.c b/tools/libxc/xc_foreign_memory.c
index 9c705b6..b205bca 100644
--- a/tools/libxc/xc_foreign_memory.c
+++ b/tools/libxc/xc_foreign_memory.c
@@ -50,10 +50,14 @@ void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
return res;
}
-/* stub for all not yet converted OSes */
-void *xc_map_foreign_bulk_compat(xc_interface *xch,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
+/*
+ * stub for all not yet converted OSes (NetBSD and Solaris). New OSes should
+ * just implement xc_map_foreign_bulk.
+ */
+#if defined(__NetBSD__) || defined(__sun__)
+void *xc_map_foreign_bulk(xc_interface *xch,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num)
{
xen_pfn_t *pfn;
unsigned int i;
@@ -90,6 +94,7 @@ void *xc_map_foreign_bulk_compat(xc_interface *xch,
return ret;
}
+#endif
/*
* Local variables:
diff --git a/tools/libxc/xc_netbsd.c b/tools/libxc/xc_netbsd.c
index 5e3b343..d7f7f31 100644
--- a/tools/libxc/xc_netbsd.c
+++ b/tools/libxc/xc_netbsd.c
@@ -67,13 +67,6 @@ int osdep_privcmd_close(xc_interface *xch)
return close(fd);
}
-void *xc_map_foreign_bulk(xc_interface *xch,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
-{
- return xc_map_foreign_bulk_compat(xch, dom, prot, arr, err, num);
-}
-
void *xc_map_foreign_batch(xc_interface *xch,
uint32_t dom, int prot,
xen_pfn_t *arr, int num)
diff --git a/tools/libxc/xc_private.h b/tools/libxc/xc_private.h
index 3ca9d29..0de907c 100644
--- a/tools/libxc/xc_private.h
+++ b/tools/libxc/xc_private.h
@@ -107,11 +107,6 @@ int osdep_privcmd_close(xc_interface *xch);
void *osdep_alloc_hypercall_buffer(xc_interface *xch, int npages);
void osdep_free_hypercall_buffer(xc_interface *xch, void *ptr, int npages);
-/* Stub for not yet converted OSes */
-void *xc_map_foreign_bulk_compat(xc_interface *xch,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num);
-
void xc_report_error(xc_interface *xch, int code, const char *fmt, ...)
__attribute__((format(printf,3,4)));
void xc_reportv(xc_interface *xch, xentoollog_logger *lg, xentoollog_level,
diff --git a/tools/libxc/xc_solaris.c b/tools/libxc/xc_solaris.c
index 18622fa..5e72dee 100644
--- a/tools/libxc/xc_solaris.c
+++ b/tools/libxc/xc_solaris.c
@@ -94,13 +94,6 @@ void *xc_map_foreign_batch(xc_interface *xch,
}
-void *xc_map_foreign_bulk(xc_interface *xch,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
-{
- return xc_map_foreign_bulk_compat(xch, dom, prot, arr, err, num);
-}
-
void *xc_map_foreign_range(xc_interface *xch,
uint32_t dom,
int size, int prot,
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 11/23] tools: Remove xc_map_foreign_batch
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (9 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 10/23] tools/libxc: drop xc_map_foreign_bulk_compat wrappers Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 12/23] tools: Implement xc_map_foreign_range(s) in terms of common helper Ian Campbell
` (12 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: George Dunlap, Ian Campbell
It can trivially be replaced by xc_map_foreign_bulk which is the
interface I want to move to going forward. All in tree users are
trivially converted by supplying the appropriate error array and
adjusting the what error handling exists (which in many cases is not
much).
This reduces the twist maze of xc_map_foreign_* by one, which will be
useful when trying to come up with an underlying stable interface.
NetBSD and Solaris implemented xc_map_foreign_bulk in terms of
xc_map_foreign_batch via a compat layer, so xc_map_foreign_batch
becomes an internal osdep for them. New OS ports should always
implement xc_map_foreign_bulk instead.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: George Dunlap <george.dunlap@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Cc: George Dunlap <george.dunlap@eu.citrix.com>
---
tools/libxc/include/xenctrl.h | 10 -------
tools/libxc/xc_foreign_memory.c | 4 ++-
tools/libxc/xc_linux_osdep.c | 59 +++--------------------------------------
tools/libxc/xc_minios.c | 22 ---------------
tools/libxc/xc_netbsd.c | 10 +++----
tools/libxc/xc_solaris.c | 6 ++---
tools/libxc/xc_vm_event.c | 25 ++++++++++++-----
tools/xenmon/xenbaked.c | 19 ++++++++++---
tools/xenpaging/pagein.c | 4 ++-
tools/xenpaging/xenpaging.c | 12 +++++----
tools/xentrace/xentrace.c | 10 ++++---
11 files changed, 63 insertions(+), 118 deletions(-)
diff --git a/tools/libxc/include/xenctrl.h b/tools/libxc/include/xenctrl.h
index bdf4f2f..685315e 100644
--- a/tools/libxc/include/xenctrl.h
+++ b/tools/libxc/include/xenctrl.h
@@ -1395,16 +1395,6 @@ void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
const xen_pfn_t *arr, int num );
/**
- * DEPRECATED - use xc_map_foreign_bulk() instead.
- *
- * Like xc_map_foreign_pages(), except it can succeeed partially.
- * When a page cannot be mapped, its PFN in @arr is or'ed with
- * 0xF0000000 to indicate the error.
- */
-void *xc_map_foreign_batch(xc_interface *xch, uint32_t dom, int prot,
- xen_pfn_t *arr, int num );
-
-/**
* Like xc_map_foreign_pages(), except it can succeed partially.
* When a page cannot be mapped, its respective field in @err is
* set to the corresponding errno value.
diff --git a/tools/libxc/xc_foreign_memory.c b/tools/libxc/xc_foreign_memory.c
index b205bca..2413e75 100644
--- a/tools/libxc/xc_foreign_memory.c
+++ b/tools/libxc/xc_foreign_memory.c
@@ -55,6 +55,8 @@ void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
* just implement xc_map_foreign_bulk.
*/
#if defined(__NetBSD__) || defined(__sun__)
+void *osdep_map_foreign_batch(xc_interface *xch, uint32_t dom, int prot,
+ xen_pfn_t *arr, int num );
void *xc_map_foreign_bulk(xc_interface *xch,
uint32_t dom, int prot,
const xen_pfn_t *arr, int *err, unsigned int num)
@@ -75,7 +77,7 @@ void *xc_map_foreign_bulk(xc_interface *xch,
}
memcpy(pfn, arr, num * sizeof(*arr));
- ret = xc_map_foreign_batch(xch, dom, prot, pfn, num);
+ ret = osdep_map_foreign_batch(xch, dom, prot, pfn, num);
if (ret) {
for (i = 0; i < num; ++i)
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index 7f27966..1404790 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -85,8 +85,8 @@ int osdep_privcmd_close(xc_interface *xch)
return close(fd);
}
-static int xc_map_foreign_batch_single(int fd, uint32_t dom,
- xen_pfn_t *mfn, unsigned long addr)
+static int map_foreign_batch_single(int fd, uint32_t dom,
+ xen_pfn_t *mfn, unsigned long addr)
{
privcmd_mmapbatch_t ioctlx;
int rc;
@@ -107,59 +107,6 @@ static int xc_map_foreign_batch_single(int fd, uint32_t dom,
return rc;
}
-void *xc_map_foreign_batch(xc_interface *xch,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
-{
- int fd = xch->privcmdfd;
- privcmd_mmapbatch_t ioctlx;
- void *addr;
- int rc;
-
- addr = mmap(NULL, num << XC_PAGE_SHIFT, prot, MAP_SHARED, fd, 0);
- if ( addr == MAP_FAILED )
- {
- PERROR("xc_map_foreign_batch: mmap failed");
- return NULL;
- }
-
- ioctlx.num = num;
- ioctlx.dom = dom;
- ioctlx.addr = (unsigned long)addr;
- ioctlx.arr = arr;
-
- rc = ioctl(fd, IOCTL_PRIVCMD_MMAPBATCH, &ioctlx);
- if ( (rc < 0) && (errno == ENOENT) )
- {
- int i;
-
- for ( i = 0; i < num; i++ )
- {
- if ( (arr[i] & PRIVCMD_MMAPBATCH_MFN_ERROR) ==
- PRIVCMD_MMAPBATCH_PAGED_ERROR )
- {
- unsigned long paged_addr = (unsigned long)addr + (i << XC_PAGE_SHIFT);
- rc = xc_map_foreign_batch_single(fd, dom, &arr[i],
- paged_addr);
- if ( rc < 0 )
- goto out;
- }
- }
- }
-
- out:
- if ( rc < 0 )
- {
- int saved_errno = errno;
- PERROR("xc_map_foreign_batch: ioctl failed");
- (void)munmap(addr, num << XC_PAGE_SHIFT);
- errno = saved_errno;
- return NULL;
- }
-
- return addr;
-}
-
/*
* Retry mmap of all paged gfns in batches
* retuns < 0 on fatal error
@@ -299,7 +246,7 @@ void *xc_map_foreign_bulk(xc_interface *xch,
err[i] = rc ?: -EINVAL;
continue;
}
- rc = xc_map_foreign_batch_single(fd, dom, pfn + i,
+ rc = map_foreign_batch_single(fd, dom, pfn + i,
(unsigned long)addr + ((unsigned long)i<<XC_PAGE_SHIFT));
if ( rc < 0 )
{
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index e3c8241..3ea3124 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -73,28 +73,6 @@ void *xc_map_foreign_bulk(xc_interface *xch,
return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
}
-void *xc_map_foreign_batch(xc_interface *xch,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
-{
- unsigned long pt_prot = 0;
- int err[num];
- int i;
- unsigned long addr;
-
- if (prot & PROT_READ)
- pt_prot = L1_PROT_RO;
- if (prot & PROT_WRITE)
- pt_prot = L1_PROT;
-
- addr = (unsigned long) map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
- for (i = 0; i < num; i++) {
- if (err[i])
- arr[i] |= 0xF0000000;
- }
- return (void *) addr;
-}
-
void *xc_map_foreign_range(xc_interface *xch,
uint32_t dom,
int size, int prot,
diff --git a/tools/libxc/xc_netbsd.c b/tools/libxc/xc_netbsd.c
index d7f7f31..6a7ff9f 100644
--- a/tools/libxc/xc_netbsd.c
+++ b/tools/libxc/xc_netbsd.c
@@ -67,16 +67,16 @@ int osdep_privcmd_close(xc_interface *xch)
return close(fd);
}
-void *xc_map_foreign_batch(xc_interface *xch,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
+void *osdep_map_foreign_batch(xc_interface *xch,
+ uint32_t dom, int prot,
+ xen_pfn_t *arr, int num)
{
int fd = xch->privcmdfd;
privcmd_mmapbatch_t ioctlx;
void *addr;
addr = mmap(NULL, num*XC_PAGE_SIZE, prot, MAP_ANON | MAP_SHARED, -1, 0);
if ( addr == MAP_FAILED ) {
- PERROR("xc_map_foreign_batch: mmap failed");
+ PERROR("osdep_map_foreign_batch: mmap failed");
return NULL;
}
@@ -87,7 +87,7 @@ void *xc_map_foreign_batch(xc_interface *xch,
if ( ioctl(fd, IOCTL_PRIVCMD_MMAPBATCH, &ioctlx) < 0 )
{
int saved_errno = errno;
- PERROR("xc_map_foreign_batch: ioctl failed");
+ PERROR("osdep_map_foreign_batch: ioctl failed");
(void)munmap(addr, num*XC_PAGE_SIZE);
errno = saved_errno;
return NULL;
diff --git a/tools/libxc/xc_solaris.c b/tools/libxc/xc_solaris.c
index 5e72dee..2df7a06 100644
--- a/tools/libxc/xc_solaris.c
+++ b/tools/libxc/xc_solaris.c
@@ -67,9 +67,9 @@ int osdep_privcmd_close(xc_interface *xch)
return close(fd);
}
-void *xc_map_foreign_batch(xc_interface *xch,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
+void *osdep_map_foreign_batch(xc_interface *xch,
+ uint32_t dom, int prot,
+ xen_pfn_t *arr, int num)
{
int fd = xch->privcmdfd;
privcmd_mmapbatch_t ioctlx;
diff --git a/tools/libxc/xc_vm_event.c b/tools/libxc/xc_vm_event.c
index 2fef96a..d74d0f0 100644
--- a/tools/libxc/xc_vm_event.c
+++ b/tools/libxc/xc_vm_event.c
@@ -45,6 +45,7 @@ void *xc_vm_event_enable(xc_interface *xch, domid_t domain_id, int param,
void *ring_page = NULL;
uint64_t pfn;
xen_pfn_t ring_pfn, mmap_pfn;
+ int err;
unsigned int op, mode;
int rc1, rc2, saved_errno;
@@ -72,9 +73,9 @@ void *xc_vm_event_enable(xc_interface *xch, domid_t domain_id, int param,
ring_pfn = pfn;
mmap_pfn = pfn;
- ring_page = xc_map_foreign_batch(xch, domain_id, PROT_READ | PROT_WRITE,
- &mmap_pfn, 1);
- if ( mmap_pfn & XEN_DOMCTL_PFINFO_XTAB )
+ ring_page = xc_map_foreign_bulk(xch, domain_id, PROT_READ | PROT_WRITE,
+ &mmap_pfn, &err, 1);
+ if ( !ring_page )
{
/* Map failed, populate ring page */
rc1 = xc_domain_populate_physmap_exact(xch, domain_id, 1, 0, 0,
@@ -86,11 +87,11 @@ void *xc_vm_event_enable(xc_interface *xch, domid_t domain_id, int param,
}
mmap_pfn = ring_pfn;
- ring_page = xc_map_foreign_batch(xch, domain_id, PROT_READ | PROT_WRITE,
- &mmap_pfn, 1);
- if ( mmap_pfn & XEN_DOMCTL_PFINFO_XTAB )
+ ring_page = xc_map_foreign_bulk(xch, domain_id, PROT_READ | PROT_WRITE,
+ &mmap_pfn, &err, 1);
+ if ( !ring_page )
{
- PERROR("Could not map the ring page\n");
+ PERROR("Could not map the ring page: %d\n", err);
goto out;
}
}
@@ -156,3 +157,13 @@ void *xc_vm_event_enable(xc_interface *xch, domid_t domain_id, int param,
return ring_page;
}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/xenmon/xenbaked.c b/tools/xenmon/xenbaked.c
index e4602ef..5cfc6af 100644
--- a/tools/xenmon/xenbaked.c
+++ b/tools/xenmon/xenbaked.c
@@ -407,14 +407,15 @@ static struct t_struct *map_tbufs(unsigned long tbufs_mfn, unsigned int num,
+ tbufs.t_info->mfn_offset[i];
int j;
xen_pfn_t pfn_list[tbufs.t_info->tbuf_size];
+ int err_list[tbufs.t_info->tbuf_size];
for ( j=0; j<tbufs.t_info->tbuf_size; j++)
pfn_list[j] = (xen_pfn_t)mfn_list[j];
- tbufs.meta[i] = xc_map_foreign_batch(xc_handle, DOMID_XEN,
- PROT_READ | PROT_WRITE,
- pfn_list,
- tbufs.t_info->tbuf_size);
+ tbufs.meta[i] = xc_map_foreign_bulk(xc_handle, DOMID_XEN,
+ PROT_READ | PROT_WRITE,
+ pfn_list, err_list,
+ tbufs.t_info->tbuf_size);
if ( tbufs.meta[i] == NULL )
{
PERROR("Failed to map cpu buffer!");
@@ -1175,3 +1176,13 @@ static int process_record(int cpu, struct t_rec *r)
return 4 + (r->cycles_included ? 8 : 0) + (r->extra_u32 * 4);
}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/xenpaging/pagein.c b/tools/xenpaging/pagein.c
index 7cb0f33..18699ec 100644
--- a/tools/xenpaging/pagein.c
+++ b/tools/xenpaging/pagein.c
@@ -23,6 +23,7 @@ static void *page_in(void *arg)
void *page;
int i, num;
xen_pfn_t gfns[XENPAGING_PAGEIN_QUEUE_SIZE];
+ int errs[XENPAGING_PAGEIN_QUEUE_SIZE];
while (1)
{
@@ -42,7 +43,8 @@ static void *page_in(void *arg)
pthread_mutex_unlock(&page_in_mutex);
/* Ignore errors */
- page = xc_map_foreign_pages(pia->xch, pia->dom, PROT_READ, gfns, num);
+ page = xc_map_foreign_bulk(pia->xch, pia->dom, PROT_READ,
+ gfns, errs, num);
if (page)
munmap(page, PAGE_SIZE * num);
}
diff --git a/tools/xenpaging/xenpaging.c b/tools/xenpaging/xenpaging.c
index df99c6a..18b27a4 100644
--- a/tools/xenpaging/xenpaging.c
+++ b/tools/xenpaging/xenpaging.c
@@ -281,6 +281,7 @@ static struct xenpaging *xenpaging_init(int argc, char *argv[])
char *p;
int rc;
unsigned long ring_pfn, mmap_pfn;
+ int mmap_err;
/* Allocate memory */
paging = calloc(1, sizeof(struct xenpaging));
@@ -342,8 +343,8 @@ static struct xenpaging *xenpaging_init(int argc, char *argv[])
HVM_PARAM_PAGING_RING_PFN, &ring_pfn);
mmap_pfn = ring_pfn;
paging->vm_event.ring_page =
- xc_map_foreign_batch(xch, paging->vm_event.domain_id,
- PROT_READ | PROT_WRITE, &mmap_pfn, 1);
+ xc_map_foreign_bulk(xch, paging->vm_event.domain_id,
+ PROT_READ | PROT_WRITE, &mmap_pfn, &mmap_err, 1);
if ( mmap_pfn & XEN_DOMCTL_PFINFO_XTAB )
{
/* Map failed, populate ring page */
@@ -358,11 +359,12 @@ static struct xenpaging *xenpaging_init(int argc, char *argv[])
mmap_pfn = ring_pfn;
paging->vm_event.ring_page =
- xc_map_foreign_batch(xch, paging->vm_event.domain_id,
- PROT_READ | PROT_WRITE, &mmap_pfn, 1);
+ xc_map_foreign_bulk(xch, paging->vm_event.domain_id,
+ PROT_READ | PROT_WRITE,
+ &mmap_pfn, &mmap_err, 1);
if ( mmap_pfn & XEN_DOMCTL_PFINFO_XTAB )
{
- PERROR("Could not map the ring page\n");
+ PERROR("Could not map the ring page: %d\n", mmap_err);
goto err;
}
}
diff --git a/tools/xentrace/xentrace.c b/tools/xentrace/xentrace.c
index c970d42..f66a77c 100644
--- a/tools/xentrace/xentrace.c
+++ b/tools/xentrace/xentrace.c
@@ -504,14 +504,15 @@ static struct t_struct *map_tbufs(unsigned long tbufs_mfn, unsigned int num,
+ tbufs.t_info->mfn_offset[i];
int j;
xen_pfn_t pfn_list[tbufs.t_info->tbuf_size];
+ int pfn_err[tbufs.t_info->tbuf_size];
for ( j=0; j<tbufs.t_info->tbuf_size; j++)
pfn_list[j] = (xen_pfn_t)mfn_list[j];
- tbufs.meta[i] = xc_map_foreign_batch(xc_handle, DOMID_XEN,
- PROT_READ | PROT_WRITE,
- pfn_list,
- tbufs.t_info->tbuf_size);
+ tbufs.meta[i] = xc_map_foreign_bulk(xc_handle, DOMID_XEN,
+ PROT_READ | PROT_WRITE,
+ pfn_list, pfn_err,
+ tbufs.t_info->tbuf_size);
if ( tbufs.meta[i] == NULL )
{
PERROR("Failed to map cpu buffer!");
@@ -1221,6 +1222,7 @@ int main(int argc, char **argv)
return ret;
}
+
/*
* Local variables:
* mode: C
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 12/23] tools: Implement xc_map_foreign_range(s) in terms of common helper
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (10 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 11/23] tools: Remove xc_map_foreign_batch Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory Ian Campbell
` (11 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
Both Linux and FreeBSD already implemented these functions using
identical helpers based on xc_map_foreign_pages. Make one copy of
these common helpers and switch all OSes to use them, even those which
previously had a specific lower level implementation of this
functionality.
This is makes two fewer low level interfaces to think about.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libxc/xc_foreign_memory.c | 50 +++++++++++++++++++++++++++++
tools/libxc/xc_freebsd_osdep.c | 50 -----------------------------
tools/libxc/xc_linux_osdep.c | 49 -----------------------------
tools/libxc/xc_minios.c | 43 -------------------------
tools/libxc/xc_netbsd.c | 70 -----------------------------------------
tools/libxc/xc_solaris.c | 67 ---------------------------------------
6 files changed, 50 insertions(+), 279 deletions(-)
diff --git a/tools/libxc/xc_foreign_memory.c b/tools/libxc/xc_foreign_memory.c
index 2413e75..d1130e6 100644
--- a/tools/libxc/xc_foreign_memory.c
+++ b/tools/libxc/xc_foreign_memory.c
@@ -50,6 +50,56 @@ void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
return res;
}
+void *xc_map_foreign_range(xc_interface *xch,
+ uint32_t dom, int size, int prot,
+ unsigned long mfn)
+{
+ xen_pfn_t *arr;
+ int num;
+ int i;
+ void *ret;
+
+ num = (size + XC_PAGE_SIZE - 1) >> XC_PAGE_SHIFT;
+ arr = calloc(num, sizeof(xen_pfn_t));
+ if ( arr == NULL )
+ return NULL;
+
+ for ( i = 0; i < num; i++ )
+ arr[i] = mfn + i;
+
+ ret = xc_map_foreign_pages(xch, dom, prot, arr, num);
+ free(arr);
+ return ret;
+}
+
+void *xc_map_foreign_ranges(xc_interface *xch,
+ uint32_t dom, size_t size,
+ int prot, size_t chunksize,
+ privcmd_mmap_entry_t entries[],
+ int nentries)
+{
+ xen_pfn_t *arr;
+ int num_per_entry;
+ int num;
+ int i;
+ int j;
+ void *ret;
+
+ num_per_entry = chunksize >> XC_PAGE_SHIFT;
+ num = num_per_entry * nentries;
+ arr = calloc(num, sizeof(xen_pfn_t));
+ if ( arr == NULL )
+ return NULL;
+
+ for ( i = 0; i < nentries; i++ )
+ for ( j = 0; j < num_per_entry; j++ )
+ arr[i * num_per_entry + j] = entries[i].mfn + j;
+
+ ret = xc_map_foreign_pages(xch, dom, prot, arr, num);
+ free(arr);
+ return ret;
+}
+
/*
* stub for all not yet converted OSes (NetBSD and Solaris). New OSes should
* just implement xc_map_foreign_bulk.
diff --git a/tools/libxc/xc_freebsd_osdep.c b/tools/libxc/xc_freebsd_osdep.c
index 6b440ee..7745d28 100644
--- a/tools/libxc/xc_freebsd_osdep.c
+++ b/tools/libxc/xc_freebsd_osdep.c
@@ -125,56 +125,6 @@ void *xc_map_foreign_bulk(xc_interface *xch,
return addr;
}
-void *xc_map_foreign_range(xc_interface *xch,
- uint32_t dom, int size, int prot,
- unsigned long mfn)
-{
- xen_pfn_t *arr;
- int num;
- int i;
- void *ret;
-
- num = (size + XC_PAGE_SIZE - 1) >> XC_PAGE_SHIFT;
- arr = calloc(num, sizeof(xen_pfn_t));
- if ( arr == NULL )
- return NULL;
-
- for ( i = 0; i < num; i++ )
- arr[i] = mfn + i;
-
- ret = xc_map_foreign_pages(xch, dom, prot, arr, num);
- free(arr);
- return ret;
-}
-
-void *xc_map_foreign_ranges(xc_interface *xch,
- uint32_t dom, size_t size,
- int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[],
- int nentries)
-{
- xen_pfn_t *arr;
- int num_per_entry;
- int num;
- int i;
- int j;
- void *ret;
-
- num_per_entry = chunksize >> XC_PAGE_SHIFT;
- num = num_per_entry * nentries;
- arr = calloc(num, sizeof(xen_pfn_t));
- if ( arr == NULL )
- return NULL;
-
- for ( i = 0; i < nentries; i++ )
- for ( j = 0; j < num_per_entry; j++ )
- arr[i * num_per_entry + j] = entries[i].mfn + j;
-
- ret = xc_map_foreign_pages(xch, dom, prot, arr, num);
- free(arr);
- return ret;
-}
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libxc/xc_linux_osdep.c
index 1404790..ae1dd9e 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libxc/xc_linux_osdep.c
@@ -284,55 +284,6 @@ void *xc_map_foreign_bulk(xc_interface *xch,
return addr;
}
-void *xc_map_foreign_range(xc_interface *xch,
- uint32_t dom, int size, int prot,
- unsigned long mfn)
-{
- xen_pfn_t *arr;
- int num;
- int i;
- void *ret;
-
- num = (size + XC_PAGE_SIZE - 1) >> XC_PAGE_SHIFT;
- arr = calloc(num, sizeof(xen_pfn_t));
- if ( arr == NULL )
- return NULL;
-
- for ( i = 0; i < num; i++ )
- arr[i] = mfn + i;
-
- ret = xc_map_foreign_pages(xch, dom, prot, arr, num);
- free(arr);
- return ret;
-}
-
-void *xc_map_foreign_ranges(xc_interface *xch,
- uint32_t dom, size_t size, int prot,
- size_t chunksize, privcmd_mmap_entry_t entries[],
- int nentries)
-{
- xen_pfn_t *arr;
- int num_per_entry;
- int num;
- int i;
- int j;
- void *ret;
-
- num_per_entry = chunksize >> XC_PAGE_SHIFT;
- num = num_per_entry * nentries;
- arr = calloc(num, sizeof(xen_pfn_t));
- if ( arr == NULL )
- return NULL;
-
- for ( i = 0; i < nentries; i++ )
- for ( j = 0; j < num_per_entry; j++ )
- arr[i * num_per_entry + j] = entries[i].mfn + j;
-
- ret = xc_map_foreign_pages(xch, dom, prot, arr, num);
- free(arr);
- return ret;
-}
-
/*
* Local variables:
* mode: C
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index 3ea3124..d8ca948 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -73,49 +73,6 @@ void *xc_map_foreign_bulk(xc_interface *xch,
return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
}
-void *xc_map_foreign_range(xc_interface *xch,
- uint32_t dom,
- int size, int prot,
- unsigned long mfn)
-{
- unsigned long pt_prot = 0;
-
- if (prot & PROT_READ)
- pt_prot = L1_PROT_RO;
- if (prot & PROT_WRITE)
- pt_prot = L1_PROT;
-
- assert(!(size % getpagesize()));
- return map_frames_ex(&mfn, size / getpagesize(), 0, 1, 1, dom, NULL, pt_prot);
-}
-
-void *xc_map_foreign_ranges(xc_interface *xch,
- uint32_t dom,
- size_t size, int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[], int nentries)
-{
- unsigned long *mfns;
- int i, j, n;
- unsigned long pt_prot = 0;
- void *ret;
-
- if (prot & PROT_READ)
- pt_prot = L1_PROT_RO;
- if (prot & PROT_WRITE)
- pt_prot = L1_PROT;
-
- mfns = malloc((size / XC_PAGE_SIZE) * sizeof(*mfns));
-
- n = 0;
- for (i = 0; i < nentries; i++)
- for (j = 0; j < chunksize / XC_PAGE_SIZE; j++)
- mfns[n++] = entries[i].mfn + j;
-
- ret = map_frames_ex(mfns, n, 1, 0, 1, dom, NULL, pt_prot);
- free(mfns);
- return ret;
-}
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
diff --git a/tools/libxc/xc_netbsd.c b/tools/libxc/xc_netbsd.c
index 6a7ff9f..3470bc4 100644
--- a/tools/libxc/xc_netbsd.c
+++ b/tools/libxc/xc_netbsd.c
@@ -96,76 +96,6 @@ void *osdep_map_foreign_batch(xc_interface *xch,
}
-void *xc_map_foreign_range(xc_interface *xch,
- uint32_t dom,
- int size, int prot,
- unsigned long mfn)
-{
- int fd = xch->privcmdfd;
- privcmd_mmap_t ioctlx;
- privcmd_mmap_entry_t entry;
- void *addr;
- addr = mmap(NULL, size, prot, MAP_ANON | MAP_SHARED, -1, 0);
- if ( addr == MAP_FAILED ) {
- PERROR("xc_map_foreign_range: mmap failed");
- return NULL;
- }
-
- ioctlx.num=1;
- ioctlx.dom=dom;
- ioctlx.entry=&entry;
- entry.va=(unsigned long) addr;
- entry.mfn=mfn;
- entry.npages=(size+XC_PAGE_SIZE-1)>>XC_PAGE_SHIFT;
- if ( ioctl(fd, IOCTL_PRIVCMD_MMAP, &ioctlx) < 0 )
- {
- int saved_errno = errno;
- PERROR("xc_map_foreign_range: ioctl failed");
- (void)munmap(addr, size);
- errno = saved_errno;
- return NULL;
- }
- return addr;
-}
-
-void *xc_map_foreign_ranges(xc_interface *xch,
- uint32_t dom,
- size_t size, int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[], int nentries)
-{
- int fd = xch->privcmdfd;
- privcmd_mmap_t ioctlx;
- int i, rc;
- void *addr;
-
- addr = mmap(NULL, size, prot, MAP_ANON | MAP_SHARED, -1, 0);
- if (addr == MAP_FAILED)
- goto mmap_failed;
-
- for (i = 0; i < nentries; i++) {
- entries[i].va = (uintptr_t)addr + (i * chunksize);
- entries[i].npages = chunksize >> XC_PAGE_SHIFT;
- }
-
- ioctlx.num = nentries;
- ioctlx.dom = dom;
- ioctlx.entry = entries;
-
- rc = ioctl(fd, IOCTL_PRIVCMD_MMAP, &ioctlx);
- if (rc)
- goto ioctl_failed;
-
- return addr;
-
-ioctl_failed:
- rc = munmap(addr, size);
- if (rc == -1)
- ERROR("%s: error in error path\n", __FUNCTION__);
-
-mmap_failed:
- return NULL;
-}
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
diff --git a/tools/libxc/xc_solaris.c b/tools/libxc/xc_solaris.c
index 2df7a06..d686867 100644
--- a/tools/libxc/xc_solaris.c
+++ b/tools/libxc/xc_solaris.c
@@ -94,73 +94,6 @@ void *osdep_map_foreign_batch(xc_interface *xch,
}
-void *xc_map_foreign_range(xc_interface *xch,
- uint32_t dom,
- int size, int prot,
- unsigned long mfn)
-{
- int fd = xch->privcmdfd;
- privcmd_mmap_t ioctlx;
- privcmd_mmap_entry_t entry;
- void *addr;
- addr = mmap(NULL, size, prot, MAP_SHARED, fd, 0);
- if ( addr == MAP_FAILED )
- return NULL;
-
- ioctlx.num=1;
- ioctlx.dom=dom;
- ioctlx.entry=&entry;
- entry.va=(unsigned long) addr;
- entry.mfn=mfn;
- entry.npages=(size+XC_PAGE_SIZE-1)>>XC_PAGE_SHIFT;
- if ( ioctl(fd, IOCTL_PRIVCMD_MMAP, &ioctlx) < 0 )
- {
- int saved_errno = errno;
- (void)munmap(addr, size);
- errno = saved_errno;
- return NULL;
- }
- return addr;
-}
-
-void *xc_map_foreign_ranges(xc_interface *xch,
- uint32_t dom,
- size_t size, int prot, size_t chunksize,
- privcmd_mmap_entry_t entries[], int nentries)
-{
- int fd = xch->privcmdfd;
- privcmd_mmap_t ioctlx;
- int i, rc;
- void *addr;
-
- addr = mmap(NULL, size, prot, MAP_SHARED, fd, 0);
- if (addr == MAP_FAILED)
- goto mmap_failed;
-
- for (i = 0; i < nentries; i++) {
- entries[i].va = (uintptr_t)addr + (i * chunksize);
- entries[i].npages = chunksize >> XC_PAGE_SHIFT;
- }
-
- ioctlx.num = nentries;
- ioctlx.dom = dom;
- ioctlx.entry = entries;
-
- rc = ioctl(fd, IOCTL_PRIVCMD_MMAP, &ioctlx);
- if (rc)
- goto ioctl_failed;
-
- return addr;
-
-ioctl_failed:
- rc = munmap(addr, size);
- if (rc == -1)
- PERROR("%s: error in error path", __FUNCTION__);
-
-mmap_failed:
- return NULL;
-}
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (11 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 12/23] tools: Implement xc_map_foreign_range(s) in terms of common helper Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 15:13 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap Ian Campbell
` (10 subsequent siblings)
23 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
libxenforeignmemory will provide a stable API and ABI for mapping
foreign domain memory (subject to appropriate privileges).
The new library exposes an interface equivalent to
xc_map_foreign_memory_bulk, which all the other
xc_map_foreign_memory_* functions (which remain in libxc) are
implemented in terms of.
Upon request (via #define XC_WANT_COMPAT_MAP_FOREIGN_API) libxenctrl
will provide a compat API for the old names. This is used by qemu-xen
and qemu-trad as well as various in tree things (which required
de-dupping various #includes in some too to get the #define before the
first).
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
Must be applied with:
- "qemu-xen-traditional: qemu-xen-traditional: Add
libxenforeignmemory to rpath-link" and a corresponding QEMU_TAG
update folded here.
- "mini-os: mini-os: Include libxenforeignmemory with libxc" and a
corresponding bump to MINIOS_UPSTREAM_REVISION folded in here.
v5: Allow close(NULL)
---
.gitignore | 2 +
stubdom/Makefile | 20 ++++-
tools/Makefile | 2 +
tools/Rules.mk | 11 ++-
tools/console/daemon/utils.c | 1 -
tools/console/daemon/utils.h | 1 +
tools/libs/Makefile | 1 +
tools/libs/foreignmemory/Makefile | 67 +++++++++++++++
tools/libs/foreignmemory/compat.c | 72 ++++++++++++++++
tools/libs/foreignmemory/core.c | 71 ++++++++++++++++
.../foreignmemory/freebsd.c} | 33 +++-----
.../libs/foreignmemory/include/xenforeignmemory.h | 71 ++++++++++++++++
tools/libs/foreignmemory/libxenforeignmemory.map | 7 ++
.../foreignmemory/linux.c} | 70 +++++++---------
tools/libs/foreignmemory/minios.c | 62 ++++++++++++++
tools/libs/foreignmemory/netbsd.c | 95 ++++++++++++++++++++++
tools/libs/foreignmemory/private.h | 47 +++++++++++
tools/libs/foreignmemory/solaris.c | 94 +++++++++++++++++++++
tools/libxc/Makefile | 8 +-
tools/libxc/include/xenctrl.h | 26 ------
tools/libxc/include/xenctrl_compat.h | 36 ++++++++
tools/libxc/xc_foreign_memory.c | 49 +----------
tools/libxc/xc_minios.c | 29 -------
tools/libxc/xc_netbsd.c | 73 -----------------
tools/libxc/xc_private.c | 13 +--
tools/libxc/xc_private.h | 11 ++-
tools/libxc/xc_solaris.c | 73 -----------------
tools/libxc/xc_sr_restore.c | 4 +-
tools/libxc/xc_sr_save.c | 4 +-
tools/libxc/xc_vm_event.c | 10 ++-
tools/libxc/xg_private.h | 3 +-
tools/libxl/libxl_internal.h | 1 +
tools/misc/xen-mfndump.c | 1 +
tools/ocaml/libs/xc/xenctrl_stubs.c | 1 +
tools/python/xen/lowlevel/xc/xc.c | 2 +-
tools/xenmon/xenbaked.c | 1 +
tools/xenpaging/pagein.c | 1 -
tools/xenpaging/xenpaging.c | 1 -
tools/xenpaging/xenpaging.h | 2 +
tools/xenstore/xenstored_core.c | 1 -
tools/xenstore/xenstored_core.h | 1 +
tools/xentrace/xenctx.c | 3 +-
tools/xentrace/xentrace.c | 1 +
43 files changed, 743 insertions(+), 339 deletions(-)
create mode 100644 tools/libs/foreignmemory/Makefile
create mode 100644 tools/libs/foreignmemory/compat.c
create mode 100644 tools/libs/foreignmemory/core.c
rename tools/{libxc/xc_freebsd_osdep.c => libs/foreignmemory/freebsd.c} (76%)
create mode 100644 tools/libs/foreignmemory/include/xenforeignmemory.h
create mode 100644 tools/libs/foreignmemory/libxenforeignmemory.map
rename tools/{libxc/xc_linux_osdep.c => libs/foreignmemory/linux.c} (83%)
create mode 100644 tools/libs/foreignmemory/minios.c
create mode 100644 tools/libs/foreignmemory/netbsd.c
create mode 100644 tools/libs/foreignmemory/private.h
create mode 100644 tools/libs/foreignmemory/solaris.c
diff --git a/.gitignore b/.gitignore
index 2899852..2549337 100644
--- a/.gitignore
+++ b/.gitignore
@@ -62,6 +62,7 @@ stubdom/libxentoollog-*
stubdom/libxenevtchn-*
stubdom/libxengnttab-*
stubdom/libxencall-*
+stubdom/libxenforeignmemory-*
stubdom/libxc-*
stubdom/lwip-*
stubdom/mini-os-*
@@ -92,6 +93,7 @@ tools/libs/toollog/headers.chk
tools/libs/evtchn/headers.chk
tools/libs/gnttab/headers.chk
tools/libs/call/headers.chk
+tools/libs/foreignmemory/headers.chk
tools/blktap2/daemon/blktapctrl
tools/blktap2/drivers/img2qcow
tools/blktap2/drivers/lock-util
diff --git a/stubdom/Makefile b/stubdom/Makefile
index 24f0e0f..dc47586 100644
--- a/stubdom/Makefile
+++ b/stubdom/Makefile
@@ -336,6 +336,12 @@ mk-headers-$(XEN_TARGET_ARCH): $(IOEMU_LINKFARM_TARGET)
ln -sf $(XEN_ROOT)/tools/libs/call/include/*.h include/ && \
ln -sf $(XEN_ROOT)/tools/libs/call/*.c . && \
ln -sf $(XEN_ROOT)/tools/libs/call/Makefile . )
+ mkdir -p libs-$(XEN_TARGET_ARCH)/foreignmemory/include
+ [ -h libs-$(XEN_TARGET_ARCH)/foreignmemory/Makefile ] || ( cd libs-$(XEN_TARGET_ARCH)/foreignmemory && \
+ ln -sf $(XEN_ROOT)/tools/libs/foreignmemory/*.h . && \
+ ln -sf $(XEN_ROOT)/tools/libs/foreignmemory/include/*.h include/ && \
+ ln -sf $(XEN_ROOT)/tools/libs/foreignmemory/*.c . && \
+ ln -sf $(XEN_ROOT)/tools/libs/foreignmemory/Makefile . )
mkdir -p libxc-$(XEN_TARGET_ARCH)
[ -h libxc-$(XEN_TARGET_ARCH)/Makefile ] || ( cd libxc-$(XEN_TARGET_ARCH) && \
ln -sf $(XEN_ROOT)/tools/libxc/*.h . && \
@@ -401,7 +407,17 @@ libs-$(XEN_TARGET_ARCH)/call/libxencall.a: $(NEWLIB_STAMPFILE)
$(MAKE) -C $(XEN_ROOT)/tools/include
$(MAKE) DESTDIR= -C $(MINI_OS) links
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/call
-APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/call -whole-archive -lxencall -no-whole-archive
+
+#######
+# libxenforeignmemory
+#######
+
+.PHONY: libxenforeignmemory
+libxenforeignmemory: libs-$(XEN_TARGET_ARCH)/foreignmemory/libxenforeignmemory.a
+libs-$(XEN_TARGET_ARCH)/foreignmemory/libxenforeignmemory.a: $(NEWLIB_STAMPFILE)
+ $(MAKE) -C $(XEN_ROOT)/tools/include
+ $(MAKE) DESTDIR= -C $(MINI_OS) links
+ CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= -C libs-$(XEN_TARGET_ARCH)/foreignmemory
#######
# libxc
@@ -409,7 +425,7 @@ APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/call -whole-archi
.PHONY: libxc
libxc: libxc-$(XEN_TARGET_ARCH)/libxenctrl.a libxc-$(XEN_TARGET_ARCH)/libxenguest.a
-libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog libxenevtchn libxengnttab libxencall cross-zlib
+libxc-$(XEN_TARGET_ARCH)/libxenctrl.a: libxentoollog libxenevtchn libxengnttab libxencall libxenforeignmemory cross-zlib
$(MAKE) -C $(XEN_ROOT)/tools/include
$(MAKE) DESTDIR= -C $(MINI_OS) links
CPPFLAGS="$(TARGET_CPPFLAGS)" CFLAGS="$(TARGET_CFLAGS)" $(MAKE) DESTDIR= CONFIG_LIBXC_MINIOS=y -C libxc-$(XEN_TARGET_ARCH)
diff --git a/tools/Makefile b/tools/Makefile
index e5bbc98..c03f6db 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -250,6 +250,7 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
--source-path=$$source \
--extra-cflags="-DXC_WANT_COMPAT_EVTCHN_API=1 \
-DXC_WANT_COMPAT_GNTTAB_API=1 \
+ -DXC_WANT_COMPAT_MAP_FOREIGN_API=1 \
-I$(XEN_ROOT)/tools/include \
-I$(XEN_ROOT)/tools/libs/toollog/include \
-I$(XEN_ROOT)/tools/libs/evtchn/include \
@@ -264,6 +265,7 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/evtchn \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/gnttab \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/call \
+ -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/foreignmemory \
$(QEMU_UPSTREAM_RPATH)" \
--bindir=$(LIBEXEC_BIN) \
--datadir=$(SHAREDIR)/qemu-xen \
diff --git a/tools/Rules.mk b/tools/Rules.mk
index 39426f4..134ddbc 100644
--- a/tools/Rules.mk
+++ b/tools/Rules.mk
@@ -14,6 +14,7 @@ XEN_LIBXENTOOLLOG = $(XEN_ROOT)/tools/libs/toollog
XEN_LIBXENEVTCHN = $(XEN_ROOT)/tools/libs/evtchn
XEN_LIBXENGNTTAB = $(XEN_ROOT)/tools/libs/gnttab
XEN_LIBXENCALL = $(XEN_ROOT)/tools/libs/call
+XEN_LIBXENFOREIGNMEMORY = $(XEN_ROOT)/tools/libs/foreignmemory
XEN_LIBXC = $(XEN_ROOT)/tools/libxc
XEN_XENLIGHT = $(XEN_ROOT)/tools/libxl
XEN_XENSTORE = $(XEN_ROOT)/tools/xenstore
@@ -103,12 +104,16 @@ CFLAGS_libxencall = -I$(XEN_LIBXENCALL)/include $(CFLAGS_xeninclude)
LDLIBS_libxencall = $(XEN_LIBXENCALL)/libxencall$(libextension)
SHLIB_libxencall = -Wl,-rpath-link=$(XEN_LIBXENCALL)
-CFLAGS_libxenctrl = -I$(XEN_LIBXC)/include $(CFLAGS_libxentoollog) $(CFLAGS_xeninclude)
-SHDEPS_libxenctrl = $(SHLIB_libxentoollog) $(SHLIB_libxenevtchn) $(SHLIB_libxengnttab) $(SHLIB_libxengntshr) $(SHLIB_libxencall)
+CFLAGS_libxenforeignmemory = -I$(XEN_LIBXENFOREIGNMEMORY)/include $(CFLAGS_xeninclude)
+LDLIBS_libxenforeignmemory = $(XEN_LIBXENFOREIGNMEMORY)/libxenforeignmemory$(libextension)
+SHLIB_libxenforeignmemory = -Wl,-rpath-link=$(XEN_LIBXENFOREIGNMEMORY)
+
+CFLAGS_libxenctrl = -I$(XEN_LIBXC)/include $(CFLAGS_libxentoollog) $(CFLAGS_libxenforeignmemory) $(CFLAGS_xeninclude)
+SHDEPS_libxenctrl = $(SHLIB_libxentoollog) $(SHLIB_libxenevtchn) $(SHLIB_libxengnttab) $(SHLIB_libxengntshr) $(SHLIB_libxencall) $(SHLIB_libxenforeignmemory)
LDLIBS_libxenctrl = $(SHDEPS_libxenctrl) $(XEN_LIBXC)/libxenctrl$(libextension)
SHLIB_libxenctrl = $(SHDEPS_libxenctrl) -Wl,-rpath-link=$(XEN_LIBXC)
-CFLAGS_libxenguest = -I$(XEN_LIBXC)/include $(CFLAGS_libxenevtchn) $(CFLAGS_xeninclude)
+CFLAGS_libxenguest = -I$(XEN_LIBXC)/include $(CFLAGS_libxenevtchn) $(CFLAGS_libxenforeignmemory) $(CFLAGS_xeninclude)
SHDEPS_libxenguest = $(SHLIB_libxenevtchn)
LDLIBS_libxenguest = $(SHDEPS_libxenguest) $(XEN_LIBXC)/libxenguest$(libextension)
SHLIB_libxenguest = $(SHDEPS_libxenguest) -Wl,-rpath-link=$(XEN_LIBXC)
diff --git a/tools/console/daemon/utils.c b/tools/console/daemon/utils.c
index 644f6af..97d7798 100644
--- a/tools/console/daemon/utils.c
+++ b/tools/console/daemon/utils.c
@@ -33,7 +33,6 @@
#include <string.h>
#include <signal.h>
-#include "xenctrl.h"
#include "utils.h"
struct xs_handle *xs;
diff --git a/tools/console/daemon/utils.h b/tools/console/daemon/utils.h
index 1295822..a010ac4 100644
--- a/tools/console/daemon/utils.h
+++ b/tools/console/daemon/utils.h
@@ -23,6 +23,7 @@
#include <stdbool.h>
#include <syslog.h>
#include <stdio.h>
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
#include <xenctrl.h>
#include <xenstore.h>
diff --git a/tools/libs/Makefile b/tools/libs/Makefile
index f4f5d57..dc5726d 100644
--- a/tools/libs/Makefile
+++ b/tools/libs/Makefile
@@ -6,5 +6,6 @@ SUBDIRS-y += toollog
SUBDIRS-y += evtchn
SUBDIRS-y += gnttab
SUBDIRS-y += call
+SUBDIRS-y += foreignmemory
all clean install distclean: %: subdirs-%
diff --git a/tools/libs/foreignmemory/Makefile b/tools/libs/foreignmemory/Makefile
new file mode 100644
index 0000000..530c3a9
--- /dev/null
+++ b/tools/libs/foreignmemory/Makefile
@@ -0,0 +1,67 @@
+XEN_ROOT = $(CURDIR)/../../..
+include $(XEN_ROOT)/tools/Rules.mk
+
+MAJOR = 1
+MINOR = 0
+SHLIB_LDFLAGS += -Wl,--version-script=libxenforeignmemory.map
+
+CFLAGS += -Werror -Wmissing-prototypes
+CFLAGS += -I./include $(CFLAGS_xeninclude)
+CFLAGS += $(CFLAGS_libxentoollog)
+
+SRCS-y += core.c
+SRCS-$(CONFIG_Linux) += linux.c
+SRCS-$(CONFIG_FreeBSD) += freebsd.c
+SRCS-$(CONFIG_SunOS) += compat.c solaris.c
+SRCS-$(CONFIG_NetBSD) += compat.c netbsd.c
+SRCS-$(CONFIG_MiniOS) += minios.c
+
+LIB_OBJS := $(patsubst %.c,%.o,$(SRCS-y))
+PIC_OBJS := $(patsubst %.c,%.opic,$(SRCS-y))
+
+LIB := libxenforeignmemory.a
+ifneq ($(nosharedlibs),y)
+LIB += libxenforeignmemory.so
+endif
+
+.PHONY: all
+all: build
+
+.PHONY: build
+build:
+ $(MAKE) libs
+
+.PHONY: libs
+libs: headers.chk $(LIB)
+
+headers.chk: $(wildcard include/*.h)
+
+libxenforeignmemory.a: $(LIB_OBJS)
+ $(AR) rc $@ $^
+
+libxenforeignmemory.so: libxenforeignmemory.so.$(MAJOR)
+ $(SYMLINK_SHLIB) $< $@
+libxenforeignmemory.so.$(MAJOR): libxenforeignmemory.so.$(MAJOR).$(MINOR)
+ $(SYMLINK_SHLIB) $< $@
+
+libxenforeignmemory.so.$(MAJOR).$(MINOR): $(PIC_OBJS) libxenforeignmemory.map
+ $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenforeignmemory.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $(PIC_OBJS) $(LDLIBS_libxentoollog) $(APPEND_LDFLAGS)
+
+.PHONY: install
+install: build
+ $(INSTALL_DIR) $(DESTDIR)$(libdir)
+ $(INSTALL_DIR) $(DESTDIR)$(includedir)
+ $(INSTALL_SHLIB) libxenforeignmemory.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)
+ $(INSTALL_DATA) libxenforeignmemory.a $(DESTDIR)$(libdir)
+ $(SYMLINK_SHLIB) libxenforeignmemory.so.$(MAJOR).$(MINOR) $(DESTDIR)$(libdir)/libxenforeignmemory.so.$(MAJOR)
+ $(SYMLINK_SHLIB) libxenforeignmemory.so.$(MAJOR) $(DESTDIR)$(libdir)/libxenforeignmemory.so
+ $(INSTALL_DATA) include/xenforeignmemory.h $(DESTDIR)$(includedir)
+
+.PHONY: TAGS
+TAGS:
+ etags -t *.c *.h
+
+.PHONY: clean
+clean:
+ rm -rf *.rpm $(LIB) *~ $(DEPS) $(LIB_OBJS) $(PIC_OBJS)
+ rm -f headers.chk
diff --git a/tools/libs/foreignmemory/compat.c b/tools/libs/foreignmemory/compat.c
new file mode 100644
index 0000000..df9702e
--- /dev/null
+++ b/tools/libs/foreignmemory/compat.c
@@ -0,0 +1,72 @@
+/* Compatibility functions for mapping foreign domain's memory.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_foreign_memory.c
+ */
+#if !defined(__NetBSD__) && !defined(__sun__)
+#error Please implement xc_map_foreign_bulk directly for new ports.
+#endif
+
+#include "private.h"
+
+void *xc_map_foreign_bulk(xenforeignmem_handle *fmem,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num)
+{
+ xen_pfn_t *pfn;
+ unsigned int i;
+ void *ret;
+
+ if ((int)num <= 0) {
+ errno = EINVAL;
+ return NULL;
+ }
+
+ pfn = malloc(num * sizeof(*pfn));
+ if (!pfn) {
+ errno = ENOMEM;
+ return NULL;
+ }
+
+ memcpy(pfn, arr, num * sizeof(*arr));
+ ret = osdep_map_foreign_batch(fmem, dom, prot, pfn, num);
+
+ if (ret) {
+ for (i = 0; i < num; ++i)
+ switch (pfn[i] ^ arr[i]) {
+ case 0:
+ err[i] = 0;
+ break;
+ default:
+ err[i] = -EINVAL;
+ break;
+ }
+ } else
+ memset(err, 0, num * sizeof(*err));
+
+ free(pfn);
+
+ return ret;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/foreignmemory/core.c b/tools/libs/foreignmemory/core.c
new file mode 100644
index 0000000..70a5e4d
--- /dev/null
+++ b/tools/libs/foreignmemory/core.c
@@ -0,0 +1,71 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <stdlib.h>
+
+#include "private.h"
+
+xenforeignmemory_handle *xenforeignmemory_open(xentoollog_logger *logger,
+ unsigned open_flags)
+{
+ xenforeignmemory_handle *fmem = malloc(sizeof(*fmem));
+ int rc;
+
+ if (!fmem) return NULL;
+
+ fmem->logger = logger;
+ fmem->logger_tofree = NULL;
+
+ if (!fmem->logger) {
+ fmem->logger = fmem->logger_tofree =
+ (xentoollog_logger*)
+ xtl_createlogger_stdiostream(stderr, XTL_PROGRESS, 0);
+ if (!fmem->logger) goto err;
+ }
+
+ rc = osdep_xenforeignmemory_open(fmem);
+ if ( rc < 0 ) goto err;
+
+ return fmem;
+
+err:
+ osdep_xenforeignmemory_close(fmem);
+ xtl_logger_destroy(fmem->logger_tofree);
+ free(fmem);
+ return NULL;
+}
+
+int xenforeignmemory_close(xenforeignmemory_handle *fmem)
+{
+ int rc;
+
+ if ( !fmem )
+ return 0;
+
+ rc = osdep_xenforeignmemory_close(fmem);
+ xtl_logger_destroy(fmem->logger_tofree);
+ free(fmem);
+ return rc;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libxc/xc_freebsd_osdep.c b/tools/libs/foreignmemory/freebsd.c
similarity index 76%
rename from tools/libxc/xc_freebsd_osdep.c
rename to tools/libs/foreignmemory/freebsd.c
index 7745d28..a7e0f6b 100644
--- a/tools/libxc/xc_freebsd_osdep.c
+++ b/tools/libs/foreignmemory/freebsd.c
@@ -3,9 +3,6 @@
* Copyright 2006 Sun Microsystems, Inc. All rights reserved.
* Use is subject to license terms.
*
- * xc_gnttab functions:
- * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
- *
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation;
@@ -30,14 +27,11 @@
#include <sys/mman.h>
#include <sys/ioctl.h>
-#include <xen/memory.h>
-
-#include "xc_private.h"
+#include "private.h"
#define PRIVCMD_DEV "/dev/xen/privcmd"
-/*------------------------- Privcmd device interface -------------------------*/
-int osdep_privcmd_open(xc_interface *xch)
+int osdep_xenforeignmemory_open(xenforeignmemory_handle *fmem)
{
int flags, saved_errno;
int fd = open(PRIVCMD_DEV, O_RDWR);
@@ -46,7 +40,7 @@ int osdep_privcmd_open(xc_interface *xch)
{
PERROR("Could not obtain handle on privileged command interface "
PRIVCMD_DEV);
- return -1
+ return -1;
}
/*
@@ -69,7 +63,7 @@ int osdep_privcmd_open(xc_interface *xch)
goto error;
}
- xch->privcmdfd = fd;
+ fmem->fd = fd;
return 0;
error:
@@ -80,26 +74,25 @@ int osdep_privcmd_open(xc_interface *xch)
return -1;
}
-int osdep_privcmd_close(xc_interface *xch)
+int osdep_xenforeignmemory_close(xenforeignmemory_handle *fmem)
{
- int fd = xch->privcmdfd;
+ int fd = fmem->fd;
if ( fd == -1 )
return 0;
return close(fd);
}
-/*----------------------- Privcmd foreign map interface ----------------------*/
-void *xc_map_foreign_bulk(xc_interface *xch,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err,
- unsigned int num)
+void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err,
+ unsigned int num)
{
- int fd = xch->privcmdfd;
+ int fd = fmem->fd;
privcmd_mmapbatch_t ioctlx;
void *addr;
int rc;
- addr = mmap(NULL, num << XC_PAGE_SHIFT, prot, MAP_SHARED, fd, 0);
+ addr = mmap(NULL, num << PAGE_SHIFT, prot, MAP_SHARED, fd, 0);
if ( addr == MAP_FAILED )
{
PERROR("xc_map_foreign_bulk: mmap failed");
@@ -117,7 +110,7 @@ void *xc_map_foreign_bulk(xc_interface *xch,
{
int saved_errno = errno;
PERROR("xc_map_foreign_bulk: ioctl failed");
- (void)munmap(addr, num << XC_PAGE_SHIFT);
+ (void)munmap(addr, num << PAGE_SHIFT);
errno = saved_errno;
return NULL;
}
diff --git a/tools/libs/foreignmemory/include/xenforeignmemory.h b/tools/libs/foreignmemory/include/xenforeignmemory.h
new file mode 100644
index 0000000..0909585
--- /dev/null
+++ b/tools/libs/foreignmemory/include/xenforeignmemory.h
@@ -0,0 +1,71 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+#ifndef XENFOREIGNMEMORY_H
+#define XENFOREIGNMEMORY_H
+
+/*
+ * This library allows you to map foreign domain memory, subject to
+ * permissions for both the process and the domain in which the
+ * process runs.
+ */
+
+#include <stdint.h>
+#include <stddef.h>
+
+#include <xen/xen.h>
+
+/* Callers who don't care don't need to #include <xentoollog.h> */
+typedef struct xentoollog_logger xentoollog_logger;
+
+typedef struct xenforeignmemory_handle xenforeignmemory_handle;
+
+/*
+ * Return a handle onto the hypercall driver. Logs errors.
+ */
+xenforeignmemory_handle *xenforeignmemory_open(xentoollog_logger *logger,
+ unsigned open_flags);
+
+/*
+ * Close a handle previously allocated with xenforeignmemory_open().
+ */
+int xenforeignmemory_close(xenforeignmemory_handle *xmem);
+
+/*
+ * Maps a range within one domain to a local address range. Mappings
+ * should be unmapped with munmap and should follow the same rules as mmap
+ * regarding page alignment.
+ *
+ * prot is as for mmap(2).
+ *
+ * Can partially succeed. When a page cannot be mapped, its respective
+ * field in @err is set to the corresponding errno value.
+ *
+ * Returns NULL if no pages can be mapped.
+ */
+void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
+ int prot, const xen_pfn_t *arr, int *err,
+ unsigned int num);
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/foreignmemory/libxenforeignmemory.map b/tools/libs/foreignmemory/libxenforeignmemory.map
new file mode 100644
index 0000000..11f0d2b
--- /dev/null
+++ b/tools/libs/foreignmemory/libxenforeignmemory.map
@@ -0,0 +1,7 @@
+VERS_1.0 {
+ global:
+ xenforeignmemory_open;
+ xenforeignmemory_close;
+ xenforeignmemory_map;
+ local: *; /* Do not expose anything by default */
+};
diff --git a/tools/libxc/xc_linux_osdep.c b/tools/libs/foreignmemory/linux.c
similarity index 83%
rename from tools/libxc/xc_linux_osdep.c
rename to tools/libs/foreignmemory/linux.c
index ae1dd9e..01cd42e 100644
--- a/tools/libxc/xc_linux_osdep.c
+++ b/tools/libs/foreignmemory/linux.c
@@ -1,11 +1,4 @@
- /******************************************************************************
- *
- * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
- * Use is subject to license terms.
- *
- * xc_gnttab functions:
- * Copyright (c) 2007-2008, D G Murray <Derek.Murray@cl.cam.ac.uk>
- *
+/*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation;
@@ -18,27 +11,26 @@
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_linus_osdep.c:
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
*/
+#include <alloca.h>
#include <errno.h>
#include <fcntl.h>
-#include <stdint.h>
-#include <stdlib.h>
-#include <string.h>
#include <unistd.h>
+#include <string.h>
#include <sys/mman.h>
#include <sys/ioctl.h>
-#include <xen/memory.h>
-
-#include "xenctrl.h"
-
-#include "xc_private.h"
+#include "private.h"
#define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & ~((1UL<<(_w))-1))
-int osdep_privcmd_open(xc_interface *xch)
+int osdep_xenforeignmemory_open(xenforeignmemory_handle *fmem)
{
int flags, saved_errno;
int fd = open("/proc/xen/privcmd", O_RDWR);
@@ -67,7 +59,7 @@ int osdep_privcmd_open(xc_interface *xch)
goto error;
}
- xch->privcmdfd = fd;
+ fmem->fd = fd;
return 0;
error:
@@ -77,9 +69,9 @@ int osdep_privcmd_open(xc_interface *xch)
return -1;
}
-int osdep_privcmd_close(xc_interface *xch)
+int osdep_xenforeignmemory_close(xenforeignmemory_handle *fmem)
{
- int fd = xch->privcmdfd;
+ int fd = fmem->fd;
if (fd == -1)
return 0;
return close(fd);
@@ -121,7 +113,7 @@ static int retry_paged(int fd, uint32_t dom, void *addr,
{
privcmd_mmapbatch_v2_t ioctlx;
int rc, paged = 0, i = 0;
-
+
do
{
/* Skip gfns not in paging state */
@@ -136,10 +128,10 @@ static int retry_paged(int fd, uint32_t dom, void *addr,
/* At least one gfn is still in paging state */
ioctlx.num = 1;
ioctlx.dom = dom;
- ioctlx.addr = (unsigned long)addr + ((unsigned long)i<<XC_PAGE_SHIFT);
+ ioctlx.addr = (unsigned long)addr + ((unsigned long)i<<PAGE_SHIFT);
ioctlx.arr = arr + i;
ioctlx.err = err + i;
-
+
/* Assemble a batch of requests */
while ( ++i < num )
{
@@ -147,34 +139,34 @@ static int retry_paged(int fd, uint32_t dom, void *addr,
break;
ioctlx.num++;
}
-
+
/* Send request and abort on fatal error */
rc = ioctl(fd, IOCTL_PRIVCMD_MMAPBATCH_V2, &ioctlx);
if ( rc < 0 && errno != ENOENT )
goto out;
} while ( i < num );
-
+
rc = paged;
out:
return rc;
}
-void *xc_map_foreign_bulk(xc_interface *xch,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
+void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num)
{
- int fd = xch->privcmdfd;
+ int fd = fmem->fd;
privcmd_mmapbatch_v2_t ioctlx;
void *addr;
unsigned int i;
int rc;
- addr = mmap(NULL, (unsigned long)num << XC_PAGE_SHIFT, prot, MAP_SHARED,
+ addr = mmap(NULL, (unsigned long)num << PAGE_SHIFT, prot, MAP_SHARED,
fd, 0);
if ( addr == MAP_FAILED )
{
- PERROR("xc_map_foreign_bulk: mmap failed");
+ PERROR("mmap failed");
return NULL;
}
@@ -203,9 +195,9 @@ void *xc_map_foreign_bulk(xc_interface *xch,
*/
privcmd_mmapbatch_t ioctlx;
xen_pfn_t *pfn;
- unsigned int pfn_arr_size = ROUNDUP((num * sizeof(*pfn)), XC_PAGE_SHIFT);
+ unsigned int pfn_arr_size = ROUNDUP((num * sizeof(*pfn)), PAGE_SHIFT);
- if ( pfn_arr_size <= XC_PAGE_SIZE )
+ if ( pfn_arr_size <= PAGE_SIZE )
pfn = alloca(num * sizeof(*pfn));
else
{
@@ -213,8 +205,8 @@ void *xc_map_foreign_bulk(xc_interface *xch,
MAP_PRIVATE | MAP_ANON | MAP_POPULATE, -1, 0);
if ( pfn == MAP_FAILED )
{
- PERROR("xc_map_foreign_bulk: mmap of pfn array failed");
- (void)munmap(addr, (unsigned long)num << XC_PAGE_SHIFT);
+ PERROR("mmap of pfn array failed");
+ (void)munmap(addr, (unsigned long)num << PAGE_SHIFT);
return NULL;
}
}
@@ -247,7 +239,7 @@ void *xc_map_foreign_bulk(xc_interface *xch,
continue;
}
rc = map_foreign_batch_single(fd, dom, pfn + i,
- (unsigned long)addr + ((unsigned long)i<<XC_PAGE_SHIFT));
+ (unsigned long)addr + ((unsigned long)i<<PAGE_SHIFT));
if ( rc < 0 )
{
rc = -errno;
@@ -259,7 +251,7 @@ void *xc_map_foreign_bulk(xc_interface *xch,
break;
}
- if ( pfn_arr_size > XC_PAGE_SIZE )
+ if ( pfn_arr_size > PAGE_SIZE )
munmap(pfn, pfn_arr_size);
if ( rc == -ENOENT && i == num )
@@ -275,8 +267,8 @@ void *xc_map_foreign_bulk(xc_interface *xch,
{
int saved_errno = errno;
- PERROR("xc_map_foreign_bulk: ioctl failed");
- (void)munmap(addr, (unsigned long)num << XC_PAGE_SHIFT);
+ PERROR("ioctl failed");
+ (void)munmap(addr, (unsigned long)num << PAGE_SHIFT);
errno = saved_errno;
return NULL;
}
diff --git a/tools/libs/foreignmemory/minios.c b/tools/libs/foreignmemory/minios.c
new file mode 100644
index 0000000..981d801
--- /dev/null
+++ b/tools/libs/foreignmemory/minios.c
@@ -0,0 +1,62 @@
+/*
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Split out from xc_minios.c
+ *
+ * Copyright 2007-2008 Samuel Thibault <samuel.thibault@eu.citrix.com>.
+ */
+
+#include <mini-os/types.h>
+#include <mini-os/os.h>
+#include <mini-os/mm.h>
+#include <mini-os/lib.h>
+
+#include <errno.h>
+
+#include <sys/mman.h>
+
+#include "private.h"
+
+int osdep_xenforeignmemory_open(xenforeignmemory_handle *fmem)
+{
+ /* No fd required */
+ return 0;
+}
+
+int osdep_xenforeignmemory_close(xenforeignmemory_handle *fmem)
+{
+ return 0;
+}
+
+void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num)
+{
+ unsigned long pt_prot = 0;
+ if (prot & PROT_READ)
+ pt_prot = L1_PROT_RO;
+ if (prot & PROT_WRITE)
+ pt_prot = L1_PROT;
+ return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/foreignmemory/netbsd.c b/tools/libs/foreignmemory/netbsd.c
new file mode 100644
index 0000000..e84d5ec
--- /dev/null
+++ b/tools/libs/foreignmemory/netbsd.c
@@ -0,0 +1,95 @@
+/******************************************************************************
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <unistd.h>
+#include <fcntl.h>
+#include <sys/mman.h>
+
+#include "private.h"
+
+int osdep_xenforeignmemory_open(xenforeignmemory_handle *fmem)
+{
+ int flags, saved_errno;
+ int fd = open("/kern/xen/privcmd", O_RDWR);
+
+ if ( fd == -1 )
+ {
+ PERROR("Could not obtain handle on privileged command interface");
+ return -1;
+ }
+
+ /* Although we return the file handle as the 'xc handle' the API
+ does not specify / guarentee that this integer is in fact
+ a file handle. Thus we must take responsiblity to ensure
+ it doesn't propagate (ie leak) outside the process */
+ if ( (flags = fcntl(fd, F_GETFD)) < 0 )
+ {
+ PERROR("Could not get file handle flags");
+ goto error;
+ }
+ flags |= FD_CLOEXEC;
+ if ( fcntl(fd, F_SETFD, flags) < 0 )
+ {
+ PERROR("Could not set file handle flags");
+ goto error;
+ }
+
+ fmem->fd = fd;
+ return 0;
+
+ error:
+ saved_errno = errno;
+ close(fd);
+ errno = saved_errno;
+ return -1;
+}
+
+int osdep_xenforeignmemory_close(xenforeignmemory_handle *fmem)
+{
+ int fd = fmem->fd;
+ return close(fd);
+}
+
+void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
+ int prot, xen_pfn_t *arr, int num)
+{
+ int fd = fmem->fd;
+ privcmd_mmapbatch_t ioctlx;
+ void *addr;
+ addr = mmap(NULL, num*XC_PAGE_SIZE, prot, MAP_ANON | MAP_SHARED, -1, 0);
+ if ( addr == MAP_FAILED ) {
+ PERROR("osdep_map_foreign_batch: mmap failed");
+ return NULL;
+ }
+
+ ioctlx.num=num;
+ ioctlx.dom=dom;
+ ioctlx.addr=(unsigned long)addr;
+ ioctlx.arr=arr;
+ if ( ioctl(fd, IOCTL_PRIVCMD_MMAPBATCH, &ioctlx) < 0 )
+ {
+ int saved_errno = errno;
+ PERROR("osdep_map_foreign_batch: ioctl failed");
+ (void)munmap(addr, num*XC_PAGE_SIZE);
+ errno = saved_errno;
+ return NULL;
+ }
+ return addr;
+
+}
diff --git a/tools/libs/foreignmemory/private.h b/tools/libs/foreignmemory/private.h
new file mode 100644
index 0000000..54dd7d5
--- /dev/null
+++ b/tools/libs/foreignmemory/private.h
@@ -0,0 +1,47 @@
+#ifndef XENFOREIGNMEMORY_PRIVATE_H
+#define XENFOREIGNMEMORY_PRIVATE_H
+
+#include <xentoollog.h>
+
+#include <xenforeignmemory.h>
+
+#include <xen/xen.h>
+#include <xen/sys/privcmd.h>
+
+#ifndef PAGE_SHIFT /* Mini-os, Yukk */
+#define PAGE_SHIFT 12
+#endif
+#ifndef __MINIOS__ /* Yukk */
+#define PAGE_SIZE (1UL << PAGE_SHIFT)
+#define PAGE_MASK (~(PAGE_SIZE-1))
+#endif
+
+struct xenforeignmemory_handle {
+ xentoollog_logger *logger, *logger_tofree;
+ unsigned flags;
+ int fd;
+};
+
+int osdep_xenforeignmemory_open(xenforeignmemory_handle *fmem);
+int osdep_xenforeignmemory_close(xenforeignmemory_handle *fmem);
+
+#if defined(__NetBSD__) || defined(__sun__)
+/* Strictly compat for those two only only */
+void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
+ int prot, xen_pfn_t *arr, int num);
+#endif
+
+#define PERROR(_f...) \
+ xtl_log(fmem->logger, XTL_ERROR, errno, "xenforeignmemory", _f)
+
+#endif
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/foreignmemory/solaris.c b/tools/libs/foreignmemory/solaris.c
new file mode 100644
index 0000000..1d27b50
--- /dev/null
+++ b/tools/libs/foreignmemory/solaris.c
@@ -0,0 +1,94 @@
+/******************************************************************************
+ *
+ * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation;
+ * version 2.1 of the License.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "xc_private.h"
+
+#include <xen/memory.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <malloc.h>
+
+int osdep_xenforeignmemory_open(xenforeignmemory_handle *fmem)
+{
+ int flags, saved_errno;
+ int fd = open("/dev/xen/privcmd", O_RDWR);
+
+ if ( fd == -1 )
+ {
+ PERROR("Could not obtain handle on privileged command interface");
+ return -1;
+ }
+
+ /* Although we return the file handle as the 'xc handle' the API
+ does not specify / guarentee that this integer is in fact
+ a file handle. Thus we must take responsiblity to ensure
+ it doesn't propagate (ie leak) outside the process */
+ if ( (flags = fcntl(fd, F_GETFD)) < 0 )
+ {
+ PERROR("Could not get file handle flags");
+ goto error;
+ }
+ flags |= FD_CLOEXEC;
+ if ( fcntl(fd, F_SETFD, flags) < 0 )
+ {
+ PERROR("Could not set file handle flags");
+ goto error;
+ }
+
+ fmem->fd = fd;
+ return 0;
+
+ error:
+ saved_errno = errno;
+ close(fd);
+ errno = saved_errno;
+ return -1;
+}
+
+int osdep_xenforeignmemory_close(xenforeignmemory_handle *fmem)
+{
+ int fd = fmem->fd;
+ return close(fd);
+}
+
+void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
+ int prot, xen_pfn_t *arr, int num)
+{
+ int fd = fmem->fd;
+ privcmd_mmapbatch_t ioctlx;
+ void *addr;
+ addr = mmap(NULL, num*XC_PAGE_SIZE, prot, MAP_SHARED, fd, 0);
+ if ( addr == MAP_FAILED )
+ return NULL;
+
+ ioctlx.num=num;
+ ioctlx.dom=dom;
+ ioctlx.addr=(unsigned long)addr;
+ ioctlx.arr=arr;
+ if ( ioctl(fd, IOCTL_PRIVCMD_MMAPBATCH, &ioctlx) < 0 )
+ {
+ int saved_errno = errno;
+ PERROR("XXXXXXXX");
+ (void)munmap(addr, num*XC_PAGE_SIZE);
+ errno = saved_errno;
+ return NULL;
+ }
+ return addr;
+
+}
diff --git a/tools/libxc/Makefile b/tools/libxc/Makefile
index a122f73..ca8bab7 100644
--- a/tools/libxc/Makefile
+++ b/tools/libxc/Makefile
@@ -42,8 +42,8 @@ CTRL_SRCS-y += xc_kexec.c
CTRL_SRCS-y += xc_resource.c
CTRL_SRCS-$(CONFIG_X86) += xc_psr.c
CTRL_SRCS-$(CONFIG_X86) += xc_pagetab.c
-CTRL_SRCS-$(CONFIG_Linux) += xc_linux.c xc_linux_osdep.c
-CTRL_SRCS-$(CONFIG_FreeBSD) += xc_freebsd.c xc_freebsd_osdep.c
+CTRL_SRCS-$(CONFIG_Linux) += xc_linux.c
+CTRL_SRCS-$(CONFIG_FreeBSD) += xc_freebsd.c
CTRL_SRCS-$(CONFIG_SunOS) += xc_solaris.c
CTRL_SRCS-$(CONFIG_NetBSD) += xc_netbsd.c
CTRL_SRCS-$(CONFIG_NetBSDRump) += xc_netbsd.c
@@ -125,7 +125,7 @@ $(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS): CFLAGS += -include $(XEN_ROOT)/tools/config.
# libxenguest includes xc_private.h, so needs this despite not using
# this functionality directly.
$(CTRL_LIB_OBJS) $(GUEST_LIB_OBJS) \
-$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS): CFLAGS += $(CFLAGS_libxencall)
+$(CTRL_PIC_OBJS) $(GUEST_PIC_OBJS): CFLAGS += $(CFLAGS_libxencall) $(CFLAGS_libxenforeignmemory)
$(CTRL_LIB_OBJS) $(CTRL_PIC_OBJS): CFLAGS += $(CFLAGS_libxengnttab) $(CFLAGS_libxengntshr)
@@ -207,7 +207,7 @@ libxenctrl.so.$(MAJOR): libxenctrl.so.$(MAJOR).$(MINOR)
$(SYMLINK_SHLIB) $< $@
libxenctrl.so.$(MAJOR).$(MINOR): $(CTRL_PIC_OBJS)
- $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_libxengntshr) $(LDLIBS_libxencall) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
+ $(CC) $(LDFLAGS) $(PTHREAD_LDFLAGS) -Wl,$(SONAME_LDFLAG) -Wl,libxenctrl.so.$(MAJOR) $(SHLIB_LDFLAGS) -o $@ $^ $(LDLIBS_libxentoollog) $(LDLIBS_libxenevtchn) $(LDLIBS_libxengnttab) $(LDLIBS_libxengntshr) $(LDLIBS_libxencall) $(LDLIBS_libxenforeignmemory) $(PTHREAD_LIBS) $(APPEND_LDFLAGS)
# libxenguest
diff --git a/tools/libxc/include/xenctrl.h b/tools/libxc/include/xenctrl.h
index 685315e..463e63b 100644
--- a/tools/libxc/include/xenctrl.h
+++ b/tools/libxc/include/xenctrl.h
@@ -1377,32 +1377,6 @@ int xc_lockprof_query(xc_interface *xch,
void *xc_memalign(xc_interface *xch, size_t alignment, size_t size);
/**
- * Memory maps a range within one domain to a local address range. Mappings
- * should be unmapped with munmap and should follow the same rules as mmap
- * regarding page alignment. Returns NULL on failure.
- *
- * @parm xch a handle on an open hypervisor interface
- * @parm dom the domain to map memory from
- * @parm size the amount of memory to map (in multiples of page size)
- * @parm prot same flag as in mmap().
- * @parm mfn the frame address to map.
- */
-void *xc_map_foreign_range(xc_interface *xch, uint32_t dom,
- int size, int prot,
- unsigned long mfn );
-
-void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
- const xen_pfn_t *arr, int num );
-
-/**
- * Like xc_map_foreign_pages(), except it can succeed partially.
- * When a page cannot be mapped, its respective field in @err is
- * set to the corresponding errno value.
- */
-void *xc_map_foreign_bulk(xc_interface *xch, uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num);
-
-/**
* Translates a virtual address in the context of a given domain and
* vcpu returning the GFN containing the address (that is, an MFN for
* PV guests, a PFN for HVM guests). Returns 0 for failure.
diff --git a/tools/libxc/include/xenctrl_compat.h b/tools/libxc/include/xenctrl_compat.h
index d99fa11..54c23a4 100644
--- a/tools/libxc/include/xenctrl_compat.h
+++ b/tools/libxc/include/xenctrl_compat.h
@@ -12,6 +12,42 @@
#ifndef XENCTRL_COMPAT_H
#define XENCTRL_COMPAT_H
+#ifdef XC_WANT_COMPAT_MAP_FOREIGN_API
+/**
+ * Memory maps a range within one domain to a local address range. Mappings
+ * should be unmapped with munmap and should follow the same rules as mmap
+ * regarding page alignment. Returns NULL on failure.
+ *
+ * @parm xch a handle on an open hypervisor interface
+ * @parm dom the domain to map memory from
+ * @parm size the amount of memory to map (in multiples of page size)
+ * @parm prot same flag as in mmap().
+ * @parm mfn the frame address to map.
+ */
+void *xc_map_foreign_range(xc_interface *xch, uint32_t dom,
+ int size, int prot,
+ unsigned long mfn );
+
+void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
+ const xen_pfn_t *arr, int num );
+
+/* Nothing within the library itself other than the compat wrapper
+ * itself should be using this, everything inside has access to
+ * xenforeignmemory_map().
+ */
+#if !defined(XC_INTERNAL_COMPAT_MAP_FOREIGN_API) || \
+ defined(XC_BUILDING_COMPAT_MAP_FOREIGN_API)
+/**
+ * Like xc_map_foreign_pages(), except it can succeed partially.
+ * When a page cannot be mapped, its respective field in @err is
+ * set to the corresponding errno value.
+ */
+void *xc_map_foreign_bulk(xc_interface *xch, uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, unsigned int num);
+#endif
+
+#endif
+
#ifdef XC_WANT_COMPAT_EVTCHN_API
typedef struct xenevtchn_handle xc_evtchn;
diff --git a/tools/libxc/xc_foreign_memory.c b/tools/libxc/xc_foreign_memory.c
index d1130e6..1737c10 100644
--- a/tools/libxc/xc_foreign_memory.c
+++ b/tools/libxc/xc_foreign_memory.c
@@ -17,6 +17,7 @@
* License along with this library; If not, see <http://www.gnu.org/licenses/>.
*/
+#define XC_BUILDING_COMPAT_MAP_FOREIGN_API
#include "xc_private.h"
void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
@@ -34,7 +35,7 @@ void *xc_map_foreign_pages(xc_interface *xch, uint32_t dom, int prot,
if (!err)
return NULL;
- res = xc_map_foreign_bulk(xch, dom, prot, arr, err, num);
+ res = xenforeignmemory_map(xch->fmem, dom, prot, arr, err, num);
if (res) {
for (i = 0; i < num; i++) {
if (err[i]) {
@@ -100,53 +101,11 @@ void *xc_map_foreign_ranges(xc_interface *xch,
return ret;
}
-/*
- * stub for all not yet converted OSes (NetBSD and Solaris). New OSes should
- * just implement xc_map_foreign_bulk.
- */
-#if defined(__NetBSD__) || defined(__sun__)
-void *osdep_map_foreign_batch(xc_interface *xch, uint32_t dom, int prot,
- xen_pfn_t *arr, int num );
-void *xc_map_foreign_bulk(xc_interface *xch,
- uint32_t dom, int prot,
+void *xc_map_foreign_bulk(xc_interface *xch, uint32_t dom, int prot,
const xen_pfn_t *arr, int *err, unsigned int num)
{
- xen_pfn_t *pfn;
- unsigned int i;
- void *ret;
-
- if ((int)num <= 0) {
- errno = EINVAL;
- return NULL;
- }
-
- pfn = malloc(num * sizeof(*pfn));
- if (!pfn) {
- errno = ENOMEM;
- return NULL;
- }
-
- memcpy(pfn, arr, num * sizeof(*arr));
- ret = osdep_map_foreign_batch(xch, dom, prot, pfn, num);
-
- if (ret) {
- for (i = 0; i < num; ++i)
- switch (pfn[i] ^ arr[i]) {
- case 0:
- err[i] = 0;
- break;
- default:
- err[i] = -EINVAL;
- break;
- }
- } else
- memset(err, 0, num * sizeof(*err));
-
- free(pfn);
-
- return ret;
+ return xenforeignmemory_map(xch->fmem, dom, prot, arr, err, num);
}
-#endif
/*
* Local variables:
diff --git a/tools/libxc/xc_minios.c b/tools/libxc/xc_minios.c
index d8ca948..1799daa 100644
--- a/tools/libxc/xc_minios.c
+++ b/tools/libxc/xc_minios.c
@@ -39,40 +39,11 @@ void minios_interface_close_fd(int fd);
extern void minios_interface_close_fd(int fd);
-int osdep_privcmd_open(xc_interface *xch)
-{
- int fd = alloc_fd(FTYPE_XC);
-
- if ( fd == -1)
- return -1;
-
- xch->privcmdfd = fd;
- return 0;
-}
-
-int osdep_privcmd_close(xc_interface *xch)
-{
- int fd = xch->privcmdfd;
- return close(fd);
-}
-
void minios_interface_close_fd(int fd)
{
files[fd].type = FTYPE_NONE;
}
-void *xc_map_foreign_bulk(xc_interface *xch,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
-{
- unsigned long pt_prot = 0;
- if (prot & PROT_READ)
- pt_prot = L1_PROT_RO;
- if (prot & PROT_WRITE)
- pt_prot = L1_PROT;
- return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
-}
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
diff --git a/tools/libxc/xc_netbsd.c b/tools/libxc/xc_netbsd.c
index 3470bc4..3197993 100644
--- a/tools/libxc/xc_netbsd.c
+++ b/tools/libxc/xc_netbsd.c
@@ -22,79 +22,6 @@
#include <unistd.h>
#include <fcntl.h>
#include <malloc.h>
-#include <sys/mman.h>
-
-int osdep_privcmd_open(xc_interface *xch)
-{
- int flags, saved_errno;
- int fd = open("/kern/xen/privcmd", O_RDWR);
-
- if ( fd == -1 )
- {
- PERROR("Could not obtain handle on privileged command interface");
- return -1;
- }
-
- /* Although we return the file handle as the 'xc handle' the API
- does not specify / guarentee that this integer is in fact
- a file handle. Thus we must take responsiblity to ensure
- it doesn't propagate (ie leak) outside the process */
- if ( (flags = fcntl(fd, F_GETFD)) < 0 )
- {
- PERROR("Could not get file handle flags");
- goto error;
- }
- flags |= FD_CLOEXEC;
- if ( fcntl(fd, F_SETFD, flags) < 0 )
- {
- PERROR("Could not set file handle flags");
- goto error;
- }
-
- xch->privcmdfd = fd;
- return 0;
-
- error:
- saved_errno = errno;
- close(fd);
- errno = saved_errno;
- return -1;
-}
-
-int osdep_privcmd_close(xc_interface *xch)
-{
- int fd = xch->privcmdfd;
- return close(fd);
-}
-
-void *osdep_map_foreign_batch(xc_interface *xch,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
-{
- int fd = xch->privcmdfd;
- privcmd_mmapbatch_t ioctlx;
- void *addr;
- addr = mmap(NULL, num*XC_PAGE_SIZE, prot, MAP_ANON | MAP_SHARED, -1, 0);
- if ( addr == MAP_FAILED ) {
- PERROR("osdep_map_foreign_batch: mmap failed");
- return NULL;
- }
-
- ioctlx.num=num;
- ioctlx.dom=dom;
- ioctlx.addr=(unsigned long)addr;
- ioctlx.arr=arr;
- if ( ioctl(fd, IOCTL_PRIVCMD_MMAPBATCH, &ioctlx) < 0 )
- {
- int saved_errno = errno;
- PERROR("osdep_map_foreign_batch: ioctl failed");
- (void)munmap(addr, num*XC_PAGE_SIZE);
- errno = saved_errno;
- return NULL;
- }
- return addr;
-
-}
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
diff --git a/tools/libxc/xc_private.c b/tools/libxc/xc_private.c
index b3a660e..5690a89 100644
--- a/tools/libxc/xc_private.c
+++ b/tools/libxc/xc_private.c
@@ -58,11 +58,12 @@ struct xc_interface_core *xc_interface_open(xentoollog_logger *logger,
if (open_flags & XC_OPENFLAG_DUMMY)
return xch; /* We are done */
- if ( osdep_privcmd_open(xch) < 0 )
- goto err;
-
xch->xcall = xencall_open(xch->error_handler,
open_flags & XC_OPENFLAG_NON_REENTRANT ? XENCALL_OPENFLAG_NON_REENTRANT : 0U);
+ if ( xch->xcall == NULL )
+ goto err;
+
+ xch->fmem = xenforeignmemory_open(xch->error_handler, 0);
if ( xch->xcall == NULL )
goto err;
@@ -70,7 +71,7 @@ struct xc_interface_core *xc_interface_open(xentoollog_logger *logger,
return xch;
err:
- osdep_privcmd_close(xch);
+ xencall_close(xch->xcall);
xtl_logger_destroy(xch->error_handler_tofree);
if (xch != &xch_buf) free(xch);
return NULL;
@@ -86,8 +87,8 @@ int xc_interface_close(xc_interface *xch)
rc = xencall_close(xch->xcall);
if (rc) PERROR("Could not close xencall interface");
- rc = osdep_privcmd_close(xch);
- if (rc) PERROR("Could not close hypervisor interface");
+ rc = xenforeignmemory_close(xch->fmem);
+ if (rc) PERROR("Could not close foreign memory interface");
xtl_logger_destroy(xch->dombuild_logger_tofree);
xtl_logger_destroy(xch->error_handler_tofree);
diff --git a/tools/libxc/xc_private.h b/tools/libxc/xc_private.h
index 0de907c..0b1104e 100644
--- a/tools/libxc/xc_private.h
+++ b/tools/libxc/xc_private.h
@@ -29,8 +29,13 @@
#include <sys/ioctl.h>
#include "_paths.h"
+
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
+#define XC_INTERNAL_COMPAT_MAP_FOREIGN_API
#include "xenctrl.h"
+
#include <xencall.h>
+#include <xenforeignmemory.h>
#include <xen/sys/privcmd.h>
@@ -94,11 +99,11 @@ struct xc_interface_core {
FILE *dombuild_logger_file;
const char *currently_progress_reporting;
- /* Privcmd interface */
- int privcmdfd;
-
/* Hypercall interface */
xencall_handle *xcall;
+
+ /* Foreign mappings */
+ xenforeignmemory_handle *fmem;
};
int osdep_privcmd_open(xc_interface *xch);
diff --git a/tools/libxc/xc_solaris.c b/tools/libxc/xc_solaris.c
index d686867..5128f3f 100644
--- a/tools/libxc/xc_solaris.c
+++ b/tools/libxc/xc_solaris.c
@@ -19,81 +19,8 @@
#include "xc_private.h"
-#include <xen/memory.h>
-#include <unistd.h>
-#include <fcntl.h>
#include <malloc.h>
-int osdep_privcmd_open(xc_interface *xch)
-{
- int flags, saved_errno;
- int fd = open("/dev/xen/privcmd", O_RDWR);
-
- if ( fd == -1 )
- {
- PERROR("Could not obtain handle on privileged command interface");
- return -1;
- }
-
- /* Although we return the file handle as the 'xc handle' the API
- does not specify / guarentee that this integer is in fact
- a file handle. Thus we must take responsiblity to ensure
- it doesn't propagate (ie leak) outside the process */
- if ( (flags = fcntl(fd, F_GETFD)) < 0 )
- {
- PERROR("Could not get file handle flags");
- goto error;
- }
- flags |= FD_CLOEXEC;
- if ( fcntl(fd, F_SETFD, flags) < 0 )
- {
- PERROR("Could not set file handle flags");
- goto error;
- }
-
- xch->privcmdfd = fd;
- return 0;
-
- error:
- saved_errno = errno;
- close(fd);
- errno = saved_errno;
- return -1;
-}
-
-int osdep_privcmd_close(xc_interface *xch)
-{
- int fd = xch->privcmdfd;
- return close(fd);
-}
-
-void *osdep_map_foreign_batch(xc_interface *xch,
- uint32_t dom, int prot,
- xen_pfn_t *arr, int num)
-{
- int fd = xch->privcmdfd;
- privcmd_mmapbatch_t ioctlx;
- void *addr;
- addr = mmap(NULL, num*XC_PAGE_SIZE, prot, MAP_SHARED, fd, 0);
- if ( addr == MAP_FAILED )
- return NULL;
-
- ioctlx.num=num;
- ioctlx.dom=dom;
- ioctlx.addr=(unsigned long)addr;
- ioctlx.arr=arr;
- if ( ioctl(fd, IOCTL_PRIVCMD_MMAPBATCH, &ioctlx) < 0 )
- {
- int saved_errno = errno;
- PERROR("XXXXXXXX");
- (void)munmap(addr, num*XC_PAGE_SIZE);
- errno = saved_errno;
- return NULL;
- }
- return addr;
-
-}
-
/* Optionally flush file to disk and discard page cache */
void discard_file_cache(xc_interface *xch, int fd, int flush)
{
diff --git a/tools/libxc/xc_sr_restore.c b/tools/libxc/xc_sr_restore.c
index f48e7fc..2e41af8 100644
--- a/tools/libxc/xc_sr_restore.c
+++ b/tools/libxc/xc_sr_restore.c
@@ -316,8 +316,8 @@ static int process_page_data(struct xc_sr_context *ctx, unsigned count,
if ( nr_pages == 0 )
goto done;
- mapping = guest_page = xc_map_foreign_bulk(
- xch, ctx->domid, PROT_READ | PROT_WRITE,
+ mapping = guest_page = xenforeignmemory_map(xch->fmem,
+ ctx->domid, PROT_READ | PROT_WRITE,
mfns, map_errs, nr_pages);
if ( !mapping )
{
diff --git a/tools/libxc/xc_sr_save.c b/tools/libxc/xc_sr_save.c
index d1bdf72..a5ac517 100644
--- a/tools/libxc/xc_sr_save.c
+++ b/tools/libxc/xc_sr_save.c
@@ -153,8 +153,8 @@ static int write_batch(struct xc_sr_context *ctx)
if ( nr_pages > 0 )
{
- guest_mapping = xc_map_foreign_bulk(
- xch, ctx->domid, PROT_READ, mfns, errors, nr_pages);
+ guest_mapping = xenforeignmemory_map(xch->fmem,
+ ctx->domid, PROT_READ, mfns, errors, nr_pages);
if ( !guest_mapping )
{
PERROR("Failed to map guest pages");
diff --git a/tools/libxc/xc_vm_event.c b/tools/libxc/xc_vm_event.c
index d74d0f0..df21f9c 100644
--- a/tools/libxc/xc_vm_event.c
+++ b/tools/libxc/xc_vm_event.c
@@ -73,8 +73,9 @@ void *xc_vm_event_enable(xc_interface *xch, domid_t domain_id, int param,
ring_pfn = pfn;
mmap_pfn = pfn;
- ring_page = xc_map_foreign_bulk(xch, domain_id, PROT_READ | PROT_WRITE,
- &mmap_pfn, &err, 1);
+ ring_page = xenforeignmemory_map(xch->fmem,
+ domain_id, PROT_READ | PROT_WRITE,
+ &mmap_pfn, &err, 1);
if ( !ring_page )
{
/* Map failed, populate ring page */
@@ -87,8 +88,9 @@ void *xc_vm_event_enable(xc_interface *xch, domid_t domain_id, int param,
}
mmap_pfn = ring_pfn;
- ring_page = xc_map_foreign_bulk(xch, domain_id, PROT_READ | PROT_WRITE,
- &mmap_pfn, &err, 1);
+ ring_page = xenforeignmemory_map(xch->fmem,
+ domain_id, PROT_READ | PROT_WRITE,
+ &mmap_pfn, &err, 1);
if ( !ring_page )
{
PERROR("Could not map the ring page: %d\n", err);
diff --git a/tools/libxc/xg_private.h b/tools/libxc/xg_private.h
index 5544897..16da428 100644
--- a/tools/libxc/xg_private.h
+++ b/tools/libxc/xg_private.h
@@ -26,9 +26,8 @@
#include <sys/types.h>
#include <sys/stat.h>
-#include "xenctrl.h"
-#include "xenguest.h"
#include "xc_private.h"
+#include "xenguest.h"
#include <xen/memory.h>
#include <xen/elfnote.h>
diff --git a/tools/libxl/libxl_internal.h b/tools/libxl/libxl_internal.h
index 3caa40d..3ce46da 100644
--- a/tools/libxl/libxl_internal.h
+++ b/tools/libxl/libxl_internal.h
@@ -50,6 +50,7 @@
#include <xenevtchn.h>
#include <xenstore.h>
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
#include <xenctrl.h>
#include <xenguest.h>
diff --git a/tools/misc/xen-mfndump.c b/tools/misc/xen-mfndump.c
index ceeeaa9..1b22ad7 100644
--- a/tools/misc/xen-mfndump.c
+++ b/tools/misc/xen-mfndump.c
@@ -1,3 +1,4 @@
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
#include <xenctrl.h>
#include <xc_private.h>
#include <xc_core.h>
diff --git a/tools/ocaml/libs/xc/xenctrl_stubs.c b/tools/ocaml/libs/xc/xenctrl_stubs.c
index 29055cc..d43ed78 100644
--- a/tools/ocaml/libs/xc/xenctrl_stubs.c
+++ b/tools/ocaml/libs/xc/xenctrl_stubs.c
@@ -29,6 +29,7 @@
#include <stdint.h>
#include <string.h>
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
#include <xenctrl.h>
#include "mmap_stubs.h"
diff --git a/tools/python/xen/lowlevel/xc/xc.c b/tools/python/xen/lowlevel/xc/xc.c
index 26290a3..4ff3f1c 100644
--- a/tools/python/xen/lowlevel/xc/xc.c
+++ b/tools/python/xen/lowlevel/xc/xc.c
@@ -5,6 +5,7 @@
*/
#include <Python.h>
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
#include <xenctrl.h>
#include <xenguest.h>
#include <zlib.h>
@@ -17,7 +18,6 @@
#include <netdb.h>
#include <arpa/inet.h>
-#include "xenctrl.h"
#include <xen/elfnote.h>
#include <xen/tmem.h>
#include "xc_dom.h"
diff --git a/tools/xenmon/xenbaked.c b/tools/xenmon/xenbaked.c
index 5cfc6af..bf018aa 100644
--- a/tools/xenmon/xenbaked.c
+++ b/tools/xenmon/xenbaked.c
@@ -38,6 +38,7 @@
#include <errno.h>
#include <signal.h>
#include <xenevtchn.h>
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
#include <xenctrl.h>
#include <xen/xen.h>
#include <string.h>
diff --git a/tools/xenpaging/pagein.c b/tools/xenpaging/pagein.c
index 18699ec..0ea8680 100644
--- a/tools/xenpaging/pagein.c
+++ b/tools/xenpaging/pagein.c
@@ -1,6 +1,5 @@
/* Trigger a page-in in a separate thread-of-execution to avoid deadlock */
#include <pthread.h>
-#include <xc_private.h>
#include "xenpaging.h"
struct page_in_args {
diff --git a/tools/xenpaging/xenpaging.c b/tools/xenpaging/xenpaging.c
index 18b27a4..e2c283d 100644
--- a/tools/xenpaging/xenpaging.c
+++ b/tools/xenpaging/xenpaging.c
@@ -27,7 +27,6 @@
#include <signal.h>
#include <unistd.h>
#include <poll.h>
-#include <xc_private.h>
#include <xenstore.h>
#include <getopt.h>
diff --git a/tools/xenpaging/xenpaging.h b/tools/xenpaging/xenpaging.h
index d0f8d20..d6c8ee5 100644
--- a/tools/xenpaging/xenpaging.h
+++ b/tools/xenpaging/xenpaging.h
@@ -25,6 +25,8 @@
#include <xenevtchn.h>
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
+#include <xenctrl.h>
#include <xc_private.h>
#include <xen/event_channel.h>
#include <xen/vm_event.h>
diff --git a/tools/xenstore/xenstored_core.c b/tools/xenstore/xenstored_core.c
index f62c192..51fb0b3 100644
--- a/tools/xenstore/xenstored_core.c
+++ b/tools/xenstore/xenstored_core.c
@@ -50,7 +50,6 @@
#include "xenstored_watch.h"
#include "xenstored_transaction.h"
#include "xenstored_domain.h"
-#include "xenctrl.h"
#include "tdb.h"
#include "hashtable.h"
diff --git a/tools/xenstore/xenstored_core.h b/tools/xenstore/xenstored_core.h
index 624737d..3a497f7 100644
--- a/tools/xenstore/xenstored_core.h
+++ b/tools/xenstore/xenstored_core.h
@@ -19,6 +19,7 @@
#ifndef _XENSTORED_CORE_H
#define _XENSTORED_CORE_H
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
#include <xenctrl.h>
#include <xengnttab.h>
diff --git a/tools/xentrace/xenctx.c b/tools/xentrace/xenctx.c
index 4ed120a..e647179 100644
--- a/tools/xentrace/xenctx.c
+++ b/tools/xentrace/xenctx.c
@@ -25,7 +25,8 @@
#include <getopt.h>
#include <limits.h>
-#include "xenctrl.h"
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
+#include <xenctrl.h>
#include <xen/foreign/x86_32.h>
#include <xen/foreign/x86_64.h>
#include <xen/hvm/save.h>
diff --git a/tools/xentrace/xentrace.c b/tools/xentrace/xentrace.c
index f66a77c..28c2b3c 100644
--- a/tools/xentrace/xentrace.c
+++ b/tools/xentrace/xentrace.c
@@ -30,6 +30,7 @@
#include <xen/xen.h>
#include <xen/trace.h>
+#define XC_WANT_COMPAT_MAP_FOREIGN_API
#include <xenevtchn.h>
#include <xenctrl.h>
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (12 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 15:14 ` Ian Jackson
2015-11-13 16:16 ` Andrew Cooper
2015-11-09 12:00 ` [PATCH XEN v5 15/23] foreignmemory: use size_t for size arguments Ian Campbell
` (9 subsequent siblings)
23 siblings, 2 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
And require it be used instead of direct munmap.
This will allow e.g. Valgrind hooks to help track incorrect use of
foreign mappings.
Switch all uses of xenforeignmemory_map to use
xenforeignmemory_unmap, not that foreign mappings via the libxc compat
xc_map_foreign_* interface will not take advantage of this and will
need converting.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
v4: xenforeignmemory_unmap takes pages not bytes, adjust callers.
---
tools/libs/foreignmemory/freebsd.c | 6 ++++++
tools/libs/foreignmemory/include/xenforeignmemory.h | 7 +++++--
tools/libs/foreignmemory/libxenforeignmemory.map | 1 +
tools/libs/foreignmemory/linux.c | 6 ++++++
tools/libs/foreignmemory/minios.c | 6 ++++++
tools/libs/foreignmemory/netbsd.c | 6 ++++++
tools/libs/foreignmemory/solaris.c | 6 ++++++
tools/libxc/xc_sr_restore.c | 2 +-
tools/libxc/xc_sr_save.c | 2 +-
tools/libxc/xc_vm_event.c | 2 +-
10 files changed, 39 insertions(+), 5 deletions(-)
diff --git a/tools/libs/foreignmemory/freebsd.c b/tools/libs/foreignmemory/freebsd.c
index a7e0f6b..f9e74fa 100644
--- a/tools/libs/foreignmemory/freebsd.c
+++ b/tools/libs/foreignmemory/freebsd.c
@@ -118,6 +118,12 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
return addr;
}
+int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
+ void *addr, unsigned int num)
+{
+ return munmap(addr, num << PAGE_SHIFT);
+}
+
/*
* Local variables:
* mode: C
diff --git a/tools/libs/foreignmemory/include/xenforeignmemory.h b/tools/libs/foreignmemory/include/xenforeignmemory.h
index 0909585..1bcdf6a 100644
--- a/tools/libs/foreignmemory/include/xenforeignmemory.h
+++ b/tools/libs/foreignmemory/include/xenforeignmemory.h
@@ -44,8 +44,8 @@ int xenforeignmemory_close(xenforeignmemory_handle *xmem);
/*
* Maps a range within one domain to a local address range. Mappings
- * should be unmapped with munmap and should follow the same rules as mmap
- * regarding page alignment.
+ * must be unmapped with xenforeignmemory_unmap and should follow the
+ * same rules as mmap regarding page alignment.
*
* prot is as for mmap(2).
*
@@ -58,6 +58,9 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
int prot, const xen_pfn_t *arr, int *err,
unsigned int num);
+int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
+ void *addr, unsigned int num);
+
#endif
/*
diff --git a/tools/libs/foreignmemory/libxenforeignmemory.map b/tools/libs/foreignmemory/libxenforeignmemory.map
index 11f0d2b..df206b3 100644
--- a/tools/libs/foreignmemory/libxenforeignmemory.map
+++ b/tools/libs/foreignmemory/libxenforeignmemory.map
@@ -3,5 +3,6 @@ VERS_1.0 {
xenforeignmemory_open;
xenforeignmemory_close;
xenforeignmemory_map;
+ xenforeignmemory_unmap;
local: *; /* Do not expose anything by default */
};
diff --git a/tools/libs/foreignmemory/linux.c b/tools/libs/foreignmemory/linux.c
index 01cd42e..86a5a97 100644
--- a/tools/libs/foreignmemory/linux.c
+++ b/tools/libs/foreignmemory/linux.c
@@ -276,6 +276,12 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
return addr;
}
+int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
+ void *addr, unsigned int num)
+{
+ return munmap(addr, (unsigned long)num << PAGE_SHIFT);
+}
+
/*
* Local variables:
* mode: C
diff --git a/tools/libs/foreignmemory/minios.c b/tools/libs/foreignmemory/minios.c
index 981d801..dbb152f 100644
--- a/tools/libs/foreignmemory/minios.c
+++ b/tools/libs/foreignmemory/minios.c
@@ -51,6 +51,12 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
}
+int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
+ void *addr, unsigned int num)
+{
+ return munmap(addr, num << PAGE_SHIFT);
+}
+
/*
* Local variables:
* mode: C
diff --git a/tools/libs/foreignmemory/netbsd.c b/tools/libs/foreignmemory/netbsd.c
index e84d5ec..4d68b52 100644
--- a/tools/libs/foreignmemory/netbsd.c
+++ b/tools/libs/foreignmemory/netbsd.c
@@ -93,3 +93,9 @@ void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
return addr;
}
+
+int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
+ void *addr, unsigned int num)
+{
+ return munmap(addr, num*XC_PAGE_SIZE);
+}
diff --git a/tools/libs/foreignmemory/solaris.c b/tools/libs/foreignmemory/solaris.c
index 1d27b50..3f8e705 100644
--- a/tools/libs/foreignmemory/solaris.c
+++ b/tools/libs/foreignmemory/solaris.c
@@ -92,3 +92,9 @@ void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
return addr;
}
+
+int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
+ void *addr, unsigned int num)
+{
+ return munmap(addr, num*XC_PAGE_SIZE);
+}
diff --git a/tools/libxc/xc_sr_restore.c b/tools/libxc/xc_sr_restore.c
index 2e41af8..c96b8c2 100644
--- a/tools/libxc/xc_sr_restore.c
+++ b/tools/libxc/xc_sr_restore.c
@@ -378,7 +378,7 @@ static int process_page_data(struct xc_sr_context *ctx, unsigned count,
err:
if ( mapping )
- munmap(mapping, nr_pages * PAGE_SIZE);
+ xenforeignmemory_unmap(xch->fmem, mapping, nr_pages);
free(map_errs);
free(mfns);
diff --git a/tools/libxc/xc_sr_save.c b/tools/libxc/xc_sr_save.c
index a5ac517..ac393ab 100644
--- a/tools/libxc/xc_sr_save.c
+++ b/tools/libxc/xc_sr_save.c
@@ -263,7 +263,7 @@ static int write_batch(struct xc_sr_context *ctx)
err:
free(rec_pfns);
if ( guest_mapping )
- munmap(guest_mapping, nr_pages_mapped * PAGE_SIZE);
+ xenforeignmemory_unmap(xch->fmem, guest_mapping, nr_pages_mapped);
for ( i = 0; local_pages && i < nr_pfns; ++i )
free(local_pages[i]);
free(iov);
diff --git a/tools/libxc/xc_vm_event.c b/tools/libxc/xc_vm_event.c
index df21f9c..e388278 100644
--- a/tools/libxc/xc_vm_event.c
+++ b/tools/libxc/xc_vm_event.c
@@ -151,7 +151,7 @@ void *xc_vm_event_enable(xc_interface *xch, domid_t domain_id, int param,
}
if ( ring_page )
- munmap(ring_page, XC_PAGE_SIZE);
+ xenforeignmemory_unmap(xch->fmem, ring_page, 1);
ring_page = NULL;
errno = saved_errno;
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 15/23] foreignmemory: use size_t for size arguments.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (13 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 15:15 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 16/23] tools/libs/evtchn: Review and update doc comments Ian Campbell
` (8 subsequent siblings)
23 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
Surprisingly it appears no callers need updating.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
v4: New patch
---
tools/libs/foreignmemory/compat.c | 2 +-
tools/libs/foreignmemory/freebsd.c | 4 ++--
.../libs/foreignmemory/include/xenforeignmemory.h | 4 ++--
tools/libs/foreignmemory/linux.c | 25 +++++++++++-----------
tools/libs/foreignmemory/minios.c | 6 +++---
tools/libs/foreignmemory/netbsd.c | 2 +-
tools/libs/foreignmemory/solaris.c | 2 +-
7 files changed, 23 insertions(+), 22 deletions(-)
diff --git a/tools/libs/foreignmemory/compat.c b/tools/libs/foreignmemory/compat.c
index df9702e..46c4b55 100644
--- a/tools/libs/foreignmemory/compat.c
+++ b/tools/libs/foreignmemory/compat.c
@@ -23,7 +23,7 @@
void *xc_map_foreign_bulk(xenforeignmem_handle *fmem,
uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
+ const xen_pfn_t *arr, int *err, size_t num)
{
xen_pfn_t *pfn;
unsigned int i;
diff --git a/tools/libs/foreignmemory/freebsd.c b/tools/libs/foreignmemory/freebsd.c
index f9e74fa..0d71713 100644
--- a/tools/libs/foreignmemory/freebsd.c
+++ b/tools/libs/foreignmemory/freebsd.c
@@ -85,7 +85,7 @@ int osdep_xenforeignmemory_close(xenforeignmemory_handle *fmem)
void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
uint32_t dom, int prot,
const xen_pfn_t *arr, int *err,
- unsigned int num)
+ size_t num)
{
int fd = fmem->fd;
privcmd_mmapbatch_t ioctlx;
@@ -119,7 +119,7 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
}
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
- void *addr, unsigned int num)
+ void *addr, size_t num)
{
return munmap(addr, num << PAGE_SHIFT);
}
diff --git a/tools/libs/foreignmemory/include/xenforeignmemory.h b/tools/libs/foreignmemory/include/xenforeignmemory.h
index 1bcdf6a..99ec883 100644
--- a/tools/libs/foreignmemory/include/xenforeignmemory.h
+++ b/tools/libs/foreignmemory/include/xenforeignmemory.h
@@ -56,10 +56,10 @@ int xenforeignmemory_close(xenforeignmemory_handle *xmem);
*/
void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
int prot, const xen_pfn_t *arr, int *err,
- unsigned int num);
+ size_t pages);
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
- void *addr, unsigned int num);
+ void *addr, size_t pages);
#endif
diff --git a/tools/libs/foreignmemory/linux.c b/tools/libs/foreignmemory/linux.c
index 86a5a97..3c08679 100644
--- a/tools/libs/foreignmemory/linux.c
+++ b/tools/libs/foreignmemory/linux.c
@@ -109,10 +109,11 @@ static int map_foreign_batch_single(int fd, uint32_t dom,
* This will keep the request ring full and avoids delays.
*/
static int retry_paged(int fd, uint32_t dom, void *addr,
- const xen_pfn_t *arr, int *err, unsigned int num)
+ const xen_pfn_t *arr, int *err, size_t num)
{
privcmd_mmapbatch_v2_t ioctlx;
- int rc, paged = 0, i = 0;
+ int rc, paged = 0;
+ size_t i = 0;
do
{
@@ -128,7 +129,7 @@ static int retry_paged(int fd, uint32_t dom, void *addr,
/* At least one gfn is still in paging state */
ioctlx.num = 1;
ioctlx.dom = dom;
- ioctlx.addr = (unsigned long)addr + ((unsigned long)i<<PAGE_SHIFT);
+ ioctlx.addr = (unsigned long)addr + (i<<PAGE_SHIFT);
ioctlx.arr = arr + i;
ioctlx.err = err + i;
@@ -153,16 +154,16 @@ out:
}
void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, size_t num)
{
int fd = fmem->fd;
privcmd_mmapbatch_v2_t ioctlx;
void *addr;
- unsigned int i;
+ size_t i;
int rc;
- addr = mmap(NULL, (unsigned long)num << PAGE_SHIFT, prot, MAP_SHARED,
+ addr = mmap(NULL, num << PAGE_SHIFT, prot, MAP_SHARED,
fd, 0);
if ( addr == MAP_FAILED )
{
@@ -206,7 +207,7 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
if ( pfn == MAP_FAILED )
{
PERROR("mmap of pfn array failed");
- (void)munmap(addr, (unsigned long)num << PAGE_SHIFT);
+ (void)munmap(addr, num << PAGE_SHIFT);
return NULL;
}
}
@@ -239,7 +240,7 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
continue;
}
rc = map_foreign_batch_single(fd, dom, pfn + i,
- (unsigned long)addr + ((unsigned long)i<<PAGE_SHIFT));
+ (unsigned long)addr + (i<<PAGE_SHIFT));
if ( rc < 0 )
{
rc = -errno;
@@ -268,7 +269,7 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
int saved_errno = errno;
PERROR("ioctl failed");
- (void)munmap(addr, (unsigned long)num << PAGE_SHIFT);
+ (void)munmap(addr, num << PAGE_SHIFT);
errno = saved_errno;
return NULL;
}
@@ -277,9 +278,9 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
}
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
- void *addr, unsigned int num)
+ void *addr, size_t num)
{
- return munmap(addr, (unsigned long)num << PAGE_SHIFT);
+ return munmap(addr, num << PAGE_SHIFT);
}
/*
diff --git a/tools/libs/foreignmemory/minios.c b/tools/libs/foreignmemory/minios.c
index dbb152f..3d3c711 100644
--- a/tools/libs/foreignmemory/minios.c
+++ b/tools/libs/foreignmemory/minios.c
@@ -40,8 +40,8 @@ int osdep_xenforeignmemory_close(xenforeignmemory_handle *fmem)
}
void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
- uint32_t dom, int prot,
- const xen_pfn_t *arr, int *err, unsigned int num)
+ uint32_t dom, int prot,
+ const xen_pfn_t *arr, int *err, size_t num)
{
unsigned long pt_prot = 0;
if (prot & PROT_READ)
@@ -52,7 +52,7 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
}
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
- void *addr, unsigned int num)
+ void *addr, size_t num)
{
return munmap(addr, num << PAGE_SHIFT);
}
diff --git a/tools/libs/foreignmemory/netbsd.c b/tools/libs/foreignmemory/netbsd.c
index 4d68b52..747d5b4 100644
--- a/tools/libs/foreignmemory/netbsd.c
+++ b/tools/libs/foreignmemory/netbsd.c
@@ -95,7 +95,7 @@ void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
}
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
- void *addr, unsigned int num)
+ void *addr, size_t num)
{
return munmap(addr, num*XC_PAGE_SIZE);
}
diff --git a/tools/libs/foreignmemory/solaris.c b/tools/libs/foreignmemory/solaris.c
index 3f8e705..359e664 100644
--- a/tools/libs/foreignmemory/solaris.c
+++ b/tools/libs/foreignmemory/solaris.c
@@ -94,7 +94,7 @@ void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
}
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
- void *addr, unsigned int num)
+ void *addr, size_t num)
{
return munmap(addr, num*XC_PAGE_SIZE);
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 16/23] tools/libs/evtchn: Review and update doc comments.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (14 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 15/23] foreignmemory: use size_t for size arguments Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 15:15 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 17/23] tools/libs: Clean up hard tabs Ian Campbell
` (7 subsequent siblings)
23 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
Remove the reference to pre-4.1, since this is now a new library.
Fixup references to xc.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libs/evtchn/include/xenevtchn.h | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/tools/libs/evtchn/include/xenevtchn.h b/tools/libs/evtchn/include/xenevtchn.h
index 3380fa3..60da2a3 100644
--- a/tools/libs/evtchn/include/xenevtchn.h
+++ b/tools/libs/evtchn/include/xenevtchn.h
@@ -46,11 +46,9 @@ typedef struct xentoollog_logger xentoollog_logger;
* which case errno will be set appropriately.
*
* Note:
- * After fork a child process must not use any opened xc evtchn
+ * After fork a child process must not use any opened evtchn
* handle inherited from their parent. They must open a new handle if
- * they want to interact with xc.
- *
- * Before Xen pre-4.1 this function would sometimes report errors with perror.
+ * they want to interact with evtchn.
*/
/* Currently no flags are defined */
xenevtchn_handle *xenevtchn_open(xentoollog_logger *logger, unsigned open_flags);
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 17/23] tools/libs: Clean up hard tabs.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (15 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 16/23] tools/libs/evtchn: Review and update doc comments Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 18/23] tools/libs/gnttab: Review and update doc comments Ian Campbell
` (6 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
These were wrong in the context of libxc before this code was
extracted, clean them up.
Also add some emacs magic blocks
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libs/call/buffer.c | 4 ++--
tools/libs/call/core.c | 2 +-
tools/libs/call/minios.c | 4 ++--
tools/libs/evtchn/minios.c | 28 ++++++++++++++--------------
tools/libs/evtchn/netbsd.c | 12 ++++++------
tools/libs/foreignmemory/freebsd.c | 2 +-
tools/libs/foreignmemory/minios.c | 6 +++---
tools/libs/foreignmemory/netbsd.c | 12 +++++++++++-
tools/libs/foreignmemory/private.h | 2 +-
tools/libs/foreignmemory/solaris.c | 13 +++++++++++--
tools/libs/gnttab/include/xengnttab.h | 4 ++--
tools/libs/gnttab/linux.c | 6 +++---
tools/libs/toollog/xtl_logger_stdio.c | 2 +-
13 files changed, 58 insertions(+), 39 deletions(-)
diff --git a/tools/libs/call/buffer.c b/tools/libs/call/buffer.c
index fbbd7ff..da27135 100644
--- a/tools/libs/call/buffer.c
+++ b/tools/libs/call/buffer.c
@@ -20,7 +20,7 @@
#include "private.h"
#define DBGPRINTF(_m...) \
- xtl_log(xcall->logger, XTL_DEBUG, -1, "xencall:buffer", _m)
+ xtl_log(xcall->logger, XTL_DEBUG, -1, "xencall:buffer", _m)
#define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & ~((1UL<<(_w))-1))
@@ -86,7 +86,7 @@ static int cache_free(xencall_handle *xcall, void *p, int nr_pages)
xcall->buffer_current_allocations--;
if ( nr_pages == 1 &&
- xcall->buffer_cache_nr < BUFFER_CACHE_SIZE )
+ xcall->buffer_cache_nr < BUFFER_CACHE_SIZE )
{
xcall->buffer_cache[xcall->buffer_cache_nr++] = p;
rc = 1;
diff --git a/tools/libs/call/core.c b/tools/libs/call/core.c
index a342871..bbf88de 100644
--- a/tools/libs/call/core.c
+++ b/tools/libs/call/core.c
@@ -19,7 +19,7 @@
xencall_handle *xencall_open(xentoollog_logger *logger, unsigned open_flags)
{
- xencall_handle *xcall = malloc(sizeof(*xcall));
+ xencall_handle *xcall = malloc(sizeof(*xcall));
int rc;
if (!xcall) return NULL;
diff --git a/tools/libs/call/minios.c b/tools/libs/call/minios.c
index 1cdc073..524fa04 100644
--- a/tools/libs/call/minios.c
+++ b/tools/libs/call/minios.c
@@ -50,8 +50,8 @@ int osdep_hypercall(xencall_handle *xcall, privcmd_hypercall_t *hypercall)
ret = HYPERVISOR_multicall(&call, 1);
if (ret < 0) {
- errno = -ret;
- return -1;
+ errno = -ret;
+ return -1;
}
if ((long) call.result < 0) {
errno = - (long) call.result;
diff --git a/tools/libs/evtchn/minios.c b/tools/libs/evtchn/minios.c
index b839cd0..b4b5f14 100644
--- a/tools/libs/evtchn/minios.c
+++ b/tools/libs/evtchn/minios.c
@@ -103,8 +103,8 @@ int xenevtchn_notify(xenevtchn_handle *xce, evtchn_port_t port)
ret = notify_remote_via_evtchn(port);
if (ret < 0) {
- errno = -ret;
- ret = -1;
+ errno = -ret;
+ ret = -1;
}
return ret;
}
@@ -138,16 +138,16 @@ evtchn_port_or_error_t xenevtchn_bind_unbound_port(xenevtchn_handle *xce, int do
assert(get_current() == main_thread);
port_info = port_alloc(fd);
if (port_info == NULL)
- return -1;
+ return -1;
printf("xenevtchn_bind_unbound_port(%d)", domid);
ret = evtchn_alloc_unbound(domid, evtchn_handler, (void*)(intptr_t)fd, &port);
printf(" = %d\n", ret);
if (ret < 0) {
- port_dealloc(port_info);
- errno = -ret;
- return -1;
+ port_dealloc(port_info);
+ errno = -ret;
+ return -1;
}
port_info->bound = 1;
port_info->port = port;
@@ -166,16 +166,16 @@ evtchn_port_or_error_t xenevtchn_bind_interdomain(xenevtchn_handle *xce, int dom
assert(get_current() == main_thread);
port_info = port_alloc(fd);
if (port_info == NULL)
- return -1;
+ return -1;
printf("xenevtchn_bind_interdomain(%d, %"PRId32")", domid, remote_port);
ret = evtchn_bind_interdomain(domid, remote_port, evtchn_handler, (void*)(intptr_t)fd, &local_port);
printf(" = %d\n", ret);
if (ret < 0) {
- port_dealloc(port_info);
- errno = -ret;
- return -1;
+ port_dealloc(port_info);
+ errno = -ret;
+ return -1;
}
port_info->bound = 1;
port_info->port = local_port;
@@ -208,15 +208,15 @@ evtchn_port_or_error_t xenevtchn_bind_virq(xenevtchn_handle *xce, unsigned int v
assert(get_current() == main_thread);
port_info = port_alloc(fd);
if (port_info == NULL)
- return -1;
+ return -1;
printf("xenevtchn_bind_virq(%d)", virq);
port = bind_virq(virq, evtchn_handler, (void*)(intptr_t)fd);
if (port < 0) {
- port_dealloc(port_info);
- errno = -port;
- return -1;
+ port_dealloc(port_info);
+ errno = -port;
+ return -1;
}
port_info->bound = 1;
port_info->port = port;
diff --git a/tools/libs/evtchn/netbsd.c b/tools/libs/evtchn/netbsd.c
index c4123fe..13b77d6 100644
--- a/tools/libs/evtchn/netbsd.c
+++ b/tools/libs/evtchn/netbsd.c
@@ -72,9 +72,9 @@ evtchn_port_or_error_t xenevtchn_bind_unbound_port(xenevtchn_handle * xce, int d
ret = ioctl(fd, IOCTL_EVTCHN_BIND_UNBOUND_PORT, &bind);
if (ret == 0)
- return bind.port;
+ return bind.port;
else
- return -1;
+ return -1;
}
evtchn_port_or_error_t xenevtchn_bind_interdomain(xenevtchn_handle *xce, int domid,
@@ -89,9 +89,9 @@ evtchn_port_or_error_t xenevtchn_bind_interdomain(xenevtchn_handle *xce, int dom
ret = ioctl(fd, IOCTL_EVTCHN_BIND_INTERDOMAIN, &bind);
if (ret == 0)
- return bind.port;
+ return bind.port;
else
- return -1;
+ return -1;
}
int xenevtchn_unbind(xenevtchn_handle *xce, evtchn_port_t port)
@@ -114,9 +114,9 @@ evtchn_port_or_error_t xenevtchn_bind_virq(xenevtchn_handle *xce, unsigned int v
err = ioctl(fd, IOCTL_EVTCHN_BIND_VIRQ, &bind);
if (err)
- return -1;
+ return -1;
else
- return bind.port;
+ return bind.port;
}
evtchn_port_or_error_t xenevtchn_pending(xenevtchn_handle *xce)
diff --git a/tools/libs/foreignmemory/freebsd.c b/tools/libs/foreignmemory/freebsd.c
index 0d71713..8793ed9 100644
--- a/tools/libs/foreignmemory/freebsd.c
+++ b/tools/libs/foreignmemory/freebsd.c
@@ -121,7 +121,7 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
void *addr, size_t num)
{
- return munmap(addr, num << PAGE_SHIFT);
+ return munmap(addr, num << PAGE_SHIFT);
}
/*
diff --git a/tools/libs/foreignmemory/minios.c b/tools/libs/foreignmemory/minios.c
index 3d3c711..015ccda 100644
--- a/tools/libs/foreignmemory/minios.c
+++ b/tools/libs/foreignmemory/minios.c
@@ -45,16 +45,16 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
{
unsigned long pt_prot = 0;
if (prot & PROT_READ)
- pt_prot = L1_PROT_RO;
+ pt_prot = L1_PROT_RO;
if (prot & PROT_WRITE)
- pt_prot = L1_PROT;
+ pt_prot = L1_PROT;
return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
}
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
void *addr, size_t num)
{
- return munmap(addr, num << PAGE_SHIFT);
+ return munmap(addr, num << PAGE_SHIFT);
}
/*
diff --git a/tools/libs/foreignmemory/netbsd.c b/tools/libs/foreignmemory/netbsd.c
index 747d5b4..a89308f 100644
--- a/tools/libs/foreignmemory/netbsd.c
+++ b/tools/libs/foreignmemory/netbsd.c
@@ -97,5 +97,15 @@ void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
void *addr, size_t num)
{
- return munmap(addr, num*XC_PAGE_SIZE);
+ return munmap(addr, num*XC_PAGE_SIZE);
}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/foreignmemory/private.h b/tools/libs/foreignmemory/private.h
index 54dd7d5..30ea892 100644
--- a/tools/libs/foreignmemory/private.h
+++ b/tools/libs/foreignmemory/private.h
@@ -32,7 +32,7 @@ void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
#endif
#define PERROR(_f...) \
- xtl_log(fmem->logger, XTL_ERROR, errno, "xenforeignmemory", _f)
+ xtl_log(fmem->logger, XTL_ERROR, errno, "xenforeignmemory", _f)
#endif
diff --git a/tools/libs/foreignmemory/solaris.c b/tools/libs/foreignmemory/solaris.c
index 359e664..ddd313a 100644
--- a/tools/libs/foreignmemory/solaris.c
+++ b/tools/libs/foreignmemory/solaris.c
@@ -90,11 +90,20 @@ void *compat_mapforeign_batch(xenforeignmem_handle *fmem, uint32_t dom,
return NULL;
}
return addr;
-
}
int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
void *addr, size_t num)
{
- return munmap(addr, num*XC_PAGE_SIZE);
+ return munmap(addr, num*XC_PAGE_SIZE);
}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/tools/libs/gnttab/include/xengnttab.h b/tools/libs/gnttab/include/xengnttab.h
index 007e6db..c810c07 100644
--- a/tools/libs/gnttab/include/xengnttab.h
+++ b/tools/libs/gnttab/include/xengnttab.h
@@ -144,7 +144,7 @@ int xengnttab_munmap(xengnttab_handle *xgt,
* of grants.
*/
int xengnttab_set_max_grants(xengnttab_handle *xgt,
- uint32_t count);
+ uint32_t count);
/*
* Grant Sharing Interface (allocating and granting pages)
@@ -162,7 +162,7 @@ typedef struct xengntdev_handle xengntshr_handle;
*
*/
xengntshr_handle *xengntshr_open(xentoollog_logger *logger,
- unsigned open_flags);
+ unsigned open_flags);
/*
* Close a handle previously allocated with xengntshr_open().
diff --git a/tools/libs/gnttab/linux.c b/tools/libs/gnttab/linux.c
index 635765b..a76f17f 100644
--- a/tools/libs/gnttab/linux.c
+++ b/tools/libs/gnttab/linux.c
@@ -294,9 +294,9 @@ void *osdep_gntshr_share_pages(xengntshr_handle *xgs,
err = ioctl(fd, IOCTL_GNTALLOC_SET_UNMAP_NOTIFY, ¬ify);
if (err) {
GSERROR(xgs->logger, "ioctl SET_UNMAP_NOTIFY failed");
- munmap(area, count * PAGE_SIZE);
- area = NULL;
- }
+ munmap(area, count * PAGE_SIZE);
+ area = NULL;
+ }
memcpy(refs, gref_info->gref_ids, count * sizeof(uint32_t));
diff --git a/tools/libs/toollog/xtl_logger_stdio.c b/tools/libs/toollog/xtl_logger_stdio.c
index f9c5bd8..52dfbf5 100644
--- a/tools/libs/toollog/xtl_logger_stdio.c
+++ b/tools/libs/toollog/xtl_logger_stdio.c
@@ -135,7 +135,7 @@ static void stdiostream_progress(struct xentoollog_logger *logger_in,
newpel = fprintf(lg->f, "%s%s" "%s: %lu/%lu %3d%%%s",
context?context:"", context?": ":"",
doing_what, done, total, percent,
- done == total ? "\n" : "");
+ done == total ? "\n" : "");
extra_erase = lg->progress_erase_len - newpel;
if (extra_erase > 0)
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 18/23] tools/libs/gnttab: Review and update doc comments.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (16 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 17/23] tools/libs: Clean up hard tabs Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 15:17 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 19/23] tools/libs/call: Update some log messages to not refer to xc Ian Campbell
` (5 subsequent siblings)
23 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
Remove some stray xc references.
While I'm not convinced by javadoc/doxygen cause the existing comments
which appear to use that syntax to have the appropriate /** marker.
Also fix a typo in a code comment.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libs/gnttab/include/xengnttab.h | 21 +++++++++++----------
tools/libs/gnttab/linux.c | 2 +-
2 files changed, 12 insertions(+), 11 deletions(-)
diff --git a/tools/libs/gnttab/include/xengnttab.h b/tools/libs/gnttab/include/xengnttab.h
index c810c07..e091fa6 100644
--- a/tools/libs/gnttab/include/xengnttab.h
+++ b/tools/libs/gnttab/include/xengnttab.h
@@ -38,9 +38,9 @@ typedef struct xengntdev_handle xengnttab_handle;
/*
* Note:
- * After fork a child process must not use any opened xc gnttab
+ * After fork a child process must not use any opened gnttab
* handle inherited from their parent. They must open a new handle if
- * they want to interact with xc.
+ * they want to interact with gnttab.
*
* Return an fd onto the grant table driver. Logs errors.
*/
@@ -52,7 +52,7 @@ xengnttab_handle *xengnttab_open(xentoollog_logger *logger, unsigned open_flags)
*/
int xengnttab_close(xengnttab_handle *xgt);
-/*
+/**
* Memory maps a grant reference from one domain to a local address range.
* Mappings should be unmapped with xengnttab_munmap. Logs errors.
*
@@ -124,7 +124,7 @@ void *xengnttab_map_grant_ref_notify(xengnttab_handle *xgt,
uint32_t notify_offset,
evtchn_port_t notify_port);
-/*
+/**
* Unmaps the @count pages starting at @start_address, which were mapped by a
* call to xengnttab_map_grant_ref or xengnttab_map_grant_refs. Never logs.
*/
@@ -132,7 +132,7 @@ int xengnttab_munmap(xengnttab_handle *xgt,
void *start_address,
uint32_t count);
-/*
+/**
* Sets the maximum number of grants that may be mapped by the given instance
* to @count. Never logs.
*
@@ -156,9 +156,9 @@ typedef struct xengntdev_handle xengntshr_handle;
* Return an fd onto the grant sharing driver. Logs errors.
*
* Note:
- * After fork a child process must not use any opened xc gntshr
+ * After fork a child process must not use any opened gntshr
* handle inherited from their parent. They must open a new handle if
- * they want to interact with xc.
+ * they want to interact with gntshr.
*
*/
xengntshr_handle *xengntshr_open(xentoollog_logger *logger,
@@ -170,7 +170,7 @@ xengntshr_handle *xengntshr_open(xentoollog_logger *logger,
*/
int xengntshr_close(xengntshr_handle *xgs);
-/*
+/**
* Creates and shares pages with another domain.
*
* @parm xgs a handle to an open grant sharing instance
@@ -183,7 +183,7 @@ int xengntshr_close(xengntshr_handle *xgs);
void *xengntshr_share_pages(xengntshr_handle *xgs, uint32_t domid,
int count, uint32_t *refs, int writable);
-/*
+/**
* Creates and shares a page with another domain, with unmap notification.
*
* @parm xgs a handle to an open grant sharing instance
@@ -199,7 +199,8 @@ void *xengntshr_share_page_notify(xengntshr_handle *xgs, uint32_t domid,
uint32_t *ref, int writable,
uint32_t notify_offset,
evtchn_port_t notify_port);
-/*
+
+/**
* Unmaps the @count pages starting at @start_address, which were mapped by a
* call to xengntshr_share_*. Never logs.
*/
diff --git a/tools/libs/gnttab/linux.c b/tools/libs/gnttab/linux.c
index a76f17f..bdf2a42 100644
--- a/tools/libs/gnttab/linux.c
+++ b/tools/libs/gnttab/linux.c
@@ -136,7 +136,7 @@ void *osdep_gnttab_grant_map(xengnttab_handle *xgt,
* swapped out. Since the paging daemon may be in the same domain, the
* hypercall cannot block without causing a deadlock.
*
- * Because there are no notificaitons when the page is swapped in, wait
+ * Because there are no notifications when the page is swapped in, wait
* a bit before retrying, and hope that the page will arrive eventually.
*/
usleep(1000);
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 19/23] tools/libs/call: Update some log messages to not refer to xc.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (17 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 18/23] tools/libs/gnttab: Review and update doc comments Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-13 16:20 ` Andrew Cooper
2015-11-09 12:00 ` [PATCH XEN v5 20/23] tools/libs/call: Avoid xc_memalign in netbsd and solaris backends Ian Campbell
` (4 subsequent siblings)
23 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libs/call/linux.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/libs/call/linux.c b/tools/libs/call/linux.c
index 906ca7e..80b505c 100644
--- a/tools/libs/call/linux.c
+++ b/tools/libs/call/linux.c
@@ -88,7 +88,7 @@ void *osdep_alloc_pages(xencall_handle *xcall, unsigned int npages)
p = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_LOCKED, -1, 0);
if ( p == MAP_FAILED )
{
- PERROR("xc_alloc_hypercall_buffer: mmap failed");
+ PERROR("alloc_pages: mmap failed");
return NULL;
}
@@ -97,7 +97,7 @@ void *osdep_alloc_pages(xencall_handle *xcall, unsigned int npages)
rc = madvise(p, npages * PAGE_SIZE, MADV_DONTFORK);
if ( rc < 0 )
{
- PERROR("xc_alloc_hypercall_buffer: madvise failed");
+ PERROR("alloc_pages: madvise failed");
goto out;
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 20/23] tools/libs/call: Avoid xc_memalign in netbsd and solaris backends
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (18 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 19/23] tools/libs/call: Update some log messages to not refer to xc Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 21/23] tools/libs/foreignmemory: Mention restrictions on fork in docs Ian Campbell
` (3 subsequent siblings)
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
These are already arch specific, so just use the appropriate
interfaces (as determined by looking at the xc_memalign backend).
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libs/call/netbsd.c | 4 ++--
tools/libs/call/solaris.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/tools/libs/call/netbsd.c b/tools/libs/call/netbsd.c
index 12008fb..c01d3de 100644
--- a/tools/libs/call/netbsd.c
+++ b/tools/libs/call/netbsd.c
@@ -74,8 +74,8 @@ void *osdep_alloc_hypercall_buffer(xencall_handle *xcall, unsigned int npages)
size_t size = npages * XC_PAGE_SIZE;
void *p;
- p = xc_memalign(xcall, XC_PAGE_SIZE, size);
- if (!p)
+ ret = posix_memalign(&p, XC_PAGE_SIZE, size);
+ if ( ret != 0 || !p )
return NULL;
if ( mlock(p, size) < 0 )
diff --git a/tools/libs/call/solaris.c b/tools/libs/call/solaris.c
index 972989d..8d9c9da 100644
--- a/tools/libs/call/solaris.c
+++ b/tools/libs/call/solaris.c
@@ -71,7 +71,7 @@ int osdep_xencall_close(xencall_handle *xcall)
void *osdep_alloc_hypercall_buffer(xencall_handle *xcall, unsigned int npages)
{
- return xc_memalign(xcall, XC_PAGE_SIZE, npages * XC_PAGE_SIZE);
+ return memalign(XC_PAGE_SIZE, npages * XC_PAGE_SIZE);
}
void osdep_free_hypercall_buffer(xencall_handle *xcall, void *ptr,
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 21/23] tools/libs/foreignmemory: Mention restrictions on fork in docs.
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (19 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 20/23] tools/libs/call: Avoid xc_memalign in netbsd and solaris backends Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 15:18 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 22/23] tools: Update CFLAGS for qemu-xen to allow it to use new libraries Ian Campbell
` (2 subsequent siblings)
23 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/libs/foreignmemory/include/xenforeignmemory.h | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/tools/libs/foreignmemory/include/xenforeignmemory.h b/tools/libs/foreignmemory/include/xenforeignmemory.h
index 99ec883..3f52417 100644
--- a/tools/libs/foreignmemory/include/xenforeignmemory.h
+++ b/tools/libs/foreignmemory/include/xenforeignmemory.h
@@ -32,7 +32,12 @@ typedef struct xentoollog_logger xentoollog_logger;
typedef struct xenforeignmemory_handle xenforeignmemory_handle;
/*
- * Return a handle onto the hypercall driver. Logs errors.
+ * Return a handle onto the foreign memory mapping driver. Logs errors.
+ *
+ * Note: After fork a child process must not use any opened handle
+ * inherited from their parent. Any existing mappings should be
+ * assumed to become invalid. Therefore children must open a new
+ * handle and make new mappings if necessary.
*/
xenforeignmemory_handle *xenforeignmemory_open(xentoollog_logger *logger,
unsigned open_flags);
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 22/23] tools: Update CFLAGS for qemu-xen to allow it to use new libraries
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (20 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 21/23] tools/libs/foreignmemory: Mention restrictions on fork in docs Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 23/23] HACK: Update Config.mk to pull all the right bits from my xenbits trees Ian Campbell
2015-11-11 15:20 ` [PATCH XEN v5 00/23] Begin to disentangle libxenctrl and provide some stable libraries Ian Jackson
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
This means adding -L for libxen{evtchn,gnttab,foreignmemory} so that
it can link them directly (rather than using the libxenctrl compat
layer exposed via -rpath-link). Also add -I for libxenforeignmemory.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
tools/Makefile | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/tools/Makefile b/tools/Makefile
index c03f6db..3575f16 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -255,12 +255,16 @@ subdir-all-qemu-xen-dir: qemu-xen-dir-find
-I$(XEN_ROOT)/tools/libs/toollog/include \
-I$(XEN_ROOT)/tools/libs/evtchn/include \
-I$(XEN_ROOT)/tools/libs/gnttab/include \
+ -I$(XEN_ROOT)/tools/libs/foreignmemory/include \
-I$(XEN_ROOT)/tools/libxc/include \
-I$(XEN_ROOT)/tools/xenstore/include \
-I$(XEN_ROOT)/tools/xenstore/compat/include \
$(EXTRA_CFLAGS_QEMU_XEN)" \
--extra-ldflags="-L$(XEN_ROOT)/tools/libxc \
-L$(XEN_ROOT)/tools/xenstore \
+ -L$(XEN_ROOT)/tools/libs/evtchn \
+ -L$(XEN_ROOT)/tools/libs/gnttab \
+ -L$(XEN_ROOT)/tools/libs/foreignmemory \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/evtchn \
-Wl,-rpath-link=$(XEN_ROOT)/tools/libs/gnttab \
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH XEN v5 23/23] HACK: Update Config.mk to pull all the right bits from my xenbits trees
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (21 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 22/23] tools: Update CFLAGS for qemu-xen to allow it to use new libraries Ian Campbell
@ 2015-11-09 12:00 ` Ian Campbell
2015-11-11 15:20 ` [PATCH XEN v5 00/23] Begin to disentangle libxenctrl and provide some stable libraries Ian Jackson
23 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:00 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
v4: Config.mk instead of .config
---
Config.mk | 20 ++++++++++----------
1 file changed, 10 insertions(+), 10 deletions(-)
diff --git a/Config.mk b/Config.mk
index 5db7ca5..7258065 100644
--- a/Config.mk
+++ b/Config.mk
@@ -242,20 +242,20 @@ endif
ifeq ($(GIT_HTTP),y)
OVMF_UPSTREAM_URL ?= http://xenbits.xen.org/git-http/ovmf.git
-QEMU_UPSTREAM_URL ?= http://xenbits.xen.org/git-http/qemu-xen.git
-QEMU_TRADITIONAL_URL ?= http://xenbits.xen.org/git-http/qemu-xen-traditional.git
+QEMU_UPSTREAM_URL ?= http://xenbits.xen.org/git-http/people/ianc/libxenctrl-split/qemu-xen.git
+QEMU_TRADITIONAL_URL ?= http://xenbits.xen.org/git-http/people/ianc/libxenctrl-split/qemu-xen-traditional.git
SEABIOS_UPSTREAM_URL ?= http://xenbits.xen.org/git-http/seabios.git
-MINIOS_UPSTREAM_URL ?= http://xenbits.xen.org/git-http/mini-os.git
+MINIOS_UPSTREAM_URL ?= http://xenbits.xen.org/git-http/people/ianc/libxenctrl-split/mini-os.git
else
OVMF_UPSTREAM_URL ?= git://xenbits.xen.org/ovmf.git
-QEMU_UPSTREAM_URL ?= git://xenbits.xen.org/qemu-xen.git
-QEMU_TRADITIONAL_URL ?= git://xenbits.xen.org/qemu-xen-traditional.git
+QEMU_UPSTREAM_URL ?= git://xenbits.xen.org/people/ianc/libxenctrl-split/qemu-xen.git
+QEMU_TRADITIONAL_URL ?= git://xenbits.xen.org/people/ianc/libxenctrl-split/qemu-xen-traditional.git
SEABIOS_UPSTREAM_URL ?= git://xenbits.xen.org/seabios.git
-MINIOS_UPSTREAM_URL ?= git://xenbits.xen.org/mini-os.git
+MINIOS_UPSTREAM_URL ?= git://xenbits.xen.org/people/ianc/libxenctrl-split/mini-os.git
endif
-OVMF_UPSTREAM_REVISION ?= af9785a9ed61daea52b47f0bf448f1f228beee1e
-QEMU_UPSTREAM_REVISION ?= master
-MINIOS_UPSTREAM_REVISION ?= 256035e01a1aa5739e34f245f3b1e9e8ee204210
+OVMF_UPSTREAM_REVISION ?= cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd
+QEMU_UPSTREAM_REVISION ?= origin/v5
+MINIOS_UPSTREAM_REVISION ?= origin/v5
# Thu Jul 23 11:08:38 2015 +0100
# arm: interrupt controller
@@ -266,7 +266,7 @@ SEABIOS_UPSTREAM_REVISION ?= rel-1.8.2
ETHERBOOT_NICS ?= rtl8139 8086100e
-QEMU_TRADITIONAL_REVISION ?= bc00cad75d8bcc3ba696992bec219c21db8406aa
+QEMU_TRADITIONAL_REVISION ?= origin/v5
# Tue Mar 11 10:19:23 2014 +0000
# block-vvfat: fix resource leaks in read_directory()
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 0/9] Begin to disentangle libxenctrl and provide some stable libraries
2015-11-09 11:59 [Qemu-devel] [Minios-devel] [PATCH v4 0/<VARIOUS>] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` Ian Campbell
` (2 subsequent siblings)
3 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
We intend to stabilise some parts of the libxenctrl interface by
splitting out some functionality into separate stable libraries.
This is the qemu-xen part of the first phase of that change.
This mail is (or is intended to be) a reply to a "0/<VARIOUS>"
super-intro mail covering all of the related patch series and which
contains more details.
Ian Campbell (9):
xen_console: correctly cleanup primary console on teardown.
xen: Switch to libxenevtchn interface for compat shims.
xen: Switch to libxengnttab interface for compat shims.
xen: Switch uses of xc_map_foreign_range into xc_map_foreign_bulk
xen: Switch uses of xc_map_foreign_pages into xc_map_foreign_bulk
xen: Switch uses of xc_map_foreign_bulk to use libxenforeignmemory
API.
xen: Use stable library interfaces when they are available.
xen: domainbuild: reopen libxenctrl interface after forking for domain
watcher.
xen: make it possible to build without the Xen PV domain builder
configure | 70 +++++++++++++++++++++++
hw/block/xen_disk.c | 38 +++++++------
hw/char/xen_console.c | 20 +++----
hw/display/xenfb.c | 22 ++++---
hw/net/xen_nic.c | 18 +++---
hw/xen/xen_backend.c | 44 +++++++-------
hw/xenpv/Makefile.objs | 4 +-
hw/xenpv/xen_domainbuild.c | 9 ++-
hw/xenpv/xen_machine_pv.c | 15 +++--
include/hw/xen/xen_backend.h | 5 +-
include/hw/xen/xen_common.h | 133 +++++++++++++++++++++++++++++++++----------
xen-common.c | 6 ++
xen-hvm.c | 53 +++++++++--------
xen-mapcache.c | 6 +-
14 files changed, 307 insertions(+), 136 deletions(-)
--
2.1.4
^ permalink raw reply [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 0/9] Begin to disentangle libxenctrl and provide some stable libraries
@ 2015-11-09 12:01 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
We intend to stabilise some parts of the libxenctrl interface by
splitting out some functionality into separate stable libraries.
This is the qemu-xen part of the first phase of that change.
This mail is (or is intended to be) a reply to a "0/<VARIOUS>"
super-intro mail covering all of the related patch series and which
contains more details.
Ian Campbell (9):
xen_console: correctly cleanup primary console on teardown.
xen: Switch to libxenevtchn interface for compat shims.
xen: Switch to libxengnttab interface for compat shims.
xen: Switch uses of xc_map_foreign_range into xc_map_foreign_bulk
xen: Switch uses of xc_map_foreign_pages into xc_map_foreign_bulk
xen: Switch uses of xc_map_foreign_bulk to use libxenforeignmemory
API.
xen: Use stable library interfaces when they are available.
xen: domainbuild: reopen libxenctrl interface after forking for domain
watcher.
xen: make it possible to build without the Xen PV domain builder
configure | 70 +++++++++++++++++++++++
hw/block/xen_disk.c | 38 +++++++------
hw/char/xen_console.c | 20 +++----
hw/display/xenfb.c | 22 ++++---
hw/net/xen_nic.c | 18 +++---
hw/xen/xen_backend.c | 44 +++++++-------
hw/xenpv/Makefile.objs | 4 +-
hw/xenpv/xen_domainbuild.c | 9 ++-
hw/xenpv/xen_machine_pv.c | 15 +++--
include/hw/xen/xen_backend.h | 5 +-
include/hw/xen/xen_common.h | 133 +++++++++++++++++++++++++++++++++----------
xen-common.c | 6 ++
xen-hvm.c | 53 +++++++++--------
xen-mapcache.c | 6 +-
14 files changed, 307 insertions(+), 136 deletions(-)
--
2.1.4
^ permalink raw reply [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 1/9] xen_console: correctly cleanup primary console on teardown.
2015-11-09 12:01 ` Ian Campbell
(?)
(?)
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
All of the work in con_disconnect applies to the primary console case
(when xendev->dev is NULL). Therefore remove the early check and bail
and allow it to fall through. All of the existing code is correctly
conditional already.
The ->dev and ->gnttabdev handles are either both set or neither. For
consistency with con_initialise() with to the former here too.
With this con_initialise and con_disconnect now mirror each other.
Fix up a hard tab in the function while editing.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
v4: New patch based on feedback to "xen: Switch uses of
xc_map_foreign_bulk to use libxenforeignmemory API."
---
hw/char/xen_console.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index eb7f450..63ade33 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -265,9 +265,6 @@ static void con_disconnect(struct XenDevice *xendev)
{
struct XenConsole *con = container_of(xendev, struct XenConsole, xendev);
- if (!xendev->dev) {
- return;
- }
if (con->chr) {
qemu_chr_add_handlers(con->chr, NULL, NULL, NULL, NULL);
qemu_chr_fe_release(con->chr);
@@ -275,12 +272,12 @@ static void con_disconnect(struct XenDevice *xendev)
xen_be_unbind_evtchn(&con->xendev);
if (con->sring) {
- if (!xendev->gnttabdev) {
+ if (!xendev->dev) {
munmap(con->sring, XC_PAGE_SIZE);
} else {
xc_gnttab_munmap(xendev->gnttabdev, con->sring, 1);
}
- con->sring = NULL;
+ con->sring = NULL;
}
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 1/9] xen_console: correctly cleanup primary console on teardown.
2015-11-09 12:01 ` Ian Campbell
(?)
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
All of the work in con_disconnect applies to the primary console case
(when xendev->dev is NULL). Therefore remove the early check and bail
and allow it to fall through. All of the existing code is correctly
conditional already.
The ->dev and ->gnttabdev handles are either both set or neither. For
consistency with con_initialise() with to the former here too.
With this con_initialise and con_disconnect now mirror each other.
Fix up a hard tab in the function while editing.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
v4: New patch based on feedback to "xen: Switch uses of
xc_map_foreign_bulk to use libxenforeignmemory API."
---
hw/char/xen_console.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index eb7f450..63ade33 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -265,9 +265,6 @@ static void con_disconnect(struct XenDevice *xendev)
{
struct XenConsole *con = container_of(xendev, struct XenConsole, xendev);
- if (!xendev->dev) {
- return;
- }
if (con->chr) {
qemu_chr_add_handlers(con->chr, NULL, NULL, NULL, NULL);
qemu_chr_fe_release(con->chr);
@@ -275,12 +272,12 @@ static void con_disconnect(struct XenDevice *xendev)
xen_be_unbind_evtchn(&con->xendev);
if (con->sring) {
- if (!xendev->gnttabdev) {
+ if (!xendev->dev) {
munmap(con->sring, XC_PAGE_SIZE);
} else {
xc_gnttab_munmap(xendev->gnttabdev, con->sring, 1);
}
- con->sring = NULL;
+ con->sring = NULL;
}
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 2/9] xen: Switch to libxenevtchn interface for compat shims.
2015-11-09 12:01 ` Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxenevtchn which provides access to event
channels.
In preparation for this switch the compatibility layer in xen_common.h
(which support building with older versions of Xen) to use what will
be the new library API. This means that the evtchn shim will disappear
for versions of Xen which include libxenevtchn.
To simplify things for the <= 4.0.0 support we wrap the int fd in a
malloc(sizeof int) such that the handle is always a pointer. This
leads to less typedef headaches and the need for
XC_HANDLER_INITIAL_VALUE etc for these interfaces.
Note that this patch does not add any support for actually using
libxenevtchn, it just adjusts the existing shims.
Note that xc_evtchn_alloc_unbound functionality remains in libxenctrl,
since that functionality is not exposed by /dev/xen/evtchn.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
v4: Ran checkpatch, fixed all errors
Allocate correct size for handle (i.e. not size of the ptr)
---
hw/xen/xen_backend.c | 31 ++++++++++++++++---------------
include/hw/xen/xen_backend.h | 2 +-
include/hw/xen/xen_common.h | 44 ++++++++++++++++++++++++++++++++++----------
xen-hvm.c | 25 +++++++++++++------------
4 files changed, 64 insertions(+), 38 deletions(-)
diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
index 2510e2e..ae2a1f0 100644
--- a/hw/xen/xen_backend.c
+++ b/hw/xen/xen_backend.c
@@ -243,19 +243,19 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev,
xendev->debug = debug;
xendev->local_port = -1;
- xendev->evtchndev = xen_xc_evtchn_open(NULL, 0);
- if (xendev->evtchndev == XC_HANDLER_INITIAL_VALUE) {
+ xendev->evtchndev = xenevtchn_open(NULL, 0);
+ if (xendev->evtchndev == NULL) {
xen_be_printf(NULL, 0, "can't open evtchn device\n");
g_free(xendev);
return NULL;
}
- fcntl(xc_evtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
+ fcntl(xenevtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
if (ops->flags & DEVOPS_FLAG_NEED_GNTDEV) {
xendev->gnttabdev = xen_xc_gnttab_open(NULL, 0);
if (xendev->gnttabdev == XC_HANDLER_INITIAL_VALUE) {
xen_be_printf(NULL, 0, "can't open gnttab device\n");
- xc_evtchn_close(xendev->evtchndev);
+ xenevtchn_close(xendev->evtchndev);
g_free(xendev);
return NULL;
}
@@ -306,8 +306,8 @@ static struct XenDevice *xen_be_del_xendev(int dom, int dev)
g_free(xendev->fe);
}
- if (xendev->evtchndev != XC_HANDLER_INITIAL_VALUE) {
- xc_evtchn_close(xendev->evtchndev);
+ if (xendev->evtchndev != NULL) {
+ xenevtchn_close(xendev->evtchndev);
}
if (xendev->gnttabdev != XC_HANDLER_INITIAL_VALUE) {
xc_gnttab_close(xendev->gnttabdev);
@@ -691,13 +691,14 @@ static void xen_be_evtchn_event(void *opaque)
struct XenDevice *xendev = opaque;
evtchn_port_t port;
- port = xc_evtchn_pending(xendev->evtchndev);
+ port = xenevtchn_pending(xendev->evtchndev);
if (port != xendev->local_port) {
- xen_be_printf(xendev, 0, "xc_evtchn_pending returned %d (expected %d)\n",
+ xen_be_printf(xendev, 0,
+ "xenevtchn_pending returned %d (expected %d)\n",
port, xendev->local_port);
return;
}
- xc_evtchn_unmask(xendev->evtchndev, port);
+ xenevtchn_unmask(xendev->evtchndev, port);
if (xendev->ops->event) {
xendev->ops->event(xendev);
@@ -740,14 +741,14 @@ int xen_be_bind_evtchn(struct XenDevice *xendev)
if (xendev->local_port != -1) {
return 0;
}
- xendev->local_port = xc_evtchn_bind_interdomain
+ xendev->local_port = xenevtchn_bind_interdomain
(xendev->evtchndev, xendev->dom, xendev->remote_port);
if (xendev->local_port == -1) {
- xen_be_printf(xendev, 0, "xc_evtchn_bind_interdomain failed\n");
+ xen_be_printf(xendev, 0, "xenevtchn_bind_interdomain failed\n");
return -1;
}
xen_be_printf(xendev, 2, "bind evtchn port %d\n", xendev->local_port);
- qemu_set_fd_handler(xc_evtchn_fd(xendev->evtchndev),
+ qemu_set_fd_handler(xenevtchn_fd(xendev->evtchndev),
xen_be_evtchn_event, NULL, xendev);
return 0;
}
@@ -757,15 +758,15 @@ void xen_be_unbind_evtchn(struct XenDevice *xendev)
if (xendev->local_port == -1) {
return;
}
- qemu_set_fd_handler(xc_evtchn_fd(xendev->evtchndev), NULL, NULL, NULL);
- xc_evtchn_unbind(xendev->evtchndev, xendev->local_port);
+ qemu_set_fd_handler(xenevtchn_fd(xendev->evtchndev), NULL, NULL, NULL);
+ xenevtchn_unbind(xendev->evtchndev, xendev->local_port);
xen_be_printf(xendev, 2, "unbind evtchn port %d\n", xendev->local_port);
xendev->local_port = -1;
}
int xen_be_send_notify(struct XenDevice *xendev)
{
- return xc_evtchn_notify(xendev->evtchndev, xendev->local_port);
+ return xenevtchn_notify(xendev->evtchndev, xendev->local_port);
}
/*
diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
index 3b4125e..a90314f 100644
--- a/include/hw/xen/xen_backend.h
+++ b/include/hw/xen/xen_backend.h
@@ -46,7 +46,7 @@ struct XenDevice {
int remote_port;
int local_port;
- XenEvtchn evtchndev;
+ xenevtchn_handle *evtchndev;
XenGnttab gnttabdev;
struct XenDevOps *ops;
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index d7fa6a4..6f7c003 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -39,17 +39,38 @@ static inline void *xc_map_foreign_bulk(int xc_handle, uint32_t dom, int prot,
#if CONFIG_XEN_CTRL_INTERFACE_VERSION < 410
typedef int XenXC;
-typedef int XenEvtchn;
+typedef int xenevtchn_handle;
typedef int XenGnttab;
# define XC_INTERFACE_FMT "%i"
# define XC_HANDLER_INITIAL_VALUE -1
-static inline XenEvtchn xen_xc_evtchn_open(void *logger,
- unsigned int open_flags)
+static inline xenevtchn_handle *xenevtchn_open(void *logger,
+ unsigned int open_flags)
{
- return xc_evtchn_open();
+ xenevtchn_handle *h = malloc(sizeof(*h));
+ if (!h) {
+ return NULL;
+ }
+ *h = xc_evtchn_open();
+ if (*h == -1) {
+ free(h);
+ h = NULL;
+ }
+ return h;
}
+static inline int xenevtchn_close(xenevtchn_handle *h)
+{
+ int rc = xc_evtchn_close(*h);
+ free(h);
+ return rc;
+}
+#define xenevtchn_fd(h) xc_evtchn_fd(*h)
+#define xenevtchn_pending(h) xc_evtchn_pending(*h)
+#define xenevtchn_notify(h, p) xc_evtchn_notify(*h, p)
+#define xenevtchn_bind_interdomain(h, d, p) xc_evtchn_bind_interdomain(*h, d, p)
+#define xenevtchn_unmask(h, p) xc_evtchn_unmask(*h, p)
+#define xenevtchn_unbind(h, p) xc_evtchn_unmask(*h, p)
static inline XenGnttab xen_xc_gnttab_open(void *logger,
unsigned int open_flags)
@@ -108,17 +129,20 @@ static inline void xs_close(struct xs_handle *xsh)
#else
typedef xc_interface *XenXC;
-typedef xc_evtchn *XenEvtchn;
+typedef xc_evtchn xenevtchn_handle;
typedef xc_gnttab *XenGnttab;
# define XC_INTERFACE_FMT "%p"
# define XC_HANDLER_INITIAL_VALUE NULL
-static inline XenEvtchn xen_xc_evtchn_open(void *logger,
- unsigned int open_flags)
-{
- return xc_evtchn_open(logger, open_flags);
-}
+#define xenevtchn_open(l, f) xc_evtchn_open(l, f);
+#define xenevtchn_close(h) xc_evtchn_close(h)
+#define xenevtchn_fd(h) xc_evtchn_fd(h)
+#define xenevtchn_pending(h) xc_evtchn_pending(h)
+#define xenevtchn_notify(h, p) xc_evtchn_notify(h, p)
+#define xenevtchn_bind_interdomain(h, d, p) xc_evtchn_bind_interdomain(h, d, p)
+#define xenevtchn_unmask(h, p) xc_evtchn_unmask(h, p)
+#define xenevtchn_unbind(h, p) xc_evtchn_unbind(h, p)
static inline XenGnttab xen_xc_gnttab_open(void *logger,
unsigned int open_flags)
diff --git a/xen-hvm.c b/xen-hvm.c
index 3d78a0c..6680782 100644
--- a/xen-hvm.c
+++ b/xen-hvm.c
@@ -109,7 +109,7 @@ typedef struct XenIOState {
/* evtchn local port for buffered io */
evtchn_port_t bufioreq_local_port;
/* the evtchn fd for polling */
- XenEvtchn xce_handle;
+ xenevtchn_handle *xce_handle;
/* which vcpu we are serving */
int send_vcpu;
@@ -714,7 +714,7 @@ static ioreq_t *cpu_get_ioreq(XenIOState *state)
int i;
evtchn_port_t port;
- port = xc_evtchn_pending(state->xce_handle);
+ port = xenevtchn_pending(state->xce_handle);
if (port == state->bufioreq_local_port) {
timer_mod(state->buffered_io_timer,
BUFFER_IO_MAX_DELAY + qemu_clock_get_ms(QEMU_CLOCK_REALTIME));
@@ -733,7 +733,7 @@ static ioreq_t *cpu_get_ioreq(XenIOState *state)
}
/* unmask the wanted port again */
- xc_evtchn_unmask(state->xce_handle, port);
+ xenevtchn_unmask(state->xce_handle, port);
/* get the io packet from shared memory */
state->send_vcpu = i;
@@ -1040,7 +1040,7 @@ static void handle_buffered_io(void *opaque)
BUFFER_IO_MAX_DELAY + qemu_clock_get_ms(QEMU_CLOCK_REALTIME));
} else {
timer_del(state->buffered_io_timer);
- xc_evtchn_unmask(state->xce_handle, state->bufioreq_local_port);
+ xenevtchn_unmask(state->xce_handle, state->bufioreq_local_port);
}
}
@@ -1084,7 +1084,8 @@ static void cpu_handle_ioreq(void *opaque)
}
req->state = STATE_IORESP_READY;
- xc_evtchn_notify(state->xce_handle, state->ioreq_local_port[state->send_vcpu]);
+ xenevtchn_notify(state->xce_handle,
+ state->ioreq_local_port[state->send_vcpu]);
}
}
@@ -1092,8 +1093,8 @@ static void xen_main_loop_prepare(XenIOState *state)
{
int evtchn_fd = -1;
- if (state->xce_handle != XC_HANDLER_INITIAL_VALUE) {
- evtchn_fd = xc_evtchn_fd(state->xce_handle);
+ if (state->xce_handle != NULL) {
+ evtchn_fd = xenevtchn_fd(state->xce_handle);
}
state->buffered_io_timer = timer_new_ms(QEMU_CLOCK_REALTIME, handle_buffered_io,
@@ -1131,7 +1132,7 @@ static void xen_exit_notifier(Notifier *n, void *data)
{
XenIOState *state = container_of(n, XenIOState, exit);
- xc_evtchn_close(state->xce_handle);
+ xenevtchn_close(state->xce_handle);
xs_daemon_close(state->xenstore);
}
@@ -1200,8 +1201,8 @@ int xen_hvm_init(PCMachineState *pcms,
state = g_malloc0(sizeof (XenIOState));
- state->xce_handle = xen_xc_evtchn_open(NULL, 0);
- if (state->xce_handle == XC_HANDLER_INITIAL_VALUE) {
+ state->xce_handle = xenevtchn_open(NULL, 0);
+ if (state->xce_handle == NULL) {
perror("xen: event channel open");
return -1;
}
@@ -1281,7 +1282,7 @@ int xen_hvm_init(PCMachineState *pcms,
/* FIXME: how about if we overflow the page here? */
for (i = 0; i < max_cpus; i++) {
- rc = xc_evtchn_bind_interdomain(state->xce_handle, xen_domid,
+ rc = xenevtchn_bind_interdomain(state->xce_handle, xen_domid,
xen_vcpu_eport(state->shared_page, i));
if (rc == -1) {
fprintf(stderr, "shared evtchn %d bind error %d\n", i, errno);
@@ -1290,7 +1291,7 @@ int xen_hvm_init(PCMachineState *pcms,
state->ioreq_local_port[i] = rc;
}
- rc = xc_evtchn_bind_interdomain(state->xce_handle, xen_domid,
+ rc = xenevtchn_bind_interdomain(state->xce_handle, xen_domid,
bufioreq_evtchn);
if (rc == -1) {
fprintf(stderr, "buffered evtchn bind error %d\n", errno);
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 2/9] xen: Switch to libxenevtchn interface for compat shims.
@ 2015-11-09 12:01 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxenevtchn which provides access to event
channels.
In preparation for this switch the compatibility layer in xen_common.h
(which support building with older versions of Xen) to use what will
be the new library API. This means that the evtchn shim will disappear
for versions of Xen which include libxenevtchn.
To simplify things for the <= 4.0.0 support we wrap the int fd in a
malloc(sizeof int) such that the handle is always a pointer. This
leads to less typedef headaches and the need for
XC_HANDLER_INITIAL_VALUE etc for these interfaces.
Note that this patch does not add any support for actually using
libxenevtchn, it just adjusts the existing shims.
Note that xc_evtchn_alloc_unbound functionality remains in libxenctrl,
since that functionality is not exposed by /dev/xen/evtchn.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
v4: Ran checkpatch, fixed all errors
Allocate correct size for handle (i.e. not size of the ptr)
---
hw/xen/xen_backend.c | 31 ++++++++++++++++---------------
include/hw/xen/xen_backend.h | 2 +-
include/hw/xen/xen_common.h | 44 ++++++++++++++++++++++++++++++++++----------
xen-hvm.c | 25 +++++++++++++------------
4 files changed, 64 insertions(+), 38 deletions(-)
diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
index 2510e2e..ae2a1f0 100644
--- a/hw/xen/xen_backend.c
+++ b/hw/xen/xen_backend.c
@@ -243,19 +243,19 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev,
xendev->debug = debug;
xendev->local_port = -1;
- xendev->evtchndev = xen_xc_evtchn_open(NULL, 0);
- if (xendev->evtchndev == XC_HANDLER_INITIAL_VALUE) {
+ xendev->evtchndev = xenevtchn_open(NULL, 0);
+ if (xendev->evtchndev == NULL) {
xen_be_printf(NULL, 0, "can't open evtchn device\n");
g_free(xendev);
return NULL;
}
- fcntl(xc_evtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
+ fcntl(xenevtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
if (ops->flags & DEVOPS_FLAG_NEED_GNTDEV) {
xendev->gnttabdev = xen_xc_gnttab_open(NULL, 0);
if (xendev->gnttabdev == XC_HANDLER_INITIAL_VALUE) {
xen_be_printf(NULL, 0, "can't open gnttab device\n");
- xc_evtchn_close(xendev->evtchndev);
+ xenevtchn_close(xendev->evtchndev);
g_free(xendev);
return NULL;
}
@@ -306,8 +306,8 @@ static struct XenDevice *xen_be_del_xendev(int dom, int dev)
g_free(xendev->fe);
}
- if (xendev->evtchndev != XC_HANDLER_INITIAL_VALUE) {
- xc_evtchn_close(xendev->evtchndev);
+ if (xendev->evtchndev != NULL) {
+ xenevtchn_close(xendev->evtchndev);
}
if (xendev->gnttabdev != XC_HANDLER_INITIAL_VALUE) {
xc_gnttab_close(xendev->gnttabdev);
@@ -691,13 +691,14 @@ static void xen_be_evtchn_event(void *opaque)
struct XenDevice *xendev = opaque;
evtchn_port_t port;
- port = xc_evtchn_pending(xendev->evtchndev);
+ port = xenevtchn_pending(xendev->evtchndev);
if (port != xendev->local_port) {
- xen_be_printf(xendev, 0, "xc_evtchn_pending returned %d (expected %d)\n",
+ xen_be_printf(xendev, 0,
+ "xenevtchn_pending returned %d (expected %d)\n",
port, xendev->local_port);
return;
}
- xc_evtchn_unmask(xendev->evtchndev, port);
+ xenevtchn_unmask(xendev->evtchndev, port);
if (xendev->ops->event) {
xendev->ops->event(xendev);
@@ -740,14 +741,14 @@ int xen_be_bind_evtchn(struct XenDevice *xendev)
if (xendev->local_port != -1) {
return 0;
}
- xendev->local_port = xc_evtchn_bind_interdomain
+ xendev->local_port = xenevtchn_bind_interdomain
(xendev->evtchndev, xendev->dom, xendev->remote_port);
if (xendev->local_port == -1) {
- xen_be_printf(xendev, 0, "xc_evtchn_bind_interdomain failed\n");
+ xen_be_printf(xendev, 0, "xenevtchn_bind_interdomain failed\n");
return -1;
}
xen_be_printf(xendev, 2, "bind evtchn port %d\n", xendev->local_port);
- qemu_set_fd_handler(xc_evtchn_fd(xendev->evtchndev),
+ qemu_set_fd_handler(xenevtchn_fd(xendev->evtchndev),
xen_be_evtchn_event, NULL, xendev);
return 0;
}
@@ -757,15 +758,15 @@ void xen_be_unbind_evtchn(struct XenDevice *xendev)
if (xendev->local_port == -1) {
return;
}
- qemu_set_fd_handler(xc_evtchn_fd(xendev->evtchndev), NULL, NULL, NULL);
- xc_evtchn_unbind(xendev->evtchndev, xendev->local_port);
+ qemu_set_fd_handler(xenevtchn_fd(xendev->evtchndev), NULL, NULL, NULL);
+ xenevtchn_unbind(xendev->evtchndev, xendev->local_port);
xen_be_printf(xendev, 2, "unbind evtchn port %d\n", xendev->local_port);
xendev->local_port = -1;
}
int xen_be_send_notify(struct XenDevice *xendev)
{
- return xc_evtchn_notify(xendev->evtchndev, xendev->local_port);
+ return xenevtchn_notify(xendev->evtchndev, xendev->local_port);
}
/*
diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
index 3b4125e..a90314f 100644
--- a/include/hw/xen/xen_backend.h
+++ b/include/hw/xen/xen_backend.h
@@ -46,7 +46,7 @@ struct XenDevice {
int remote_port;
int local_port;
- XenEvtchn evtchndev;
+ xenevtchn_handle *evtchndev;
XenGnttab gnttabdev;
struct XenDevOps *ops;
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index d7fa6a4..6f7c003 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -39,17 +39,38 @@ static inline void *xc_map_foreign_bulk(int xc_handle, uint32_t dom, int prot,
#if CONFIG_XEN_CTRL_INTERFACE_VERSION < 410
typedef int XenXC;
-typedef int XenEvtchn;
+typedef int xenevtchn_handle;
typedef int XenGnttab;
# define XC_INTERFACE_FMT "%i"
# define XC_HANDLER_INITIAL_VALUE -1
-static inline XenEvtchn xen_xc_evtchn_open(void *logger,
- unsigned int open_flags)
+static inline xenevtchn_handle *xenevtchn_open(void *logger,
+ unsigned int open_flags)
{
- return xc_evtchn_open();
+ xenevtchn_handle *h = malloc(sizeof(*h));
+ if (!h) {
+ return NULL;
+ }
+ *h = xc_evtchn_open();
+ if (*h == -1) {
+ free(h);
+ h = NULL;
+ }
+ return h;
}
+static inline int xenevtchn_close(xenevtchn_handle *h)
+{
+ int rc = xc_evtchn_close(*h);
+ free(h);
+ return rc;
+}
+#define xenevtchn_fd(h) xc_evtchn_fd(*h)
+#define xenevtchn_pending(h) xc_evtchn_pending(*h)
+#define xenevtchn_notify(h, p) xc_evtchn_notify(*h, p)
+#define xenevtchn_bind_interdomain(h, d, p) xc_evtchn_bind_interdomain(*h, d, p)
+#define xenevtchn_unmask(h, p) xc_evtchn_unmask(*h, p)
+#define xenevtchn_unbind(h, p) xc_evtchn_unmask(*h, p)
static inline XenGnttab xen_xc_gnttab_open(void *logger,
unsigned int open_flags)
@@ -108,17 +129,20 @@ static inline void xs_close(struct xs_handle *xsh)
#else
typedef xc_interface *XenXC;
-typedef xc_evtchn *XenEvtchn;
+typedef xc_evtchn xenevtchn_handle;
typedef xc_gnttab *XenGnttab;
# define XC_INTERFACE_FMT "%p"
# define XC_HANDLER_INITIAL_VALUE NULL
-static inline XenEvtchn xen_xc_evtchn_open(void *logger,
- unsigned int open_flags)
-{
- return xc_evtchn_open(logger, open_flags);
-}
+#define xenevtchn_open(l, f) xc_evtchn_open(l, f);
+#define xenevtchn_close(h) xc_evtchn_close(h)
+#define xenevtchn_fd(h) xc_evtchn_fd(h)
+#define xenevtchn_pending(h) xc_evtchn_pending(h)
+#define xenevtchn_notify(h, p) xc_evtchn_notify(h, p)
+#define xenevtchn_bind_interdomain(h, d, p) xc_evtchn_bind_interdomain(h, d, p)
+#define xenevtchn_unmask(h, p) xc_evtchn_unmask(h, p)
+#define xenevtchn_unbind(h, p) xc_evtchn_unbind(h, p)
static inline XenGnttab xen_xc_gnttab_open(void *logger,
unsigned int open_flags)
diff --git a/xen-hvm.c b/xen-hvm.c
index 3d78a0c..6680782 100644
--- a/xen-hvm.c
+++ b/xen-hvm.c
@@ -109,7 +109,7 @@ typedef struct XenIOState {
/* evtchn local port for buffered io */
evtchn_port_t bufioreq_local_port;
/* the evtchn fd for polling */
- XenEvtchn xce_handle;
+ xenevtchn_handle *xce_handle;
/* which vcpu we are serving */
int send_vcpu;
@@ -714,7 +714,7 @@ static ioreq_t *cpu_get_ioreq(XenIOState *state)
int i;
evtchn_port_t port;
- port = xc_evtchn_pending(state->xce_handle);
+ port = xenevtchn_pending(state->xce_handle);
if (port == state->bufioreq_local_port) {
timer_mod(state->buffered_io_timer,
BUFFER_IO_MAX_DELAY + qemu_clock_get_ms(QEMU_CLOCK_REALTIME));
@@ -733,7 +733,7 @@ static ioreq_t *cpu_get_ioreq(XenIOState *state)
}
/* unmask the wanted port again */
- xc_evtchn_unmask(state->xce_handle, port);
+ xenevtchn_unmask(state->xce_handle, port);
/* get the io packet from shared memory */
state->send_vcpu = i;
@@ -1040,7 +1040,7 @@ static void handle_buffered_io(void *opaque)
BUFFER_IO_MAX_DELAY + qemu_clock_get_ms(QEMU_CLOCK_REALTIME));
} else {
timer_del(state->buffered_io_timer);
- xc_evtchn_unmask(state->xce_handle, state->bufioreq_local_port);
+ xenevtchn_unmask(state->xce_handle, state->bufioreq_local_port);
}
}
@@ -1084,7 +1084,8 @@ static void cpu_handle_ioreq(void *opaque)
}
req->state = STATE_IORESP_READY;
- xc_evtchn_notify(state->xce_handle, state->ioreq_local_port[state->send_vcpu]);
+ xenevtchn_notify(state->xce_handle,
+ state->ioreq_local_port[state->send_vcpu]);
}
}
@@ -1092,8 +1093,8 @@ static void xen_main_loop_prepare(XenIOState *state)
{
int evtchn_fd = -1;
- if (state->xce_handle != XC_HANDLER_INITIAL_VALUE) {
- evtchn_fd = xc_evtchn_fd(state->xce_handle);
+ if (state->xce_handle != NULL) {
+ evtchn_fd = xenevtchn_fd(state->xce_handle);
}
state->buffered_io_timer = timer_new_ms(QEMU_CLOCK_REALTIME, handle_buffered_io,
@@ -1131,7 +1132,7 @@ static void xen_exit_notifier(Notifier *n, void *data)
{
XenIOState *state = container_of(n, XenIOState, exit);
- xc_evtchn_close(state->xce_handle);
+ xenevtchn_close(state->xce_handle);
xs_daemon_close(state->xenstore);
}
@@ -1200,8 +1201,8 @@ int xen_hvm_init(PCMachineState *pcms,
state = g_malloc0(sizeof (XenIOState));
- state->xce_handle = xen_xc_evtchn_open(NULL, 0);
- if (state->xce_handle == XC_HANDLER_INITIAL_VALUE) {
+ state->xce_handle = xenevtchn_open(NULL, 0);
+ if (state->xce_handle == NULL) {
perror("xen: event channel open");
return -1;
}
@@ -1281,7 +1282,7 @@ int xen_hvm_init(PCMachineState *pcms,
/* FIXME: how about if we overflow the page here? */
for (i = 0; i < max_cpus; i++) {
- rc = xc_evtchn_bind_interdomain(state->xce_handle, xen_domid,
+ rc = xenevtchn_bind_interdomain(state->xce_handle, xen_domid,
xen_vcpu_eport(state->shared_page, i));
if (rc == -1) {
fprintf(stderr, "shared evtchn %d bind error %d\n", i, errno);
@@ -1290,7 +1291,7 @@ int xen_hvm_init(PCMachineState *pcms,
state->ioreq_local_port[i] = rc;
}
- rc = xc_evtchn_bind_interdomain(state->xce_handle, xen_domid,
+ rc = xenevtchn_bind_interdomain(state->xce_handle, xen_domid,
bufioreq_evtchn);
if (rc == -1) {
fprintf(stderr, "buffered evtchn bind error %d\n", errno);
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 3/9] xen: Switch to libxengnttab interface for compat shims.
2015-11-09 12:01 ` Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxengnttab which provides access to grant
tables.
In preparation for this switch the compatibility layer in xen_common.h
(which support building with older versions of Xen) to use what will
be the new library API. This means that the gnttab shim will disappear
for versions of Xen which include libxengnttab.
To simplify things for the <= 4.0.0 support we wrap the int fd in a
malloc(sizeof int) such that the handle is always a pointer. This
leads to less typedef headaches and the need for
XC_HANDLER_INITIAL_VALUE etc for these interfaces.
Note that this patch does not add any support for actually using
libxengnttab, it just adjusts the existing shims.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v4: Ran checkpatch, fixed all errors
Allocate correct size for handle (i.e. not size of the ptr)
Rebase onto "xen_console: correctly cleanup primary console on
teardown."
v5: Rebase over b4f72e31b924 "hw/net/xen_nic.c: Free 'netdev->txs'
when map 'netdev->rxs' fails" which added a new gnttab_munmap.
---
hw/block/xen_disk.c | 38 ++++++++++++++++++++------------------
hw/char/xen_console.c | 4 ++--
hw/net/xen_nic.c | 18 +++++++++---------
hw/xen/xen_backend.c | 10 +++++-----
include/hw/xen/xen_backend.h | 2 +-
include/hw/xen/xen_common.h | 42 ++++++++++++++++++++++++++++++++----------
6 files changed, 69 insertions(+), 45 deletions(-)
diff --git a/hw/block/xen_disk.c b/hw/block/xen_disk.c
index 1bbc111..0e80da1 100644
--- a/hw/block/xen_disk.c
+++ b/hw/block/xen_disk.c
@@ -173,11 +173,11 @@ static gint int_cmp(gconstpointer a, gconstpointer b, gpointer user_data)
static void destroy_grant(gpointer pgnt)
{
PersistentGrant *grant = pgnt;
- XenGnttab gnt = grant->blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = grant->blkdev->xendev.gnttabdev;
- if (xc_gnttab_munmap(gnt, grant->page, 1) != 0) {
+ if (xengnttab_munmap(gnt, grant->page, 1) != 0) {
xen_be_printf(&grant->blkdev->xendev, 0,
- "xc_gnttab_munmap failed: %s\n",
+ "xengnttab_munmap failed: %s\n",
strerror(errno));
}
grant->blkdev->persistent_gnt_count--;
@@ -190,11 +190,11 @@ static void remove_persistent_region(gpointer data, gpointer dev)
{
PersistentRegion *region = data;
struct XenBlkDev *blkdev = dev;
- XenGnttab gnt = blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = blkdev->xendev.gnttabdev;
- if (xc_gnttab_munmap(gnt, region->addr, region->num) != 0) {
+ if (xengnttab_munmap(gnt, region->addr, region->num) != 0) {
xen_be_printf(&blkdev->xendev, 0,
- "xc_gnttab_munmap region %p failed: %s\n",
+ "xengnttab_munmap region %p failed: %s\n",
region->addr, strerror(errno));
}
xen_be_printf(&blkdev->xendev, 3,
@@ -329,7 +329,7 @@ err:
static void ioreq_unmap(struct ioreq *ioreq)
{
- XenGnttab gnt = ioreq->blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
int i;
if (ioreq->num_unmap == 0 || ioreq->mapped == 0) {
@@ -339,8 +339,9 @@ static void ioreq_unmap(struct ioreq *ioreq)
if (!ioreq->pages) {
return;
}
- if (xc_gnttab_munmap(gnt, ioreq->pages, ioreq->num_unmap) != 0) {
- xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
+ if (xengnttab_munmap(gnt, ioreq->pages, ioreq->num_unmap) != 0) {
+ xen_be_printf(&ioreq->blkdev->xendev, 0,
+ "xengnttab_munmap failed: %s\n",
strerror(errno));
}
ioreq->blkdev->cnt_map -= ioreq->num_unmap;
@@ -350,8 +351,9 @@ static void ioreq_unmap(struct ioreq *ioreq)
if (!ioreq->page[i]) {
continue;
}
- if (xc_gnttab_munmap(gnt, ioreq->page[i], 1) != 0) {
- xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
+ if (xengnttab_munmap(gnt, ioreq->page[i], 1) != 0) {
+ xen_be_printf(&ioreq->blkdev->xendev, 0,
+ "xengnttab_munmap failed: %s\n",
strerror(errno));
}
ioreq->blkdev->cnt_map--;
@@ -363,7 +365,7 @@ static void ioreq_unmap(struct ioreq *ioreq)
static int ioreq_map(struct ioreq *ioreq)
{
- XenGnttab gnt = ioreq->blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
uint32_t domids[BLKIF_MAX_SEGMENTS_PER_REQUEST];
uint32_t refs[BLKIF_MAX_SEGMENTS_PER_REQUEST];
void *page[BLKIF_MAX_SEGMENTS_PER_REQUEST];
@@ -414,7 +416,7 @@ static int ioreq_map(struct ioreq *ioreq)
}
if (batch_maps && new_maps) {
- ioreq->pages = xc_gnttab_map_grant_refs
+ ioreq->pages = xengnttab_map_grant_refs
(gnt, new_maps, domids, refs, ioreq->prot);
if (ioreq->pages == NULL) {
xen_be_printf(&ioreq->blkdev->xendev, 0,
@@ -430,7 +432,7 @@ static int ioreq_map(struct ioreq *ioreq)
ioreq->blkdev->cnt_map += new_maps;
} else if (new_maps) {
for (i = 0; i < new_maps; i++) {
- ioreq->page[i] = xc_gnttab_map_grant_ref
+ ioreq->page[i] = xengnttab_map_grant_ref
(gnt, domids[i], refs[i], ioreq->prot);
if (ioreq->page[i] == NULL) {
xen_be_printf(&ioreq->blkdev->xendev, 0,
@@ -763,9 +765,9 @@ static void blk_alloc(struct XenDevice *xendev)
if (xen_mode != XEN_EMULATE) {
batch_maps = 1;
}
- if (xc_gnttab_set_max_grants(xendev->gnttabdev,
+ if (xengnttab_set_max_grants(xendev->gnttabdev,
MAX_GRANTS(max_requests, BLKIF_MAX_SEGMENTS_PER_REQUEST)) < 0) {
- xen_be_printf(xendev, 0, "xc_gnttab_set_max_grants failed: %s\n",
+ xen_be_printf(xendev, 0, "xengnttab_set_max_grants failed: %s\n",
strerror(errno));
}
}
@@ -972,7 +974,7 @@ static int blk_connect(struct XenDevice *xendev)
}
}
- blkdev->sring = xc_gnttab_map_grant_ref(blkdev->xendev.gnttabdev,
+ blkdev->sring = xengnttab_map_grant_ref(blkdev->xendev.gnttabdev,
blkdev->xendev.dom,
blkdev->ring_ref,
PROT_READ | PROT_WRITE);
@@ -1037,7 +1039,7 @@ static void blk_disconnect(struct XenDevice *xendev)
xen_be_unbind_evtchn(&blkdev->xendev);
if (blkdev->sring) {
- xc_gnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
+ xengnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
blkdev->cnt_map--;
blkdev->sring = NULL;
}
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index 63ade33..8c06c19 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -233,7 +233,7 @@ static int con_initialise(struct XenDevice *xendev)
PROT_READ|PROT_WRITE,
con->ring_ref);
} else {
- con->sring = xc_gnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
+ con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
con->ring_ref,
PROT_READ|PROT_WRITE);
}
@@ -275,7 +275,7 @@ static void con_disconnect(struct XenDevice *xendev)
if (!xendev->dev) {
munmap(con->sring, XC_PAGE_SIZE);
} else {
- xc_gnttab_munmap(xendev->gnttabdev, con->sring, 1);
+ xengnttab_munmap(xendev->gnttabdev, con->sring, 1);
}
con->sring = NULL;
}
diff --git a/hw/net/xen_nic.c b/hw/net/xen_nic.c
index 0da16b4..6d69a6a 100644
--- a/hw/net/xen_nic.c
+++ b/hw/net/xen_nic.c
@@ -168,7 +168,7 @@ static void net_tx_packets(struct XenNetDev *netdev)
(txreq.flags & NETTXF_more_data) ? " more_data" : "",
(txreq.flags & NETTXF_extra_info) ? " extra_info" : "");
- page = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
+ page = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
netdev->xendev.dom,
txreq.gref, PROT_READ);
if (page == NULL) {
@@ -190,7 +190,7 @@ static void net_tx_packets(struct XenNetDev *netdev)
qemu_send_packet(qemu_get_queue(netdev->nic),
page + txreq.offset, txreq.size);
}
- xc_gnttab_munmap(netdev->xendev.gnttabdev, page, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, page, 1);
net_tx_response(netdev, &txreq, NETIF_RSP_OKAY);
}
if (!netdev->tx_work) {
@@ -260,7 +260,7 @@ static ssize_t net_rx_packet(NetClientState *nc, const uint8_t *buf, size_t size
memcpy(&rxreq, RING_GET_REQUEST(&netdev->rx_ring, rc), sizeof(rxreq));
netdev->rx_ring.req_cons = ++rc;
- page = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
+ page = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
netdev->xendev.dom,
rxreq.gref, PROT_WRITE);
if (page == NULL) {
@@ -270,7 +270,7 @@ static ssize_t net_rx_packet(NetClientState *nc, const uint8_t *buf, size_t size
return -1;
}
memcpy(page + NET_IP_ALIGN, buf, size);
- xc_gnttab_munmap(netdev->xendev.gnttabdev, page, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, page, 1);
net_rx_response(netdev, &rxreq, NETIF_RSP_OKAY, NET_IP_ALIGN, size, 0);
return size;
@@ -342,19 +342,19 @@ static int net_connect(struct XenDevice *xendev)
return -1;
}
- netdev->txs = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
+ netdev->txs = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
netdev->xendev.dom,
netdev->tx_ring_ref,
PROT_READ | PROT_WRITE);
if (!netdev->txs) {
return -1;
}
- netdev->rxs = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
+ netdev->rxs = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
netdev->xendev.dom,
netdev->rx_ring_ref,
PROT_READ | PROT_WRITE);
if (!netdev->rxs) {
- xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
netdev->txs = NULL;
return -1;
}
@@ -379,11 +379,11 @@ static void net_disconnect(struct XenDevice *xendev)
xen_be_unbind_evtchn(&netdev->xendev);
if (netdev->txs) {
- xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
netdev->txs = NULL;
}
if (netdev->rxs) {
- xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->rxs, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, netdev->rxs, 1);
netdev->rxs = NULL;
}
}
diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
index ae2a1f0..966e34f 100644
--- a/hw/xen/xen_backend.c
+++ b/hw/xen/xen_backend.c
@@ -252,15 +252,15 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev,
fcntl(xenevtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
if (ops->flags & DEVOPS_FLAG_NEED_GNTDEV) {
- xendev->gnttabdev = xen_xc_gnttab_open(NULL, 0);
- if (xendev->gnttabdev == XC_HANDLER_INITIAL_VALUE) {
+ xendev->gnttabdev = xengnttab_open(NULL, 0);
+ if (xendev->gnttabdev == NULL) {
xen_be_printf(NULL, 0, "can't open gnttab device\n");
xenevtchn_close(xendev->evtchndev);
g_free(xendev);
return NULL;
}
} else {
- xendev->gnttabdev = XC_HANDLER_INITIAL_VALUE;
+ xendev->gnttabdev = NULL;
}
QTAILQ_INSERT_TAIL(&xendevs, xendev, next);
@@ -309,8 +309,8 @@ static struct XenDevice *xen_be_del_xendev(int dom, int dev)
if (xendev->evtchndev != NULL) {
xenevtchn_close(xendev->evtchndev);
}
- if (xendev->gnttabdev != XC_HANDLER_INITIAL_VALUE) {
- xc_gnttab_close(xendev->gnttabdev);
+ if (xendev->gnttabdev != NULL) {
+ xengnttab_close(xendev->gnttabdev);
}
QTAILQ_REMOVE(&xendevs, xendev, next);
diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
index a90314f..8e8857b 100644
--- a/include/hw/xen/xen_backend.h
+++ b/include/hw/xen/xen_backend.h
@@ -47,7 +47,7 @@ struct XenDevice {
int local_port;
xenevtchn_handle *evtchndev;
- XenGnttab gnttabdev;
+ xengnttab_handle *gnttabdev;
struct XenDevOps *ops;
QTAILQ_ENTRY(XenDevice) next;
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index 6f7c003..dde6b7f 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -40,7 +40,7 @@ static inline void *xc_map_foreign_bulk(int xc_handle, uint32_t dom, int prot,
typedef int XenXC;
typedef int xenevtchn_handle;
-typedef int XenGnttab;
+typedef int xengnttab_handle;
# define XC_INTERFACE_FMT "%i"
# define XC_HANDLER_INITIAL_VALUE -1
@@ -72,11 +72,31 @@ static inline int xenevtchn_close(xenevtchn_handle *h)
#define xenevtchn_unmask(h, p) xc_evtchn_unmask(*h, p)
#define xenevtchn_unbind(h, p) xc_evtchn_unmask(*h, p)
-static inline XenGnttab xen_xc_gnttab_open(void *logger,
- unsigned int open_flags)
+static inline xengnttab_handle *xengnttab_open(void *logger,
+ unsigned int open_flags)
{
- return xc_gnttab_open();
+ xengnttab_handle *h = malloc(sizeof(*h));
+ if (!h) {
+ return NULL;
+ }
+ *h = xc_gnttab_open();
+ if (*h == -1) {
+ free(h);
+ h = NULL;
+ }
+ return h;
}
+static inline int xengnttab_close(xengnttab_handle *h)
+{
+ int rc = xc_gnttab_close(*h);
+ free(h);
+ return rc;
+}
+#define xengnttab_set_max_grants(h, n) xc_gnttab_set_max_grants(*h, n)
+#define xengnttab_map_grant_ref(h, d, r, p) xc_gnttab_map_grant_ref(*h, d, r, p)
+#define xengnttab_map_grant_refs(h, c, d, r, p) \
+ xc_gnttab_map_grant_refs(*h, c, d, r, p)
+#define xengnttab_munmap(h, a, n) xc_gnttab_munmap(*h, a, n)
static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
unsigned int open_flags)
@@ -130,7 +150,7 @@ static inline void xs_close(struct xs_handle *xsh)
typedef xc_interface *XenXC;
typedef xc_evtchn xenevtchn_handle;
-typedef xc_gnttab *XenGnttab;
+typedef xc_gnttab xengnttab_handle;
# define XC_INTERFACE_FMT "%p"
# define XC_HANDLER_INITIAL_VALUE NULL
@@ -144,11 +164,13 @@ typedef xc_gnttab *XenGnttab;
#define xenevtchn_unmask(h, p) xc_evtchn_unmask(h, p)
#define xenevtchn_unbind(h, p) xc_evtchn_unbind(h, p)
-static inline XenGnttab xen_xc_gnttab_open(void *logger,
- unsigned int open_flags)
-{
- return xc_gnttab_open(logger, open_flags);
-}
+#define xengnttab_open(l, f) xc_gnttab_open(l, f)
+#define xengnttab_close(h) xc_gnttab_close(h)
+#define xengnttab_set_max_grants(h, n) xc_gnttab_set_max_grants(h, n)
+#define xengnttab_map_grant_ref(h, d, r, p) xc_gnttab_map_grant_ref(h, d, r, p)
+#define xengnttab_munmap(h, a, n) xc_gnttab_munmap(h, a, n)
+#define xengnttab_map_grant_refs(h, c, d, r, p) \
+ xc_gnttab_map_grant_refs(h, c, d, r, p)
static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
unsigned int open_flags)
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 3/9] xen: Switch to libxengnttab interface for compat shims.
@ 2015-11-09 12:01 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxengnttab which provides access to grant
tables.
In preparation for this switch the compatibility layer in xen_common.h
(which support building with older versions of Xen) to use what will
be the new library API. This means that the gnttab shim will disappear
for versions of Xen which include libxengnttab.
To simplify things for the <= 4.0.0 support we wrap the int fd in a
malloc(sizeof int) such that the handle is always a pointer. This
leads to less typedef headaches and the need for
XC_HANDLER_INITIAL_VALUE etc for these interfaces.
Note that this patch does not add any support for actually using
libxengnttab, it just adjusts the existing shims.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v4: Ran checkpatch, fixed all errors
Allocate correct size for handle (i.e. not size of the ptr)
Rebase onto "xen_console: correctly cleanup primary console on
teardown."
v5: Rebase over b4f72e31b924 "hw/net/xen_nic.c: Free 'netdev->txs'
when map 'netdev->rxs' fails" which added a new gnttab_munmap.
---
hw/block/xen_disk.c | 38 ++++++++++++++++++++------------------
hw/char/xen_console.c | 4 ++--
hw/net/xen_nic.c | 18 +++++++++---------
hw/xen/xen_backend.c | 10 +++++-----
include/hw/xen/xen_backend.h | 2 +-
include/hw/xen/xen_common.h | 42 ++++++++++++++++++++++++++++++++----------
6 files changed, 69 insertions(+), 45 deletions(-)
diff --git a/hw/block/xen_disk.c b/hw/block/xen_disk.c
index 1bbc111..0e80da1 100644
--- a/hw/block/xen_disk.c
+++ b/hw/block/xen_disk.c
@@ -173,11 +173,11 @@ static gint int_cmp(gconstpointer a, gconstpointer b, gpointer user_data)
static void destroy_grant(gpointer pgnt)
{
PersistentGrant *grant = pgnt;
- XenGnttab gnt = grant->blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = grant->blkdev->xendev.gnttabdev;
- if (xc_gnttab_munmap(gnt, grant->page, 1) != 0) {
+ if (xengnttab_munmap(gnt, grant->page, 1) != 0) {
xen_be_printf(&grant->blkdev->xendev, 0,
- "xc_gnttab_munmap failed: %s\n",
+ "xengnttab_munmap failed: %s\n",
strerror(errno));
}
grant->blkdev->persistent_gnt_count--;
@@ -190,11 +190,11 @@ static void remove_persistent_region(gpointer data, gpointer dev)
{
PersistentRegion *region = data;
struct XenBlkDev *blkdev = dev;
- XenGnttab gnt = blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = blkdev->xendev.gnttabdev;
- if (xc_gnttab_munmap(gnt, region->addr, region->num) != 0) {
+ if (xengnttab_munmap(gnt, region->addr, region->num) != 0) {
xen_be_printf(&blkdev->xendev, 0,
- "xc_gnttab_munmap region %p failed: %s\n",
+ "xengnttab_munmap region %p failed: %s\n",
region->addr, strerror(errno));
}
xen_be_printf(&blkdev->xendev, 3,
@@ -329,7 +329,7 @@ err:
static void ioreq_unmap(struct ioreq *ioreq)
{
- XenGnttab gnt = ioreq->blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
int i;
if (ioreq->num_unmap == 0 || ioreq->mapped == 0) {
@@ -339,8 +339,9 @@ static void ioreq_unmap(struct ioreq *ioreq)
if (!ioreq->pages) {
return;
}
- if (xc_gnttab_munmap(gnt, ioreq->pages, ioreq->num_unmap) != 0) {
- xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
+ if (xengnttab_munmap(gnt, ioreq->pages, ioreq->num_unmap) != 0) {
+ xen_be_printf(&ioreq->blkdev->xendev, 0,
+ "xengnttab_munmap failed: %s\n",
strerror(errno));
}
ioreq->blkdev->cnt_map -= ioreq->num_unmap;
@@ -350,8 +351,9 @@ static void ioreq_unmap(struct ioreq *ioreq)
if (!ioreq->page[i]) {
continue;
}
- if (xc_gnttab_munmap(gnt, ioreq->page[i], 1) != 0) {
- xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
+ if (xengnttab_munmap(gnt, ioreq->page[i], 1) != 0) {
+ xen_be_printf(&ioreq->blkdev->xendev, 0,
+ "xengnttab_munmap failed: %s\n",
strerror(errno));
}
ioreq->blkdev->cnt_map--;
@@ -363,7 +365,7 @@ static void ioreq_unmap(struct ioreq *ioreq)
static int ioreq_map(struct ioreq *ioreq)
{
- XenGnttab gnt = ioreq->blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
uint32_t domids[BLKIF_MAX_SEGMENTS_PER_REQUEST];
uint32_t refs[BLKIF_MAX_SEGMENTS_PER_REQUEST];
void *page[BLKIF_MAX_SEGMENTS_PER_REQUEST];
@@ -414,7 +416,7 @@ static int ioreq_map(struct ioreq *ioreq)
}
if (batch_maps && new_maps) {
- ioreq->pages = xc_gnttab_map_grant_refs
+ ioreq->pages = xengnttab_map_grant_refs
(gnt, new_maps, domids, refs, ioreq->prot);
if (ioreq->pages == NULL) {
xen_be_printf(&ioreq->blkdev->xendev, 0,
@@ -430,7 +432,7 @@ static int ioreq_map(struct ioreq *ioreq)
ioreq->blkdev->cnt_map += new_maps;
} else if (new_maps) {
for (i = 0; i < new_maps; i++) {
- ioreq->page[i] = xc_gnttab_map_grant_ref
+ ioreq->page[i] = xengnttab_map_grant_ref
(gnt, domids[i], refs[i], ioreq->prot);
if (ioreq->page[i] == NULL) {
xen_be_printf(&ioreq->blkdev->xendev, 0,
@@ -763,9 +765,9 @@ static void blk_alloc(struct XenDevice *xendev)
if (xen_mode != XEN_EMULATE) {
batch_maps = 1;
}
- if (xc_gnttab_set_max_grants(xendev->gnttabdev,
+ if (xengnttab_set_max_grants(xendev->gnttabdev,
MAX_GRANTS(max_requests, BLKIF_MAX_SEGMENTS_PER_REQUEST)) < 0) {
- xen_be_printf(xendev, 0, "xc_gnttab_set_max_grants failed: %s\n",
+ xen_be_printf(xendev, 0, "xengnttab_set_max_grants failed: %s\n",
strerror(errno));
}
}
@@ -972,7 +974,7 @@ static int blk_connect(struct XenDevice *xendev)
}
}
- blkdev->sring = xc_gnttab_map_grant_ref(blkdev->xendev.gnttabdev,
+ blkdev->sring = xengnttab_map_grant_ref(blkdev->xendev.gnttabdev,
blkdev->xendev.dom,
blkdev->ring_ref,
PROT_READ | PROT_WRITE);
@@ -1037,7 +1039,7 @@ static void blk_disconnect(struct XenDevice *xendev)
xen_be_unbind_evtchn(&blkdev->xendev);
if (blkdev->sring) {
- xc_gnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
+ xengnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
blkdev->cnt_map--;
blkdev->sring = NULL;
}
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index 63ade33..8c06c19 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -233,7 +233,7 @@ static int con_initialise(struct XenDevice *xendev)
PROT_READ|PROT_WRITE,
con->ring_ref);
} else {
- con->sring = xc_gnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
+ con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
con->ring_ref,
PROT_READ|PROT_WRITE);
}
@@ -275,7 +275,7 @@ static void con_disconnect(struct XenDevice *xendev)
if (!xendev->dev) {
munmap(con->sring, XC_PAGE_SIZE);
} else {
- xc_gnttab_munmap(xendev->gnttabdev, con->sring, 1);
+ xengnttab_munmap(xendev->gnttabdev, con->sring, 1);
}
con->sring = NULL;
}
diff --git a/hw/net/xen_nic.c b/hw/net/xen_nic.c
index 0da16b4..6d69a6a 100644
--- a/hw/net/xen_nic.c
+++ b/hw/net/xen_nic.c
@@ -168,7 +168,7 @@ static void net_tx_packets(struct XenNetDev *netdev)
(txreq.flags & NETTXF_more_data) ? " more_data" : "",
(txreq.flags & NETTXF_extra_info) ? " extra_info" : "");
- page = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
+ page = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
netdev->xendev.dom,
txreq.gref, PROT_READ);
if (page == NULL) {
@@ -190,7 +190,7 @@ static void net_tx_packets(struct XenNetDev *netdev)
qemu_send_packet(qemu_get_queue(netdev->nic),
page + txreq.offset, txreq.size);
}
- xc_gnttab_munmap(netdev->xendev.gnttabdev, page, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, page, 1);
net_tx_response(netdev, &txreq, NETIF_RSP_OKAY);
}
if (!netdev->tx_work) {
@@ -260,7 +260,7 @@ static ssize_t net_rx_packet(NetClientState *nc, const uint8_t *buf, size_t size
memcpy(&rxreq, RING_GET_REQUEST(&netdev->rx_ring, rc), sizeof(rxreq));
netdev->rx_ring.req_cons = ++rc;
- page = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
+ page = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
netdev->xendev.dom,
rxreq.gref, PROT_WRITE);
if (page == NULL) {
@@ -270,7 +270,7 @@ static ssize_t net_rx_packet(NetClientState *nc, const uint8_t *buf, size_t size
return -1;
}
memcpy(page + NET_IP_ALIGN, buf, size);
- xc_gnttab_munmap(netdev->xendev.gnttabdev, page, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, page, 1);
net_rx_response(netdev, &rxreq, NETIF_RSP_OKAY, NET_IP_ALIGN, size, 0);
return size;
@@ -342,19 +342,19 @@ static int net_connect(struct XenDevice *xendev)
return -1;
}
- netdev->txs = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
+ netdev->txs = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
netdev->xendev.dom,
netdev->tx_ring_ref,
PROT_READ | PROT_WRITE);
if (!netdev->txs) {
return -1;
}
- netdev->rxs = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
+ netdev->rxs = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
netdev->xendev.dom,
netdev->rx_ring_ref,
PROT_READ | PROT_WRITE);
if (!netdev->rxs) {
- xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
netdev->txs = NULL;
return -1;
}
@@ -379,11 +379,11 @@ static void net_disconnect(struct XenDevice *xendev)
xen_be_unbind_evtchn(&netdev->xendev);
if (netdev->txs) {
- xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
netdev->txs = NULL;
}
if (netdev->rxs) {
- xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->rxs, 1);
+ xengnttab_munmap(netdev->xendev.gnttabdev, netdev->rxs, 1);
netdev->rxs = NULL;
}
}
diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
index ae2a1f0..966e34f 100644
--- a/hw/xen/xen_backend.c
+++ b/hw/xen/xen_backend.c
@@ -252,15 +252,15 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev,
fcntl(xenevtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
if (ops->flags & DEVOPS_FLAG_NEED_GNTDEV) {
- xendev->gnttabdev = xen_xc_gnttab_open(NULL, 0);
- if (xendev->gnttabdev == XC_HANDLER_INITIAL_VALUE) {
+ xendev->gnttabdev = xengnttab_open(NULL, 0);
+ if (xendev->gnttabdev == NULL) {
xen_be_printf(NULL, 0, "can't open gnttab device\n");
xenevtchn_close(xendev->evtchndev);
g_free(xendev);
return NULL;
}
} else {
- xendev->gnttabdev = XC_HANDLER_INITIAL_VALUE;
+ xendev->gnttabdev = NULL;
}
QTAILQ_INSERT_TAIL(&xendevs, xendev, next);
@@ -309,8 +309,8 @@ static struct XenDevice *xen_be_del_xendev(int dom, int dev)
if (xendev->evtchndev != NULL) {
xenevtchn_close(xendev->evtchndev);
}
- if (xendev->gnttabdev != XC_HANDLER_INITIAL_VALUE) {
- xc_gnttab_close(xendev->gnttabdev);
+ if (xendev->gnttabdev != NULL) {
+ xengnttab_close(xendev->gnttabdev);
}
QTAILQ_REMOVE(&xendevs, xendev, next);
diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
index a90314f..8e8857b 100644
--- a/include/hw/xen/xen_backend.h
+++ b/include/hw/xen/xen_backend.h
@@ -47,7 +47,7 @@ struct XenDevice {
int local_port;
xenevtchn_handle *evtchndev;
- XenGnttab gnttabdev;
+ xengnttab_handle *gnttabdev;
struct XenDevOps *ops;
QTAILQ_ENTRY(XenDevice) next;
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index 6f7c003..dde6b7f 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -40,7 +40,7 @@ static inline void *xc_map_foreign_bulk(int xc_handle, uint32_t dom, int prot,
typedef int XenXC;
typedef int xenevtchn_handle;
-typedef int XenGnttab;
+typedef int xengnttab_handle;
# define XC_INTERFACE_FMT "%i"
# define XC_HANDLER_INITIAL_VALUE -1
@@ -72,11 +72,31 @@ static inline int xenevtchn_close(xenevtchn_handle *h)
#define xenevtchn_unmask(h, p) xc_evtchn_unmask(*h, p)
#define xenevtchn_unbind(h, p) xc_evtchn_unmask(*h, p)
-static inline XenGnttab xen_xc_gnttab_open(void *logger,
- unsigned int open_flags)
+static inline xengnttab_handle *xengnttab_open(void *logger,
+ unsigned int open_flags)
{
- return xc_gnttab_open();
+ xengnttab_handle *h = malloc(sizeof(*h));
+ if (!h) {
+ return NULL;
+ }
+ *h = xc_gnttab_open();
+ if (*h == -1) {
+ free(h);
+ h = NULL;
+ }
+ return h;
}
+static inline int xengnttab_close(xengnttab_handle *h)
+{
+ int rc = xc_gnttab_close(*h);
+ free(h);
+ return rc;
+}
+#define xengnttab_set_max_grants(h, n) xc_gnttab_set_max_grants(*h, n)
+#define xengnttab_map_grant_ref(h, d, r, p) xc_gnttab_map_grant_ref(*h, d, r, p)
+#define xengnttab_map_grant_refs(h, c, d, r, p) \
+ xc_gnttab_map_grant_refs(*h, c, d, r, p)
+#define xengnttab_munmap(h, a, n) xc_gnttab_munmap(*h, a, n)
static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
unsigned int open_flags)
@@ -130,7 +150,7 @@ static inline void xs_close(struct xs_handle *xsh)
typedef xc_interface *XenXC;
typedef xc_evtchn xenevtchn_handle;
-typedef xc_gnttab *XenGnttab;
+typedef xc_gnttab xengnttab_handle;
# define XC_INTERFACE_FMT "%p"
# define XC_HANDLER_INITIAL_VALUE NULL
@@ -144,11 +164,13 @@ typedef xc_gnttab *XenGnttab;
#define xenevtchn_unmask(h, p) xc_evtchn_unmask(h, p)
#define xenevtchn_unbind(h, p) xc_evtchn_unbind(h, p)
-static inline XenGnttab xen_xc_gnttab_open(void *logger,
- unsigned int open_flags)
-{
- return xc_gnttab_open(logger, open_flags);
-}
+#define xengnttab_open(l, f) xc_gnttab_open(l, f)
+#define xengnttab_close(h) xc_gnttab_close(h)
+#define xengnttab_set_max_grants(h, n) xc_gnttab_set_max_grants(h, n)
+#define xengnttab_map_grant_ref(h, d, r, p) xc_gnttab_map_grant_ref(h, d, r, p)
+#define xengnttab_munmap(h, a, n) xc_gnttab_munmap(h, a, n)
+#define xengnttab_map_grant_refs(h, c, d, r, p) \
+ xc_gnttab_map_grant_refs(h, c, d, r, p)
static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
unsigned int open_flags)
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 4/9] xen: Switch uses of xc_map_foreign_range into xc_map_foreign_bulk
2015-11-09 12:01 ` Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxenforeignmemory which provides access to
privileged foreign mappings and which will provide an interface
equivalent to xc_map_foreign_bulk.
In preparation for this switch all uses of xc_map_foreign_range to
xc_map_foreign_bulk. This is trivial because size was always
XC_PAGE_SIZE so the necessary adjustments are trivial:
* Pass &mfn (an array of length 1) instead of mfn. The function
takes a pointer to const, so there is no possibily of mfn changing
due to this change.
* Add a local err variable to receive the errors and pass &err
(again, an array of length 1)
* Adjust any existing logging to include err too
* Pass nr_pages=1 instead of size=XC_PAGE_SIZE
There is one wrinkle in xen_console.c:con_initialise() where
con->ring_ref is an int but can in some code paths (when !xendev->dev)
be treated as an mfn. I think this is an existing latent truncation
hazard on platforms where xen_pfn_t is 64-bit and int is 32-bit (e.g.
amd64, both arm* variants). I'm unsure under what circumstances
xendev->dev can be NULL or if anything elsewhere ensures the value
fits into an int. For now I just use a temporary xen_pfn_t to in
effect upcast the pointer from int* to xen_pfn_t*.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
v4: Ran checkpatch, fixed all errors
Mention the const-ness of the mfn array in the commit log
---
hw/char/xen_console.c | 9 +++++----
hw/display/xenfb.c | 6 +++---
xen-hvm.c | 30 ++++++++++++++++--------------
3 files changed, 24 insertions(+), 21 deletions(-)
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index 8c06c19..11c6472 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -219,6 +219,7 @@ static int con_initialise(struct XenDevice *xendev)
{
struct XenConsole *con = container_of(xendev, struct XenConsole, xendev);
int limit;
+ int err;
if (xenstore_read_int(con->console, "ring-ref", &con->ring_ref) == -1)
return -1;
@@ -228,10 +229,10 @@ static int con_initialise(struct XenDevice *xendev)
con->buffer.max_capacity = limit;
if (!xendev->dev) {
- con->sring = xc_map_foreign_range(xen_xc, con->xendev.dom,
- XC_PAGE_SIZE,
- PROT_READ|PROT_WRITE,
- con->ring_ref);
+ xen_pfn_t mfn = con->ring_ref;
+ con->sring = xc_map_foreign_bulk(xen_xc, con->xendev.dom,
+ PROT_READ|PROT_WRITE,
+ &mfn, &err, 1);
} else {
con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
con->ring_ref,
diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index 5e324ef..e913bf6 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -96,6 +96,7 @@ struct XenFB {
static int common_bind(struct common *c)
{
uint64_t mfn;
+ int err;
if (xenstore_read_fe_uint64(&c->xendev, "page-ref", &mfn) == -1)
return -1;
@@ -104,9 +105,8 @@ static int common_bind(struct common *c)
if (xenstore_read_fe_int(&c->xendev, "event-channel", &c->xendev.remote_port) == -1)
return -1;
- c->page = xc_map_foreign_range(xen_xc, c->xendev.dom,
- XC_PAGE_SIZE,
- PROT_READ | PROT_WRITE, mfn);
+ c->page = xc_map_foreign_bulk(xen_xc, c->xendev.dom,
+ PROT_READ | PROT_WRITE, &mfn, &err, 1);
if (c->page == NULL)
return -1;
diff --git a/xen-hvm.c b/xen-hvm.c
index 6680782..afc4fc8 100644
--- a/xen-hvm.c
+++ b/xen-hvm.c
@@ -1193,7 +1193,7 @@ static void xen_wakeup_notifier(Notifier *notifier, void *data)
int xen_hvm_init(PCMachineState *pcms,
MemoryRegion **ram_memory)
{
- int i, rc;
+ int i, rc, map_err;
xen_pfn_t ioreq_pfn;
xen_pfn_t bufioreq_pfn;
evtchn_port_t bufioreq_evtchn;
@@ -1240,33 +1240,35 @@ int xen_hvm_init(PCMachineState *pcms,
DPRINTF("buffered io page at pfn %lx\n", bufioreq_pfn);
DPRINTF("buffered io evtchn is %x\n", bufioreq_evtchn);
- state->shared_page = xc_map_foreign_range(xen_xc, xen_domid, XC_PAGE_SIZE,
- PROT_READ|PROT_WRITE, ioreq_pfn);
+ state->shared_page = xc_map_foreign_bulk(xen_xc, xen_domid,
+ PROT_READ|PROT_WRITE,
+ &ioreq_pfn, &map_err, 1);
if (state->shared_page == NULL) {
- hw_error("map shared IO page returned error %d handle=" XC_INTERFACE_FMT,
- errno, xen_xc);
+ hw_error(
+ "map shared IO page returned error %d(%d) handle=" XC_INTERFACE_FMT,
+ errno, map_err, xen_xc);
}
rc = xen_get_vmport_regs_pfn(xen_xc, xen_domid, &ioreq_pfn);
if (!rc) {
DPRINTF("shared vmport page at pfn %lx\n", ioreq_pfn);
state->shared_vmport_page =
- xc_map_foreign_range(xen_xc, xen_domid, XC_PAGE_SIZE,
- PROT_READ|PROT_WRITE, ioreq_pfn);
+ xc_map_foreign_bulk(xen_xc, xen_domid, PROT_READ|PROT_WRITE,
+ &ioreq_pfn, &map_err, 1);
if (state->shared_vmport_page == NULL) {
- hw_error("map shared vmport IO page returned error %d handle="
- XC_INTERFACE_FMT, errno, xen_xc);
+ hw_error("map shared vmport IO page returned error %d (%d) handle="
+ XC_INTERFACE_FMT, errno, map_err, xen_xc);
}
} else if (rc != -ENOSYS) {
hw_error("get vmport regs pfn returned error %d, rc=%d", errno, rc);
}
- state->buffered_io_page = xc_map_foreign_range(xen_xc, xen_domid,
- XC_PAGE_SIZE,
- PROT_READ|PROT_WRITE,
- bufioreq_pfn);
+ state->buffered_io_page = xc_map_foreign_bulk(xen_xc, xen_domid,
+ PROT_READ|PROT_WRITE,
+ &bufioreq_pfn,
+ &map_err, 1);
if (state->buffered_io_page == NULL) {
- hw_error("map buffered IO page returned error %d", errno);
+ hw_error("map buffered IO page returned error %d (%d)", errno, map_err);
}
/* Note: cpus is empty at this point in init */
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 4/9] xen: Switch uses of xc_map_foreign_range into xc_map_foreign_bulk
@ 2015-11-09 12:01 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxenforeignmemory which provides access to
privileged foreign mappings and which will provide an interface
equivalent to xc_map_foreign_bulk.
In preparation for this switch all uses of xc_map_foreign_range to
xc_map_foreign_bulk. This is trivial because size was always
XC_PAGE_SIZE so the necessary adjustments are trivial:
* Pass &mfn (an array of length 1) instead of mfn. The function
takes a pointer to const, so there is no possibily of mfn changing
due to this change.
* Add a local err variable to receive the errors and pass &err
(again, an array of length 1)
* Adjust any existing logging to include err too
* Pass nr_pages=1 instead of size=XC_PAGE_SIZE
There is one wrinkle in xen_console.c:con_initialise() where
con->ring_ref is an int but can in some code paths (when !xendev->dev)
be treated as an mfn. I think this is an existing latent truncation
hazard on platforms where xen_pfn_t is 64-bit and int is 32-bit (e.g.
amd64, both arm* variants). I'm unsure under what circumstances
xendev->dev can be NULL or if anything elsewhere ensures the value
fits into an int. For now I just use a temporary xen_pfn_t to in
effect upcast the pointer from int* to xen_pfn_t*.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
v4: Ran checkpatch, fixed all errors
Mention the const-ness of the mfn array in the commit log
---
hw/char/xen_console.c | 9 +++++----
hw/display/xenfb.c | 6 +++---
xen-hvm.c | 30 ++++++++++++++++--------------
3 files changed, 24 insertions(+), 21 deletions(-)
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index 8c06c19..11c6472 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -219,6 +219,7 @@ static int con_initialise(struct XenDevice *xendev)
{
struct XenConsole *con = container_of(xendev, struct XenConsole, xendev);
int limit;
+ int err;
if (xenstore_read_int(con->console, "ring-ref", &con->ring_ref) == -1)
return -1;
@@ -228,10 +229,10 @@ static int con_initialise(struct XenDevice *xendev)
con->buffer.max_capacity = limit;
if (!xendev->dev) {
- con->sring = xc_map_foreign_range(xen_xc, con->xendev.dom,
- XC_PAGE_SIZE,
- PROT_READ|PROT_WRITE,
- con->ring_ref);
+ xen_pfn_t mfn = con->ring_ref;
+ con->sring = xc_map_foreign_bulk(xen_xc, con->xendev.dom,
+ PROT_READ|PROT_WRITE,
+ &mfn, &err, 1);
} else {
con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
con->ring_ref,
diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index 5e324ef..e913bf6 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -96,6 +96,7 @@ struct XenFB {
static int common_bind(struct common *c)
{
uint64_t mfn;
+ int err;
if (xenstore_read_fe_uint64(&c->xendev, "page-ref", &mfn) == -1)
return -1;
@@ -104,9 +105,8 @@ static int common_bind(struct common *c)
if (xenstore_read_fe_int(&c->xendev, "event-channel", &c->xendev.remote_port) == -1)
return -1;
- c->page = xc_map_foreign_range(xen_xc, c->xendev.dom,
- XC_PAGE_SIZE,
- PROT_READ | PROT_WRITE, mfn);
+ c->page = xc_map_foreign_bulk(xen_xc, c->xendev.dom,
+ PROT_READ | PROT_WRITE, &mfn, &err, 1);
if (c->page == NULL)
return -1;
diff --git a/xen-hvm.c b/xen-hvm.c
index 6680782..afc4fc8 100644
--- a/xen-hvm.c
+++ b/xen-hvm.c
@@ -1193,7 +1193,7 @@ static void xen_wakeup_notifier(Notifier *notifier, void *data)
int xen_hvm_init(PCMachineState *pcms,
MemoryRegion **ram_memory)
{
- int i, rc;
+ int i, rc, map_err;
xen_pfn_t ioreq_pfn;
xen_pfn_t bufioreq_pfn;
evtchn_port_t bufioreq_evtchn;
@@ -1240,33 +1240,35 @@ int xen_hvm_init(PCMachineState *pcms,
DPRINTF("buffered io page at pfn %lx\n", bufioreq_pfn);
DPRINTF("buffered io evtchn is %x\n", bufioreq_evtchn);
- state->shared_page = xc_map_foreign_range(xen_xc, xen_domid, XC_PAGE_SIZE,
- PROT_READ|PROT_WRITE, ioreq_pfn);
+ state->shared_page = xc_map_foreign_bulk(xen_xc, xen_domid,
+ PROT_READ|PROT_WRITE,
+ &ioreq_pfn, &map_err, 1);
if (state->shared_page == NULL) {
- hw_error("map shared IO page returned error %d handle=" XC_INTERFACE_FMT,
- errno, xen_xc);
+ hw_error(
+ "map shared IO page returned error %d(%d) handle=" XC_INTERFACE_FMT,
+ errno, map_err, xen_xc);
}
rc = xen_get_vmport_regs_pfn(xen_xc, xen_domid, &ioreq_pfn);
if (!rc) {
DPRINTF("shared vmport page at pfn %lx\n", ioreq_pfn);
state->shared_vmport_page =
- xc_map_foreign_range(xen_xc, xen_domid, XC_PAGE_SIZE,
- PROT_READ|PROT_WRITE, ioreq_pfn);
+ xc_map_foreign_bulk(xen_xc, xen_domid, PROT_READ|PROT_WRITE,
+ &ioreq_pfn, &map_err, 1);
if (state->shared_vmport_page == NULL) {
- hw_error("map shared vmport IO page returned error %d handle="
- XC_INTERFACE_FMT, errno, xen_xc);
+ hw_error("map shared vmport IO page returned error %d (%d) handle="
+ XC_INTERFACE_FMT, errno, map_err, xen_xc);
}
} else if (rc != -ENOSYS) {
hw_error("get vmport regs pfn returned error %d, rc=%d", errno, rc);
}
- state->buffered_io_page = xc_map_foreign_range(xen_xc, xen_domid,
- XC_PAGE_SIZE,
- PROT_READ|PROT_WRITE,
- bufioreq_pfn);
+ state->buffered_io_page = xc_map_foreign_bulk(xen_xc, xen_domid,
+ PROT_READ|PROT_WRITE,
+ &bufioreq_pfn,
+ &map_err, 1);
if (state->buffered_io_page == NULL) {
- hw_error("map buffered IO page returned error %d", errno);
+ hw_error("map buffered IO page returned error %d (%d)", errno, map_err);
}
/* Note: cpus is empty at this point in init */
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 5/9] xen: Switch uses of xc_map_foreign_pages into xc_map_foreign_bulk
2015-11-09 12:01 ` Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxenforeignmemory which provides access to
privileged foreign mappings and which will provide an interface
equivalent to xc_map_foreign_bulk.
In preparation for this switch both uses of xc_map_foreign_pages
(which both happen to be in xenfb_map_fb) to xc_map_foreign_bulk. This
simply requires allocating and passing a new err array (the same one
for both calls).
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
v4: Fix indentation
---
hw/display/xenfb.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index e913bf6..71de556 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -433,6 +433,7 @@ static int xenfb_map_fb(struct XenFB *xenfb)
int n_fbdirs;
xen_pfn_t *pgmfns = NULL;
xen_pfn_t *fbmfns = NULL;
+ int *errs = NULL;
void *map, *pd;
int mode, ret = -1;
@@ -492,17 +493,18 @@ static int xenfb_map_fb(struct XenFB *xenfb)
pgmfns = g_malloc0(sizeof(xen_pfn_t) * n_fbdirs);
fbmfns = g_malloc0(sizeof(xen_pfn_t) * xenfb->fbpages);
+ errs = g_malloc0(sizeof(int) * n_fbdirs);
xenfb_copy_mfns(mode, n_fbdirs, pgmfns, pd);
- map = xc_map_foreign_pages(xen_xc, xenfb->c.xendev.dom,
- PROT_READ, pgmfns, n_fbdirs);
+ map = xc_map_foreign_bulk(xen_xc, xenfb->c.xendev.dom,
+ PROT_READ, pgmfns, errs, n_fbdirs);
if (map == NULL)
goto out;
xenfb_copy_mfns(mode, xenfb->fbpages, fbmfns, map);
munmap(map, n_fbdirs * XC_PAGE_SIZE);
- xenfb->pixels = xc_map_foreign_pages(xen_xc, xenfb->c.xendev.dom,
- PROT_READ, fbmfns, xenfb->fbpages);
+ xenfb->pixels = xc_map_foreign_bulk(xen_xc, xenfb->c.xendev.dom,
+ PROT_READ, fbmfns, errs, xenfb->fbpages);
if (xenfb->pixels == NULL)
goto out;
@@ -511,6 +513,7 @@ static int xenfb_map_fb(struct XenFB *xenfb)
out:
g_free(pgmfns);
g_free(fbmfns);
+ g_free(errs);
return ret;
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 5/9] xen: Switch uses of xc_map_foreign_pages into xc_map_foreign_bulk
@ 2015-11-09 12:01 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxenforeignmemory which provides access to
privileged foreign mappings and which will provide an interface
equivalent to xc_map_foreign_bulk.
In preparation for this switch both uses of xc_map_foreign_pages
(which both happen to be in xenfb_map_fb) to xc_map_foreign_bulk. This
simply requires allocating and passing a new err array (the same one
for both calls).
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
v4: Fix indentation
---
hw/display/xenfb.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index e913bf6..71de556 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -433,6 +433,7 @@ static int xenfb_map_fb(struct XenFB *xenfb)
int n_fbdirs;
xen_pfn_t *pgmfns = NULL;
xen_pfn_t *fbmfns = NULL;
+ int *errs = NULL;
void *map, *pd;
int mode, ret = -1;
@@ -492,17 +493,18 @@ static int xenfb_map_fb(struct XenFB *xenfb)
pgmfns = g_malloc0(sizeof(xen_pfn_t) * n_fbdirs);
fbmfns = g_malloc0(sizeof(xen_pfn_t) * xenfb->fbpages);
+ errs = g_malloc0(sizeof(int) * n_fbdirs);
xenfb_copy_mfns(mode, n_fbdirs, pgmfns, pd);
- map = xc_map_foreign_pages(xen_xc, xenfb->c.xendev.dom,
- PROT_READ, pgmfns, n_fbdirs);
+ map = xc_map_foreign_bulk(xen_xc, xenfb->c.xendev.dom,
+ PROT_READ, pgmfns, errs, n_fbdirs);
if (map == NULL)
goto out;
xenfb_copy_mfns(mode, xenfb->fbpages, fbmfns, map);
munmap(map, n_fbdirs * XC_PAGE_SIZE);
- xenfb->pixels = xc_map_foreign_pages(xen_xc, xenfb->c.xendev.dom,
- PROT_READ, fbmfns, xenfb->fbpages);
+ xenfb->pixels = xc_map_foreign_bulk(xen_xc, xenfb->c.xendev.dom,
+ PROT_READ, fbmfns, errs, xenfb->fbpages);
if (xenfb->pixels == NULL)
goto out;
@@ -511,6 +513,7 @@ static int xenfb_map_fb(struct XenFB *xenfb)
out:
g_free(pgmfns);
g_free(fbmfns);
+ g_free(errs);
return ret;
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 6/9] xen: Switch uses of xc_map_foreign_bulk to use libxenforeignmemory API.
2015-11-09 12:01 ` Ian Campbell
` (6 preceding siblings ...)
(?)
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxenforeignmemory which provides access to
privileged foreign mappings and which will provide an interface
equivalent to xc_map_foreign_bulk.
In preparation for adding support for libxenforeignmemory add support
to the <=4.0 and <=4.6 compat code in xen_common.h to allow us to
switch to using the new API. These shims will disappear for versions
of Xen which include libxenforeignmemory.
Since libxenforeignmemory will have its own handle type but for <= 4.6
the functionality is provided by using a libxenctrl handle we
introduce a new global xen_fmem alongside the existing xen_xc. In fact
we make xen_fmem a pointer to the existing xen_xc, which then works
correctly with both <=4.0 (xc handle is an int) and <=4.6 (xc handle
is a pointer). In the latter case xen_fmem is actually a double
indirect pointer, but it all falls out in the wash.
Unlike libxenctrl libxenforeignmemory has an explicit unmap function,
rather than just specifying that munmap should be used, so the unmap
paths are updated to use xenforeignmemory_unmap, which is a shim for
munmap on these versions of xen. The mappings in xen-hvm.c do not
appear to be unmapped (which makes sense for a qemu-dm process)
In fb_disconnect this results in a change from simply mmap over the
existing mapping (with an implicit munmap) to expliclty unmapping with
xenforeignmemory_unmap and then mapping the required anonymous memory
in the same hole. I don't think this is a problem since any other
thread which was racily touching this region would already be running
the risk of hitting the mapping halfway through the call. If this is
thought to be a problem then we could consider adding an extra API to
the libxenforeignmemory interface to replace a foreign mapping with
anonymous shared memory, but I'd prefer not to.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
I noticed in xen_console.c that the decision to use a foreign
privileged memory mapping vs a grant dev is made using different
variables in con_initialise vs con_disconnect. The former uses
xendev->dev while the latter uses xendev->gnttabdev. Is this a latent
bug?
v4: Rebase onto "xen_console: correctly cleanup primary console on
teardown."
xenforeignmemory_unmap takes pages not bytes
Compat wrapper for xenforeignmemory_open instead of ifdef in code.
Run check patch and fix most issues. I did not fix:
ERROR: do not initialise globals to 0 or NULL
+xenforeignmemory_handle *xen_fmem = NULL;
=> This is consistent with all of the existing declarations.
ERROR: need consistent spacing around '*' (ctx:WxV)
+typedef xc_interface *xenforeignmemory_handle;
=> I think this is a false +ve since this is a pointer "*" not a multiple "*".
---
hw/char/xen_console.c | 8 ++++----
hw/display/xenfb.c | 15 ++++++++-------
hw/xen/xen_backend.c | 3 ++-
include/hw/xen/xen_backend.h | 1 +
include/hw/xen/xen_common.h | 12 ++++++++++++
xen-common.c | 6 ++++++
xen-hvm.c | 18 +++++++++---------
xen-mapcache.c | 6 +++---
8 files changed, 45 insertions(+), 24 deletions(-)
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index 11c6472..24f3a40 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -230,9 +230,9 @@ static int con_initialise(struct XenDevice *xendev)
if (!xendev->dev) {
xen_pfn_t mfn = con->ring_ref;
- con->sring = xc_map_foreign_bulk(xen_xc, con->xendev.dom,
- PROT_READ|PROT_WRITE,
- &mfn, &err, 1);
+ con->sring = xenforeignmemory_map(xen_fmem, con->xendev.dom,
+ PROT_READ|PROT_WRITE,
+ &mfn, &err, 1);
} else {
con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
con->ring_ref,
@@ -274,7 +274,7 @@ static void con_disconnect(struct XenDevice *xendev)
if (con->sring) {
if (!xendev->dev) {
- munmap(con->sring, XC_PAGE_SIZE);
+ xenforeignmemory_unmap(xen_fmem, con->sring, 1);
} else {
xengnttab_munmap(xendev->gnttabdev, con->sring, 1);
}
diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index 71de556..706f6ba 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -105,8 +105,8 @@ static int common_bind(struct common *c)
if (xenstore_read_fe_int(&c->xendev, "event-channel", &c->xendev.remote_port) == -1)
return -1;
- c->page = xc_map_foreign_bulk(xen_xc, c->xendev.dom,
- PROT_READ | PROT_WRITE, &mfn, &err, 1);
+ c->page = xenforeignmemory_map(xen_fmem, c->xendev.dom,
+ PROT_READ | PROT_WRITE, &mfn, &err, 1);
if (c->page == NULL)
return -1;
@@ -121,7 +121,7 @@ static void common_unbind(struct common *c)
{
xen_be_unbind_evtchn(&c->xendev);
if (c->page) {
- munmap(c->page, XC_PAGE_SIZE);
+ xenforeignmemory_unmap(xen_fmem, c->page, 1);
c->page = NULL;
}
}
@@ -496,14 +496,14 @@ static int xenfb_map_fb(struct XenFB *xenfb)
errs = g_malloc0(sizeof(int) * n_fbdirs);
xenfb_copy_mfns(mode, n_fbdirs, pgmfns, pd);
- map = xc_map_foreign_bulk(xen_xc, xenfb->c.xendev.dom,
- PROT_READ, pgmfns, errs, n_fbdirs);
+ map = xenforeignmemory_map(xen_fmem, xenfb->c.xendev.dom,
+ PROT_READ, pgmfns, errs, n_fbdirs);
if (map == NULL)
goto out;
xenfb_copy_mfns(mode, xenfb->fbpages, fbmfns, map);
- munmap(map, n_fbdirs * XC_PAGE_SIZE);
+ xenforeignmemory_unmap(xen_fmem, map, n_fbdirs);
- xenfb->pixels = xc_map_foreign_bulk(xen_xc, xenfb->c.xendev.dom,
+ xenfb->pixels = xenforeignmemory_map(xen_fmem, xenfb->c.xendev.dom,
PROT_READ, fbmfns, errs, xenfb->fbpages);
if (xenfb->pixels == NULL)
goto out;
@@ -912,6 +912,7 @@ static void fb_disconnect(struct XenDevice *xendev)
* Replacing the framebuffer with anonymous shared memory
* instead. This releases the guest pages and keeps qemu happy.
*/
+ xenforeignmemory_unmap(xen_fmem, fb->pixels, fb->fbpages);
fb->pixels = mmap(fb->pixels, fb->fbpages * XC_PAGE_SIZE,
PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANON,
-1, 0);
diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
index 966e34f..caa459c 100644
--- a/hw/xen/xen_backend.c
+++ b/hw/xen/xen_backend.c
@@ -45,6 +45,7 @@
/* public */
XenXC xen_xc = XC_HANDLER_INITIAL_VALUE;
+xenforeignmemory_handle *xen_fmem = NULL;
struct xs_handle *xenstore = NULL;
const char *xen_protocol;
@@ -717,7 +718,7 @@ int xen_be_init(void)
qemu_set_fd_handler(xs_fileno(xenstore), xenstore_update, NULL, NULL);
- if (xen_xc == XC_HANDLER_INITIAL_VALUE) {
+ if (xen_xc == XC_HANDLER_INITIAL_VALUE || xen_fmem == NULL) {
/* Check if xen_init() have been called */
goto err;
}
diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
index 8e8857b..e0d52ee 100644
--- a/include/hw/xen/xen_backend.h
+++ b/include/hw/xen/xen_backend.h
@@ -57,6 +57,7 @@ struct XenDevice {
/* variables */
extern XenXC xen_xc;
+extern xenforeignmemory_handle *xen_fmem;
extern struct xs_handle *xenstore;
extern const char *xen_protocol;
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index dde6b7f..323c76c 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -41,6 +41,7 @@ static inline void *xc_map_foreign_bulk(int xc_handle, uint32_t dom, int prot,
typedef int XenXC;
typedef int xenevtchn_handle;
typedef int xengnttab_handle;
+typedef int xenforeignmemory_handle;
# define XC_INTERFACE_FMT "%i"
# define XC_HANDLER_INITIAL_VALUE -1
@@ -104,6 +105,11 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
return xc_interface_open();
}
+#define xenforeignmemory_open(l, f) &xen_xc
+#define xenforeignmemory_map(h, d, p, a, e, n) \
+ xc_map_foreign_bulk(*h, d, p, a, e, n)
+#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
+
static inline int xc_fd(int xen_xc)
{
return xen_xc;
@@ -149,6 +155,7 @@ static inline void xs_close(struct xs_handle *xsh)
#else
typedef xc_interface *XenXC;
+typedef xc_interface *xenforeignmemory_handle;
typedef xc_evtchn xenevtchn_handle;
typedef xc_gnttab xengnttab_handle;
@@ -178,6 +185,11 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
return xc_interface_open(logger, dombuild_logger, open_flags);
}
+#define xenforeignmemory_open(l, f) &xen_xc
+#define xenforeignmemory_map(h, d, p, a, e, n) \
+ xc_map_foreign_bulk(*h, d, p, a, e, n)
+#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
+
/* FIXME There is now way to have the xen fd */
static inline int xc_fd(xc_interface *xen_xc)
{
diff --git a/xen-common.c b/xen-common.c
index 0dcdbc3..6cd2959 100644
--- a/xen-common.c
+++ b/xen-common.c
@@ -118,6 +118,12 @@ static int xen_init(MachineState *ms)
xen_be_printf(NULL, 0, "can't open xen interface\n");
return -1;
}
+ xen_fmem = xenforeignmemory_open(0, 0);
+ if (xen_fmem == NULL) {
+ xen_be_printf(NULL, 0, "can't open xen fmem interface\n");
+ xc_interface_close(xen_xc);
+ return -1;
+ }
qemu_add_vm_change_state_handler(xen_change_state_handler, NULL);
global_state_set_optional();
diff --git a/xen-hvm.c b/xen-hvm.c
index afc4fc8..2c09ba1 100644
--- a/xen-hvm.c
+++ b/xen-hvm.c
@@ -1240,9 +1240,9 @@ int xen_hvm_init(PCMachineState *pcms,
DPRINTF("buffered io page at pfn %lx\n", bufioreq_pfn);
DPRINTF("buffered io evtchn is %x\n", bufioreq_evtchn);
- state->shared_page = xc_map_foreign_bulk(xen_xc, xen_domid,
- PROT_READ|PROT_WRITE,
- &ioreq_pfn, &map_err, 1);
+ state->shared_page = xenforeignmemory_map(xen_fmem, xen_domid,
+ PROT_READ|PROT_WRITE,
+ &ioreq_pfn, &map_err, 1);
if (state->shared_page == NULL) {
hw_error(
"map shared IO page returned error %d(%d) handle=" XC_INTERFACE_FMT,
@@ -1253,8 +1253,8 @@ int xen_hvm_init(PCMachineState *pcms,
if (!rc) {
DPRINTF("shared vmport page at pfn %lx\n", ioreq_pfn);
state->shared_vmport_page =
- xc_map_foreign_bulk(xen_xc, xen_domid, PROT_READ|PROT_WRITE,
- &ioreq_pfn, &map_err, 1);
+ xenforeignmemory_map(xen_fmem, xen_domid, PROT_READ|PROT_WRITE,
+ &ioreq_pfn, &map_err, 1);
if (state->shared_vmport_page == NULL) {
hw_error("map shared vmport IO page returned error %d (%d) handle="
XC_INTERFACE_FMT, errno, map_err, xen_xc);
@@ -1263,10 +1263,10 @@ int xen_hvm_init(PCMachineState *pcms,
hw_error("get vmport regs pfn returned error %d, rc=%d", errno, rc);
}
- state->buffered_io_page = xc_map_foreign_bulk(xen_xc, xen_domid,
- PROT_READ|PROT_WRITE,
- &bufioreq_pfn,
- &map_err, 1);
+ state->buffered_io_page = xenforeignmemory_map(xen_fmem, xen_domid,
+ PROT_READ|PROT_WRITE,
+ &bufioreq_pfn,
+ &map_err, 1);
if (state->buffered_io_page == NULL) {
hw_error("map buffered IO page returned error %d (%d)", errno, map_err);
}
diff --git a/xen-mapcache.c b/xen-mapcache.c
index 97fece2..7c7e8e7 100644
--- a/xen-mapcache.c
+++ b/xen-mapcache.c
@@ -176,10 +176,10 @@ static void xen_remap_bucket(MapCacheEntry *entry,
pfns[i] = (address_index << (MCACHE_BUCKET_SHIFT-XC_PAGE_SHIFT)) + i;
}
- vaddr_base = xc_map_foreign_bulk(xen_xc, xen_domid, PROT_READ|PROT_WRITE,
- pfns, err, nb_pfn);
+ vaddr_base = xenforeignmemory_map(xen_fmem, xen_domid, PROT_READ|PROT_WRITE,
+ pfns, err, nb_pfn);
if (vaddr_base == NULL) {
- perror("xc_map_foreign_bulk");
+ perror("xenforeignmemory_map");
exit(-1);
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 6/9] xen: Switch uses of xc_map_foreign_bulk to use libxenforeignmemory API.
2015-11-09 12:01 ` Ian Campbell
` (7 preceding siblings ...)
(?)
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
One such library will be libxenforeignmemory which provides access to
privileged foreign mappings and which will provide an interface
equivalent to xc_map_foreign_bulk.
In preparation for adding support for libxenforeignmemory add support
to the <=4.0 and <=4.6 compat code in xen_common.h to allow us to
switch to using the new API. These shims will disappear for versions
of Xen which include libxenforeignmemory.
Since libxenforeignmemory will have its own handle type but for <= 4.6
the functionality is provided by using a libxenctrl handle we
introduce a new global xen_fmem alongside the existing xen_xc. In fact
we make xen_fmem a pointer to the existing xen_xc, which then works
correctly with both <=4.0 (xc handle is an int) and <=4.6 (xc handle
is a pointer). In the latter case xen_fmem is actually a double
indirect pointer, but it all falls out in the wash.
Unlike libxenctrl libxenforeignmemory has an explicit unmap function,
rather than just specifying that munmap should be used, so the unmap
paths are updated to use xenforeignmemory_unmap, which is a shim for
munmap on these versions of xen. The mappings in xen-hvm.c do not
appear to be unmapped (which makes sense for a qemu-dm process)
In fb_disconnect this results in a change from simply mmap over the
existing mapping (with an implicit munmap) to expliclty unmapping with
xenforeignmemory_unmap and then mapping the required anonymous memory
in the same hole. I don't think this is a problem since any other
thread which was racily touching this region would already be running
the risk of hitting the mapping halfway through the call. If this is
thought to be a problem then we could consider adding an extra API to
the libxenforeignmemory interface to replace a foreign mapping with
anonymous shared memory, but I'd prefer not to.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
I noticed in xen_console.c that the decision to use a foreign
privileged memory mapping vs a grant dev is made using different
variables in con_initialise vs con_disconnect. The former uses
xendev->dev while the latter uses xendev->gnttabdev. Is this a latent
bug?
v4: Rebase onto "xen_console: correctly cleanup primary console on
teardown."
xenforeignmemory_unmap takes pages not bytes
Compat wrapper for xenforeignmemory_open instead of ifdef in code.
Run check patch and fix most issues. I did not fix:
ERROR: do not initialise globals to 0 or NULL
+xenforeignmemory_handle *xen_fmem = NULL;
=> This is consistent with all of the existing declarations.
ERROR: need consistent spacing around '*' (ctx:WxV)
+typedef xc_interface *xenforeignmemory_handle;
=> I think this is a false +ve since this is a pointer "*" not a multiple "*".
---
hw/char/xen_console.c | 8 ++++----
hw/display/xenfb.c | 15 ++++++++-------
hw/xen/xen_backend.c | 3 ++-
include/hw/xen/xen_backend.h | 1 +
include/hw/xen/xen_common.h | 12 ++++++++++++
xen-common.c | 6 ++++++
xen-hvm.c | 18 +++++++++---------
xen-mapcache.c | 6 +++---
8 files changed, 45 insertions(+), 24 deletions(-)
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index 11c6472..24f3a40 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -230,9 +230,9 @@ static int con_initialise(struct XenDevice *xendev)
if (!xendev->dev) {
xen_pfn_t mfn = con->ring_ref;
- con->sring = xc_map_foreign_bulk(xen_xc, con->xendev.dom,
- PROT_READ|PROT_WRITE,
- &mfn, &err, 1);
+ con->sring = xenforeignmemory_map(xen_fmem, con->xendev.dom,
+ PROT_READ|PROT_WRITE,
+ &mfn, &err, 1);
} else {
con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
con->ring_ref,
@@ -274,7 +274,7 @@ static void con_disconnect(struct XenDevice *xendev)
if (con->sring) {
if (!xendev->dev) {
- munmap(con->sring, XC_PAGE_SIZE);
+ xenforeignmemory_unmap(xen_fmem, con->sring, 1);
} else {
xengnttab_munmap(xendev->gnttabdev, con->sring, 1);
}
diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index 71de556..706f6ba 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -105,8 +105,8 @@ static int common_bind(struct common *c)
if (xenstore_read_fe_int(&c->xendev, "event-channel", &c->xendev.remote_port) == -1)
return -1;
- c->page = xc_map_foreign_bulk(xen_xc, c->xendev.dom,
- PROT_READ | PROT_WRITE, &mfn, &err, 1);
+ c->page = xenforeignmemory_map(xen_fmem, c->xendev.dom,
+ PROT_READ | PROT_WRITE, &mfn, &err, 1);
if (c->page == NULL)
return -1;
@@ -121,7 +121,7 @@ static void common_unbind(struct common *c)
{
xen_be_unbind_evtchn(&c->xendev);
if (c->page) {
- munmap(c->page, XC_PAGE_SIZE);
+ xenforeignmemory_unmap(xen_fmem, c->page, 1);
c->page = NULL;
}
}
@@ -496,14 +496,14 @@ static int xenfb_map_fb(struct XenFB *xenfb)
errs = g_malloc0(sizeof(int) * n_fbdirs);
xenfb_copy_mfns(mode, n_fbdirs, pgmfns, pd);
- map = xc_map_foreign_bulk(xen_xc, xenfb->c.xendev.dom,
- PROT_READ, pgmfns, errs, n_fbdirs);
+ map = xenforeignmemory_map(xen_fmem, xenfb->c.xendev.dom,
+ PROT_READ, pgmfns, errs, n_fbdirs);
if (map == NULL)
goto out;
xenfb_copy_mfns(mode, xenfb->fbpages, fbmfns, map);
- munmap(map, n_fbdirs * XC_PAGE_SIZE);
+ xenforeignmemory_unmap(xen_fmem, map, n_fbdirs);
- xenfb->pixels = xc_map_foreign_bulk(xen_xc, xenfb->c.xendev.dom,
+ xenfb->pixels = xenforeignmemory_map(xen_fmem, xenfb->c.xendev.dom,
PROT_READ, fbmfns, errs, xenfb->fbpages);
if (xenfb->pixels == NULL)
goto out;
@@ -912,6 +912,7 @@ static void fb_disconnect(struct XenDevice *xendev)
* Replacing the framebuffer with anonymous shared memory
* instead. This releases the guest pages and keeps qemu happy.
*/
+ xenforeignmemory_unmap(xen_fmem, fb->pixels, fb->fbpages);
fb->pixels = mmap(fb->pixels, fb->fbpages * XC_PAGE_SIZE,
PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANON,
-1, 0);
diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
index 966e34f..caa459c 100644
--- a/hw/xen/xen_backend.c
+++ b/hw/xen/xen_backend.c
@@ -45,6 +45,7 @@
/* public */
XenXC xen_xc = XC_HANDLER_INITIAL_VALUE;
+xenforeignmemory_handle *xen_fmem = NULL;
struct xs_handle *xenstore = NULL;
const char *xen_protocol;
@@ -717,7 +718,7 @@ int xen_be_init(void)
qemu_set_fd_handler(xs_fileno(xenstore), xenstore_update, NULL, NULL);
- if (xen_xc == XC_HANDLER_INITIAL_VALUE) {
+ if (xen_xc == XC_HANDLER_INITIAL_VALUE || xen_fmem == NULL) {
/* Check if xen_init() have been called */
goto err;
}
diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
index 8e8857b..e0d52ee 100644
--- a/include/hw/xen/xen_backend.h
+++ b/include/hw/xen/xen_backend.h
@@ -57,6 +57,7 @@ struct XenDevice {
/* variables */
extern XenXC xen_xc;
+extern xenforeignmemory_handle *xen_fmem;
extern struct xs_handle *xenstore;
extern const char *xen_protocol;
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index dde6b7f..323c76c 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -41,6 +41,7 @@ static inline void *xc_map_foreign_bulk(int xc_handle, uint32_t dom, int prot,
typedef int XenXC;
typedef int xenevtchn_handle;
typedef int xengnttab_handle;
+typedef int xenforeignmemory_handle;
# define XC_INTERFACE_FMT "%i"
# define XC_HANDLER_INITIAL_VALUE -1
@@ -104,6 +105,11 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
return xc_interface_open();
}
+#define xenforeignmemory_open(l, f) &xen_xc
+#define xenforeignmemory_map(h, d, p, a, e, n) \
+ xc_map_foreign_bulk(*h, d, p, a, e, n)
+#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
+
static inline int xc_fd(int xen_xc)
{
return xen_xc;
@@ -149,6 +155,7 @@ static inline void xs_close(struct xs_handle *xsh)
#else
typedef xc_interface *XenXC;
+typedef xc_interface *xenforeignmemory_handle;
typedef xc_evtchn xenevtchn_handle;
typedef xc_gnttab xengnttab_handle;
@@ -178,6 +185,11 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
return xc_interface_open(logger, dombuild_logger, open_flags);
}
+#define xenforeignmemory_open(l, f) &xen_xc
+#define xenforeignmemory_map(h, d, p, a, e, n) \
+ xc_map_foreign_bulk(*h, d, p, a, e, n)
+#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
+
/* FIXME There is now way to have the xen fd */
static inline int xc_fd(xc_interface *xen_xc)
{
diff --git a/xen-common.c b/xen-common.c
index 0dcdbc3..6cd2959 100644
--- a/xen-common.c
+++ b/xen-common.c
@@ -118,6 +118,12 @@ static int xen_init(MachineState *ms)
xen_be_printf(NULL, 0, "can't open xen interface\n");
return -1;
}
+ xen_fmem = xenforeignmemory_open(0, 0);
+ if (xen_fmem == NULL) {
+ xen_be_printf(NULL, 0, "can't open xen fmem interface\n");
+ xc_interface_close(xen_xc);
+ return -1;
+ }
qemu_add_vm_change_state_handler(xen_change_state_handler, NULL);
global_state_set_optional();
diff --git a/xen-hvm.c b/xen-hvm.c
index afc4fc8..2c09ba1 100644
--- a/xen-hvm.c
+++ b/xen-hvm.c
@@ -1240,9 +1240,9 @@ int xen_hvm_init(PCMachineState *pcms,
DPRINTF("buffered io page at pfn %lx\n", bufioreq_pfn);
DPRINTF("buffered io evtchn is %x\n", bufioreq_evtchn);
- state->shared_page = xc_map_foreign_bulk(xen_xc, xen_domid,
- PROT_READ|PROT_WRITE,
- &ioreq_pfn, &map_err, 1);
+ state->shared_page = xenforeignmemory_map(xen_fmem, xen_domid,
+ PROT_READ|PROT_WRITE,
+ &ioreq_pfn, &map_err, 1);
if (state->shared_page == NULL) {
hw_error(
"map shared IO page returned error %d(%d) handle=" XC_INTERFACE_FMT,
@@ -1253,8 +1253,8 @@ int xen_hvm_init(PCMachineState *pcms,
if (!rc) {
DPRINTF("shared vmport page at pfn %lx\n", ioreq_pfn);
state->shared_vmport_page =
- xc_map_foreign_bulk(xen_xc, xen_domid, PROT_READ|PROT_WRITE,
- &ioreq_pfn, &map_err, 1);
+ xenforeignmemory_map(xen_fmem, xen_domid, PROT_READ|PROT_WRITE,
+ &ioreq_pfn, &map_err, 1);
if (state->shared_vmport_page == NULL) {
hw_error("map shared vmport IO page returned error %d (%d) handle="
XC_INTERFACE_FMT, errno, map_err, xen_xc);
@@ -1263,10 +1263,10 @@ int xen_hvm_init(PCMachineState *pcms,
hw_error("get vmport regs pfn returned error %d, rc=%d", errno, rc);
}
- state->buffered_io_page = xc_map_foreign_bulk(xen_xc, xen_domid,
- PROT_READ|PROT_WRITE,
- &bufioreq_pfn,
- &map_err, 1);
+ state->buffered_io_page = xenforeignmemory_map(xen_fmem, xen_domid,
+ PROT_READ|PROT_WRITE,
+ &bufioreq_pfn,
+ &map_err, 1);
if (state->buffered_io_page == NULL) {
hw_error("map buffered IO page returned error %d (%d)", errno, map_err);
}
diff --git a/xen-mapcache.c b/xen-mapcache.c
index 97fece2..7c7e8e7 100644
--- a/xen-mapcache.c
+++ b/xen-mapcache.c
@@ -176,10 +176,10 @@ static void xen_remap_bucket(MapCacheEntry *entry,
pfns[i] = (address_index << (MCACHE_BUCKET_SHIFT-XC_PAGE_SHIFT)) + i;
}
- vaddr_base = xc_map_foreign_bulk(xen_xc, xen_domid, PROT_READ|PROT_WRITE,
- pfns, err, nb_pfn);
+ vaddr_base = xenforeignmemory_map(xen_fmem, xen_domid, PROT_READ|PROT_WRITE,
+ pfns, err, nb_pfn);
if (vaddr_base == NULL) {
- perror("xc_map_foreign_bulk");
+ perror("xenforeignmemory_map");
exit(-1);
}
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 7/9] xen: Use stable library interfaces when they are available.
2015-11-09 12:01 ` Ian Campbell
` (9 preceding siblings ...)
(?)
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-10 14:00 ` Stefano Stabellini
2015-11-10 14:00 ` Stefano Stabellini
-1 siblings, 2 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
Specifically libxenevtchn, libxengnttab and libxenforeignmemory.
Previous patches have already laid the groundwork for using these by
switching the existing compatibility shims to reflect the intefaces to
these libraries.
So all which remains is to update configure to detect the libraries
and enable their use. Although they are notionally independent we take
an all or nothing approach to the three libraries since they were
added at the same time.
The only non-obvious bit is that we now open a proper xenforeignmemory
handle for xen_fmem instead of reusing the xen_xc handle.
Build tested with 4.0 .. 4.6 (inclusive) and the patches targetting
4.7 which adds these libraries.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v5: Remove ifdef check when undeffing the compat stuff.
s/now way/no way/
v4: xenforeignmemory_open is now a compat wrapper, so no ifdef.
Simplify configury by asserting that interface version 470 will
always have the libraries (lack of them makes it 460).
Ran checkpatch and fixed everything except:
ERROR: need consistent spacing around '*' (ctx:WxV)
+typedef xc_interface *XenXC;
Which I think is a false +ve.
Simpler configure stuff
---
configure | 55 +++++++++++++++++++++++++++++++++++++++++++++
include/hw/xen/xen_common.h | 36 +++++++++++++++++++++++++++--
2 files changed, 89 insertions(+), 2 deletions(-)
diff --git a/configure b/configure
index b4a3488..0253cc9 100755
--- a/configure
+++ b/configure
@@ -1898,6 +1898,7 @@ fi
if test "$xen" != "no" ; then
xen_libs="-lxenstore -lxenctrl -lxenguest"
+ xen_stable_libs="-lxenforeignmemory -lxengnttab -lxenevtchn"
# First we test whether Xen headers and libraries are available.
# If no, we are done and there is no Xen support.
@@ -1920,6 +1921,57 @@ EOF
# Xen unstable
elif
cat > $TMPC <<EOF &&
+/*
+ * If we have stable libs the we don't want the libxc compat
+ * layers, regardless of what CFLAGS we may have been given.
+ */
+#undef XC_WANT_COMPAT_EVTCHN_API
+#undef XC_WANT_COMPAT_GNTTAB_API
+#undef XC_WANT_COMPAT_MAP_FOREIGN_API
+#include <xenctrl.h>
+#include <xenstore.h>
+#include <xenevtchn.h>
+#include <xengnttab.h>
+#include <xenforeignmemory.h>
+#include <stdint.h>
+#include <xen/hvm/hvm_info_table.h>
+#if !defined(HVM_MAX_VCPUS)
+# error HVM_MAX_VCPUS not defined
+#endif
+int main(void) {
+ xc_interface *xc;
+ xenforeignmemory_handle *xfmem;
+ xenevtchn_handle *xe;
+ xengnttab_handle *xg;
+
+ xs_daemon_open();
+
+ xc = xc_interface_open(0, 0, 0);
+ xc_hvm_set_mem_type(0, 0, HVMMEM_ram_ro, 0, 0);
+ xc_domain_add_to_physmap(0, 0, XENMAPSPACE_gmfn, 0, 0);
+ xc_hvm_inject_msi(xc, 0, 0xf0000000, 0x00000000);
+ xc_hvm_create_ioreq_server(xc, 0, HVM_IOREQSRV_BUFIOREQ_ATOMIC, NULL);
+
+ xfmem = xenforeignmemory_open(0, 0);
+ xenforeignmemory_map(xfmem, 0, 0, 0, 0, 0);
+
+ xe = xenevtchn_open(0, 0);
+ xenevtchn_fd(xe);
+
+ xg = xengnttab_open(0, 0);
+ xengnttab_map_grant_ref(xg, 0, 0, 0);
+
+ return 0;
+}
+EOF
+ compile_prog "" "$xen_libs $xen_stable_libs"
+ then
+ xen_ctrl_version=470
+ xen=yes
+
+ # Xen 4.6
+ elif
+ cat > $TMPC <<EOF &&
#include <xenctrl.h>
#include <xenstore.h>
#include <stdint.h>
@@ -2096,6 +2148,9 @@ EOF
fi
if test "$xen" = yes; then
+ if test $xen_ctrl_version -ge 470 ; then
+ libs_softmmu="$xen_stable_libs $libs_softmmu"
+ fi
libs_softmmu="$xen_libs $libs_softmmu"
fi
fi
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index 323c76c..a758ac4 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -6,6 +6,15 @@
#include <stddef.h>
#include <inttypes.h>
+/*
+ * If we have new enough libxenctrl then we do not want/need these compat
+ * interfaces, despite what the user supplied cflags might say. They
+ * must be undefined before including xenctrl.h
+ */
+#undef XC_WANT_COMPAT_EVTCHN_API
+#undef XC_WANT_COMPAT_GNTTAB_API
+#undef XC_WANT_COMPAT_MAP_FOREIGN_API
+
#include <xenctrl.h>
#if CONFIG_XEN_CTRL_INTERFACE_VERSION < 420
# include <xs.h>
@@ -151,8 +160,8 @@ static inline void xs_close(struct xs_handle *xsh)
}
-/* Xen 4.1 */
-#else
+/* Xen 4.1 thru 4.6 */
+#elif CONFIG_XEN_CTRL_INTERFACE_VERSION < 470
typedef xc_interface *XenXC;
typedef xc_interface *xenforeignmemory_handle;
@@ -190,11 +199,34 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
xc_map_foreign_bulk(*h, d, p, a, e, n)
#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
+/* FIXME There is no way to have the xen fd */
+static inline int xc_fd(xc_interface *xen_xc)
+{
+ return -1;
+}
+#else /* CONFIG_XEN_CTRL_INTERFACE_VERSION >= 470 */
+
+typedef xc_interface *XenXC;
+
+# define XC_INTERFACE_FMT "%p"
+# define XC_HANDLER_INITIAL_VALUE NULL
+
+#include <xenevtchn.h>
+#include <xengnttab.h>
+#include <xenforeignmemory.h>
+
+static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
+ unsigned int open_flags)
+{
+ return xc_interface_open(logger, dombuild_logger, open_flags);
+}
+
/* FIXME There is now way to have the xen fd */
static inline int xc_fd(xc_interface *xen_xc)
{
return -1;
}
+
#endif
/* Xen before 4.2 */
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 7/9] xen: Use stable library interfaces when they are available.
2015-11-09 12:01 ` Ian Campbell
` (8 preceding siblings ...)
(?)
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
In Xen 4.7 we are refactoring parts libxenctrl into a number of
separate libraries which will provide backward and forward API and ABI
compatiblity.
Specifically libxenevtchn, libxengnttab and libxenforeignmemory.
Previous patches have already laid the groundwork for using these by
switching the existing compatibility shims to reflect the intefaces to
these libraries.
So all which remains is to update configure to detect the libraries
and enable their use. Although they are notionally independent we take
an all or nothing approach to the three libraries since they were
added at the same time.
The only non-obvious bit is that we now open a proper xenforeignmemory
handle for xen_fmem instead of reusing the xen_xc handle.
Build tested with 4.0 .. 4.6 (inclusive) and the patches targetting
4.7 which adds these libraries.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v5: Remove ifdef check when undeffing the compat stuff.
s/now way/no way/
v4: xenforeignmemory_open is now a compat wrapper, so no ifdef.
Simplify configury by asserting that interface version 470 will
always have the libraries (lack of them makes it 460).
Ran checkpatch and fixed everything except:
ERROR: need consistent spacing around '*' (ctx:WxV)
+typedef xc_interface *XenXC;
Which I think is a false +ve.
Simpler configure stuff
---
configure | 55 +++++++++++++++++++++++++++++++++++++++++++++
include/hw/xen/xen_common.h | 36 +++++++++++++++++++++++++++--
2 files changed, 89 insertions(+), 2 deletions(-)
diff --git a/configure b/configure
index b4a3488..0253cc9 100755
--- a/configure
+++ b/configure
@@ -1898,6 +1898,7 @@ fi
if test "$xen" != "no" ; then
xen_libs="-lxenstore -lxenctrl -lxenguest"
+ xen_stable_libs="-lxenforeignmemory -lxengnttab -lxenevtchn"
# First we test whether Xen headers and libraries are available.
# If no, we are done and there is no Xen support.
@@ -1920,6 +1921,57 @@ EOF
# Xen unstable
elif
cat > $TMPC <<EOF &&
+/*
+ * If we have stable libs the we don't want the libxc compat
+ * layers, regardless of what CFLAGS we may have been given.
+ */
+#undef XC_WANT_COMPAT_EVTCHN_API
+#undef XC_WANT_COMPAT_GNTTAB_API
+#undef XC_WANT_COMPAT_MAP_FOREIGN_API
+#include <xenctrl.h>
+#include <xenstore.h>
+#include <xenevtchn.h>
+#include <xengnttab.h>
+#include <xenforeignmemory.h>
+#include <stdint.h>
+#include <xen/hvm/hvm_info_table.h>
+#if !defined(HVM_MAX_VCPUS)
+# error HVM_MAX_VCPUS not defined
+#endif
+int main(void) {
+ xc_interface *xc;
+ xenforeignmemory_handle *xfmem;
+ xenevtchn_handle *xe;
+ xengnttab_handle *xg;
+
+ xs_daemon_open();
+
+ xc = xc_interface_open(0, 0, 0);
+ xc_hvm_set_mem_type(0, 0, HVMMEM_ram_ro, 0, 0);
+ xc_domain_add_to_physmap(0, 0, XENMAPSPACE_gmfn, 0, 0);
+ xc_hvm_inject_msi(xc, 0, 0xf0000000, 0x00000000);
+ xc_hvm_create_ioreq_server(xc, 0, HVM_IOREQSRV_BUFIOREQ_ATOMIC, NULL);
+
+ xfmem = xenforeignmemory_open(0, 0);
+ xenforeignmemory_map(xfmem, 0, 0, 0, 0, 0);
+
+ xe = xenevtchn_open(0, 0);
+ xenevtchn_fd(xe);
+
+ xg = xengnttab_open(0, 0);
+ xengnttab_map_grant_ref(xg, 0, 0, 0);
+
+ return 0;
+}
+EOF
+ compile_prog "" "$xen_libs $xen_stable_libs"
+ then
+ xen_ctrl_version=470
+ xen=yes
+
+ # Xen 4.6
+ elif
+ cat > $TMPC <<EOF &&
#include <xenctrl.h>
#include <xenstore.h>
#include <stdint.h>
@@ -2096,6 +2148,9 @@ EOF
fi
if test "$xen" = yes; then
+ if test $xen_ctrl_version -ge 470 ; then
+ libs_softmmu="$xen_stable_libs $libs_softmmu"
+ fi
libs_softmmu="$xen_libs $libs_softmmu"
fi
fi
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index 323c76c..a758ac4 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -6,6 +6,15 @@
#include <stddef.h>
#include <inttypes.h>
+/*
+ * If we have new enough libxenctrl then we do not want/need these compat
+ * interfaces, despite what the user supplied cflags might say. They
+ * must be undefined before including xenctrl.h
+ */
+#undef XC_WANT_COMPAT_EVTCHN_API
+#undef XC_WANT_COMPAT_GNTTAB_API
+#undef XC_WANT_COMPAT_MAP_FOREIGN_API
+
#include <xenctrl.h>
#if CONFIG_XEN_CTRL_INTERFACE_VERSION < 420
# include <xs.h>
@@ -151,8 +160,8 @@ static inline void xs_close(struct xs_handle *xsh)
}
-/* Xen 4.1 */
-#else
+/* Xen 4.1 thru 4.6 */
+#elif CONFIG_XEN_CTRL_INTERFACE_VERSION < 470
typedef xc_interface *XenXC;
typedef xc_interface *xenforeignmemory_handle;
@@ -190,11 +199,34 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
xc_map_foreign_bulk(*h, d, p, a, e, n)
#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
+/* FIXME There is no way to have the xen fd */
+static inline int xc_fd(xc_interface *xen_xc)
+{
+ return -1;
+}
+#else /* CONFIG_XEN_CTRL_INTERFACE_VERSION >= 470 */
+
+typedef xc_interface *XenXC;
+
+# define XC_INTERFACE_FMT "%p"
+# define XC_HANDLER_INITIAL_VALUE NULL
+
+#include <xenevtchn.h>
+#include <xengnttab.h>
+#include <xenforeignmemory.h>
+
+static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
+ unsigned int open_flags)
+{
+ return xc_interface_open(logger, dombuild_logger, open_flags);
+}
+
/* FIXME There is now way to have the xen fd */
static inline int xc_fd(xc_interface *xen_xc)
{
return -1;
}
+
#endif
/* Xen before 4.2 */
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 8/9] xen: domainbuild: reopen libxenctrl interface after forking for domain watcher.
2015-11-09 12:01 ` Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
Using an existing libxenctrl handle after a fork was never
particularly safe (especially if foreign mappings existed at the time
of the fork) and the xc fd has been unavailable for many releases.
Reopen the handle after fork and therefore do away with xc_fd().
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
The fact that xc_fd hasn't been useful since at least Xen 4.1 makes me
question the utility of this domainbuild in QEMU. Perhaps we should
just nuke it?
---
hw/xenpv/xen_domainbuild.c | 9 ++++++---
include/hw/xen/xen_common.h | 17 -----------------
2 files changed, 6 insertions(+), 20 deletions(-)
diff --git a/hw/xenpv/xen_domainbuild.c b/hw/xenpv/xen_domainbuild.c
index c0ab753..3e8422f 100644
--- a/hw/xenpv/xen_domainbuild.c
+++ b/hw/xenpv/xen_domainbuild.c
@@ -174,12 +174,15 @@ static int xen_domain_watcher(void)
for (i = 3; i < n; i++) {
if (i == fd[0])
continue;
- if (i == xc_fd(xen_xc)) {
- continue;
- }
close(i);
}
+ /*
+ * Reopen xc interface, since the original is unsafe after fork
+ * and was closed above.
+ */
+ xen_xc = xc_interface_open(0, 0, 0);
+
/* ignore term signals */
signal(SIGINT, SIG_IGN);
signal(SIGTERM, SIG_IGN);
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index a758ac4..76650bc 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -119,12 +119,6 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
xc_map_foreign_bulk(*h, d, p, a, e, n)
#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
-static inline int xc_fd(int xen_xc)
-{
- return xen_xc;
-}
-
-
static inline int xc_domain_populate_physmap_exact
(XenXC xc_handle, uint32_t domid, unsigned long nr_extents,
unsigned int extent_order, unsigned int mem_flags, xen_pfn_t *extent_start)
@@ -199,11 +193,6 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
xc_map_foreign_bulk(*h, d, p, a, e, n)
#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
-/* FIXME There is no way to have the xen fd */
-static inline int xc_fd(xc_interface *xen_xc)
-{
- return -1;
-}
#else /* CONFIG_XEN_CTRL_INTERFACE_VERSION >= 470 */
typedef xc_interface *XenXC;
@@ -221,12 +210,6 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
return xc_interface_open(logger, dombuild_logger, open_flags);
}
-/* FIXME There is now way to have the xen fd */
-static inline int xc_fd(xc_interface *xen_xc)
-{
- return -1;
-}
-
#endif
/* Xen before 4.2 */
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 8/9] xen: domainbuild: reopen libxenctrl interface after forking for domain watcher.
@ 2015-11-09 12:01 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
Using an existing libxenctrl handle after a fork was never
particularly safe (especially if foreign mappings existed at the time
of the fork) and the xc fd has been unavailable for many releases.
Reopen the handle after fork and therefore do away with xc_fd().
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
---
The fact that xc_fd hasn't been useful since at least Xen 4.1 makes me
question the utility of this domainbuild in QEMU. Perhaps we should
just nuke it?
---
hw/xenpv/xen_domainbuild.c | 9 ++++++---
include/hw/xen/xen_common.h | 17 -----------------
2 files changed, 6 insertions(+), 20 deletions(-)
diff --git a/hw/xenpv/xen_domainbuild.c b/hw/xenpv/xen_domainbuild.c
index c0ab753..3e8422f 100644
--- a/hw/xenpv/xen_domainbuild.c
+++ b/hw/xenpv/xen_domainbuild.c
@@ -174,12 +174,15 @@ static int xen_domain_watcher(void)
for (i = 3; i < n; i++) {
if (i == fd[0])
continue;
- if (i == xc_fd(xen_xc)) {
- continue;
- }
close(i);
}
+ /*
+ * Reopen xc interface, since the original is unsafe after fork
+ * and was closed above.
+ */
+ xen_xc = xc_interface_open(0, 0, 0);
+
/* ignore term signals */
signal(SIGINT, SIG_IGN);
signal(SIGTERM, SIG_IGN);
diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
index a758ac4..76650bc 100644
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -119,12 +119,6 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
xc_map_foreign_bulk(*h, d, p, a, e, n)
#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
-static inline int xc_fd(int xen_xc)
-{
- return xen_xc;
-}
-
-
static inline int xc_domain_populate_physmap_exact
(XenXC xc_handle, uint32_t domid, unsigned long nr_extents,
unsigned int extent_order, unsigned int mem_flags, xen_pfn_t *extent_start)
@@ -199,11 +193,6 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
xc_map_foreign_bulk(*h, d, p, a, e, n)
#define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
-/* FIXME There is no way to have the xen fd */
-static inline int xc_fd(xc_interface *xen_xc)
-{
- return -1;
-}
#else /* CONFIG_XEN_CTRL_INTERFACE_VERSION >= 470 */
typedef xc_interface *XenXC;
@@ -221,12 +210,6 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
return xc_interface_open(logger, dombuild_logger, open_flags);
}
-/* FIXME There is now way to have the xen fd */
-static inline int xc_fd(xc_interface *xen_xc)
-{
- return -1;
-}
-
#endif
/* Xen before 4.2 */
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [Qemu-devel] [PATCH QEMU-XEN v5 9/9] xen: make it possible to build without the Xen PV domain builder
2015-11-09 12:01 ` Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
-1 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
Until the previous patch this relied on xc_fd(), which was only
implemented for Xen 4.0 and earlier.
Given this wasn't working since Xen 4.0 I have marked this as disabled
by default.
Removing this support drops the use of a bunch of symbols from
libxenctrl, specifically:
- xc_domain_create
- xc_domain_destroy
- xc_domain_getinfo
- xc_domain_max_vcpus
- xc_domain_setmaxmem
- xc_domain_unpause
- xc_evtchn_alloc_unbound
- xc_linux_build
This is another step towards only using Xen libraries which provide a
stable inteface.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v5: XEN_CREATE entirely wihtin CONFIG_XEN_PV_DOMAIN_BUILD ifdef.
Simplify configure'ry.
v4: Fixed all checkpatch errors.
Disabled by default.
---
configure | 15 +++++++++++++++
hw/xenpv/Makefile.objs | 4 +++-
hw/xenpv/xen_machine_pv.c | 15 +++++++++++----
3 files changed, 29 insertions(+), 5 deletions(-)
diff --git a/configure b/configure
index 0253cc9..4ab6d9c 100755
--- a/configure
+++ b/configure
@@ -247,6 +247,7 @@ vnc_jpeg=""
vnc_png=""
xen=""
xen_ctrl_version=""
+xen_pv_domain_build="no"
xen_pci_passthrough=""
linux_aio=""
cap_ng=""
@@ -917,6 +918,10 @@ for opt do
;;
--enable-xen-pci-passthrough) xen_pci_passthrough="yes"
;;
+ --disable-xen-pv-domain-build) xen_pv_domain_build="no"
+ ;;
+ --enable-xen-pv-domain-build) xen_pv_domain_build="yes"
+ ;;
--disable-brlapi) brlapi="no"
;;
--enable-brlapi) brlapi="yes"
@@ -2172,6 +2177,12 @@ if test "$xen_pci_passthrough" != "no"; then
fi
fi
+if test "$xen_pv_domain_build" = "yes" &&
+ test "$xen" != "yes"; then
+ error_exit "User requested Xen PV domain builder support" \
+ "which requires Xen support."
+fi
+
##########################################
# libtool probe
@@ -4762,6 +4773,7 @@ fi
echo "xen support $xen"
if test "$xen" = "yes" ; then
echo "xen ctrl version $xen_ctrl_version"
+ echo "pv dom build $xen_pv_domain_build"
fi
echo "brlapi support $brlapi"
echo "bluez support $bluez"
@@ -5130,6 +5142,9 @@ fi
if test "$xen" = "yes" ; then
echo "CONFIG_XEN_BACKEND=y" >> $config_host_mak
echo "CONFIG_XEN_CTRL_INTERFACE_VERSION=$xen_ctrl_version" >> $config_host_mak
+ if test "$xen_pv_domain_build" = "yes" ; then
+ echo "CONFIG_XEN_PV_DOMAIN_BUILD=y" >> $config_host_mak
+ fi
fi
if test "$linux_aio" = "yes" ; then
echo "CONFIG_LINUX_AIO=y" >> $config_host_mak
diff --git a/hw/xenpv/Makefile.objs b/hw/xenpv/Makefile.objs
index 49f6e9e..bbf5873 100644
--- a/hw/xenpv/Makefile.objs
+++ b/hw/xenpv/Makefile.objs
@@ -1,2 +1,4 @@
# Xen PV machine support
-obj-$(CONFIG_XEN) += xen_domainbuild.o xen_machine_pv.o
+obj-$(CONFIG_XEN) += xen_machine_pv.o
+# Xen PV machine builder support
+obj-$(CONFIG_XEN_PV_DOMAIN_BUILD) += xen_domainbuild.o
diff --git a/hw/xenpv/xen_machine_pv.c b/hw/xenpv/xen_machine_pv.c
index 23d6ef0..3250b94 100644
--- a/hw/xenpv/xen_machine_pv.c
+++ b/hw/xenpv/xen_machine_pv.c
@@ -30,9 +30,6 @@
static void xen_init_pv(MachineState *machine)
{
- const char *kernel_filename = machine->kernel_filename;
- const char *kernel_cmdline = machine->kernel_cmdline;
- const char *initrd_filename = machine->initrd_filename;
DriveInfo *dinfo;
int i;
@@ -46,17 +43,27 @@ static void xen_init_pv(MachineState *machine)
case XEN_ATTACH:
/* nothing to do, xend handles everything */
break;
- case XEN_CREATE:
+#ifdef CONFIG_XEN_PV_DOMAIN_BUILD
+ case XEN_CREATE: {
+ const char *kernel_filename = machine->kernel_filename;
+ const char *kernel_cmdline = machine->kernel_cmdline;
+ const char *initrd_filename = machine->initrd_filename;
if (xen_domain_build_pv(kernel_filename, initrd_filename,
kernel_cmdline) < 0) {
fprintf(stderr, "xen pv domain creation failed\n");
exit(1);
}
break;
+ }
+#endif
case XEN_EMULATE:
fprintf(stderr, "xen emulation not implemented (yet)\n");
exit(1);
break;
+ default:
+ fprintf(stderr, "unhandled xen_mode %d\n", xen_mode);
+ exit(1);
+ break;
}
xen_be_register("console", &xen_console_ops);
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN v5 9/9] xen: make it possible to build without the Xen PV domain builder
@ 2015-11-09 12:01 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: Ian Campbell, qemu-devel, stefano.stabellini
Until the previous patch this relied on xc_fd(), which was only
implemented for Xen 4.0 and earlier.
Given this wasn't working since Xen 4.0 I have marked this as disabled
by default.
Removing this support drops the use of a bunch of symbols from
libxenctrl, specifically:
- xc_domain_create
- xc_domain_destroy
- xc_domain_getinfo
- xc_domain_max_vcpus
- xc_domain_setmaxmem
- xc_domain_unpause
- xc_evtchn_alloc_unbound
- xc_linux_build
This is another step towards only using Xen libraries which provide a
stable inteface.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v5: XEN_CREATE entirely wihtin CONFIG_XEN_PV_DOMAIN_BUILD ifdef.
Simplify configure'ry.
v4: Fixed all checkpatch errors.
Disabled by default.
---
configure | 15 +++++++++++++++
hw/xenpv/Makefile.objs | 4 +++-
hw/xenpv/xen_machine_pv.c | 15 +++++++++++----
3 files changed, 29 insertions(+), 5 deletions(-)
diff --git a/configure b/configure
index 0253cc9..4ab6d9c 100755
--- a/configure
+++ b/configure
@@ -247,6 +247,7 @@ vnc_jpeg=""
vnc_png=""
xen=""
xen_ctrl_version=""
+xen_pv_domain_build="no"
xen_pci_passthrough=""
linux_aio=""
cap_ng=""
@@ -917,6 +918,10 @@ for opt do
;;
--enable-xen-pci-passthrough) xen_pci_passthrough="yes"
;;
+ --disable-xen-pv-domain-build) xen_pv_domain_build="no"
+ ;;
+ --enable-xen-pv-domain-build) xen_pv_domain_build="yes"
+ ;;
--disable-brlapi) brlapi="no"
;;
--enable-brlapi) brlapi="yes"
@@ -2172,6 +2177,12 @@ if test "$xen_pci_passthrough" != "no"; then
fi
fi
+if test "$xen_pv_domain_build" = "yes" &&
+ test "$xen" != "yes"; then
+ error_exit "User requested Xen PV domain builder support" \
+ "which requires Xen support."
+fi
+
##########################################
# libtool probe
@@ -4762,6 +4773,7 @@ fi
echo "xen support $xen"
if test "$xen" = "yes" ; then
echo "xen ctrl version $xen_ctrl_version"
+ echo "pv dom build $xen_pv_domain_build"
fi
echo "brlapi support $brlapi"
echo "bluez support $bluez"
@@ -5130,6 +5142,9 @@ fi
if test "$xen" = "yes" ; then
echo "CONFIG_XEN_BACKEND=y" >> $config_host_mak
echo "CONFIG_XEN_CTRL_INTERFACE_VERSION=$xen_ctrl_version" >> $config_host_mak
+ if test "$xen_pv_domain_build" = "yes" ; then
+ echo "CONFIG_XEN_PV_DOMAIN_BUILD=y" >> $config_host_mak
+ fi
fi
if test "$linux_aio" = "yes" ; then
echo "CONFIG_LINUX_AIO=y" >> $config_host_mak
diff --git a/hw/xenpv/Makefile.objs b/hw/xenpv/Makefile.objs
index 49f6e9e..bbf5873 100644
--- a/hw/xenpv/Makefile.objs
+++ b/hw/xenpv/Makefile.objs
@@ -1,2 +1,4 @@
# Xen PV machine support
-obj-$(CONFIG_XEN) += xen_domainbuild.o xen_machine_pv.o
+obj-$(CONFIG_XEN) += xen_machine_pv.o
+# Xen PV machine builder support
+obj-$(CONFIG_XEN_PV_DOMAIN_BUILD) += xen_domainbuild.o
diff --git a/hw/xenpv/xen_machine_pv.c b/hw/xenpv/xen_machine_pv.c
index 23d6ef0..3250b94 100644
--- a/hw/xenpv/xen_machine_pv.c
+++ b/hw/xenpv/xen_machine_pv.c
@@ -30,9 +30,6 @@
static void xen_init_pv(MachineState *machine)
{
- const char *kernel_filename = machine->kernel_filename;
- const char *kernel_cmdline = machine->kernel_cmdline;
- const char *initrd_filename = machine->initrd_filename;
DriveInfo *dinfo;
int i;
@@ -46,17 +43,27 @@ static void xen_init_pv(MachineState *machine)
case XEN_ATTACH:
/* nothing to do, xend handles everything */
break;
- case XEN_CREATE:
+#ifdef CONFIG_XEN_PV_DOMAIN_BUILD
+ case XEN_CREATE: {
+ const char *kernel_filename = machine->kernel_filename;
+ const char *kernel_cmdline = machine->kernel_cmdline;
+ const char *initrd_filename = machine->initrd_filename;
if (xen_domain_build_pv(kernel_filename, initrd_filename,
kernel_cmdline) < 0) {
fprintf(stderr, "xen pv domain creation failed\n");
exit(1);
}
break;
+ }
+#endif
case XEN_EMULATE:
fprintf(stderr, "xen emulation not implemented (yet)\n");
exit(1);
break;
+ default:
+ fprintf(stderr, "unhandled xen_mode %d\n", xen_mode);
+ exit(1);
+ break;
}
xen_be_register("console", &xen_console_ops);
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries
2015-11-09 11:59 [Qemu-devel] [Minios-devel] [PATCH v4 0/<VARIOUS>] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
2015-11-09 12:01 ` Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 1/5] qemu-xen-traditional: Use xentoollog as a separate library Ian Campbell
` (5 more replies)
2015-11-09 12:01 ` [PATCH MINI-OS " Ian Campbell
3 siblings, 6 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
We intend to stabilise some parts of the libxenctrl interface by
splitting out some functionality into separate stable libraries.
This is the qemu-xen-traditional part of the first phase of that change.
This mail is (or is intended to be) a reply to a "0/<VARIOUS>"
super-intro mail covering all of the related patch series and which
contains more details.
Ian Campbell (5):
qemu-xen-traditional: Use xentoollog as a separate library
qemu-xen-traditional: Use libxenevtchn
qemu-xen-traditional: Use libxengnttab
qemu-xen-traditional: Add libxencall to rpath-link
qemu-xen-traditional: Add libxenforeignmemory to rpath-link
hw/xen_backend.c | 30 +++++++++++++++---------------
hw/xen_backend.h | 4 ++--
hw/xen_common.h | 2 ++
hw/xen_console.c | 4 ++--
hw/xen_disk.c | 24 ++++++++++++------------
i386-dm/helper2.c | 19 ++++++++++---------
xen-hooks.mak | 9 +++++++++
7 files changed, 52 insertions(+), 40 deletions(-)
--
2.1.4
^ permalink raw reply [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN-TRADITIONAL v5 1/5] qemu-xen-traditional: Use xentoollog as a separate library
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 2/5] qemu-xen-traditional: Use libxenevtchn Ian Campbell
` (4 subsequent siblings)
5 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
This has been split out of libxenctrl, so the build needs to be able
to find the header and the library. QEMU does not use xtl_* itself so
-rpath-link is sufficient to allow linking to libxenctrl.so (which
links against libxentoollog).
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
---
v3: Library moved to tools/libs/
---
xen-hooks.mak | 2 ++
1 file changed, 2 insertions(+)
diff --git a/xen-hooks.mak b/xen-hooks.mak
index bc7f1f1..bf7f027 100644
--- a/xen-hooks.mak
+++ b/xen-hooks.mak
@@ -1,3 +1,4 @@
+CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/toollog/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/libxc/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/xenstore/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/include
@@ -19,6 +20,7 @@ CFLAGS += $(CMDLINE_CFLAGS)
LIBS += -L$(XEN_ROOT)/tools/libxc -lxenctrl -lxenguest
LIBS += -L$(XEN_ROOT)/tools/xenstore -lxenstore
+LIBS += -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog
LDFLAGS := $(CFLAGS) $(LDFLAGS)
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN-TRADITIONAL v5 2/5] qemu-xen-traditional: Use libxenevtchn
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 1/5] qemu-xen-traditional: Use xentoollog as a separate library Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-11 15:19 ` Ian Jackson
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 3/5] qemu-xen-traditional: Use libxengnttab Ian Campbell
` (3 subsequent siblings)
5 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
/dev/xen/evtchn related wrappers have been moved out of libxenctrl
into their own library.
Note that i386-dm/helper2.c's xc_interface * was always really an
xc_evtchn *, it's just they used to be typedefs to the same thing...
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Library moved to tools/libs/
---
hw/xen_backend.c | 26 +++++++++++++-------------
hw/xen_backend.h | 2 +-
hw/xen_common.h | 1 +
i386-dm/helper2.c | 19 ++++++++++---------
xen-hooks.mak | 2 ++
5 files changed, 27 insertions(+), 23 deletions(-)
diff --git a/hw/xen_backend.c b/hw/xen_backend.c
index 92b3506..40f6887 100644
--- a/hw/xen_backend.c
+++ b/hw/xen_backend.c
@@ -208,19 +208,19 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev,
xendev->debug = debug;
xendev->local_port = -1;
- xendev->evtchndev = xc_evtchn_open(NULL, 0);
+ xendev->evtchndev = xenevtchn_open(NULL, 0);
if (xendev->evtchndev == NULL) {
xen_be_printf(NULL, 0, "can't open evtchn device\n");
qemu_free(xendev);
return NULL;
}
- fcntl(xc_evtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
+ fcntl(xenevtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
if (ops->flags & DEVOPS_FLAG_NEED_GNTDEV) {
xendev->gnttabdev = xc_gnttab_open(NULL, 0);
if (xendev->gnttabdev == NULL) {
xen_be_printf(NULL, 0, "can't open gnttab device\n");
- xc_evtchn_close(xendev->evtchndev);
+ xenevtchn_close(xendev->evtchndev);
qemu_free(xendev);
return NULL;
}
@@ -268,7 +268,7 @@ static struct XenDevice *xen_be_del_xendev(int dom, int dev)
}
if (xendev->evtchndev != NULL)
- xc_evtchn_close(xendev->evtchndev);
+ xenevtchn_close(xendev->evtchndev);
if (xendev->gnttabdev != NULL)
xc_gnttab_close(xendev->gnttabdev);
@@ -630,13 +630,13 @@ static void xen_be_evtchn_event(void *opaque)
struct XenDevice *xendev = opaque;
evtchn_port_t port;
- port = xc_evtchn_pending(xendev->evtchndev);
+ port = xenevtchn_pending(xendev->evtchndev);
if (port != xendev->local_port) {
- xen_be_printf(xendev, 0, "xc_evtchn_pending returned %d (expected %d)\n",
+ xen_be_printf(xendev, 0, "xenevtchn_pending returned %d (expected %d)\n",
port, xendev->local_port);
return;
}
- xc_evtchn_unmask(xendev->evtchndev, port);
+ xenevtchn_unmask(xendev->evtchndev, port);
if (xendev->ops->event)
xendev->ops->event(xendev);
@@ -683,14 +683,14 @@ int xen_be_bind_evtchn(struct XenDevice *xendev)
{
if (xendev->local_port != -1)
return 0;
- xendev->local_port = xc_evtchn_bind_interdomain
+ xendev->local_port = xenevtchn_bind_interdomain
(xendev->evtchndev, xendev->dom, xendev->remote_port);
if (xendev->local_port == -1) {
- xen_be_printf(xendev, 0, "xc_evtchn_bind_interdomain failed\n");
+ xen_be_printf(xendev, 0, "xenevtchn_bind_interdomain failed\n");
return -1;
}
xen_be_printf(xendev, 2, "bind evtchn port %d\n", xendev->local_port);
- qemu_set_fd_handler(xc_evtchn_fd(xendev->evtchndev),
+ qemu_set_fd_handler(xenevtchn_fd(xendev->evtchndev),
xen_be_evtchn_event, NULL, xendev);
return 0;
}
@@ -699,15 +699,15 @@ void xen_be_unbind_evtchn(struct XenDevice *xendev)
{
if (xendev->local_port == -1)
return;
- qemu_set_fd_handler(xc_evtchn_fd(xendev->evtchndev), NULL, NULL, NULL);
- xc_evtchn_unbind(xendev->evtchndev, xendev->local_port);
+ qemu_set_fd_handler(xenevtchn_fd(xendev->evtchndev), NULL, NULL, NULL);
+ xenevtchn_unbind(xendev->evtchndev, xendev->local_port);
xen_be_printf(xendev, 2, "unbind evtchn port %d\n", xendev->local_port);
xendev->local_port = -1;
}
int xen_be_send_notify(struct XenDevice *xendev)
{
- return xc_evtchn_notify(xendev->evtchndev, xendev->local_port);
+ return xenevtchn_notify(xendev->evtchndev, xendev->local_port);
}
/*
diff --git a/hw/xen_backend.h b/hw/xen_backend.h
index e421391..60f18f8 100644
--- a/hw/xen_backend.h
+++ b/hw/xen_backend.h
@@ -44,7 +44,7 @@ struct XenDevice {
int remote_port;
int local_port;
- xc_evtchn *evtchndev;
+ xenevtchn_handle *evtchndev;
xc_gnttab *gnttabdev;
struct XenDevOps *ops;
diff --git a/hw/xen_common.h b/hw/xen_common.h
index a615052..cee908f 100644
--- a/hw/xen_common.h
+++ b/hw/xen_common.h
@@ -4,6 +4,7 @@
#include <stddef.h>
#include <inttypes.h>
+#include <xenevtchn.h>
#include <xenctrl.h>
#include <xenstore.h>
#include <xen/io/xenbus.h>
diff --git a/i386-dm/helper2.c b/i386-dm/helper2.c
index ede3105..2706f2e 100644
--- a/i386-dm/helper2.c
+++ b/i386-dm/helper2.c
@@ -48,6 +48,7 @@
#include <limits.h>
#include <fcntl.h>
+#include <xenevtchn.h>
#include <xenctrl.h>
#include <xen/hvm/ioreq.h>
#include <xen/hvm/hvm_info_table.h>
@@ -104,7 +105,7 @@ buffered_iopage_t *buffered_io_page = NULL;
QEMUTimer *buffered_io_timer;
/* the evtchn fd for polling */
-xc_interface *xce_handle = NULL;
+xenevtchn_handle *xce_handle = NULL;
/* which vcpu we are serving */
int send_vcpu = 0;
@@ -141,7 +142,7 @@ CPUX86State *cpu_x86_init(const char *cpu_model)
cpu_single_env = env;
- xce_handle = xc_evtchn_open(NULL, 0);
+ xce_handle = xenevtchn_open(NULL, 0);
if (xce_handle == NULL) {
perror("open");
return NULL;
@@ -149,7 +150,7 @@ CPUX86State *cpu_x86_init(const char *cpu_model)
/* FIXME: how about if we overflow the page here? */
for (i = 0; i < vcpus; i++) {
- rc = xc_evtchn_bind_interdomain(
+ rc = xenevtchn_bind_interdomain(
xce_handle, domid, shared_page->vcpu_ioreq[i].vp_eport);
if (rc == -1) {
fprintf(logfile, "bind interdomain ioctl error %d\n", errno);
@@ -164,7 +165,7 @@ CPUX86State *cpu_x86_init(const char *cpu_model)
errno);
return NULL;
}
- rc = xc_evtchn_bind_interdomain(xce_handle, domid, (uint32_t)bufioreq_evtchn);
+ rc = xenevtchn_bind_interdomain(xce_handle, domid, (uint32_t)bufioreq_evtchn);
if (rc == -1) {
fprintf(logfile, "bind interdomain ioctl error %d\n", errno);
return NULL;
@@ -278,7 +279,7 @@ static ioreq_t *cpu_get_ioreq(void)
int i;
evtchn_port_t port;
- port = xc_evtchn_pending(xce_handle);
+ port = xenevtchn_pending(xce_handle);
if (port == bufioreq_local_port) {
qemu_mod_timer(buffered_io_timer,
BUFFER_IO_MAX_DELAY + qemu_get_clock(rt_clock));
@@ -296,7 +297,7 @@ static ioreq_t *cpu_get_ioreq(void)
}
// unmask the wanted port again
- xc_evtchn_unmask(xce_handle, port);
+ xenevtchn_unmask(xce_handle, port);
//get the io packet from shared memory
send_vcpu = i;
@@ -539,7 +540,7 @@ static void handle_buffered_io(void *opaque)
BUFFER_IO_MAX_DELAY + qemu_get_clock(rt_clock));
} else {
qemu_del_timer(buffered_io_timer);
- xc_evtchn_unmask(xce_handle, bufioreq_local_port);
+ xenevtchn_unmask(xce_handle, bufioreq_local_port);
}
}
@@ -583,7 +584,7 @@ static void cpu_handle_ioreq(void *opaque)
}
req->state = STATE_IORESP_READY;
- xc_evtchn_notify(xce_handle, ioreq_local_port[send_vcpu]);
+ xenevtchn_notify(xce_handle, ioreq_local_port[send_vcpu]);
}
}
@@ -592,7 +593,7 @@ int xen_pause_requested;
int main_loop(void)
{
CPUState *env = cpu_single_env;
- int evtchn_fd = xce_handle == NULL ? -1 : xc_evtchn_fd(xce_handle);
+ int evtchn_fd = xce_handle == NULL ? -1 : xenevtchn_fd(xce_handle);
char *qemu_file;
fd_set fds;
diff --git a/xen-hooks.mak b/xen-hooks.mak
index bf7f027..18259d4 100644
--- a/xen-hooks.mak
+++ b/xen-hooks.mak
@@ -1,4 +1,5 @@
CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/toollog/include
+CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/evtchn/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/libxc/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/xenstore/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/include
@@ -18,6 +19,7 @@ endif
CFLAGS += $(CMDLINE_CFLAGS)
+LIBS += -L$(XEN_ROOT)/tools/libs/evtchn -lxenevtchn
LIBS += -L$(XEN_ROOT)/tools/libxc -lxenctrl -lxenguest
LIBS += -L$(XEN_ROOT)/tools/xenstore -lxenstore
LIBS += -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN-TRADITIONAL v5 3/5] qemu-xen-traditional: Use libxengnttab
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 1/5] qemu-xen-traditional: Use xentoollog as a separate library Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 2/5] qemu-xen-traditional: Use libxenevtchn Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-11 15:20 ` Ian Jackson
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 4/5] qemu-xen-traditional: Add libxencall to rpath-link Ian Campbell
` (2 subsequent siblings)
5 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
/dev/xen/gntdev related wrappers have been moved out of libxenctrl
into their own library.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Library moved to tools/libs/
---
hw/xen_backend.c | 4 ++--
hw/xen_backend.h | 2 +-
hw/xen_common.h | 1 +
hw/xen_console.c | 4 ++--
hw/xen_disk.c | 24 ++++++++++++------------
xen-hooks.mak | 2 ++
6 files changed, 20 insertions(+), 17 deletions(-)
diff --git a/hw/xen_backend.c b/hw/xen_backend.c
index 40f6887..97d25da 100644
--- a/hw/xen_backend.c
+++ b/hw/xen_backend.c
@@ -217,7 +217,7 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev,
fcntl(xenevtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
if (ops->flags & DEVOPS_FLAG_NEED_GNTDEV) {
- xendev->gnttabdev = xc_gnttab_open(NULL, 0);
+ xendev->gnttabdev = xengnttab_open(NULL, 0);
if (xendev->gnttabdev == NULL) {
xen_be_printf(NULL, 0, "can't open gnttab device\n");
xenevtchn_close(xendev->evtchndev);
@@ -270,7 +270,7 @@ static struct XenDevice *xen_be_del_xendev(int dom, int dev)
if (xendev->evtchndev != NULL)
xenevtchn_close(xendev->evtchndev);
if (xendev->gnttabdev != NULL)
- xc_gnttab_close(xendev->gnttabdev);
+ xengnttab_close(xendev->gnttabdev);
TAILQ_REMOVE(&xendevs, xendev, next);
qemu_free(xendev);
diff --git a/hw/xen_backend.h b/hw/xen_backend.h
index 60f18f8..ba1e12f 100644
--- a/hw/xen_backend.h
+++ b/hw/xen_backend.h
@@ -45,7 +45,7 @@ struct XenDevice {
int local_port;
xenevtchn_handle *evtchndev;
- xc_gnttab *gnttabdev;
+ xengnttab_handle *gnttabdev;
struct XenDevOps *ops;
TAILQ_ENTRY(XenDevice) next;
diff --git a/hw/xen_common.h b/hw/xen_common.h
index cee908f..cc48892 100644
--- a/hw/xen_common.h
+++ b/hw/xen_common.h
@@ -5,6 +5,7 @@
#include <inttypes.h>
#include <xenevtchn.h>
+#include <xengnttab.h>
#include <xenctrl.h>
#include <xenstore.h>
#include <xen/io/xenbus.h>
diff --git a/hw/xen_console.c b/hw/xen_console.c
index 80beb31..76c4e93 100644
--- a/hw/xen_console.c
+++ b/hw/xen_console.c
@@ -230,7 +230,7 @@ static int con_initialise(struct XenDevice *xendev)
PROT_READ|PROT_WRITE,
con->ring_ref);
else
- con->sring = xc_gnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
+ con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
con->ring_ref,
PROT_READ|PROT_WRITE);
if (!con->sring)
@@ -263,7 +263,7 @@ static void con_disconnect(struct XenDevice *xendev)
if (!xendev->gnttabdev)
munmap(con->sring, XC_PAGE_SIZE);
else
- xc_gnttab_munmap(xendev->gnttabdev, con->sring, 1);
+ xengnttab_munmap(xendev->gnttabdev, con->sring, 1);
con->sring = NULL;
}
}
diff --git a/hw/xen_disk.c b/hw/xen_disk.c
index 250d806..6ccf073 100644
--- a/hw/xen_disk.c
+++ b/hw/xen_disk.c
@@ -266,7 +266,7 @@ err:
static void ioreq_unmap(struct ioreq *ioreq)
{
- xc_gnttab *gnt = ioreq->blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
int i;
if (ioreq->v.niov == 0)
@@ -274,8 +274,8 @@ static void ioreq_unmap(struct ioreq *ioreq)
if (batch_maps) {
if (!ioreq->pages)
return;
- if (xc_gnttab_munmap(gnt, ioreq->pages, ioreq->v.niov) != 0)
- xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
+ if (xengnttab_munmap(gnt, ioreq->pages, ioreq->v.niov) != 0)
+ xen_be_printf(&ioreq->blkdev->xendev, 0, "xengnttab_munmap failed: %s\n",
strerror(errno));
ioreq->blkdev->cnt_map -= ioreq->v.niov;
ioreq->pages = NULL;
@@ -283,8 +283,8 @@ static void ioreq_unmap(struct ioreq *ioreq)
for (i = 0; i < ioreq->v.niov; i++) {
if (!ioreq->page[i])
continue;
- if (xc_gnttab_munmap(gnt, ioreq->page[i], 1) != 0)
- xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
+ if (xengnttab_munmap(gnt, ioreq->page[i], 1) != 0)
+ xen_be_printf(&ioreq->blkdev->xendev, 0, "xengnttab_munmap failed: %s\n",
strerror(errno));
ioreq->blkdev->cnt_map--;
ioreq->page[i] = NULL;
@@ -294,13 +294,13 @@ static void ioreq_unmap(struct ioreq *ioreq)
static int ioreq_map(struct ioreq *ioreq)
{
- xc_gnttab *gnt = ioreq->blkdev->xendev.gnttabdev;
+ xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
int i;
if (ioreq->v.niov == 0)
return 0;
if (batch_maps) {
- ioreq->pages = xc_gnttab_map_grant_refs
+ ioreq->pages = xengnttab_map_grant_refs
(gnt, ioreq->v.niov, ioreq->domids, ioreq->refs, ioreq->prot);
if (ioreq->pages == NULL) {
xen_be_printf(&ioreq->blkdev->xendev, 0,
@@ -314,7 +314,7 @@ static int ioreq_map(struct ioreq *ioreq)
ioreq->blkdev->cnt_map += ioreq->v.niov;
} else {
for (i = 0; i < ioreq->v.niov; i++) {
- ioreq->page[i] = xc_gnttab_map_grant_ref
+ ioreq->page[i] = xengnttab_map_grant_ref
(gnt, ioreq->domids[i], ioreq->refs[i], ioreq->prot);
if (ioreq->page[i] == NULL) {
xen_be_printf(&ioreq->blkdev->xendev, 0,
@@ -611,9 +611,9 @@ static void blk_alloc(struct XenDevice *xendev)
blkdev->bh = qemu_bh_new(blk_bh, blkdev);
if (xen_mode != XEN_EMULATE)
batch_maps = 1;
- if (xc_gnttab_set_max_grants(xendev->gnttabdev,
+ if (xengnttab_set_max_grants(xendev->gnttabdev,
MAX_GRANTS(max_requests, BLKIF_MAX_SEGMENTS_PER_REQUEST)) < 0) {
- xen_be_printf(xendev, 0, "xc_gnttab_set_max_grants failed: %s\n",
+ xen_be_printf(xendev, 0, "xengnttab_set_max_grants failed: %s\n",
strerror(errno));
}
}
@@ -734,7 +734,7 @@ static int blk_connect(struct XenDevice *xendev)
blkdev->protocol = BLKIF_PROTOCOL_X86_64;
}
- blkdev->sring = xc_gnttab_map_grant_ref(blkdev->xendev.gnttabdev,
+ blkdev->sring = xengnttab_map_grant_ref(blkdev->xendev.gnttabdev,
blkdev->xendev.dom,
blkdev->ring_ref,
PROT_READ | PROT_WRITE);
@@ -787,7 +787,7 @@ static void blk_disconnect(struct XenDevice *xendev)
xen_be_unbind_evtchn(&blkdev->xendev);
if (blkdev->sring) {
- xc_gnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
+ xengnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
blkdev->cnt_map--;
blkdev->sring = NULL;
}
diff --git a/xen-hooks.mak b/xen-hooks.mak
index 18259d4..179a6b7 100644
--- a/xen-hooks.mak
+++ b/xen-hooks.mak
@@ -1,5 +1,6 @@
CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/toollog/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/evtchn/include
+CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/gnttab/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/libxc/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/xenstore/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/include
@@ -20,6 +21,7 @@ endif
CFLAGS += $(CMDLINE_CFLAGS)
LIBS += -L$(XEN_ROOT)/tools/libs/evtchn -lxenevtchn
+LIBS += -L$(XEN_ROOT)/tools/libs/gnttab -lxengnttab
LIBS += -L$(XEN_ROOT)/tools/libxc -lxenctrl -lxenguest
LIBS += -L$(XEN_ROOT)/tools/xenstore -lxenstore
LIBS += -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN-TRADITIONAL v5 4/5] qemu-xen-traditional: Add libxencall to rpath-link
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
` (2 preceding siblings ...)
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 3/5] qemu-xen-traditional: Use libxengnttab Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-11 15:20 ` Ian Jackson
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 5/5] qemu-xen-traditional: Add libxenforeignmemory " Ian Campbell
2015-11-11 15:23 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Jackson
5 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
libxenctrl links against this library
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Library moved to tools/libs
---
xen-hooks.mak | 1 +
1 file changed, 1 insertion(+)
diff --git a/xen-hooks.mak b/xen-hooks.mak
index 179a6b7..229d642 100644
--- a/xen-hooks.mak
+++ b/xen-hooks.mak
@@ -25,6 +25,7 @@ LIBS += -L$(XEN_ROOT)/tools/libs/gnttab -lxengnttab
LIBS += -L$(XEN_ROOT)/tools/libxc -lxenctrl -lxenguest
LIBS += -L$(XEN_ROOT)/tools/xenstore -lxenstore
LIBS += -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog
+LIBS += -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/call
LDFLAGS := $(CFLAGS) $(LDFLAGS)
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH QEMU-XEN-TRADITIONAL v5 5/5] qemu-xen-traditional: Add libxenforeignmemory to rpath-link
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
` (3 preceding siblings ...)
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 4/5] qemu-xen-traditional: Add libxencall to rpath-link Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-11 15:21 ` Ian Jackson
2015-11-11 15:23 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Jackson
5 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel; +Cc: Ian Campbell
libxenctrl links against this library.
Also, request the compat xc_map_foreign API from libxc.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Library moved to tools/libs/
---
xen-hooks.mak | 2 ++
1 file changed, 2 insertions(+)
diff --git a/xen-hooks.mak b/xen-hooks.mak
index 229d642..c1ea4be 100644
--- a/xen-hooks.mak
+++ b/xen-hooks.mak
@@ -1,6 +1,7 @@
CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/toollog/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/evtchn/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/libs/gnttab/include
+CPPFLAGS+= -DXC_WANT_COMPAT_MAP_FOREIGN_API
CPPFLAGS+= -I$(XEN_ROOT)/tools/libxc/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/xenstore/include
CPPFLAGS+= -I$(XEN_ROOT)/tools/include
@@ -26,6 +27,7 @@ LIBS += -L$(XEN_ROOT)/tools/libxc -lxenctrl -lxenguest
LIBS += -L$(XEN_ROOT)/tools/xenstore -lxenstore
LIBS += -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/toollog
LIBS += -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/call
+LIBS += -Wl,-rpath-link=$(XEN_ROOT)/tools/libs/foreignmemory
LDFLAGS := $(CFLAGS) $(LDFLAGS)
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH MINI-OS v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries
2015-11-09 11:59 [Qemu-devel] [Minios-devel] [PATCH v4 0/<VARIOUS>] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
` (2 preceding siblings ...)
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 1/5] mini-os: Include libxentoollog with libxc Ian Campbell
` (4 more replies)
3 siblings, 5 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: minios-devel, samuel.thibault, Ian Campbell
We intend to stabilise some parts of the libxenctrl interface by
splitting out some functionality into separate stable libraries.
This is the mini-os part of the first phase of that change.
This mail is (or is intended to be) a reply to a "0/<VARIOUS>"
super-intro mail covering all of the related patch series and which
contains more details.
Ian Campbell (5):
mini-os: Include libxentoollog with libxc
mini-os: Include libxenevtchn with libxc
mini-os: Include libxengnttab with libxc
mini-os: Include libxencall with libxc
mini-os: Include libxenforeignmemory with libxc
Makefile | 5 +++++
1 file changed, 5 insertions(+)
--
2.1.4
^ permalink raw reply [flat|nested] 111+ messages in thread
* [PATCH MINI-OS v5 1/5] mini-os: Include libxentoollog with libxc
2015-11-09 12:01 ` [PATCH MINI-OS " Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 2/5] mini-os: Include libxenevtchn " Ian Campbell
` (3 subsequent siblings)
4 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: minios-devel, samuel.thibault, Ian Campbell
libxentoollog has just been split out from libxc. From mini-os's point
of view we don't care about the distinction, so keep things simple by
just including libxentoollog if libxc is enabled.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
v2: Adjust for libs/$lib layout.
---
Makefile | 1 +
1 file changed, 1 insertion(+)
diff --git a/Makefile b/Makefile
index 2cb5e51..daee46c 100644
--- a/Makefile
+++ b/Makefile
@@ -165,6 +165,7 @@ OBJS := $(filter-out $(OBJ_DIR)/lwip%.o $(LWO), $(OBJS))
ifeq ($(libc),y)
ifeq ($(CONFIG_XC),y)
+APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/toollog -whole-archive -lxentoollog -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libxc-$(MINIOS_TARGET_ARCH) -whole-archive -lxenguest -lxenctrl -no-whole-archive
endif
APP_LDLIBS += -lpci
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH MINI-OS v5 2/5] mini-os: Include libxenevtchn with libxc
2015-11-09 12:01 ` [PATCH MINI-OS " Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 1/5] mini-os: Include libxentoollog with libxc Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 3/5] mini-os: Include libxengnttab " Ian Campbell
` (2 subsequent siblings)
4 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: minios-devel, samuel.thibault, Ian Campbell
libxenevtchn has just been split out from libxc. From mini-os's point
of view we don't care about the distinction, so keep things simple by
just including libxenevtchn if libxc is enabled.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
v2: Adjust for libs/$lib layout.
---
Makefile | 1 +
1 file changed, 1 insertion(+)
diff --git a/Makefile b/Makefile
index daee46c..d1d8dc4 100644
--- a/Makefile
+++ b/Makefile
@@ -166,6 +166,7 @@ OBJS := $(filter-out $(OBJ_DIR)/lwip%.o $(LWO), $(OBJS))
ifeq ($(libc),y)
ifeq ($(CONFIG_XC),y)
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/toollog -whole-archive -lxentoollog -no-whole-archive
+APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/evtchn -whole-archive -lxenevtchn -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libxc-$(MINIOS_TARGET_ARCH) -whole-archive -lxenguest -lxenctrl -no-whole-archive
endif
APP_LDLIBS += -lpci
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH MINI-OS v5 3/5] mini-os: Include libxengnttab with libxc
2015-11-09 12:01 ` [PATCH MINI-OS " Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 1/5] mini-os: Include libxentoollog with libxc Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 2/5] mini-os: Include libxenevtchn " Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 4/5] mini-os: Include libxencall " Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 5/5] mini-os: Include libxenforeignmemory " Ian Campbell
4 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: minios-devel, samuel.thibault, Ian Campbell
libxengnttab has just been split out from libxc. From mini-os's point
of view we don't care about the distinction, so keep things simple by
just including libxengnttab if libxc is enabled.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
v2: Adjust for libs/$lib layout.
---
Makefile | 1 +
1 file changed, 1 insertion(+)
diff --git a/Makefile b/Makefile
index d1d8dc4..521f647 100644
--- a/Makefile
+++ b/Makefile
@@ -167,6 +167,7 @@ ifeq ($(libc),y)
ifeq ($(CONFIG_XC),y)
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/toollog -whole-archive -lxentoollog -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/evtchn -whole-archive -lxenevtchn -no-whole-archive
+APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/gnttab -whole-archive -lxengnttab -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libxc-$(MINIOS_TARGET_ARCH) -whole-archive -lxenguest -lxenctrl -no-whole-archive
endif
APP_LDLIBS += -lpci
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH MINI-OS v5 4/5] mini-os: Include libxencall with libxc
2015-11-09 12:01 ` [PATCH MINI-OS " Ian Campbell
` (2 preceding siblings ...)
2015-11-09 12:01 ` [PATCH MINI-OS v5 3/5] mini-os: Include libxengnttab " Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 5/5] mini-os: Include libxenforeignmemory " Ian Campbell
4 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: minios-devel, samuel.thibault, Ian Campbell
libxencall has just been split out from libxc. From mini-os's point
of view we don't care about the distinction, so keep things simple by
just including libxencall if libxc is enabled.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
v2: Adjust for libs/$lib layout.
---
Makefile | 1 +
1 file changed, 1 insertion(+)
diff --git a/Makefile b/Makefile
index 521f647..c900540 100644
--- a/Makefile
+++ b/Makefile
@@ -168,6 +168,7 @@ ifeq ($(CONFIG_XC),y)
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/toollog -whole-archive -lxentoollog -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/evtchn -whole-archive -lxenevtchn -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/gnttab -whole-archive -lxengnttab -no-whole-archive
+APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/call -whole-archive -lxencall -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libxc-$(MINIOS_TARGET_ARCH) -whole-archive -lxenguest -lxenctrl -no-whole-archive
endif
APP_LDLIBS += -lpci
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* [PATCH MINI-OS v5 5/5] mini-os: Include libxenforeignmemory with libxc
2015-11-09 12:01 ` [PATCH MINI-OS " Ian Campbell
` (3 preceding siblings ...)
2015-11-09 12:01 ` [PATCH MINI-OS v5 4/5] mini-os: Include libxencall " Ian Campbell
@ 2015-11-09 12:01 ` Ian Campbell
4 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-09 12:01 UTC (permalink / raw)
To: ian.jackson, wei.liu2, xen-devel
Cc: minios-devel, samuel.thibault, Ian Campbell
libxenforeignmemory has just been split out from libxc. From mini-os's
point of view we don't care about the distinction, so keep things
simple by just including libxenforeignmemory if libxc is enabled.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Acked-by: Wei Liu <wei.liu2@citrix.com>
---
v2: Adjust for libs/$lib layout.
---
Makefile | 1 +
1 file changed, 1 insertion(+)
diff --git a/Makefile b/Makefile
index c900540..cfe015a 100644
--- a/Makefile
+++ b/Makefile
@@ -169,6 +169,7 @@ APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/toollog -whole-ar
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/evtchn -whole-archive -lxenevtchn -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/gnttab -whole-archive -lxengnttab -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/call -whole-archive -lxencall -no-whole-archive
+APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libs-$(MINIOS_TARGET_ARCH)/foreignmemory -whole-archive -lxenforeignmemory -no-whole-archive
APP_LDLIBS += -L$(XEN_ROOT)/stubdom/libxc-$(MINIOS_TARGET_ARCH) -whole-archive -lxenguest -lxenctrl -no-whole-archive
endif
APP_LDLIBS += -lpci
--
2.1.4
^ permalink raw reply related [flat|nested] 111+ messages in thread
* Re: [Qemu-devel] [PATCH QEMU-XEN v5 9/9] xen: make it possible to build without the Xen PV domain builder
2015-11-09 12:01 ` Ian Campbell
(?)
(?)
@ 2015-11-10 12:48 ` Stefano Stabellini
-1 siblings, 0 replies; 111+ messages in thread
From: Stefano Stabellini @ 2015-11-10 12:48 UTC (permalink / raw)
To: Ian Campbell
Cc: wei.liu2, stefano.stabellini, ian.jackson, qemu-devel, xen-devel
On Mon, 9 Nov 2015, Ian Campbell wrote:
> Until the previous patch this relied on xc_fd(), which was only
> implemented for Xen 4.0 and earlier.
>
> Given this wasn't working since Xen 4.0 I have marked this as disabled
> by default.
>
> Removing this support drops the use of a bunch of symbols from
> libxenctrl, specifically:
>
> - xc_domain_create
> - xc_domain_destroy
> - xc_domain_getinfo
> - xc_domain_max_vcpus
> - xc_domain_setmaxmem
> - xc_domain_unpause
> - xc_evtchn_alloc_unbound
> - xc_linux_build
>
> This is another step towards only using Xen libraries which provide a
> stable inteface.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> v5: XEN_CREATE entirely wihtin CONFIG_XEN_PV_DOMAIN_BUILD ifdef.
> Simplify configure'ry.
>
> v4: Fixed all checkpatch errors.
> Disabled by default.
> ---
> configure | 15 +++++++++++++++
> hw/xenpv/Makefile.objs | 4 +++-
> hw/xenpv/xen_machine_pv.c | 15 +++++++++++----
> 3 files changed, 29 insertions(+), 5 deletions(-)
>
> diff --git a/configure b/configure
> index 0253cc9..4ab6d9c 100755
> --- a/configure
> +++ b/configure
> @@ -247,6 +247,7 @@ vnc_jpeg=""
> vnc_png=""
> xen=""
> xen_ctrl_version=""
> +xen_pv_domain_build="no"
> xen_pci_passthrough=""
> linux_aio=""
> cap_ng=""
> @@ -917,6 +918,10 @@ for opt do
> ;;
> --enable-xen-pci-passthrough) xen_pci_passthrough="yes"
> ;;
> + --disable-xen-pv-domain-build) xen_pv_domain_build="no"
> + ;;
> + --enable-xen-pv-domain-build) xen_pv_domain_build="yes"
> + ;;
> --disable-brlapi) brlapi="no"
> ;;
> --enable-brlapi) brlapi="yes"
> @@ -2172,6 +2177,12 @@ if test "$xen_pci_passthrough" != "no"; then
> fi
> fi
>
> +if test "$xen_pv_domain_build" = "yes" &&
> + test "$xen" != "yes"; then
> + error_exit "User requested Xen PV domain builder support" \
> + "which requires Xen support."
> +fi
> +
> ##########################################
> # libtool probe
>
> @@ -4762,6 +4773,7 @@ fi
> echo "xen support $xen"
> if test "$xen" = "yes" ; then
> echo "xen ctrl version $xen_ctrl_version"
> + echo "pv dom build $xen_pv_domain_build"
> fi
> echo "brlapi support $brlapi"
> echo "bluez support $bluez"
> @@ -5130,6 +5142,9 @@ fi
> if test "$xen" = "yes" ; then
> echo "CONFIG_XEN_BACKEND=y" >> $config_host_mak
> echo "CONFIG_XEN_CTRL_INTERFACE_VERSION=$xen_ctrl_version" >> $config_host_mak
> + if test "$xen_pv_domain_build" = "yes" ; then
> + echo "CONFIG_XEN_PV_DOMAIN_BUILD=y" >> $config_host_mak
> + fi
> fi
> if test "$linux_aio" = "yes" ; then
> echo "CONFIG_LINUX_AIO=y" >> $config_host_mak
> diff --git a/hw/xenpv/Makefile.objs b/hw/xenpv/Makefile.objs
> index 49f6e9e..bbf5873 100644
> --- a/hw/xenpv/Makefile.objs
> +++ b/hw/xenpv/Makefile.objs
> @@ -1,2 +1,4 @@
> # Xen PV machine support
> -obj-$(CONFIG_XEN) += xen_domainbuild.o xen_machine_pv.o
> +obj-$(CONFIG_XEN) += xen_machine_pv.o
> +# Xen PV machine builder support
> +obj-$(CONFIG_XEN_PV_DOMAIN_BUILD) += xen_domainbuild.o
> diff --git a/hw/xenpv/xen_machine_pv.c b/hw/xenpv/xen_machine_pv.c
> index 23d6ef0..3250b94 100644
> --- a/hw/xenpv/xen_machine_pv.c
> +++ b/hw/xenpv/xen_machine_pv.c
> @@ -30,9 +30,6 @@
>
> static void xen_init_pv(MachineState *machine)
> {
> - const char *kernel_filename = machine->kernel_filename;
> - const char *kernel_cmdline = machine->kernel_cmdline;
> - const char *initrd_filename = machine->initrd_filename;
> DriveInfo *dinfo;
> int i;
>
> @@ -46,17 +43,27 @@ static void xen_init_pv(MachineState *machine)
> case XEN_ATTACH:
> /* nothing to do, xend handles everything */
> break;
> - case XEN_CREATE:
> +#ifdef CONFIG_XEN_PV_DOMAIN_BUILD
> + case XEN_CREATE: {
> + const char *kernel_filename = machine->kernel_filename;
> + const char *kernel_cmdline = machine->kernel_cmdline;
> + const char *initrd_filename = machine->initrd_filename;
> if (xen_domain_build_pv(kernel_filename, initrd_filename,
> kernel_cmdline) < 0) {
> fprintf(stderr, "xen pv domain creation failed\n");
> exit(1);
> }
> break;
> + }
> +#endif
> case XEN_EMULATE:
> fprintf(stderr, "xen emulation not implemented (yet)\n");
> exit(1);
> break;
> + default:
> + fprintf(stderr, "unhandled xen_mode %d\n", xen_mode);
> + exit(1);
> + break;
> }
>
> xen_be_register("console", &xen_console_ops);
> --
> 2.1.4
>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH QEMU-XEN v5 9/9] xen: make it possible to build without the Xen PV domain builder
2015-11-09 12:01 ` Ian Campbell
(?)
@ 2015-11-10 12:48 ` Stefano Stabellini
-1 siblings, 0 replies; 111+ messages in thread
From: Stefano Stabellini @ 2015-11-10 12:48 UTC (permalink / raw)
To: Ian Campbell
Cc: wei.liu2, stefano.stabellini, ian.jackson, qemu-devel, xen-devel
On Mon, 9 Nov 2015, Ian Campbell wrote:
> Until the previous patch this relied on xc_fd(), which was only
> implemented for Xen 4.0 and earlier.
>
> Given this wasn't working since Xen 4.0 I have marked this as disabled
> by default.
>
> Removing this support drops the use of a bunch of symbols from
> libxenctrl, specifically:
>
> - xc_domain_create
> - xc_domain_destroy
> - xc_domain_getinfo
> - xc_domain_max_vcpus
> - xc_domain_setmaxmem
> - xc_domain_unpause
> - xc_evtchn_alloc_unbound
> - xc_linux_build
>
> This is another step towards only using Xen libraries which provide a
> stable inteface.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> v5: XEN_CREATE entirely wihtin CONFIG_XEN_PV_DOMAIN_BUILD ifdef.
> Simplify configure'ry.
>
> v4: Fixed all checkpatch errors.
> Disabled by default.
> ---
> configure | 15 +++++++++++++++
> hw/xenpv/Makefile.objs | 4 +++-
> hw/xenpv/xen_machine_pv.c | 15 +++++++++++----
> 3 files changed, 29 insertions(+), 5 deletions(-)
>
> diff --git a/configure b/configure
> index 0253cc9..4ab6d9c 100755
> --- a/configure
> +++ b/configure
> @@ -247,6 +247,7 @@ vnc_jpeg=""
> vnc_png=""
> xen=""
> xen_ctrl_version=""
> +xen_pv_domain_build="no"
> xen_pci_passthrough=""
> linux_aio=""
> cap_ng=""
> @@ -917,6 +918,10 @@ for opt do
> ;;
> --enable-xen-pci-passthrough) xen_pci_passthrough="yes"
> ;;
> + --disable-xen-pv-domain-build) xen_pv_domain_build="no"
> + ;;
> + --enable-xen-pv-domain-build) xen_pv_domain_build="yes"
> + ;;
> --disable-brlapi) brlapi="no"
> ;;
> --enable-brlapi) brlapi="yes"
> @@ -2172,6 +2177,12 @@ if test "$xen_pci_passthrough" != "no"; then
> fi
> fi
>
> +if test "$xen_pv_domain_build" = "yes" &&
> + test "$xen" != "yes"; then
> + error_exit "User requested Xen PV domain builder support" \
> + "which requires Xen support."
> +fi
> +
> ##########################################
> # libtool probe
>
> @@ -4762,6 +4773,7 @@ fi
> echo "xen support $xen"
> if test "$xen" = "yes" ; then
> echo "xen ctrl version $xen_ctrl_version"
> + echo "pv dom build $xen_pv_domain_build"
> fi
> echo "brlapi support $brlapi"
> echo "bluez support $bluez"
> @@ -5130,6 +5142,9 @@ fi
> if test "$xen" = "yes" ; then
> echo "CONFIG_XEN_BACKEND=y" >> $config_host_mak
> echo "CONFIG_XEN_CTRL_INTERFACE_VERSION=$xen_ctrl_version" >> $config_host_mak
> + if test "$xen_pv_domain_build" = "yes" ; then
> + echo "CONFIG_XEN_PV_DOMAIN_BUILD=y" >> $config_host_mak
> + fi
> fi
> if test "$linux_aio" = "yes" ; then
> echo "CONFIG_LINUX_AIO=y" >> $config_host_mak
> diff --git a/hw/xenpv/Makefile.objs b/hw/xenpv/Makefile.objs
> index 49f6e9e..bbf5873 100644
> --- a/hw/xenpv/Makefile.objs
> +++ b/hw/xenpv/Makefile.objs
> @@ -1,2 +1,4 @@
> # Xen PV machine support
> -obj-$(CONFIG_XEN) += xen_domainbuild.o xen_machine_pv.o
> +obj-$(CONFIG_XEN) += xen_machine_pv.o
> +# Xen PV machine builder support
> +obj-$(CONFIG_XEN_PV_DOMAIN_BUILD) += xen_domainbuild.o
> diff --git a/hw/xenpv/xen_machine_pv.c b/hw/xenpv/xen_machine_pv.c
> index 23d6ef0..3250b94 100644
> --- a/hw/xenpv/xen_machine_pv.c
> +++ b/hw/xenpv/xen_machine_pv.c
> @@ -30,9 +30,6 @@
>
> static void xen_init_pv(MachineState *machine)
> {
> - const char *kernel_filename = machine->kernel_filename;
> - const char *kernel_cmdline = machine->kernel_cmdline;
> - const char *initrd_filename = machine->initrd_filename;
> DriveInfo *dinfo;
> int i;
>
> @@ -46,17 +43,27 @@ static void xen_init_pv(MachineState *machine)
> case XEN_ATTACH:
> /* nothing to do, xend handles everything */
> break;
> - case XEN_CREATE:
> +#ifdef CONFIG_XEN_PV_DOMAIN_BUILD
> + case XEN_CREATE: {
> + const char *kernel_filename = machine->kernel_filename;
> + const char *kernel_cmdline = machine->kernel_cmdline;
> + const char *initrd_filename = machine->initrd_filename;
> if (xen_domain_build_pv(kernel_filename, initrd_filename,
> kernel_cmdline) < 0) {
> fprintf(stderr, "xen pv domain creation failed\n");
> exit(1);
> }
> break;
> + }
> +#endif
> case XEN_EMULATE:
> fprintf(stderr, "xen emulation not implemented (yet)\n");
> exit(1);
> break;
> + default:
> + fprintf(stderr, "unhandled xen_mode %d\n", xen_mode);
> + exit(1);
> + break;
> }
>
> xen_be_register("console", &xen_console_ops);
> --
> 2.1.4
>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [Qemu-devel] [PATCH QEMU-XEN v5 7/9] xen: Use stable library interfaces when they are available.
2015-11-09 12:01 ` [Qemu-devel] " Ian Campbell
@ 2015-11-10 14:00 ` Stefano Stabellini
2015-11-10 14:00 ` Stefano Stabellini
1 sibling, 0 replies; 111+ messages in thread
From: Stefano Stabellini @ 2015-11-10 14:00 UTC (permalink / raw)
To: Ian Campbell
Cc: wei.liu2, stefano.stabellini, ian.jackson, qemu-devel, xen-devel
On Mon, 9 Nov 2015, Ian Campbell wrote:
> In Xen 4.7 we are refactoring parts libxenctrl into a number of
> separate libraries which will provide backward and forward API and ABI
> compatiblity.
>
> Specifically libxenevtchn, libxengnttab and libxenforeignmemory.
>
> Previous patches have already laid the groundwork for using these by
> switching the existing compatibility shims to reflect the intefaces to
> these libraries.
>
> So all which remains is to update configure to detect the libraries
> and enable their use. Although they are notionally independent we take
> an all or nothing approach to the three libraries since they were
> added at the same time.
>
> The only non-obvious bit is that we now open a proper xenforeignmemory
> handle for xen_fmem instead of reusing the xen_xc handle.
>
> Build tested with 4.0 .. 4.6 (inclusive) and the patches targetting
> 4.7 which adds these libraries.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Two formatting issues below:
> v5: Remove ifdef check when undeffing the compat stuff.
> s/now way/no way/
>
> v4: xenforeignmemory_open is now a compat wrapper, so no ifdef.
>
> Simplify configury by asserting that interface version 470 will
> always have the libraries (lack of them makes it 460).
>
> Ran checkpatch and fixed everything except:
>
> ERROR: need consistent spacing around '*' (ctx:WxV)
> +typedef xc_interface *XenXC;
>
> Which I think is a false +ve.
>
> Simpler configure stuff
> ---
> configure | 55 +++++++++++++++++++++++++++++++++++++++++++++
> include/hw/xen/xen_common.h | 36 +++++++++++++++++++++++++++--
> 2 files changed, 89 insertions(+), 2 deletions(-)
>
> diff --git a/configure b/configure
> index b4a3488..0253cc9 100755
> --- a/configure
> +++ b/configure
> @@ -1898,6 +1898,7 @@ fi
>
> if test "$xen" != "no" ; then
> xen_libs="-lxenstore -lxenctrl -lxenguest"
> + xen_stable_libs="-lxenforeignmemory -lxengnttab -lxenevtchn"
>
> # First we test whether Xen headers and libraries are available.
> # If no, we are done and there is no Xen support.
> @@ -1920,6 +1921,57 @@ EOF
> # Xen unstable
> elif
> cat > $TMPC <<EOF &&
> +/*
> + * If we have stable libs the we don't want the libxc compat
> + * layers, regardless of what CFLAGS we may have been given.
> + */
> +#undef XC_WANT_COMPAT_EVTCHN_API
> +#undef XC_WANT_COMPAT_GNTTAB_API
> +#undef XC_WANT_COMPAT_MAP_FOREIGN_API
> +#include <xenctrl.h>
> +#include <xenstore.h>
> +#include <xenevtchn.h>
> +#include <xengnttab.h>
> +#include <xenforeignmemory.h>
> +#include <stdint.h>
> +#include <xen/hvm/hvm_info_table.h>
> +#if !defined(HVM_MAX_VCPUS)
> +# error HVM_MAX_VCPUS not defined
> +#endif
> +int main(void) {
> + xc_interface *xc;
> + xenforeignmemory_handle *xfmem;
> + xenevtchn_handle *xe;
> + xengnttab_handle *xg;
> +
> + xs_daemon_open();
> +
> + xc = xc_interface_open(0, 0, 0);
> + xc_hvm_set_mem_type(0, 0, HVMMEM_ram_ro, 0, 0);
> + xc_domain_add_to_physmap(0, 0, XENMAPSPACE_gmfn, 0, 0);
> + xc_hvm_inject_msi(xc, 0, 0xf0000000, 0x00000000);
> + xc_hvm_create_ioreq_server(xc, 0, HVM_IOREQSRV_BUFIOREQ_ATOMIC, NULL);
> +
> + xfmem = xenforeignmemory_open(0, 0);
> + xenforeignmemory_map(xfmem, 0, 0, 0, 0, 0);
> +
> + xe = xenevtchn_open(0, 0);
> + xenevtchn_fd(xe);
> +
> + xg = xengnttab_open(0, 0);
> + xengnttab_map_grant_ref(xg, 0, 0, 0);
> +
> + return 0;
> +}
> +EOF
> + compile_prog "" "$xen_libs $xen_stable_libs"
> + then
> + xen_ctrl_version=470
> + xen=yes
> +
> + # Xen 4.6
> + elif
> + cat > $TMPC <<EOF &&
> #include <xenctrl.h>
> #include <xenstore.h>
> #include <stdint.h>
> @@ -2096,6 +2148,9 @@ EOF
> fi
>
> if test "$xen" = yes; then
> + if test $xen_ctrl_version -ge 470 ; then
> + libs_softmmu="$xen_stable_libs $libs_softmmu"
tab
> + fi
> libs_softmmu="$xen_libs $libs_softmmu"
> fi
> fi
> diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
> index 323c76c..a758ac4 100644
> --- a/include/hw/xen/xen_common.h
> +++ b/include/hw/xen/xen_common.h
> @@ -6,6 +6,15 @@
> #include <stddef.h>
> #include <inttypes.h>
>
> +/*
> + * If we have new enough libxenctrl then we do not want/need these compat
> + * interfaces, despite what the user supplied cflags might say. They
> + * must be undefined before including xenctrl.h
> + */
> +#undef XC_WANT_COMPAT_EVTCHN_API
> +#undef XC_WANT_COMPAT_GNTTAB_API
> +#undef XC_WANT_COMPAT_MAP_FOREIGN_API
> +
> #include <xenctrl.h>
> #if CONFIG_XEN_CTRL_INTERFACE_VERSION < 420
> # include <xs.h>
> @@ -151,8 +160,8 @@ static inline void xs_close(struct xs_handle *xsh)
> }
>
>
> -/* Xen 4.1 */
> -#else
> +/* Xen 4.1 thru 4.6 */
> +#elif CONFIG_XEN_CTRL_INTERFACE_VERSION < 470
>
> typedef xc_interface *XenXC;
> typedef xc_interface *xenforeignmemory_handle;
> @@ -190,11 +199,34 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
> xc_map_foreign_bulk(*h, d, p, a, e, n)
> #define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
>
> +/* FIXME There is no way to have the xen fd */
> +static inline int xc_fd(xc_interface *xen_xc)
> +{
> + return -1;
> +}
> +#else /* CONFIG_XEN_CTRL_INTERFACE_VERSION >= 470 */
> +
> +typedef xc_interface *XenXC;
> +
> +# define XC_INTERFACE_FMT "%p"
> +# define XC_HANDLER_INITIAL_VALUE NULL
> +
> +#include <xenevtchn.h>
> +#include <xengnttab.h>
> +#include <xenforeignmemory.h>
> +
> +static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
> + unsigned int open_flags)
> +{
> + return xc_interface_open(logger, dombuild_logger, open_flags);
> +}
> +
> /* FIXME There is now way to have the xen fd */
> static inline int xc_fd(xc_interface *xen_xc)
> {
> return -1;
> }
> +
stray line
> #endif
>
> /* Xen before 4.2 */
> --
> 2.1.4
>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH QEMU-XEN v5 7/9] xen: Use stable library interfaces when they are available.
2015-11-09 12:01 ` [Qemu-devel] " Ian Campbell
2015-11-10 14:00 ` Stefano Stabellini
@ 2015-11-10 14:00 ` Stefano Stabellini
1 sibling, 0 replies; 111+ messages in thread
From: Stefano Stabellini @ 2015-11-10 14:00 UTC (permalink / raw)
To: Ian Campbell
Cc: wei.liu2, stefano.stabellini, ian.jackson, qemu-devel, xen-devel
On Mon, 9 Nov 2015, Ian Campbell wrote:
> In Xen 4.7 we are refactoring parts libxenctrl into a number of
> separate libraries which will provide backward and forward API and ABI
> compatiblity.
>
> Specifically libxenevtchn, libxengnttab and libxenforeignmemory.
>
> Previous patches have already laid the groundwork for using these by
> switching the existing compatibility shims to reflect the intefaces to
> these libraries.
>
> So all which remains is to update configure to detect the libraries
> and enable their use. Although they are notionally independent we take
> an all or nothing approach to the three libraries since they were
> added at the same time.
>
> The only non-obvious bit is that we now open a proper xenforeignmemory
> handle for xen_fmem instead of reusing the xen_xc handle.
>
> Build tested with 4.0 .. 4.6 (inclusive) and the patches targetting
> 4.7 which adds these libraries.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Two formatting issues below:
> v5: Remove ifdef check when undeffing the compat stuff.
> s/now way/no way/
>
> v4: xenforeignmemory_open is now a compat wrapper, so no ifdef.
>
> Simplify configury by asserting that interface version 470 will
> always have the libraries (lack of them makes it 460).
>
> Ran checkpatch and fixed everything except:
>
> ERROR: need consistent spacing around '*' (ctx:WxV)
> +typedef xc_interface *XenXC;
>
> Which I think is a false +ve.
>
> Simpler configure stuff
> ---
> configure | 55 +++++++++++++++++++++++++++++++++++++++++++++
> include/hw/xen/xen_common.h | 36 +++++++++++++++++++++++++++--
> 2 files changed, 89 insertions(+), 2 deletions(-)
>
> diff --git a/configure b/configure
> index b4a3488..0253cc9 100755
> --- a/configure
> +++ b/configure
> @@ -1898,6 +1898,7 @@ fi
>
> if test "$xen" != "no" ; then
> xen_libs="-lxenstore -lxenctrl -lxenguest"
> + xen_stable_libs="-lxenforeignmemory -lxengnttab -lxenevtchn"
>
> # First we test whether Xen headers and libraries are available.
> # If no, we are done and there is no Xen support.
> @@ -1920,6 +1921,57 @@ EOF
> # Xen unstable
> elif
> cat > $TMPC <<EOF &&
> +/*
> + * If we have stable libs the we don't want the libxc compat
> + * layers, regardless of what CFLAGS we may have been given.
> + */
> +#undef XC_WANT_COMPAT_EVTCHN_API
> +#undef XC_WANT_COMPAT_GNTTAB_API
> +#undef XC_WANT_COMPAT_MAP_FOREIGN_API
> +#include <xenctrl.h>
> +#include <xenstore.h>
> +#include <xenevtchn.h>
> +#include <xengnttab.h>
> +#include <xenforeignmemory.h>
> +#include <stdint.h>
> +#include <xen/hvm/hvm_info_table.h>
> +#if !defined(HVM_MAX_VCPUS)
> +# error HVM_MAX_VCPUS not defined
> +#endif
> +int main(void) {
> + xc_interface *xc;
> + xenforeignmemory_handle *xfmem;
> + xenevtchn_handle *xe;
> + xengnttab_handle *xg;
> +
> + xs_daemon_open();
> +
> + xc = xc_interface_open(0, 0, 0);
> + xc_hvm_set_mem_type(0, 0, HVMMEM_ram_ro, 0, 0);
> + xc_domain_add_to_physmap(0, 0, XENMAPSPACE_gmfn, 0, 0);
> + xc_hvm_inject_msi(xc, 0, 0xf0000000, 0x00000000);
> + xc_hvm_create_ioreq_server(xc, 0, HVM_IOREQSRV_BUFIOREQ_ATOMIC, NULL);
> +
> + xfmem = xenforeignmemory_open(0, 0);
> + xenforeignmemory_map(xfmem, 0, 0, 0, 0, 0);
> +
> + xe = xenevtchn_open(0, 0);
> + xenevtchn_fd(xe);
> +
> + xg = xengnttab_open(0, 0);
> + xengnttab_map_grant_ref(xg, 0, 0, 0);
> +
> + return 0;
> +}
> +EOF
> + compile_prog "" "$xen_libs $xen_stable_libs"
> + then
> + xen_ctrl_version=470
> + xen=yes
> +
> + # Xen 4.6
> + elif
> + cat > $TMPC <<EOF &&
> #include <xenctrl.h>
> #include <xenstore.h>
> #include <stdint.h>
> @@ -2096,6 +2148,9 @@ EOF
> fi
>
> if test "$xen" = yes; then
> + if test $xen_ctrl_version -ge 470 ; then
> + libs_softmmu="$xen_stable_libs $libs_softmmu"
tab
> + fi
> libs_softmmu="$xen_libs $libs_softmmu"
> fi
> fi
> diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
> index 323c76c..a758ac4 100644
> --- a/include/hw/xen/xen_common.h
> +++ b/include/hw/xen/xen_common.h
> @@ -6,6 +6,15 @@
> #include <stddef.h>
> #include <inttypes.h>
>
> +/*
> + * If we have new enough libxenctrl then we do not want/need these compat
> + * interfaces, despite what the user supplied cflags might say. They
> + * must be undefined before including xenctrl.h
> + */
> +#undef XC_WANT_COMPAT_EVTCHN_API
> +#undef XC_WANT_COMPAT_GNTTAB_API
> +#undef XC_WANT_COMPAT_MAP_FOREIGN_API
> +
> #include <xenctrl.h>
> #if CONFIG_XEN_CTRL_INTERFACE_VERSION < 420
> # include <xs.h>
> @@ -151,8 +160,8 @@ static inline void xs_close(struct xs_handle *xsh)
> }
>
>
> -/* Xen 4.1 */
> -#else
> +/* Xen 4.1 thru 4.6 */
> +#elif CONFIG_XEN_CTRL_INTERFACE_VERSION < 470
>
> typedef xc_interface *XenXC;
> typedef xc_interface *xenforeignmemory_handle;
> @@ -190,11 +199,34 @@ static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
> xc_map_foreign_bulk(*h, d, p, a, e, n)
> #define xenforeignmemory_unmap(h, p, s) munmap(p, s * XC_PAGE_SIZE)
>
> +/* FIXME There is no way to have the xen fd */
> +static inline int xc_fd(xc_interface *xen_xc)
> +{
> + return -1;
> +}
> +#else /* CONFIG_XEN_CTRL_INTERFACE_VERSION >= 470 */
> +
> +typedef xc_interface *XenXC;
> +
> +# define XC_INTERFACE_FMT "%p"
> +# define XC_HANDLER_INITIAL_VALUE NULL
> +
> +#include <xenevtchn.h>
> +#include <xengnttab.h>
> +#include <xenforeignmemory.h>
> +
> +static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
> + unsigned int open_flags)
> +{
> + return xc_interface_open(logger, dombuild_logger, open_flags);
> +}
> +
> /* FIXME There is now way to have the xen fd */
> static inline int xc_fd(xc_interface *xen_xc)
> {
> return -1;
> }
> +
stray line
> #endif
>
> /* Xen before 4.2 */
> --
> 2.1.4
>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [Qemu-devel] [PATCH QEMU-XEN v5 3/9] xen: Switch to libxengnttab interface for compat shims.
2015-11-09 12:01 ` Ian Campbell
(?)
(?)
@ 2015-11-10 14:02 ` Stefano Stabellini
-1 siblings, 0 replies; 111+ messages in thread
From: Stefano Stabellini @ 2015-11-10 14:02 UTC (permalink / raw)
To: Ian Campbell
Cc: wei.liu2, stefano.stabellini, ian.jackson, qemu-devel, xen-devel
On Mon, 9 Nov 2015, Ian Campbell wrote:
> In Xen 4.7 we are refactoring parts libxenctrl into a number of
> separate libraries which will provide backward and forward API and ABI
> compatiblity.
>
> One such library will be libxengnttab which provides access to grant
> tables.
>
> In preparation for this switch the compatibility layer in xen_common.h
> (which support building with older versions of Xen) to use what will
> be the new library API. This means that the gnttab shim will disappear
> for versions of Xen which include libxengnttab.
>
> To simplify things for the <= 4.0.0 support we wrap the int fd in a
> malloc(sizeof int) such that the handle is always a pointer. This
> leads to less typedef headaches and the need for
> XC_HANDLER_INITIAL_VALUE etc for these interfaces.
>
> Note that this patch does not add any support for actually using
> libxengnttab, it just adjusts the existing shims.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> v4: Ran checkpatch, fixed all errors
> Allocate correct size for handle (i.e. not size of the ptr)
> Rebase onto "xen_console: correctly cleanup primary console on
> teardown."
>
> v5: Rebase over b4f72e31b924 "hw/net/xen_nic.c: Free 'netdev->txs'
> when map 'netdev->rxs' fails" which added a new gnttab_munmap.
> ---
> hw/block/xen_disk.c | 38 ++++++++++++++++++++------------------
> hw/char/xen_console.c | 4 ++--
> hw/net/xen_nic.c | 18 +++++++++---------
> hw/xen/xen_backend.c | 10 +++++-----
> include/hw/xen/xen_backend.h | 2 +-
> include/hw/xen/xen_common.h | 42 ++++++++++++++++++++++++++++++++----------
> 6 files changed, 69 insertions(+), 45 deletions(-)
>
> diff --git a/hw/block/xen_disk.c b/hw/block/xen_disk.c
> index 1bbc111..0e80da1 100644
> --- a/hw/block/xen_disk.c
> +++ b/hw/block/xen_disk.c
> @@ -173,11 +173,11 @@ static gint int_cmp(gconstpointer a, gconstpointer b, gpointer user_data)
> static void destroy_grant(gpointer pgnt)
> {
> PersistentGrant *grant = pgnt;
> - XenGnttab gnt = grant->blkdev->xendev.gnttabdev;
> + xengnttab_handle *gnt = grant->blkdev->xendev.gnttabdev;
>
> - if (xc_gnttab_munmap(gnt, grant->page, 1) != 0) {
> + if (xengnttab_munmap(gnt, grant->page, 1) != 0) {
> xen_be_printf(&grant->blkdev->xendev, 0,
> - "xc_gnttab_munmap failed: %s\n",
> + "xengnttab_munmap failed: %s\n",
> strerror(errno));
> }
> grant->blkdev->persistent_gnt_count--;
> @@ -190,11 +190,11 @@ static void remove_persistent_region(gpointer data, gpointer dev)
> {
> PersistentRegion *region = data;
> struct XenBlkDev *blkdev = dev;
> - XenGnttab gnt = blkdev->xendev.gnttabdev;
> + xengnttab_handle *gnt = blkdev->xendev.gnttabdev;
>
> - if (xc_gnttab_munmap(gnt, region->addr, region->num) != 0) {
> + if (xengnttab_munmap(gnt, region->addr, region->num) != 0) {
> xen_be_printf(&blkdev->xendev, 0,
> - "xc_gnttab_munmap region %p failed: %s\n",
> + "xengnttab_munmap region %p failed: %s\n",
> region->addr, strerror(errno));
> }
> xen_be_printf(&blkdev->xendev, 3,
> @@ -329,7 +329,7 @@ err:
>
> static void ioreq_unmap(struct ioreq *ioreq)
> {
> - XenGnttab gnt = ioreq->blkdev->xendev.gnttabdev;
> + xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
> int i;
>
> if (ioreq->num_unmap == 0 || ioreq->mapped == 0) {
> @@ -339,8 +339,9 @@ static void ioreq_unmap(struct ioreq *ioreq)
> if (!ioreq->pages) {
> return;
> }
> - if (xc_gnttab_munmap(gnt, ioreq->pages, ioreq->num_unmap) != 0) {
> - xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
> + if (xengnttab_munmap(gnt, ioreq->pages, ioreq->num_unmap) != 0) {
> + xen_be_printf(&ioreq->blkdev->xendev, 0,
> + "xengnttab_munmap failed: %s\n",
> strerror(errno));
> }
> ioreq->blkdev->cnt_map -= ioreq->num_unmap;
> @@ -350,8 +351,9 @@ static void ioreq_unmap(struct ioreq *ioreq)
> if (!ioreq->page[i]) {
> continue;
> }
> - if (xc_gnttab_munmap(gnt, ioreq->page[i], 1) != 0) {
> - xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
> + if (xengnttab_munmap(gnt, ioreq->page[i], 1) != 0) {
> + xen_be_printf(&ioreq->blkdev->xendev, 0,
> + "xengnttab_munmap failed: %s\n",
> strerror(errno));
> }
> ioreq->blkdev->cnt_map--;
> @@ -363,7 +365,7 @@ static void ioreq_unmap(struct ioreq *ioreq)
>
> static int ioreq_map(struct ioreq *ioreq)
> {
> - XenGnttab gnt = ioreq->blkdev->xendev.gnttabdev;
> + xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
> uint32_t domids[BLKIF_MAX_SEGMENTS_PER_REQUEST];
> uint32_t refs[BLKIF_MAX_SEGMENTS_PER_REQUEST];
> void *page[BLKIF_MAX_SEGMENTS_PER_REQUEST];
> @@ -414,7 +416,7 @@ static int ioreq_map(struct ioreq *ioreq)
> }
>
> if (batch_maps && new_maps) {
> - ioreq->pages = xc_gnttab_map_grant_refs
> + ioreq->pages = xengnttab_map_grant_refs
> (gnt, new_maps, domids, refs, ioreq->prot);
> if (ioreq->pages == NULL) {
> xen_be_printf(&ioreq->blkdev->xendev, 0,
> @@ -430,7 +432,7 @@ static int ioreq_map(struct ioreq *ioreq)
> ioreq->blkdev->cnt_map += new_maps;
> } else if (new_maps) {
> for (i = 0; i < new_maps; i++) {
> - ioreq->page[i] = xc_gnttab_map_grant_ref
> + ioreq->page[i] = xengnttab_map_grant_ref
> (gnt, domids[i], refs[i], ioreq->prot);
> if (ioreq->page[i] == NULL) {
> xen_be_printf(&ioreq->blkdev->xendev, 0,
> @@ -763,9 +765,9 @@ static void blk_alloc(struct XenDevice *xendev)
> if (xen_mode != XEN_EMULATE) {
> batch_maps = 1;
> }
> - if (xc_gnttab_set_max_grants(xendev->gnttabdev,
> + if (xengnttab_set_max_grants(xendev->gnttabdev,
> MAX_GRANTS(max_requests, BLKIF_MAX_SEGMENTS_PER_REQUEST)) < 0) {
> - xen_be_printf(xendev, 0, "xc_gnttab_set_max_grants failed: %s\n",
> + xen_be_printf(xendev, 0, "xengnttab_set_max_grants failed: %s\n",
> strerror(errno));
> }
> }
> @@ -972,7 +974,7 @@ static int blk_connect(struct XenDevice *xendev)
> }
> }
>
> - blkdev->sring = xc_gnttab_map_grant_ref(blkdev->xendev.gnttabdev,
> + blkdev->sring = xengnttab_map_grant_ref(blkdev->xendev.gnttabdev,
> blkdev->xendev.dom,
> blkdev->ring_ref,
> PROT_READ | PROT_WRITE);
> @@ -1037,7 +1039,7 @@ static void blk_disconnect(struct XenDevice *xendev)
> xen_be_unbind_evtchn(&blkdev->xendev);
>
> if (blkdev->sring) {
> - xc_gnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
> + xengnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
> blkdev->cnt_map--;
> blkdev->sring = NULL;
> }
> diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
> index 63ade33..8c06c19 100644
> --- a/hw/char/xen_console.c
> +++ b/hw/char/xen_console.c
> @@ -233,7 +233,7 @@ static int con_initialise(struct XenDevice *xendev)
> PROT_READ|PROT_WRITE,
> con->ring_ref);
> } else {
> - con->sring = xc_gnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
> + con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
> con->ring_ref,
> PROT_READ|PROT_WRITE);
> }
> @@ -275,7 +275,7 @@ static void con_disconnect(struct XenDevice *xendev)
> if (!xendev->dev) {
> munmap(con->sring, XC_PAGE_SIZE);
> } else {
> - xc_gnttab_munmap(xendev->gnttabdev, con->sring, 1);
> + xengnttab_munmap(xendev->gnttabdev, con->sring, 1);
> }
> con->sring = NULL;
> }
> diff --git a/hw/net/xen_nic.c b/hw/net/xen_nic.c
> index 0da16b4..6d69a6a 100644
> --- a/hw/net/xen_nic.c
> +++ b/hw/net/xen_nic.c
> @@ -168,7 +168,7 @@ static void net_tx_packets(struct XenNetDev *netdev)
> (txreq.flags & NETTXF_more_data) ? " more_data" : "",
> (txreq.flags & NETTXF_extra_info) ? " extra_info" : "");
>
> - page = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
> + page = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
> netdev->xendev.dom,
> txreq.gref, PROT_READ);
> if (page == NULL) {
> @@ -190,7 +190,7 @@ static void net_tx_packets(struct XenNetDev *netdev)
> qemu_send_packet(qemu_get_queue(netdev->nic),
> page + txreq.offset, txreq.size);
> }
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, page, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, page, 1);
> net_tx_response(netdev, &txreq, NETIF_RSP_OKAY);
> }
> if (!netdev->tx_work) {
> @@ -260,7 +260,7 @@ static ssize_t net_rx_packet(NetClientState *nc, const uint8_t *buf, size_t size
> memcpy(&rxreq, RING_GET_REQUEST(&netdev->rx_ring, rc), sizeof(rxreq));
> netdev->rx_ring.req_cons = ++rc;
>
> - page = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
> + page = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
> netdev->xendev.dom,
> rxreq.gref, PROT_WRITE);
> if (page == NULL) {
> @@ -270,7 +270,7 @@ static ssize_t net_rx_packet(NetClientState *nc, const uint8_t *buf, size_t size
> return -1;
> }
> memcpy(page + NET_IP_ALIGN, buf, size);
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, page, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, page, 1);
> net_rx_response(netdev, &rxreq, NETIF_RSP_OKAY, NET_IP_ALIGN, size, 0);
>
> return size;
> @@ -342,19 +342,19 @@ static int net_connect(struct XenDevice *xendev)
> return -1;
> }
>
> - netdev->txs = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
> + netdev->txs = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
> netdev->xendev.dom,
> netdev->tx_ring_ref,
> PROT_READ | PROT_WRITE);
> if (!netdev->txs) {
> return -1;
> }
> - netdev->rxs = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
> + netdev->rxs = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
> netdev->xendev.dom,
> netdev->rx_ring_ref,
> PROT_READ | PROT_WRITE);
> if (!netdev->rxs) {
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
> netdev->txs = NULL;
> return -1;
> }
> @@ -379,11 +379,11 @@ static void net_disconnect(struct XenDevice *xendev)
> xen_be_unbind_evtchn(&netdev->xendev);
>
> if (netdev->txs) {
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
> netdev->txs = NULL;
> }
> if (netdev->rxs) {
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->rxs, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, netdev->rxs, 1);
> netdev->rxs = NULL;
> }
> }
> diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
> index ae2a1f0..966e34f 100644
> --- a/hw/xen/xen_backend.c
> +++ b/hw/xen/xen_backend.c
> @@ -252,15 +252,15 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev,
> fcntl(xenevtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
>
> if (ops->flags & DEVOPS_FLAG_NEED_GNTDEV) {
> - xendev->gnttabdev = xen_xc_gnttab_open(NULL, 0);
> - if (xendev->gnttabdev == XC_HANDLER_INITIAL_VALUE) {
> + xendev->gnttabdev = xengnttab_open(NULL, 0);
> + if (xendev->gnttabdev == NULL) {
> xen_be_printf(NULL, 0, "can't open gnttab device\n");
> xenevtchn_close(xendev->evtchndev);
> g_free(xendev);
> return NULL;
> }
> } else {
> - xendev->gnttabdev = XC_HANDLER_INITIAL_VALUE;
> + xendev->gnttabdev = NULL;
> }
>
> QTAILQ_INSERT_TAIL(&xendevs, xendev, next);
> @@ -309,8 +309,8 @@ static struct XenDevice *xen_be_del_xendev(int dom, int dev)
> if (xendev->evtchndev != NULL) {
> xenevtchn_close(xendev->evtchndev);
> }
> - if (xendev->gnttabdev != XC_HANDLER_INITIAL_VALUE) {
> - xc_gnttab_close(xendev->gnttabdev);
> + if (xendev->gnttabdev != NULL) {
> + xengnttab_close(xendev->gnttabdev);
> }
>
> QTAILQ_REMOVE(&xendevs, xendev, next);
> diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
> index a90314f..8e8857b 100644
> --- a/include/hw/xen/xen_backend.h
> +++ b/include/hw/xen/xen_backend.h
> @@ -47,7 +47,7 @@ struct XenDevice {
> int local_port;
>
> xenevtchn_handle *evtchndev;
> - XenGnttab gnttabdev;
> + xengnttab_handle *gnttabdev;
>
> struct XenDevOps *ops;
> QTAILQ_ENTRY(XenDevice) next;
> diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
> index 6f7c003..dde6b7f 100644
> --- a/include/hw/xen/xen_common.h
> +++ b/include/hw/xen/xen_common.h
> @@ -40,7 +40,7 @@ static inline void *xc_map_foreign_bulk(int xc_handle, uint32_t dom, int prot,
>
> typedef int XenXC;
> typedef int xenevtchn_handle;
> -typedef int XenGnttab;
> +typedef int xengnttab_handle;
>
> # define XC_INTERFACE_FMT "%i"
> # define XC_HANDLER_INITIAL_VALUE -1
> @@ -72,11 +72,31 @@ static inline int xenevtchn_close(xenevtchn_handle *h)
> #define xenevtchn_unmask(h, p) xc_evtchn_unmask(*h, p)
> #define xenevtchn_unbind(h, p) xc_evtchn_unmask(*h, p)
>
> -static inline XenGnttab xen_xc_gnttab_open(void *logger,
> - unsigned int open_flags)
> +static inline xengnttab_handle *xengnttab_open(void *logger,
> + unsigned int open_flags)
> {
> - return xc_gnttab_open();
> + xengnttab_handle *h = malloc(sizeof(*h));
> + if (!h) {
> + return NULL;
> + }
> + *h = xc_gnttab_open();
> + if (*h == -1) {
> + free(h);
> + h = NULL;
> + }
> + return h;
> }
> +static inline int xengnttab_close(xengnttab_handle *h)
> +{
> + int rc = xc_gnttab_close(*h);
> + free(h);
> + return rc;
> +}
> +#define xengnttab_set_max_grants(h, n) xc_gnttab_set_max_grants(*h, n)
> +#define xengnttab_map_grant_ref(h, d, r, p) xc_gnttab_map_grant_ref(*h, d, r, p)
> +#define xengnttab_map_grant_refs(h, c, d, r, p) \
> + xc_gnttab_map_grant_refs(*h, c, d, r, p)
> +#define xengnttab_munmap(h, a, n) xc_gnttab_munmap(*h, a, n)
>
> static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
> unsigned int open_flags)
> @@ -130,7 +150,7 @@ static inline void xs_close(struct xs_handle *xsh)
>
> typedef xc_interface *XenXC;
> typedef xc_evtchn xenevtchn_handle;
> -typedef xc_gnttab *XenGnttab;
> +typedef xc_gnttab xengnttab_handle;
>
> # define XC_INTERFACE_FMT "%p"
> # define XC_HANDLER_INITIAL_VALUE NULL
> @@ -144,11 +164,13 @@ typedef xc_gnttab *XenGnttab;
> #define xenevtchn_unmask(h, p) xc_evtchn_unmask(h, p)
> #define xenevtchn_unbind(h, p) xc_evtchn_unbind(h, p)
>
> -static inline XenGnttab xen_xc_gnttab_open(void *logger,
> - unsigned int open_flags)
> -{
> - return xc_gnttab_open(logger, open_flags);
> -}
> +#define xengnttab_open(l, f) xc_gnttab_open(l, f)
> +#define xengnttab_close(h) xc_gnttab_close(h)
> +#define xengnttab_set_max_grants(h, n) xc_gnttab_set_max_grants(h, n)
> +#define xengnttab_map_grant_ref(h, d, r, p) xc_gnttab_map_grant_ref(h, d, r, p)
> +#define xengnttab_munmap(h, a, n) xc_gnttab_munmap(h, a, n)
> +#define xengnttab_map_grant_refs(h, c, d, r, p) \
> + xc_gnttab_map_grant_refs(h, c, d, r, p)
>
> static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
> unsigned int open_flags)
> --
> 2.1.4
>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH QEMU-XEN v5 3/9] xen: Switch to libxengnttab interface for compat shims.
2015-11-09 12:01 ` Ian Campbell
(?)
@ 2015-11-10 14:02 ` Stefano Stabellini
-1 siblings, 0 replies; 111+ messages in thread
From: Stefano Stabellini @ 2015-11-10 14:02 UTC (permalink / raw)
To: Ian Campbell
Cc: wei.liu2, stefano.stabellini, ian.jackson, qemu-devel, xen-devel
On Mon, 9 Nov 2015, Ian Campbell wrote:
> In Xen 4.7 we are refactoring parts libxenctrl into a number of
> separate libraries which will provide backward and forward API and ABI
> compatiblity.
>
> One such library will be libxengnttab which provides access to grant
> tables.
>
> In preparation for this switch the compatibility layer in xen_common.h
> (which support building with older versions of Xen) to use what will
> be the new library API. This means that the gnttab shim will disappear
> for versions of Xen which include libxengnttab.
>
> To simplify things for the <= 4.0.0 support we wrap the int fd in a
> malloc(sizeof int) such that the handle is always a pointer. This
> leads to less typedef headaches and the need for
> XC_HANDLER_INITIAL_VALUE etc for these interfaces.
>
> Note that this patch does not add any support for actually using
> libxengnttab, it just adjusts the existing shims.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> v4: Ran checkpatch, fixed all errors
> Allocate correct size for handle (i.e. not size of the ptr)
> Rebase onto "xen_console: correctly cleanup primary console on
> teardown."
>
> v5: Rebase over b4f72e31b924 "hw/net/xen_nic.c: Free 'netdev->txs'
> when map 'netdev->rxs' fails" which added a new gnttab_munmap.
> ---
> hw/block/xen_disk.c | 38 ++++++++++++++++++++------------------
> hw/char/xen_console.c | 4 ++--
> hw/net/xen_nic.c | 18 +++++++++---------
> hw/xen/xen_backend.c | 10 +++++-----
> include/hw/xen/xen_backend.h | 2 +-
> include/hw/xen/xen_common.h | 42 ++++++++++++++++++++++++++++++++----------
> 6 files changed, 69 insertions(+), 45 deletions(-)
>
> diff --git a/hw/block/xen_disk.c b/hw/block/xen_disk.c
> index 1bbc111..0e80da1 100644
> --- a/hw/block/xen_disk.c
> +++ b/hw/block/xen_disk.c
> @@ -173,11 +173,11 @@ static gint int_cmp(gconstpointer a, gconstpointer b, gpointer user_data)
> static void destroy_grant(gpointer pgnt)
> {
> PersistentGrant *grant = pgnt;
> - XenGnttab gnt = grant->blkdev->xendev.gnttabdev;
> + xengnttab_handle *gnt = grant->blkdev->xendev.gnttabdev;
>
> - if (xc_gnttab_munmap(gnt, grant->page, 1) != 0) {
> + if (xengnttab_munmap(gnt, grant->page, 1) != 0) {
> xen_be_printf(&grant->blkdev->xendev, 0,
> - "xc_gnttab_munmap failed: %s\n",
> + "xengnttab_munmap failed: %s\n",
> strerror(errno));
> }
> grant->blkdev->persistent_gnt_count--;
> @@ -190,11 +190,11 @@ static void remove_persistent_region(gpointer data, gpointer dev)
> {
> PersistentRegion *region = data;
> struct XenBlkDev *blkdev = dev;
> - XenGnttab gnt = blkdev->xendev.gnttabdev;
> + xengnttab_handle *gnt = blkdev->xendev.gnttabdev;
>
> - if (xc_gnttab_munmap(gnt, region->addr, region->num) != 0) {
> + if (xengnttab_munmap(gnt, region->addr, region->num) != 0) {
> xen_be_printf(&blkdev->xendev, 0,
> - "xc_gnttab_munmap region %p failed: %s\n",
> + "xengnttab_munmap region %p failed: %s\n",
> region->addr, strerror(errno));
> }
> xen_be_printf(&blkdev->xendev, 3,
> @@ -329,7 +329,7 @@ err:
>
> static void ioreq_unmap(struct ioreq *ioreq)
> {
> - XenGnttab gnt = ioreq->blkdev->xendev.gnttabdev;
> + xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
> int i;
>
> if (ioreq->num_unmap == 0 || ioreq->mapped == 0) {
> @@ -339,8 +339,9 @@ static void ioreq_unmap(struct ioreq *ioreq)
> if (!ioreq->pages) {
> return;
> }
> - if (xc_gnttab_munmap(gnt, ioreq->pages, ioreq->num_unmap) != 0) {
> - xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
> + if (xengnttab_munmap(gnt, ioreq->pages, ioreq->num_unmap) != 0) {
> + xen_be_printf(&ioreq->blkdev->xendev, 0,
> + "xengnttab_munmap failed: %s\n",
> strerror(errno));
> }
> ioreq->blkdev->cnt_map -= ioreq->num_unmap;
> @@ -350,8 +351,9 @@ static void ioreq_unmap(struct ioreq *ioreq)
> if (!ioreq->page[i]) {
> continue;
> }
> - if (xc_gnttab_munmap(gnt, ioreq->page[i], 1) != 0) {
> - xen_be_printf(&ioreq->blkdev->xendev, 0, "xc_gnttab_munmap failed: %s\n",
> + if (xengnttab_munmap(gnt, ioreq->page[i], 1) != 0) {
> + xen_be_printf(&ioreq->blkdev->xendev, 0,
> + "xengnttab_munmap failed: %s\n",
> strerror(errno));
> }
> ioreq->blkdev->cnt_map--;
> @@ -363,7 +365,7 @@ static void ioreq_unmap(struct ioreq *ioreq)
>
> static int ioreq_map(struct ioreq *ioreq)
> {
> - XenGnttab gnt = ioreq->blkdev->xendev.gnttabdev;
> + xengnttab_handle *gnt = ioreq->blkdev->xendev.gnttabdev;
> uint32_t domids[BLKIF_MAX_SEGMENTS_PER_REQUEST];
> uint32_t refs[BLKIF_MAX_SEGMENTS_PER_REQUEST];
> void *page[BLKIF_MAX_SEGMENTS_PER_REQUEST];
> @@ -414,7 +416,7 @@ static int ioreq_map(struct ioreq *ioreq)
> }
>
> if (batch_maps && new_maps) {
> - ioreq->pages = xc_gnttab_map_grant_refs
> + ioreq->pages = xengnttab_map_grant_refs
> (gnt, new_maps, domids, refs, ioreq->prot);
> if (ioreq->pages == NULL) {
> xen_be_printf(&ioreq->blkdev->xendev, 0,
> @@ -430,7 +432,7 @@ static int ioreq_map(struct ioreq *ioreq)
> ioreq->blkdev->cnt_map += new_maps;
> } else if (new_maps) {
> for (i = 0; i < new_maps; i++) {
> - ioreq->page[i] = xc_gnttab_map_grant_ref
> + ioreq->page[i] = xengnttab_map_grant_ref
> (gnt, domids[i], refs[i], ioreq->prot);
> if (ioreq->page[i] == NULL) {
> xen_be_printf(&ioreq->blkdev->xendev, 0,
> @@ -763,9 +765,9 @@ static void blk_alloc(struct XenDevice *xendev)
> if (xen_mode != XEN_EMULATE) {
> batch_maps = 1;
> }
> - if (xc_gnttab_set_max_grants(xendev->gnttabdev,
> + if (xengnttab_set_max_grants(xendev->gnttabdev,
> MAX_GRANTS(max_requests, BLKIF_MAX_SEGMENTS_PER_REQUEST)) < 0) {
> - xen_be_printf(xendev, 0, "xc_gnttab_set_max_grants failed: %s\n",
> + xen_be_printf(xendev, 0, "xengnttab_set_max_grants failed: %s\n",
> strerror(errno));
> }
> }
> @@ -972,7 +974,7 @@ static int blk_connect(struct XenDevice *xendev)
> }
> }
>
> - blkdev->sring = xc_gnttab_map_grant_ref(blkdev->xendev.gnttabdev,
> + blkdev->sring = xengnttab_map_grant_ref(blkdev->xendev.gnttabdev,
> blkdev->xendev.dom,
> blkdev->ring_ref,
> PROT_READ | PROT_WRITE);
> @@ -1037,7 +1039,7 @@ static void blk_disconnect(struct XenDevice *xendev)
> xen_be_unbind_evtchn(&blkdev->xendev);
>
> if (blkdev->sring) {
> - xc_gnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
> + xengnttab_munmap(blkdev->xendev.gnttabdev, blkdev->sring, 1);
> blkdev->cnt_map--;
> blkdev->sring = NULL;
> }
> diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
> index 63ade33..8c06c19 100644
> --- a/hw/char/xen_console.c
> +++ b/hw/char/xen_console.c
> @@ -233,7 +233,7 @@ static int con_initialise(struct XenDevice *xendev)
> PROT_READ|PROT_WRITE,
> con->ring_ref);
> } else {
> - con->sring = xc_gnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
> + con->sring = xengnttab_map_grant_ref(xendev->gnttabdev, con->xendev.dom,
> con->ring_ref,
> PROT_READ|PROT_WRITE);
> }
> @@ -275,7 +275,7 @@ static void con_disconnect(struct XenDevice *xendev)
> if (!xendev->dev) {
> munmap(con->sring, XC_PAGE_SIZE);
> } else {
> - xc_gnttab_munmap(xendev->gnttabdev, con->sring, 1);
> + xengnttab_munmap(xendev->gnttabdev, con->sring, 1);
> }
> con->sring = NULL;
> }
> diff --git a/hw/net/xen_nic.c b/hw/net/xen_nic.c
> index 0da16b4..6d69a6a 100644
> --- a/hw/net/xen_nic.c
> +++ b/hw/net/xen_nic.c
> @@ -168,7 +168,7 @@ static void net_tx_packets(struct XenNetDev *netdev)
> (txreq.flags & NETTXF_more_data) ? " more_data" : "",
> (txreq.flags & NETTXF_extra_info) ? " extra_info" : "");
>
> - page = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
> + page = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
> netdev->xendev.dom,
> txreq.gref, PROT_READ);
> if (page == NULL) {
> @@ -190,7 +190,7 @@ static void net_tx_packets(struct XenNetDev *netdev)
> qemu_send_packet(qemu_get_queue(netdev->nic),
> page + txreq.offset, txreq.size);
> }
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, page, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, page, 1);
> net_tx_response(netdev, &txreq, NETIF_RSP_OKAY);
> }
> if (!netdev->tx_work) {
> @@ -260,7 +260,7 @@ static ssize_t net_rx_packet(NetClientState *nc, const uint8_t *buf, size_t size
> memcpy(&rxreq, RING_GET_REQUEST(&netdev->rx_ring, rc), sizeof(rxreq));
> netdev->rx_ring.req_cons = ++rc;
>
> - page = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
> + page = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
> netdev->xendev.dom,
> rxreq.gref, PROT_WRITE);
> if (page == NULL) {
> @@ -270,7 +270,7 @@ static ssize_t net_rx_packet(NetClientState *nc, const uint8_t *buf, size_t size
> return -1;
> }
> memcpy(page + NET_IP_ALIGN, buf, size);
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, page, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, page, 1);
> net_rx_response(netdev, &rxreq, NETIF_RSP_OKAY, NET_IP_ALIGN, size, 0);
>
> return size;
> @@ -342,19 +342,19 @@ static int net_connect(struct XenDevice *xendev)
> return -1;
> }
>
> - netdev->txs = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
> + netdev->txs = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
> netdev->xendev.dom,
> netdev->tx_ring_ref,
> PROT_READ | PROT_WRITE);
> if (!netdev->txs) {
> return -1;
> }
> - netdev->rxs = xc_gnttab_map_grant_ref(netdev->xendev.gnttabdev,
> + netdev->rxs = xengnttab_map_grant_ref(netdev->xendev.gnttabdev,
> netdev->xendev.dom,
> netdev->rx_ring_ref,
> PROT_READ | PROT_WRITE);
> if (!netdev->rxs) {
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
> netdev->txs = NULL;
> return -1;
> }
> @@ -379,11 +379,11 @@ static void net_disconnect(struct XenDevice *xendev)
> xen_be_unbind_evtchn(&netdev->xendev);
>
> if (netdev->txs) {
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, netdev->txs, 1);
> netdev->txs = NULL;
> }
> if (netdev->rxs) {
> - xc_gnttab_munmap(netdev->xendev.gnttabdev, netdev->rxs, 1);
> + xengnttab_munmap(netdev->xendev.gnttabdev, netdev->rxs, 1);
> netdev->rxs = NULL;
> }
> }
> diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
> index ae2a1f0..966e34f 100644
> --- a/hw/xen/xen_backend.c
> +++ b/hw/xen/xen_backend.c
> @@ -252,15 +252,15 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev,
> fcntl(xenevtchn_fd(xendev->evtchndev), F_SETFD, FD_CLOEXEC);
>
> if (ops->flags & DEVOPS_FLAG_NEED_GNTDEV) {
> - xendev->gnttabdev = xen_xc_gnttab_open(NULL, 0);
> - if (xendev->gnttabdev == XC_HANDLER_INITIAL_VALUE) {
> + xendev->gnttabdev = xengnttab_open(NULL, 0);
> + if (xendev->gnttabdev == NULL) {
> xen_be_printf(NULL, 0, "can't open gnttab device\n");
> xenevtchn_close(xendev->evtchndev);
> g_free(xendev);
> return NULL;
> }
> } else {
> - xendev->gnttabdev = XC_HANDLER_INITIAL_VALUE;
> + xendev->gnttabdev = NULL;
> }
>
> QTAILQ_INSERT_TAIL(&xendevs, xendev, next);
> @@ -309,8 +309,8 @@ static struct XenDevice *xen_be_del_xendev(int dom, int dev)
> if (xendev->evtchndev != NULL) {
> xenevtchn_close(xendev->evtchndev);
> }
> - if (xendev->gnttabdev != XC_HANDLER_INITIAL_VALUE) {
> - xc_gnttab_close(xendev->gnttabdev);
> + if (xendev->gnttabdev != NULL) {
> + xengnttab_close(xendev->gnttabdev);
> }
>
> QTAILQ_REMOVE(&xendevs, xendev, next);
> diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
> index a90314f..8e8857b 100644
> --- a/include/hw/xen/xen_backend.h
> +++ b/include/hw/xen/xen_backend.h
> @@ -47,7 +47,7 @@ struct XenDevice {
> int local_port;
>
> xenevtchn_handle *evtchndev;
> - XenGnttab gnttabdev;
> + xengnttab_handle *gnttabdev;
>
> struct XenDevOps *ops;
> QTAILQ_ENTRY(XenDevice) next;
> diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h
> index 6f7c003..dde6b7f 100644
> --- a/include/hw/xen/xen_common.h
> +++ b/include/hw/xen/xen_common.h
> @@ -40,7 +40,7 @@ static inline void *xc_map_foreign_bulk(int xc_handle, uint32_t dom, int prot,
>
> typedef int XenXC;
> typedef int xenevtchn_handle;
> -typedef int XenGnttab;
> +typedef int xengnttab_handle;
>
> # define XC_INTERFACE_FMT "%i"
> # define XC_HANDLER_INITIAL_VALUE -1
> @@ -72,11 +72,31 @@ static inline int xenevtchn_close(xenevtchn_handle *h)
> #define xenevtchn_unmask(h, p) xc_evtchn_unmask(*h, p)
> #define xenevtchn_unbind(h, p) xc_evtchn_unmask(*h, p)
>
> -static inline XenGnttab xen_xc_gnttab_open(void *logger,
> - unsigned int open_flags)
> +static inline xengnttab_handle *xengnttab_open(void *logger,
> + unsigned int open_flags)
> {
> - return xc_gnttab_open();
> + xengnttab_handle *h = malloc(sizeof(*h));
> + if (!h) {
> + return NULL;
> + }
> + *h = xc_gnttab_open();
> + if (*h == -1) {
> + free(h);
> + h = NULL;
> + }
> + return h;
> }
> +static inline int xengnttab_close(xengnttab_handle *h)
> +{
> + int rc = xc_gnttab_close(*h);
> + free(h);
> + return rc;
> +}
> +#define xengnttab_set_max_grants(h, n) xc_gnttab_set_max_grants(*h, n)
> +#define xengnttab_map_grant_ref(h, d, r, p) xc_gnttab_map_grant_ref(*h, d, r, p)
> +#define xengnttab_map_grant_refs(h, c, d, r, p) \
> + xc_gnttab_map_grant_refs(*h, c, d, r, p)
> +#define xengnttab_munmap(h, a, n) xc_gnttab_munmap(*h, a, n)
>
> static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
> unsigned int open_flags)
> @@ -130,7 +150,7 @@ static inline void xs_close(struct xs_handle *xsh)
>
> typedef xc_interface *XenXC;
> typedef xc_evtchn xenevtchn_handle;
> -typedef xc_gnttab *XenGnttab;
> +typedef xc_gnttab xengnttab_handle;
>
> # define XC_INTERFACE_FMT "%p"
> # define XC_HANDLER_INITIAL_VALUE NULL
> @@ -144,11 +164,13 @@ typedef xc_gnttab *XenGnttab;
> #define xenevtchn_unmask(h, p) xc_evtchn_unmask(h, p)
> #define xenevtchn_unbind(h, p) xc_evtchn_unbind(h, p)
>
> -static inline XenGnttab xen_xc_gnttab_open(void *logger,
> - unsigned int open_flags)
> -{
> - return xc_gnttab_open(logger, open_flags);
> -}
> +#define xengnttab_open(l, f) xc_gnttab_open(l, f)
> +#define xengnttab_close(h) xc_gnttab_close(h)
> +#define xengnttab_set_max_grants(h, n) xc_gnttab_set_max_grants(h, n)
> +#define xengnttab_map_grant_ref(h, d, r, p) xc_gnttab_map_grant_ref(h, d, r, p)
> +#define xengnttab_munmap(h, a, n) xc_gnttab_munmap(h, a, n)
> +#define xengnttab_map_grant_refs(h, c, d, r, p) \
> + xc_gnttab_map_grant_refs(h, c, d, r, p)
>
> static inline XenXC xen_xc_interface_open(void *logger, void *dombuild_logger,
> unsigned int open_flags)
> --
> 2.1.4
>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-09 12:00 ` [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab Ian Campbell
@ 2015-11-11 14:34 ` Wei Liu
2015-11-11 15:03 ` Ian Jackson
1 sibling, 0 replies; 111+ messages in thread
From: Wei Liu @ 2015-11-11 14:34 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, ian.jackson, xen-devel
On Mon, Nov 09, 2015 at 12:00:42PM +0000, Ian Campbell wrote:
> libxengnttab will provide a stable API and ABI for accessing the
> grant table devices.
>
> The functions are moved into the xengnt{tab,shr} namespace to make a
> clean break from libxc and avoid ambiguity regarding which interfaces
> are stable.
>
> XXX consider combining into a single namespace (i.e. with
> xengnttab_handle having two open fd's in it on Linux)
>
Forgot to delete this?
I think your analysis in earlier reply is reasonable, so
Acked-by: Wei Liu <wei.liu2@citrix.com>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 04/23] tools: Refactor /dev/xen/evtchn wrappers into libxenevtchn.
2015-11-09 12:00 ` [PATCH XEN v5 04/23] tools: Refactor /dev/xen/evtchn wrappers into libxenevtchn Ian Campbell
@ 2015-11-11 14:50 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 14:50 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 04/23] tools: Refactor /dev/xen/evtchn wrappers into libxenevtchn."):
> libxenevtchn will provide a stable API and ABI for accessing the
> evtchn device.
API reviewed.
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-09 12:00 ` [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab Ian Campbell
2015-11-11 14:34 ` Wei Liu
@ 2015-11-11 15:03 ` Ian Jackson
2015-11-13 15:02 ` Ian Campbell
2015-11-23 17:05 ` Ian Campbell
1 sibling, 2 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:03 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev,shr} wrappers into libxengnttab."):
> libxengnttab will provide a stable API and ABI for accessing the
> grant table devices.
>
> The functions are moved into the xengnt{tab,shr} namespace to make a
> clean break from libxc and avoid ambiguity regarding which interfaces
> are stable.
I have reviewed the API. Mostly this produced questions...
> XXX consider combining into a single namespace (i.e. with
> xengnttab_handle having two open fd's in it on Linux)
Arguments in favour of combining:
* Some kernel might provide a single thing in /dev, and might even
have trouble coping with a process which opens it more than once.
* This is less fiddly for callers.
Arguments against combining:
* Perhaps some OSes will exist which can do only one and not the
other. (Tenuous, and a workaround is possible.)
* Slightly lower performance for a client which uses only one of the
interfaces. (Barrel-scraping, here.)
* It is more work.
I conclude it would be better to unify them.
> +/*
> + * Note:
> + * After fork a child process must not use any opened xc gnttab
> + * handle inherited from their parent. They must open a new handle if
> + * they want to interact with xc.
This should document
(a) that it is safe to close the handle
(b) that it is not safe to access grant mapped areas in the child
(c) how to reclaim the address space which is otherwise lost
> +/**
> + * Memory maps one or more grant references from one or more domains to a
> + * contiguous local address range. Mappings should be unmapped with
> + * xengnttab_munmap. Logs errors.
> + *
> + * @parm xgt a handle on an open grant table interface
> + * @parm count the number of grant references to be mapped
> + * @parm domids an array of @count domain IDs by which the corresponding @refs
> + * were granted
> + * @parm refs an array of @count grant references to be mapped
> + * @parm prot same flag as in mmap()
> + */
> +void *xengnttab_map_grant_refs(xengnttab_handle *xgt,
Should explain what happens on partial failure.
> +/**
> + * Memory maps one or more grant references from one domain to a
> + * contiguous local address range. Mappings should be unmapped with
> + * xengnttab_munmap. Logs errors.
...
> +void *xengnttab_map_domain_grant_refs(xengnttab_handle *xgt,
Should say that this is equivalent to a suitable call to
xengnttab_map_grant_refs. (Assuming that it is.)
> +/**
> + * Memory maps a grant reference from one domain to a local address range.
> + * Mappings should be unmapped with xengnttab_munmap. If notify_offset or
> + * notify_port are not -1, this version will attempt to set up an unmap
> + * notification at the given offset and event channel. When the page is
> + * unmapped, the byte at the given offset will be zeroed and a wakeup will be
> + * sent to the given event channel. Logs errors.
What happens if the unmap notification cannot be set up ?
Also "when the page is unmapped" makes it sound like you mean
xengnttab_munmap but actually I think this is when the grant is
withdrawn by the grantor ?
If the grant is withdrawn by the grantor, does the page become
unuseable immediately ? If so, how can anyone ever use this safely ?
> +/*
> + * Sets the maximum number of grants that may be mapped by the given instance
> + * to @count. Never logs.
Line wrap.
> +int xengnttab_set_max_grants(xengnttab_handle *xgt,
> + uint32_t count);
Is count in pages or grants or what ?
> +/*
> + * Return an fd onto the grant sharing driver. Logs errors.
It does not return an fd. (See also my comments on gntmap.)
> +/*
> + * Creates and shares pages with another domain.
> + *
...
> +void *xengntshr_share_pages(xengntshr_handle *xgs, uint32_t domid,
> + int count, uint32_t *refs, int writable);
Can this fail ? Can it partially succeed ?
> +/*
> + * Creates and shares a page with another domain, with unmap notification.
> + *
> + * @parm xgs a handle to an open grant sharing instance
> + * @parm domid the domain to share memory with
> + * @parm refs the grant reference of the pages (output)
> + * @parm writable true if the other domain can write to the page
> + * @parm notify_offset The byte offset in the page to use for unmap
> + * notification; -1 for none.
> + * @parm notify_port The event channel port to use for unmap notify, or -1
> + * @return local mapping of the page
> + */
> +void *xengntshr_share_page_notify(xengntshr_handle *xgs, uint32_t domid,
What is this `unmap notification' ?
> +/*
> + * Unmaps the @count pages starting at @start_address, which were mapped by a
> + * call to xengntshr_share_*. Never logs.
Linewrap in the comment.
> + */
> +int xengntshr_munmap(xengntshr_handle *xgs, void *start_address, uint32_t count);
What effect does this have on the peer ?
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-09 12:00 ` [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall Ian Campbell
@ 2015-11-11 15:08 ` Ian Jackson
2015-11-11 15:31 ` Wei Liu
2015-11-13 15:16 ` Ian Campbell
2015-11-13 15:35 ` Andrew Cooper
2015-11-13 15:49 ` Andrew Cooper
2 siblings, 2 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:08 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall."):
> libxencall will provide a stable API and ABI for calling hypercalls
> (although those hypercalls themselves may not have a stable API). As
> well as the hypercall buffer infrastructure needed in order to safely
> provide pointer arguments to hypercalls.
...
> +/*
> + * This library allows you to make arbitrary hypercalls (subject to
> + * sufficient permission for the process and the domain itself). Note
> + * that while the library interface is stable the hypercalls are
> + * subject to their own rules.
Something needs to say what the error handling is like.
Do these functions set errno ?
> +/*
> + * Call hypercalls with varying numbers of arguments.
> + */
> +int xencall0(xencall_handle *xcall, unsigned int op);
Is the return value the raw hypercall return value, or is hypervisor
do_foo returning -EFOOBAR turned into to -1/errno=EFOOBAR ?
(Hopefully the answer to this doesn't depend on the hypercall ABI...)
> +/*
> + * Allocate and free memory which is suitable for use as a pointer
> + * argument to a hypercall.
> + */
> +void *xencall_alloc_buffer_pages(xencall_handle *xcall, int nr_pages);
> +void xencall_free_buffer_pages(xencall_handle *xcall, void *p, int nr_pages);
> +
> +void *xencall_alloc_buffer(xencall_handle *xcall, size_t size);
> +void xencall_free_buffer(xencall_handle *xcall, void *p);
See above re error handling.
Can these functions be used without (a) knowing the page size
(b) a rounding macro ?
It would be best to save callers the trouble of providing those
themselves.
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
2015-11-09 12:00 ` [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory Ian Campbell
@ 2015-11-11 15:13 ` Ian Jackson
2015-11-13 15:21 ` Ian Campbell
2015-11-27 15:46 ` Ian Campbell
0 siblings, 2 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:13 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory"):
> libxenforeignmemory will provide a stable API and ABI for mapping
> foreign domain memory (subject to appropriate privileges).
>
> The new library exposes an interface equivalent to
> xc_map_foreign_memory_bulk, which all the other
> xc_map_foreign_memory_* functions (which remain in libxc) are
> implemented in terms of.
...
> diff --git a/tools/libs/foreignmemory/include/xenforeignmemory.h b/tools/libs/foreignmemory/include/xenforeignmemory.h
> +/*
> + * Return a handle onto the hypercall driver. Logs errors.
> + */
> +xenforeignmemory_handle *xenforeignmemory_open(xentoollog_logger *logger,
> + unsigned open_flags);
`onto the foreign memory driver'.
This file should probably say something about fork(). Perhaps just a
refernce to the rules in the grant map driver.
> +/*
> + * Maps a range within one domain to a local address range. Mappings
> + * should be unmapped with munmap and should follow the same rules as mmap
> + * regarding page alignment.
> + *
> + * prot is as for mmap(2).
> + *
> + * Can partially succeed. When a page cannot be mapped, its respective
> + * field in @err is set to the corresponding errno value.
> + *
> + * Returns NULL if no pages can be mapped.
...
> +void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
> + int prot, const xen_pfn_t *arr, int *err,
> + unsigned int num);
If it returns NULL, will all of *err be filled with errno values ?
Ie is all of err[] always written ?
Does this function ever set errno ?
I would write `int err_out[num]' in the prototype which is a bit
clearer about the semantics.
This API invites callers to fail to notice partial failures.
How about permitting err==NULL to mean `on partial failure, unmap
everything and return NULL setting errno' ?
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap.
2015-11-09 12:00 ` [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap Ian Campbell
@ 2015-11-11 15:14 ` Ian Jackson
2015-11-27 15:57 ` Ian Campbell
2015-11-13 16:16 ` Andrew Cooper
1 sibling, 1 reply; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:14 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap."):
> And require it be used instead of direct munmap.
>
> This will allow e.g. Valgrind hooks to help track incorrect use of
> foreign mappings.
>
> Switch all uses of xenforeignmemory_map to use
> xenforeignmemory_unmap, not that foreign mappings via the libxc compat
> xc_map_foreign_* interface will not take advantage of this and will
> need converting.
...
> +int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
> + void *addr, unsigned int num);
> +
What kind of errors might this return ?
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 15/23] foreignmemory: use size_t for size arguments.
2015-11-09 12:00 ` [PATCH XEN v5 15/23] foreignmemory: use size_t for size arguments Ian Campbell
@ 2015-11-11 15:15 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:15 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 15/23] foreignmemory: use size_t for size arguments."):
> Surprisingly it appears no callers need updating.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
> Acked-by: Wei Liu <wei.liu2@citrix.com>
This is a good API change.
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 16/23] tools/libs/evtchn: Review and update doc comments.
2015-11-09 12:00 ` [PATCH XEN v5 16/23] tools/libs/evtchn: Review and update doc comments Ian Campbell
@ 2015-11-11 15:15 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:15 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 16/23] tools/libs/evtchn: Review and update doc comments."):
> Remove the reference to pre-4.1, since this is now a new library.
>
> Fixup references to xc.
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 18/23] tools/libs/gnttab: Review and update doc comments.
2015-11-09 12:00 ` [PATCH XEN v5 18/23] tools/libs/gnttab: Review and update doc comments Ian Campbell
@ 2015-11-11 15:17 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:17 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 18/23] tools/libs/gnttab: Review and update doc comments."):
> Remove some stray xc references.
>
> While I'm not convinced by javadoc/doxygen cause the existing comments
> which appear to use that syntax to have the appropriate /** marker.
>
> Also fix a typo in a code comment.
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
While you're here you might
> diff --git a/tools/libs/gnttab/linux.c b/tools/libs/gnttab/linux.c
> index a76f17f..bdf2a42 100644
> --- a/tools/libs/gnttab/linux.c
> +++ b/tools/libs/gnttab/linux.c
> @@ -136,7 +136,7 @@ void *osdep_gnttab_grant_map(xengnttab_handle *xgt,
> * swapped out. Since the paging daemon may be in the same domain\
, the
> * hypercall cannot block without causing a deadlock.
> *
> - * Because there are no notificaitons when the page is swapped in\
, wait
> + * Because there are no notifications when the page is swapped in\
, wait
> * a bit before retrying, and hope that the page will arrive even\
tually.
> */
rewrap this.
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 21/23] tools/libs/foreignmemory: Mention restrictions on fork in docs.
2015-11-09 12:00 ` [PATCH XEN v5 21/23] tools/libs/foreignmemory: Mention restrictions on fork in docs Ian Campbell
@ 2015-11-11 15:18 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:18 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 21/23] tools/libs/foreignmemory: Mention restrictions on fork in docs."):
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
> Acked-by: Wei Liu <wei.liu2@citrix.com>
Oh, good.
> /*
> - * Return a handle onto the hypercall driver. Logs errors.
> + * Return a handle onto the foreign memory mapping driver. Logs errors.
> + *
> + * Note: After fork a child process must not use any opened handle
> + * inherited from their parent. Any existing mappings should be
> + * assumed to become invalid. Therefore children must open a new
> + * handle and make new mappings if necessary.
However, you still need to say how to reclaim the virtual address
space. (See my comments on the gntmap API.)
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH QEMU-XEN-TRADITIONAL v5 2/5] qemu-xen-traditional: Use libxenevtchn
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 2/5] qemu-xen-traditional: Use libxenevtchn Ian Campbell
@ 2015-11-11 15:19 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:19 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH QEMU-XEN-TRADITIONAL v5 2/5] qemu-xen-traditional: Use libxenevtchn"):
> /dev/xen/evtchn related wrappers have been moved out of libxenctrl
> into their own library.
>
> Note that i386-dm/helper2.c's xc_interface * was always really an
> xc_evtchn *, it's just they used to be typedefs to the same thing...
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 00/23] Begin to disentangle libxenctrl and provide some stable libraries
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
` (22 preceding siblings ...)
2015-11-09 12:00 ` [PATCH XEN v5 23/23] HACK: Update Config.mk to pull all the right bits from my xenbits trees Ian Campbell
@ 2015-11-11 15:20 ` Ian Jackson
23 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:20 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH XEN v5 00/23] Begin to disentangle libxenctrl and provide some stable libraries"):
> We intend to stabilise some parts of the libxenctrl interface by
> splitting out some functionality into separate stable libraries.
I have, I think, now reviewed all of the proposed new stable APIs.
This is good work, thanks.
I have made a few comments on the individual patches, mostly about
error handling (ah, my favourite topic).
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH QEMU-XEN-TRADITIONAL v5 3/5] qemu-xen-traditional: Use libxengnttab
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 3/5] qemu-xen-traditional: Use libxengnttab Ian Campbell
@ 2015-11-11 15:20 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:20 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH QEMU-XEN-TRADITIONAL v5 3/5] qemu-xen-traditional: Use libxengnttab"):
> /dev/xen/gntdev related wrappers have been moved out of libxenctrl
> into their own library.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH QEMU-XEN-TRADITIONAL v5 4/5] qemu-xen-traditional: Add libxencall to rpath-link
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 4/5] qemu-xen-traditional: Add libxencall to rpath-link Ian Campbell
@ 2015-11-11 15:20 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:20 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH QEMU-XEN-TRADITIONAL v5 4/5] qemu-xen-traditional: Add libxencall to rpath-link"):
> libxenctrl links against this library
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH QEMU-XEN-TRADITIONAL v5 5/5] qemu-xen-traditional: Add libxenforeignmemory to rpath-link
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 5/5] qemu-xen-traditional: Add libxenforeignmemory " Ian Campbell
@ 2015-11-11 15:21 ` Ian Jackson
0 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:21 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH QEMU-XEN-TRADITIONAL v5 5/5] qemu-xen-traditional: Add libxenforeignmemory to rpath-link"):
> libxenctrl links against this library.
>
> Also, request the compat xc_map_foreign API from libxc.
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
` (4 preceding siblings ...)
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 5/5] qemu-xen-traditional: Add libxenforeignmemory " Ian Campbell
@ 2015-11-11 15:23 ` Ian Jackson
5 siblings, 0 replies; 111+ messages in thread
From: Ian Jackson @ 2015-11-11 15:23 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("[PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries"):
> We intend to stabilise some parts of the libxenctrl interface by
> splitting out some functionality into separate stable libraries.
>
> This is the qemu-xen-traditional part of the first phase of that change.
I have given this a fairly cursory review and it all looks very dull,
as I would expect.
Thanks,
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-11 15:08 ` Ian Jackson
@ 2015-11-11 15:31 ` Wei Liu
2015-11-13 15:17 ` Ian Campbell
2015-11-13 15:16 ` Ian Campbell
1 sibling, 1 reply; 111+ messages in thread
From: Wei Liu @ 2015-11-11 15:31 UTC (permalink / raw)
To: Ian Jackson; +Cc: wei.liu2, Ian Campbell, xen-devel
On Wed, Nov 11, 2015 at 03:08:13PM +0000, Ian Jackson wrote:
> Ian Campbell writes ("[PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall."):
> > libxencall will provide a stable API and ABI for calling hypercalls
> > (although those hypercalls themselves may not have a stable API). As
> > well as the hypercall buffer infrastructure needed in order to safely
> > provide pointer arguments to hypercalls.
> ...
> > +/*
> > + * This library allows you to make arbitrary hypercalls (subject to
> > + * sufficient permission for the process and the domain itself). Note
> > + * that while the library interface is stable the hypercalls are
> > + * subject to their own rules.
>
> Something needs to say what the error handling is like.
>
> Do these functions set errno ?
>
> > +/*
> > + * Call hypercalls with varying numbers of arguments.
> > + */
> > +int xencall0(xencall_handle *xcall, unsigned int op);
>
> Is the return value the raw hypercall return value, or is hypervisor
> do_foo returning -EFOOBAR turned into to -1/errno=EFOOBAR ?
> (Hopefully the answer to this doesn't depend on the hypercall ABI...)
>
That is -1/errno=E_XEN_FOOBAR across all OSes.
(I think this stems from Linux ioctl semantics. Then NetBSD etc follow
suit.)
Never noticed before the error handling is not documented though. I
agree it should be documented. :-)
> > +/*
> > + * Allocate and free memory which is suitable for use as a pointer
> > + * argument to a hypercall.
> > + */
> > +void *xencall_alloc_buffer_pages(xencall_handle *xcall, int nr_pages);
> > +void xencall_free_buffer_pages(xencall_handle *xcall, void *p, int nr_pages);
> > +
> > +void *xencall_alloc_buffer(xencall_handle *xcall, size_t size);
> > +void xencall_free_buffer(xencall_handle *xcall, void *p);
>
> See above re error handling.
>
> Can these functions be used without (a) knowing the page size
> (b) a rounding macro ?
>
> It would be best to save callers the trouble of providing those
> themselves.
>
> Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-11 15:03 ` Ian Jackson
@ 2015-11-13 15:02 ` Ian Campbell
2015-11-13 20:38 ` Daniel De Graaf
2015-11-23 17:05 ` Ian Campbell
1 sibling, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-13 15:02 UTC (permalink / raw)
To: Ian Jackson, Daniel De Graaf; +Cc: wei.liu2, xen-devel
On Wed, 2015-11-11 at 15:03 +0000, Ian Jackson wrote:
> Ian Campbell writes ("[PATCH XEN v5 07/23] tools: Refactor
> /dev/xen/gnt{dev,shr} wrappers into libxengnttab."):
> > libxengnttab will provide a stable API and ABI for accessing the
> > grant table devices.
> >
> > The functions are moved into the xengnt{tab,shr} namespace to make a
> > clean break from libxc and avoid ambiguity regarding which interfaces
> > are stable.
>
> I have reviewed the API. Mostly this produced questions...
Yes, thanks this is the sort of feedback I was looking for before we call
this a stable interface.
Some of my answers are of the "yes, we should decide that" variety.
Daniel, I've added you because some of the questions below relate to the
notification mechanism and to gntshr which I don't really understand (not
that I really understand gnttab either).
> > XXX consider combining into a single namespace (i.e. with
> > xengnttab_handle having two open fd's in it on Linux)
>
> Arguments in favour of combining:
> * Some kernel might provide a single thing in /dev, and might even
> have trouble coping with a process which opens it more than once.
> * This is less fiddly for callers.
>
> Arguments against combining:
> * Perhaps some OSes will exist which can do only one and not the
> other. (Tenuous, and a workaround is possible.)
> * Slightly lower performance for a client which uses only one of the
> interfaces. (Barrel-scraping, here.)
> * It is more work.
>
> I conclude it would be better to unify them.
OK, I'm convinced.
> > +/*
> > + * Note:
> > + * After fork a child process must not use any opened xc gnttab
> > + * handle inherited from their parent. They must open a new handle if
> > + * they want to interact with xc.
>
> This should document
> (a) that it is safe to close the handle
> (b) that it is not safe to access grant mapped areas in the child
These two are easy.
> (c) how to reclaim the address space which is otherwise lost
The "obviously going to work, but probably not very useful" answer is you
should reclaim them before forking. That's clearly not practical in any
number of situations.
I really not sure what would happen if the child did a gnttab_close(). I
_think_ the kernel side ought to do the right thing (in that my opinion is
that really it should, and looking at the code it seems like it might).
Making a statement here that gnttab_close() is OK (and the right thing) to
do in the child might make the kernel driver insta-buggy. That might be
tolerable though, no one seems too be bothered by such bugs today and if
someone trips over them we can fix them.
> > +/**
> > + * Memory maps one or more grant references from one or more domains
> > to a
> > + * contiguous local address range. Mappings should be unmapped with
> > + * xengnttab_munmap. Logs errors.
> > + *
> > + * @parm xgt a handle on an open grant table interface
> > + * @parm count the number of grant references to be mapped
> > + * @parm domids an array of @count domain IDs by which the
> > corresponding @refs
> > + * were granted
> > + * @parm refs an array of @count grant references to be mapped
> > + * @parm prot same flag as in mmap()
> > + */
> > +void *xengnttab_map_grant_refs(xengnttab_handle *xgt,
>
> Should explain what happens on partial failure.
>
> > +/**
> > + * Memory maps one or more grant references from one domain to a
> > + * contiguous local address range. Mappings should be unmapped with
> > + * xengnttab_munmap. Logs errors.
> ...
> > +void *xengnttab_map_domain_grant_refs(xengnttab_handle *xgt,
>
> Should say that this is equivalent to a suitable call to
> xengnttab_map_grant_refs. (Assuming that it is.)
I believe so.
Which might argue for not having both, but I think avoiding the common case
of needing to provide an array full of the same domid makes it worthwhile
to have this.
> > +/**
> > + * Memory maps a grant reference from one domain to a local address
> > range.
> > + * Mappings should be unmapped with xengnttab_munmap. If notify_offset
> > or
> > + * notify_port are not -1, this version will attempt to set up an
> > unmap
> > + * notification at the given offset and event channel. When the page
> > is
> > + * unmapped, the byte at the given offset will be zeroed and a wakeup
> > will be
> > + * sent to the given event channel. Logs errors.
>
> What happens if the unmap notification cannot be set up ?
>
> Also "when the page is unmapped" makes it sound like you mean
> xengnttab_munmap but actually I think this is when the grant is
> withdrawn by the grantor ?
>
> If the grant is withdrawn by the grantor, does the page become
> unuseable immediately ? If so, how can anyone ever use this safely ?
Daniel, could you answer these ones please.
>
> > +/*
> > + * Sets the maximum number of grants that may be mapped by the given
> > instance
> > + * to @count. Never logs.
>
> Line wrap.
>
>
> > +int xengnttab_set_max_grants(xengnttab_handle *xgt,
> > + uint32_t count);
>
> Is count in pages or grants or what ?
It was hard to find this out, because the underlying ioctl doesn't exist in
the Linux side. In the end I found it is in grants as the name suggests.
AFAICT this is implemented only by linux-2.6.18-xen.hg and mini-os.git. We
could consider omitting it from the initial API. QEMU calls it but doesn't
care if it didn't work.
> > +/*
> > + * Return an fd onto the grant sharing driver. Logs errors.
>
> It does not return an fd.
Oh yes.
> (See also my comments on gntmap.)
I'm not sure which ones you mean here. If you meant gnttab then I'm still
not sure since you didn't comment on _open() there, but that's the only
commonality in the API.
>
> > +/*
> > + * Creates and shares pages with another domain.
> > + *
> ...
> > +void *xengntshr_share_pages(xengntshr_handle *xgs, uint32_t domid,
> > + int count, uint32_t *refs, int writable);
>
> Can this fail ? Can it partially succeed ?
Daniel?
>
> > +/*
> > + * Creates and shares a page with another domain, with unmap
> > notification.
> > + *
> > + * @parm xgs a handle to an open grant sharing instance
> > + * @parm domid the domain to share memory with
> > + * @parm refs the grant reference of the pages (output)
> > + * @parm writable true if the other domain can write to the page
> > + * @parm notify_offset The byte offset in the page to use for unmap
> > + * notification; -1 for none.
> > + * @parm notify_port The event channel port to use for unmap notify,
> > or -1
> > + * @return local mapping of the page
> > + */
> > +void *xengntshr_share_page_notify(xengntshr_handle *xgs, uint32_t
> > domid,
>
> What is this `unmap notification' ?
Daniel?
>
> > +/*
> > + * Unmaps the @count pages starting at @start_address, which were
> > mapped by a
> > + * call to xengntshr_share_*. Never logs.
>
> Linewrap in the comment.
>
> > + */
> > +int xengntshr_munmap(xengntshr_handle *xgs, void *start_address,
> > uint32_t count);
>
> What effect does this have on the peer ?
Daniel?
>
> Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 02/23] tools: Refactor "xentoollog" into its own library
2015-11-09 12:00 ` [PATCH XEN v5 02/23] tools: Refactor "xentoollog" into its own library Ian Campbell
@ 2015-11-13 15:05 ` Andrew Cooper
2015-11-13 15:35 ` Ian Campbell
0 siblings, 1 reply; 111+ messages in thread
From: Andrew Cooper @ 2015-11-13 15:05 UTC (permalink / raw)
To: Ian Campbell, ian.jackson, wei.liu2, xen-devel
On 09/11/15 12:00, Ian Campbell wrote:
> diff --git a/tools/ocaml/Makefile.rules b/tools/ocaml/Makefile.rules
> index 1796060..45e71f2 100644
> --- a/tools/ocaml/Makefile.rules
> +++ b/tools/ocaml/Makefile.rules
> @@ -1,18 +1,18 @@
> -ifdef V
> - ifeq ("$(origin V)", "command line")
> - BUILD_VERBOSE = $(V)
> - endif
> -endif
> -ifndef BUILD_VERBOSE
> - BUILD_VERBOSE = 0
> -endif
> -ifeq ($(BUILD_VERBOSE),1)
> +#ifdef V
> +# ifeq ("$(origin V)", "command line")
> +# BUILD_VERBOSE = $(V)
> +# endif
> +#endif
> +#ifndef BUILD_VERBOSE
> +# BUILD_VERBOSE = 0
> +#endif
> +#ifeq ($(BUILD_VERBOSE),1)
> E = @true
> Q =
> -else
> - E = @echo
> - Q = @
> -endif
> +#else
> +# E = @echo
> +# Q = @
> +#endif
This hunk looks like temporary debugging?
~Andrew
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-11 15:08 ` Ian Jackson
2015-11-11 15:31 ` Wei Liu
@ 2015-11-13 15:16 ` Ian Campbell
2015-11-13 16:56 ` Ian Campbell
2015-11-13 16:57 ` Ian Campbell
1 sibling, 2 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-13 15:16 UTC (permalink / raw)
To: Ian Jackson, Roger Pau Monne; +Cc: wei.liu2, xen-devel
On Wed, 2015-11-11 at 15:08 +0000, Ian Jackson wrote:
> Ian Campbell writes ("[PATCH XEN v5 09/23] tools: Refactor hypercall
> calling wrappers into libxencall."):
> > libxencall will provide a stable API and ABI for calling hypercalls
> > (although those hypercalls themselves may not have a stable API). As
> > well as the hypercall buffer infrastructure needed in order to safely
> > provide pointer arguments to hypercalls.
> ...
> > +/*
> > + * This library allows you to make arbitrary hypercalls (subject to
> > + * sufficient permission for the process and the domain itself). Note
> > + * that while the library interface is stable the hypercalls are
> > + * subject to their own rules.
>
> Something needs to say what the error handling is like.
>
> Do these functions set errno ?
Yes, they should have the customary "return -1 and set errno" pattern. I
shall make this explicit.
> > +/*
> > + * Call hypercalls with varying numbers of arguments.
> > + */
> > +int xencall0(xencall_handle *xcall, unsigned int op);
>
> Is the return value the raw hypercall return value, or is hypervisor
> do_foo returning -EFOOBAR turned into to -1/errno=EFOOBAR ?
> (Hopefully the answer to this doesn't depend on the hypercall ABI...)
Oh god, this looks to vary by platform :-(
Linux returns the hypercall return value directly from the ioctl as its
return value, which the usual libc stuff turns into -1/errno=EFOO on
failure.
FreeBSD has a separate hypervisor retval field in the ioctl argument struct
and does:
return (ret == 0) ? hypercall->retval : ret;
MiniOS goes for -1/errno=EFOO.
What a mess.
Either Linux+MiniOS or FreeBSD are going to have to change (i.e. in the
wrappers in this library, not the underlying mechanisms) to match the other
for consistency.
On FreeBSD hypercall->retval is in the XEN_ERRNO_* space. But I think we
probably want to map onto the FreeBSD ERRNO space, IIRC we decided on this
approach, rather than converting all our callers to the XEN_ERRNO_* space,
in a previous discussion.
If we are going to standardise on one of those then given we've gone for
-1/errno in most other places and it's the "normal" way to do things I
think we should do the same here and the freebsd driver in libxencall
should set errno.
There are other alternatives, such as having this library return the
"privcmd ioctl" result in its return value + errno and the "hypercall
result" in a separate output pointer argument. This would require, possibly
substantial, changes in the callers, but right now they are all in libxc,
which could munge that back into the current somewhat braindead scheme
allowing us to update those call paths piecemeal and allowing new users to
use the sane interface.
Thoughts?
> > +/*
> > + * Allocate and free memory which is suitable for use as a pointer
> > + * argument to a hypercall.
> > + */
> > +void *xencall_alloc_buffer_pages(xencall_handle *xcall, int nr_pages);
> > +void xencall_free_buffer_pages(xencall_handle *xcall, void *p, int
> > nr_pages);
> > +
> > +void *xencall_alloc_buffer(xencall_handle *xcall, size_t size);
> > +void xencall_free_buffer(xencall_handle *xcall, void *p);
>
> See above re error handling.
>
> Can these functions be used without (a) knowing the page size
> (b) a rounding macro ?
>
> It would be best to save callers the trouble of providing those
> themselves.
The library deals with this internally, size can be anything the caller
likes.
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-11 15:31 ` Wei Liu
@ 2015-11-13 15:17 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-13 15:17 UTC (permalink / raw)
To: Wei Liu, Ian Jackson; +Cc: xen-devel
On Wed, 2015-11-11 at 15:31 +0000, Wei Liu wrote:
> On Wed, Nov 11, 2015 at 03:08:13PM +0000, Ian Jackson wrote:
> > Ian Campbell writes ("[PATCH XEN v5 09/23] tools: Refactor hypercall
> > calling wrappers into libxencall."):
> > > libxencall will provide a stable API and ABI for calling hypercalls
> > > (although those hypercalls themselves may not have a stable API). As
> > > well as the hypercall buffer infrastructure needed in order to safely
> > > provide pointer arguments to hypercalls.
> > ...
> > > +/*
> > > + * This library allows you to make arbitrary hypercalls (subject to
> > > + * sufficient permission for the process and the domain itself).
> > > Note
> > > + * that while the library interface is stable the hypercalls are
> > > + * subject to their own rules.
> >
> > Something needs to say what the error handling is like.
> >
> > Do these functions set errno ?
> >
> > > +/*
> > > + * Call hypercalls with varying numbers of arguments.
> > > + */
> > > +int xencall0(xencall_handle *xcall, unsigned int op);
> >
> > Is the return value the raw hypercall return value, or is hypervisor
> > do_foo returning -EFOOBAR turned into to -1/errno=EFOOBAR ?
> > (Hopefully the answer to this doesn't depend on the hypercall ABI...)
> >
>
> That is -1/errno=E_XEN_FOOBAR across all OSes.
>
> (I think this stems from Linux ioctl semantics. Then NetBSD etc follow
> suit.)
I wish it were that simple :-(
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
2015-11-11 15:13 ` Ian Jackson
@ 2015-11-13 15:21 ` Ian Campbell
2015-11-13 15:27 ` Ian Jackson
2015-11-27 15:46 ` Ian Campbell
1 sibling, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-13 15:21 UTC (permalink / raw)
To: Ian Jackson; +Cc: wei.liu2, xen-devel
On Wed, 2015-11-11 at 15:13 +0000, Ian Jackson wrote:
> Ian Campbell writes ("[PATCH XEN v5 13/23] tools: Refactor foreign memory
> mapping into libxenforeignmemory"):
> > libxenforeignmemory will provide a stable API and ABI for mapping
> > foreign domain memory (subject to appropriate privileges).
> >
> > The new library exposes an interface equivalent to
> > xc_map_foreign_memory_bulk, which all the other
> > xc_map_foreign_memory_* functions (which remain in libxc) are
> > implemented in terms of.
> ...
> > diff --git a/tools/libs/foreignmemory/include/xenforeignmemory.h
> > b/tools/libs/foreignmemory/include/xenforeignmemory.h
>
> > +/*
> > + * Return a handle onto the hypercall driver. Logs errors.
> > + */
> > +xenforeignmemory_handle *xenforeignmemory_open(xentoollog_logger
> > *logger,
> > + unsigned open_flags);
>
> `onto the foreign memory driver'.
>
> This file should probably say something about fork(). Perhaps just a
> refernce to the rules in the grant map driver.
Yes, same rules I think.
>
> > +/*
> > + * Maps a range within one domain to a local address range. Mappings
> > + * should be unmapped with munmap and should follow the same rules as
> > mmap
> > + * regarding page alignment.
> > + *
> > + * prot is as for mmap(2).
> > + *
> > + * Can partially succeed. When a page cannot be mapped, its respective
> > + * field in @err is set to the corresponding errno value.
> > + *
> > + * Returns NULL if no pages can be mapped.
> ...
> > +void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t
> > dom,
> > + int prot, const xen_pfn_t *arr, int *err,
> > + unsigned int num);
>
> If it returns NULL, will all of *err be filled with errno values ?
> Ie is all of err[] always written ?
>
> Does this function ever set errno ?
I'll figure all these out and say.
> I would write `int err_out[num]' in the prototype which is a bit
> clearer about the semantics.
Are you suggesting:
void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
int prot, const xen_pfn_t *arr,
int err_out[num],
unsigned int num);
Is that (a var sized array based on another param) really allowed?
Or did you mean as comment like:
void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
int prot, const xen_pfn_t *arr,
int *err_out /*int err_out[num]*/,
unsigned
int num);
> This API invites callers to fail to notice partial failures.
>
> How about permitting err==NULL to mean `on partial failure, unmap
> everything and return NULL setting errno' ?
Not a bad idea.
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
2015-11-13 15:21 ` Ian Campbell
@ 2015-11-13 15:27 ` Ian Jackson
2015-11-27 17:18 ` Ian Campbell
0 siblings, 1 reply; 111+ messages in thread
From: Ian Jackson @ 2015-11-13 15:27 UTC (permalink / raw)
To: Ian Campbell; +Cc: wei.liu2, xen-devel
Ian Campbell writes ("Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory"):
> On Wed, 2015-11-11 at 15:13 +0000, Ian Jackson wrote:
> Are you suggesting:
>
> void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
> int prot, const xen_pfn_t *arr,
> int err_out[num],
> unsigned int num);
>
> Is that (a var sized array based on another param) really allowed?
Yes, if the array size precedes the array. Maybe you don't prefer
that.
> Or did you mean as comment like:
>
> void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
>
> int prot, const xen_pfn_t *arr,
>
> int *err_out /*int err_out[num]*/,
> unsigned
> int num);
If you don't want to swap the arguments, at the very least
int err_out[/*num*/]
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-09 12:00 ` [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall Ian Campbell
2015-11-11 15:08 ` Ian Jackson
@ 2015-11-13 15:35 ` Andrew Cooper
2015-11-13 15:49 ` Andrew Cooper
2 siblings, 0 replies; 111+ messages in thread
From: Andrew Cooper @ 2015-11-13 15:35 UTC (permalink / raw)
To: Ian Campbell, ian.jackson, wei.liu2, xen-devel
On 09/11/15 12:00, Ian Campbell wrote:
> diff --git a/tools/libs/call/include/xencall.h b/tools/libs/call/include/xencall.h
> new file mode 100644
> index 0000000..a0b3591
> --- /dev/null
> +++ b/tools/libs/call/include/xencall.h
> @@ -0,0 +1,84 @@
> +/*
> + * This library is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU Lesser General Public
> + * License as published by the Free Software Foundation;
> + * version 2.1 of the License.
> + *
> + * This library is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * Lesser General Public License for more details.
> + *
> + * You should have received a copy of the GNU Lesser General Public
> + * License along with this library; If not, see <http://www.gnu.org/licenses/>.
> + */
> +#ifndef XENCALL_H
> +#define XENCALL_H
> +
> +/*
> + * This library allows you to make arbitrary hypercalls (subject to
> + * sufficient permission for the process and the domain itself). Note
> + * that while the library interface is stable the hypercalls are
> + * subject to their own rules.
> + */
> +
> +#include <stdint.h>
> +#include <stddef.h>
> +
> +/* Callers who don't care don't need to #include <xentoollog.h> */
> +typedef struct xentoollog_logger xentoollog_logger;
> +
> +typedef struct xencall_handle xencall_handle;
> +
> +/*
> + */
> +#define XENCALL_OPENFLAG_NON_REENTRANT (1U<<0)
> +
> +/*
> + * Return a handle onto the hypercall driver. Logs errors.
> + */
> +xencall_handle *xencall_open(xentoollog_logger *logger, unsigned open_flags);
> +
> +/*
> + * Close a handle previously allocated with xencall_open().
> + */
> +int xencall_close(xencall_handle *xcall);
> +
> +/*
> + * Call hypercalls with varying numbers of arguments.
> + */
> +int xencall0(xencall_handle *xcall, unsigned int op);
> +int xencall1(xencall_handle *xcall, unsigned int op,
> + uint64_t arg1);
> +int xencall2(xencall_handle *xcall, unsigned int op,
> + uint64_t arg1, uint64_t arg2);
> +int xencall3(xencall_handle *xcall, unsigned int op,
> + uint64_t arg1, uint64_t arg2, uint64_t arg3);
> +int xencall4(xencall_handle *xcall, unsigned int op,
> + uint64_t arg1, uint64_t arg2, uint64_t arg3,
> + uint64_t arg4);
> +int xencall5(xencall_handle *xcall, unsigned int op,
> + uint64_t arg1, uint64_t arg2, uint64_t arg3,
> + uint64_t arg4, uint64_t arg5);
> +
> +/*
> + * Allocate and free memory which is suitable for use as a pointer
> + * argument to a hypercall.
> + */
> +void *xencall_alloc_buffer_pages(xencall_handle *xcall, int nr_pages);
> +void xencall_free_buffer_pages(xencall_handle *xcall, void *p, int nr_pages);
size_t nr_pages. (The osdep internals appear to use unsigned nr_pages,
just to be different).
~Andrew
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 02/23] tools: Refactor "xentoollog" into its own library
2015-11-13 15:05 ` Andrew Cooper
@ 2015-11-13 15:35 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-13 15:35 UTC (permalink / raw)
To: Andrew Cooper, ian.jackson, wei.liu2, xen-devel
On Fri, 2015-11-13 at 15:05 +0000, Andrew Cooper wrote:
> On 09/11/15 12:00, Ian Campbell wrote:
> > diff --git a/tools/ocaml/Makefile.rules b/tools/ocaml/Makefile.rules
> > index 1796060..45e71f2 100644
> > --- a/tools/ocaml/Makefile.rules
> > +++ b/tools/ocaml/Makefile.rules
> > @@ -1,18 +1,18 @@
> > -ifdef V
> > - ifeq ("$(origin V)", "command line")
> > - BUILD_VERBOSE = $(V)
> > - endif
> > -endif
> > -ifndef BUILD_VERBOSE
> > - BUILD_VERBOSE = 0
> > -endif
> > -ifeq ($(BUILD_VERBOSE),1)
> > +#ifdef V
> > +# ifeq ("$(origin V)", "command line")
> > +# BUILD_VERBOSE = $(V)
> > +# endif
> > +#endif
> > +#ifndef BUILD_VERBOSE
> > +# BUILD_VERBOSE = 0
> > +#endif
> > +#ifeq ($(BUILD_VERBOSE),1)
> > E = @true
> > Q =
> > -else
> > - E = @echo
> > - Q = @
> > -endif
> > +#else
> > +# E = @echo
> > +# Q = @
> > +#endif
>
> This hunk looks like temporary debugging?
It does rather, I've no idea why or what I was doing.
Dropped.
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-09 12:00 ` [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall Ian Campbell
2015-11-11 15:08 ` Ian Jackson
2015-11-13 15:35 ` Andrew Cooper
@ 2015-11-13 15:49 ` Andrew Cooper
2015-12-02 14:55 ` Ian Campbell
2 siblings, 1 reply; 111+ messages in thread
From: Andrew Cooper @ 2015-11-13 15:49 UTC (permalink / raw)
To: Ian Campbell, ian.jackson, wei.liu2, xen-devel
On 09/11/15 12:00, Ian Campbell wrote:
> diff --git a/tools/libs/call/linux.c b/tools/libs/call/linux.c
> new file mode 100644
> index 0000000..906ca7e
> --- /dev/null
> +++ b/tools/libs/call/linux.c
> @@ -0,0 +1,132 @@
> +/*
> + * This library is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU Lesser General Public
> + * License as published by the Free Software Foundation;
> + * version 2.1 of the License.
> + *
> + * This library is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * Lesser General Public License for more details.
> + *
> + * You should have received a copy of the GNU Lesser General Public
> + * License along with this library; If not, see <http://www.gnu.org/licenses/>.
> + *
> + * Split out from xc_linus_osdep.c:
> + *
> + * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
> + */
> +
> +#include <errno.h>
> +#include <fcntl.h>
> +#include <unistd.h>
> +
> +#include <sys/mman.h>
> +#include <sys/ioctl.h>
> +
> +#include "private.h"
> +
> +int osdep_xencall_open(xencall_handle *xcall)
> +{
> + int flags, saved_errno;
> + int fd = open("/proc/xen/privcmd", O_RDWR);
I know this is a straight copy, but these new libraries really should
start with the PVOps interface of /dev/xen/privcmd before falling back
to the older classic-linux interfaces under /proc
Ideally, once all these library improvements are done, there should be
no need to mount xenfs in a Linux PVops domain.
> +
> + if ( fd == -1 )
> + {
> + PERROR("Could not obtain handle on privileged command interface");
> + return -1;
> + }
> +
> + /* Although we return the file handle as the 'xc handle' the API
> + does not specify / guarentee that this integer is in fact
> + a file handle. Thus we must take responsiblity to ensure
> + it doesn't propagate (ie leak) outside the process */
Again, I know this is a straight copy, but it is racy. The open() call
needs O_CLOEXEC.
There might be some issues here to do with older versions of libc.
> + if ( (flags = fcntl(fd, F_GETFD)) < 0 )
> + {
> + PERROR("Could not get file handle flags");
> + goto error;
> + }
> +
> + flags |= FD_CLOEXEC;
> +
> + if ( fcntl(fd, F_SETFD, flags) < 0 )
> + {
> + PERROR("Could not set file handle flags");
> + goto error;
> + }
> +
> + xcall->fd = fd;
> + return 0;
> +
> + error:
> + saved_errno = errno;
> + close(fd);
> + errno = saved_errno;
> + return -1;
> +}
> +
> +int osdep_xencall_close(xencall_handle *xcall)
> +{
> + int fd = xcall->fd;
> + if (fd == -1)
> + return 0;
> + return close(fd);
> +}
> +
> +int osdep_hypercall(xencall_handle *xcall, privcmd_hypercall_t *hypercall)
> +{
> + return ioctl(xcall->fd, IOCTL_PRIVCMD_HYPERCALL, hypercall);
> +}
> +
> +void *osdep_alloc_pages(xencall_handle *xcall, unsigned int npages)
> +{
> + size_t size = npages * PAGE_SIZE;
> + void *p;
> + int rc, saved_errno;
> +
> + /* Address returned by mmap is page aligned. */
> + p = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_LOCKED, -1, 0);
> + if ( p == MAP_FAILED )
> + {
> + PERROR("xc_alloc_hypercall_buffer: mmap failed");
> + return NULL;
> + }
> +
> + /* Do not copy the VMA to child process on fork. Avoid the page being COW
> + on hypercall. */
> + rc = madvise(p, npages * PAGE_SIZE, MADV_DONTFORK);
This also seems racy, although it is unclear from the manpages how to
avoid the race.
~Andrew
> + if ( rc < 0 )
> + {
> + PERROR("xc_alloc_hypercall_buffer: madvise failed");
> + goto out;
> + }
> +
> + return p;
> +
> +out:
> + saved_errno = errno;
> + (void)munmap(p, size);
> + errno = saved_errno;
> + return NULL;
> +}
> +
> +void osdep_free_pages(xencall_handle *xcall, void *ptr, unsigned int npages)
> +{
> + int saved_errno = errno;
> + /* Recover the VMA flags. Maybe it's not necessary */
> + madvise(ptr, npages * PAGE_SIZE, MADV_DOFORK);
> +
> + munmap(ptr, npages * PAGE_SIZE);
> + /* We MUST propagate the hypercall errno, not unmap call's. */
> + errno = saved_errno;
> +}
> +
> +/*
> + * Local variables:
> + * mode: C
> + * c-file-style: "BSD"
> + * c-basic-offset: 4
> + * tab-width: 4
> + * indent-tabs-mode: nil
> + * End:
> + */
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap.
2015-11-09 12:00 ` [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap Ian Campbell
2015-11-11 15:14 ` Ian Jackson
@ 2015-11-13 16:16 ` Andrew Cooper
2015-11-13 16:17 ` Andrew Cooper
1 sibling, 1 reply; 111+ messages in thread
From: Andrew Cooper @ 2015-11-13 16:16 UTC (permalink / raw)
To: Ian Campbell, ian.jackson, wei.liu2, xen-devel
On 09/11/15 12:00, Ian Campbell wrote:
> diff --git a/tools/libs/foreignmemory/linux.c b/tools/libs/foreignmemory/linux.c
> index 01cd42e..86a5a97 100644
> --- a/tools/libs/foreignmemory/linux.c
> +++ b/tools/libs/foreignmemory/linux.c
> @@ -276,6 +276,12 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
> return addr;
> }
>
> +int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
> + void *addr, unsigned int num)
> +{
> + return munmap(addr, (unsigned long)num << PAGE_SHIFT);
This should be size_t to match the type expected by munmap()
> +}
> +
> /*
> * Local variables:
> * mode: C
> diff --git a/tools/libs/foreignmemory/minios.c b/tools/libs/foreignmemory/minios.c
> index 981d801..dbb152f 100644
> --- a/tools/libs/foreignmemory/minios.c
> +++ b/tools/libs/foreignmemory/minios.c
> @@ -51,6 +51,12 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
> return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
> }
>
> +int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
> + void *addr, unsigned int num)
> +{
> + return munmap(addr, num << PAGE_SHIFT);
This num (and freebsd/netbsd/solaris) needs upcasting before
multiplication, to avoid loosing the top 12 bits on 64bit platforms.
~Andrew
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap.
2015-11-13 16:16 ` Andrew Cooper
@ 2015-11-13 16:17 ` Andrew Cooper
0 siblings, 0 replies; 111+ messages in thread
From: Andrew Cooper @ 2015-11-13 16:17 UTC (permalink / raw)
To: Ian Campbell, ian.jackson, wei.liu2, xen-devel
On 13/11/15 16:16, Andrew Cooper wrote:
> On 09/11/15 12:00, Ian Campbell wrote:
>> diff --git a/tools/libs/foreignmemory/linux.c b/tools/libs/foreignmemory/linux.c
>> index 01cd42e..86a5a97 100644
>> --- a/tools/libs/foreignmemory/linux.c
>> +++ b/tools/libs/foreignmemory/linux.c
>> @@ -276,6 +276,12 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
>> return addr;
>> }
>>
>> +int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
>> + void *addr, unsigned int num)
>> +{
>> + return munmap(addr, (unsigned long)num << PAGE_SHIFT);
> This should be size_t to match the type expected by munmap()
>
>> +}
>> +
>> /*
>> * Local variables:
>> * mode: C
>> diff --git a/tools/libs/foreignmemory/minios.c b/tools/libs/foreignmemory/minios.c
>> index 981d801..dbb152f 100644
>> --- a/tools/libs/foreignmemory/minios.c
>> +++ b/tools/libs/foreignmemory/minios.c
>> @@ -51,6 +51,12 @@ void *xenforeignmemory_map(xenforeignmemory_handle *fmem,
>> return map_frames_ex(arr, num, 1, 0, 1, dom, err, pt_prot);
>> }
>>
>> +int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
>> + void *addr, unsigned int num)
>> +{
>> + return munmap(addr, num << PAGE_SHIFT);
> This num (and freebsd/netbsd/solaris) needs upcasting before
> multiplication, to avoid loosing the top 12 bits on 64bit platforms.
Sorry - ignore me. I hadn't looked at the subsequent patch while
reviewing this one.
~Andrew
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 19/23] tools/libs/call: Update some log messages to not refer to xc.
2015-11-09 12:00 ` [PATCH XEN v5 19/23] tools/libs/call: Update some log messages to not refer to xc Ian Campbell
@ 2015-11-13 16:20 ` Andrew Cooper
2015-11-23 13:10 ` Ian Campbell
0 siblings, 1 reply; 111+ messages in thread
From: Andrew Cooper @ 2015-11-13 16:20 UTC (permalink / raw)
To: Ian Campbell, ian.jackson, wei.liu2, xen-devel
On 09/11/15 12:00, Ian Campbell wrote:
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
> Acked-by: Wei Liu <wei.liu2@citrix.com>
I agree with the sentiment, but the error message refer to the
libxencall entry API, so a developer can match it back to the call in
their code.
~Andrew
> ---
> tools/libs/call/linux.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/tools/libs/call/linux.c b/tools/libs/call/linux.c
> index 906ca7e..80b505c 100644
> --- a/tools/libs/call/linux.c
> +++ b/tools/libs/call/linux.c
> @@ -88,7 +88,7 @@ void *osdep_alloc_pages(xencall_handle *xcall, unsigned int npages)
> p = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_LOCKED, -1, 0);
> if ( p == MAP_FAILED )
> {
> - PERROR("xc_alloc_hypercall_buffer: mmap failed");
> + PERROR("alloc_pages: mmap failed");
> return NULL;
> }
>
> @@ -97,7 +97,7 @@ void *osdep_alloc_pages(xencall_handle *xcall, unsigned int npages)
> rc = madvise(p, npages * PAGE_SIZE, MADV_DONTFORK);
> if ( rc < 0 )
> {
> - PERROR("xc_alloc_hypercall_buffer: madvise failed");
> + PERROR("alloc_pages: madvise failed");
> goto out;
> }
>
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-13 15:16 ` Ian Campbell
@ 2015-11-13 16:56 ` Ian Campbell
2015-11-13 16:57 ` Ian Campbell
1 sibling, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-13 16:56 UTC (permalink / raw)
To: Ian Jackson, Roger Pau Monne; +Cc: wei.liu2, xen-devel
On Fri, 2015-11-13 at 15:16 +0000, Ian Campbell wrote:
>
> If we are going to standardise on one of those then given we've gone for
> -1/errno in most other places and it's the "normal" way to do things I
> think we should do the same here and the freebsd driver in libxencall
> should set errno.
>
> There are other alternatives, such as having this library return the
> "privcmd ioctl" result in its return value + errno and the "hypercall
> result" in a separate output pointer argument. This would require, possibly
> substantial, changes in the callers, but right now they are all in libxc,
> which could munge that back into the current somewhat braindead scheme
> allowing us to update those call paths piecemeal and allowing new users to
> use the sane interface.
>
> Thoughts?
Ian and I discusses this IRL. We decided that an interface that required
the caller to check two error returns (the "privcmd ioctl result" and the
"hypercall result" was unwieldy).
As a result we've decided that -1/errno=EFOO is the best approach, but that
we should try and partition the errno space (so far as possible given the
existing hypercall interfaces) into "From hypercall" and "From OS machinery
for making hypercalls" where it matters.
For some cases the distinction may not be so important (e.g. EFAULT from
either the kernel reading the ioctl arg or the hypervisor reading the
hypercall arg). But in other cases it might be relevant (e.g. EPERM from
the hypervisor vs a permissions error relating to the privcmd fd having
been locked to a particular domain at the kernel/process level).
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-13 15:16 ` Ian Campbell
2015-11-13 16:56 ` Ian Campbell
@ 2015-11-13 16:57 ` Ian Campbell
2015-11-13 17:16 ` Roger Pau Monné
1 sibling, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-13 16:57 UTC (permalink / raw)
To: Ian Jackson, Roger Pau Monne; +Cc: wei.liu2, xen-devel
On Fri, 2015-11-13 at 15:16 +0000, Ian Campbell wrote:
>
> On FreeBSD hypercall->retval is in the XEN_ERRNO_* space. But I think we
> probably want to map onto the FreeBSD ERRNO space, IIRC we decided on this
> approach, rather than converting all our callers to the XEN_ERRNO_* space,
> in a previous discussion.
Roger, can you confirm whether or not something on FreeBSD translated the
XEN_ERRNO errors into the BSD errno space? If so, then where is it? I
looked in FreeBSD's privcmd.c and in libxc and failed to find it...
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-13 16:57 ` Ian Campbell
@ 2015-11-13 17:16 ` Roger Pau Monné
2015-11-13 17:23 ` Ian Campbell
0 siblings, 1 reply; 111+ messages in thread
From: Roger Pau Monné @ 2015-11-13 17:16 UTC (permalink / raw)
To: Ian Campbell, Ian Jackson; +Cc: wei.liu2, xen-devel
El 13/11/15 a les 17.57, Ian Campbell ha escrit:
> On Fri, 2015-11-13 at 15:16 +0000, Ian Campbell wrote:
>>
>> On FreeBSD hypercall->retval is in the XEN_ERRNO_* space. But I think we
>> probably want to map onto the FreeBSD ERRNO space, IIRC we decided on this
>> approach, rather than converting all our callers to the XEN_ERRNO_* space,
>> in a previous discussion.
>
> Roger, can you confirm whether or not something on FreeBSD translated the
> XEN_ERRNO errors into the BSD errno space? If so, then where is it? I
> looked in FreeBSD's privcmd.c and in libxc and failed to find it...
Yes, the translation is done in the privcmd driver:
https://github.com/freebsd/freebsd/blob/master/sys/dev/xen/privcmd/privcmd.c#L242
So errors returned from the ioctl (in errno) are in the BSD space. I
still have to implement the error codes distributed from the Xen public
headers, right now we have a custom Xen error translation:
https://github.com/freebsd/freebsd/blob/master/sys/xen/error.h
Roger.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-13 17:16 ` Roger Pau Monné
@ 2015-11-13 17:23 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-13 17:23 UTC (permalink / raw)
To: Roger Pau Monné, Ian Jackson; +Cc: wei.liu2, xen-devel
On Fri, 2015-11-13 at 18:16 +0100, Roger Pau Monné wrote:
> El 13/11/15 a les 17.57, Ian Campbell ha escrit:
> > On Fri, 2015-11-13 at 15:16 +0000, Ian Campbell wrote:
> > >
> > > On FreeBSD hypercall->retval is in the XEN_ERRNO_* space. But I think
> > > we
> > > probably want to map onto the FreeBSD ERRNO space, IIRC we decided on
> > > this
> > > approach, rather than converting all our callers to the XEN_ERRNO_*
> > > space,
> > > in a previous discussion.
> >
> > Roger, can you confirm whether or not something on FreeBSD translated
> > the
> > XEN_ERRNO errors into the BSD errno space? If so, then where is it? I
> > looked in FreeBSD's privcmd.c and in libxc and failed to find it...
>
> Yes, the translation is done in the privcmd driver:
>
> https://github.com/freebsd/freebsd/blob/master/sys/dev/xen/privcmd/privcm
> d.c#L242
Thanks, I was looking for a big table but it's out of line in another file
so I missed it.
> So errors returned from the ioctl (in errno) are in the BSD space. I
> still have to implement the error codes distributed from the Xen public
> headers, right now we have a custom Xen error translation:
>
> https://github.com/freebsd/freebsd/blob/master/sys/xen/error.h
>
> Roger.
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-13 15:02 ` Ian Campbell
@ 2015-11-13 20:38 ` Daniel De Graaf
2015-11-16 12:30 ` Ian Campbell
0 siblings, 1 reply; 111+ messages in thread
From: Daniel De Graaf @ 2015-11-13 20:38 UTC (permalink / raw)
To: Ian Campbell, Ian Jackson; +Cc: wei.liu2, xen-devel
On 13/11/15 10:02, Ian Campbell wrote:
> On Wed, 2015-11-11 at 15:03 +0000, Ian Jackson wrote:
>> Ian Campbell writes ("[PATCH XEN v5 07/23] tools: Refactor
>> /dev/xen/gnt{dev,shr} wrappers into libxengnttab."):
>>> libxengnttab will provide a stable API and ABI for accessing the
>>> grant table devices.
>>>
>>> The functions are moved into the xengnt{tab,shr} namespace to make a
>>> clean break from libxc and avoid ambiguity regarding which interfaces
>>> are stable.
>>
>> I have reviewed the API. Mostly this produced questions...
>
> Yes, thanks this is the sort of feedback I was looking for before we call
> this a stable interface.
>
> Some of my answers are of the "yes, we should decide that" variety.
>
> Daniel, I've added you because some of the questions below relate to the
> notification mechanism and to gntshr which I don't really understand (not
> that I really understand gnttab either).
Responses below. I also agree that unifying the interfaces would be
better.
[...]
>>> +/**
>>> + * Memory maps a grant reference from one domain to a local address
>>> range.
>>> + * Mappings should be unmapped with xengnttab_munmap. If notify_offset
>>> or
>>> + * notify_port are not -1, this version will attempt to set up an
>>> unmap
>>> + * notification at the given offset and event channel. When the page
>>> is
>>> + * unmapped, the byte at the given offset will be zeroed and a wakeup
>>> will be
>>> + * sent to the given event channel. Logs errors.
>>
>> What happens if the unmap notification cannot be set up ?
>>
>> Also "when the page is unmapped" makes it sound like you mean
>> xengnttab_munmap but actually I think this is when the grant is
>> withdrawn by the grantor ?
>>
>> If the grant is withdrawn by the grantor, does the page become
>> unuseable immediately ? If so, how can anyone ever use this safely ?
>
> Daniel, could you answer these ones please.
This is intended to allow the kernel to send a close-request notification
when the application that allocated the grant page exits without calling
a proper shutdown (i.e. it crashes, calls _exit, calls execve, etc).
Without this signal, the kernel has no way to request that the mapper of
the page release it, and since Xen has no grant revocation mechanism, the
page will likely be tied up until the process on the other side is told to
release the page through some other method.
>>> +/*
>>> + * Creates and shares pages with another domain.
>>> + *
>> ...
>>> +void *xengntshr_share_pages(xengntshr_handle *xgs, uint32_t domid,
>>> + int count, uint32_t *refs, int writable);
>>
>> Can this fail ? Can it partially succeed ?
>
> Daniel?
It can fail if you are out of pages to grant (there is a module parameter
that can be adjusted via sysfs for the maximum), or in the unlikely case
that the kernel itself is out of room in its grant mapping table (or if
the syscall itself encounters -ENOMEM).
>>> +/*
>>> + * Creates and shares a page with another domain, with unmap
>>> notification.
>>> + *
>>> + * @parm xgs a handle to an open grant sharing instance
>>> + * @parm domid the domain to share memory with
>>> + * @parm refs the grant reference of the pages (output)
>>> + * @parm writable true if the other domain can write to the page
>>> + * @parm notify_offset The byte offset in the page to use for unmap
>>> + * notification; -1 for none.
>>> + * @parm notify_port The event channel port to use for unmap notify,
>>> or -1
>>> + * @return local mapping of the page
>>> + */
>>> +void *xengntshr_share_page_notify(xengntshr_handle *xgs, uint32_t
>>> domid,
>>
>> What is this `unmap notification' ?
>
> Daniel?
As mentioned above, it is a way for the kernel to request for the other
side to unmap a page. It is probably easiest to understand if looking
at the libvchan driver: one byte of the page is a "status" byte, and
when the side using xengntshr exits or dies, this byte is set to zero
and the event channel is notified. When the peer is woken by the notify
and sees the state byte set to zero, it removes its own mappings so that
the shared pages can be freed.
>>
>>> +/*
>>> + * Unmaps the @count pages starting at @start_address, which were
>>> mapped by a
>>> + * call to xengntshr_share_*. Never logs.
>>
>> Linewrap in the comment.
>>
>>> + */
>>> +int xengntshr_munmap(xengntshr_handle *xgs, void *start_address,
>>> uint32_t count);
>>
>> What effect does this have on the peer ?
>
> Daniel?
If this removes the (final copy of the) mapping and a notify offset/port
is set, that processing happens. Otherwise, the peer cannot tell when
this is called.
--
Daniel De Graaf
National Security Agency
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-13 20:38 ` Daniel De Graaf
@ 2015-11-16 12:30 ` Ian Campbell
2015-11-24 16:32 ` Daniel De Graaf
0 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-16 12:30 UTC (permalink / raw)
To: Daniel De Graaf, Ian Jackson; +Cc: wei.liu2, xen-devel
On Fri, 2015-11-13 at 15:38 -0500, Daniel De Graaf wrote:
> On 13/11/15 10:02, Ian Campbell wrote:
> > On Wed, 2015-11-11 at 15:03 +0000, Ian Jackson wrote:
> > > Ian Campbell writes ("[PATCH XEN v5 07/23] tools: Refactor
> > > /dev/xen/gnt{dev,shr} wrappers into libxengnttab."):
> > > > libxengnttab will provide a stable API and ABI for accessing the
> > > > grant table devices.
> > > >
> > > > The functions are moved into the xengnt{tab,shr} namespace to make
> > > > a
> > > > clean break from libxc and avoid ambiguity regarding which
> > > > interfaces
> > > > are stable.
> > >
> > > I have reviewed the API. Mostly this produced questions...
> >
> > Yes, thanks this is the sort of feedback I was looking for before we
> > call
> > this a stable interface.
> >
> > Some of my answers are of the "yes, we should decide that" variety.
> >
> > Daniel, I've added you because some of the questions below relate to
> > the
> > notification mechanism and to gntshr which I don't really understand
> > (not
> > that I really understand gnttab either).
>
> Responses below.
Thanks.
> I also agree that unifying the interfaces would be
> better.
>
> [...]
> > > > +/**
> > > > + * Memory maps a grant reference from one domain to a local
> > > > address
> > > > range.
> > > > + * Mappings should be unmapped with xengnttab_munmap. If
> > > > notify_offset
> > > > or
> > > > + * notify_port are not -1, this version will attempt to set up an
> > > > unmap
> > > > + * notification at the given offset and event channel. When the
> > > > page
> > > > is
> > > > + * unmapped, the byte at the given offset will be zeroed and a
> > > > wakeup
> > > > will be
> > > > + * sent to the given event channel. Logs errors.
> > >
> > > What happens if the unmap notification cannot be set up ?
> > >
> > > Also "when the page is unmapped" makes it sound like you mean
> > > xengnttab_munmap but actually I think this is when the grant is
> > > withdrawn by the grantor ?
> > >
> > > If the grant is withdrawn by the grantor, does the page become
> > > unuseable immediately ? If so, how can anyone ever use this safely ?
> >
> > Daniel, could you answer these ones please.
>
> This is intended to allow the kernel to send a close-request notification
> when the application that allocated the grant page exits without calling
> a proper shutdown (i.e. it crashes, calls _exit, calls execve, etc).
That is the kernel of the grantor or grantee process? It sounds like
grantor tells grantee (who would then be expected to unmap?)
Who actually does the unmap, the grantee process or their kernel? I suppose
the process, which is expected to be watching for the notification and is
required to do the unmap itself. IOW the "munmap notification" is a request
to please munmap, not a notification that something has been unmapped out
from beneath the calling process.
What happens if the unmap notification cannot be set up? Does the call fail
(and unmap what it has done) or does it succeed?
I think the answer to the other two questions depend on the clarifications
above, but I think it is the case that nothing is unmapped automatically,
all that this does is give you a result from evtchn_poll etc with the
expectation that the caller will call xengnttab_munmap in a controlled way
by themselves.
> Without this signal, the kernel has no way to request that the mapper of
> the page release it, and since Xen has no grant revocation mechanism, the
> page will likely be tied up until the process on the other side is told to
> release the page through some other method.
>
> > > > +/*
> > > > + * Creates and shares pages with another domain.
> > > > + *
> > > ...
> > > > +void *xengntshr_share_pages(xengntshr_handle *xgs, uint32_t domid,
> > > > + int count, uint32_t *refs, int
> > > > writable);
> > >
> > > Can this fail ? Can it partially succeed ?
> >
> > Daniel?
>
> It can fail if you are out of pages to grant (there is a module parameter
> that can be adjusted via sysfs for the maximum), or in the unlikely case
> that the kernel itself is out of room in its grant mapping table (or if
> the syscall itself encounters -ENOMEM).
Does it either completely succeed or undo partial work, or does it return
partial success somehow?
>
> > > > +/*
> > > > + * Creates and shares a page with another domain, with unmap
> > > > notification.
> > > > + *
> > > > + * @parm xgs a handle to an open grant sharing instance
> > > > + * @parm domid the domain to share memory with
> > > > + * @parm refs the grant reference of the pages (output)
> > > > + * @parm writable true if the other domain can write to the page
> > > > + * @parm notify_offset The byte offset in the page to use for
> > > > unmap
> > > > + * notification; -1 for none.
> > > > + * @parm notify_port The event channel port to use for unmap
> > > > notify,
> > > > or -1
> > > > + * @return local mapping of the page
> > > > + */
> > > > +void *xengntshr_share_page_notify(xengntshr_handle *xgs, uint32_t
> > > > domid,
> > >
> > > What is this `unmap notification' ?
> >
> > Daniel?
>
> As mentioned above, it is a way for the kernel to request for the other
> side to unmap a page. It is probably easiest to understand if looking
> at the libvchan driver: one byte of the page is a "status" byte, and
> when the side using xengntshr exits or dies, this byte is set to zero
> and the event channel is notified. When the peer is woken by the notify
> and sees the state byte set to zero, it removes its own mappings so that
> the shared pages can be freed.
This is the sending end of the notification requested by the caller of
xengnttab_map_grant_ref_notify I think?
>
> > >
> > > > +/*
> > > > + * Unmaps the @count pages starting at @start_address, which were
> > > > mapped by a
> > > > + * call to xengntshr_share_*. Never logs.
> > >
> > > Linewrap in the comment.
> > >
> > > > + */
> > > > +int xengntshr_munmap(xengntshr_handle *xgs, void *start_address,
> > > > uint32_t count);
> > >
> > > What effect does this have on the peer ?
> >
> > Daniel?
>
> If this removes the (final copy of the) mapping and a notify offset/port
> is set, that processing happens. Otherwise, the peer cannot tell when
> this is called.
So this will always success (I think?) but the underlying page cannot be
freed until the other end unmaps it (whether because of the notification of
some other reason).
What is the status of the memory at start_address between the call to
xengntshr_munmap and the otherend doing xengnttab_munmap?
What is the status of the mapping made by the peer via
xengnttab_map_grant_ref_notify in that same interval?
Who is responsible for reclaiming the underlying memory, the kernel or the
process?
Thanks,
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 19/23] tools/libs/call: Update some log messages to not refer to xc.
2015-11-13 16:20 ` Andrew Cooper
@ 2015-11-23 13:10 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-23 13:10 UTC (permalink / raw)
To: Andrew Cooper, ian.jackson, wei.liu2, xen-devel
On Fri, 2015-11-13 at 16:20 +0000, Andrew Cooper wrote:
> On 09/11/15 12:00, Ian Campbell wrote:
> > Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
> > Acked-by: Wei Liu <wei.liu2@citrix.com>
>
> I agree with the sentiment, but the error message refer to the
> libxencall entry API, so a developer can match it back to the call in
> their code.
The PERROR will prepend "xencall" onto the message.
This osdep_alloc_pages can be called from either xencall_alloc_buffer_pages
or xencall_alloc_buffer, so I don't think it is possible to make them more
specific than that "xencall: alloc_pages". (The old message was similarly
misleading in that regard)
Ian.
>
> ~Andrew
>
> > ---
> > tools/libs/call/linux.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/tools/libs/call/linux.c b/tools/libs/call/linux.c
> > index 906ca7e..80b505c 100644
> > --- a/tools/libs/call/linux.c
> > +++ b/tools/libs/call/linux.c
> > @@ -88,7 +88,7 @@ void *osdep_alloc_pages(xencall_handle *xcall,
> > unsigned int npages)
> > p = mmap(NULL, size, PROT_READ|PROT_WRITE,
> > MAP_PRIVATE|MAP_ANONYMOUS|MAP_LOCKED, -1, 0);
> > if ( p == MAP_FAILED )
> > {
> > - PERROR("xc_alloc_hypercall_buffer: mmap failed");
> > + PERROR("alloc_pages: mmap failed");
> > return NULL;
> > }
> >
> > @@ -97,7 +97,7 @@ void *osdep_alloc_pages(xencall_handle *xcall,
> > unsigned int npages)
> > rc = madvise(p, npages * PAGE_SIZE, MADV_DONTFORK);
> > if ( rc < 0 )
> > {
> > - PERROR("xc_alloc_hypercall_buffer: madvise failed");
> > + PERROR("alloc_pages: madvise failed");
> > goto out;
> > }
> >
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-11 15:03 ` Ian Jackson
2015-11-13 15:02 ` Ian Campbell
@ 2015-11-23 17:05 ` Ian Campbell
2015-11-24 9:34 ` Ian Campbell
1 sibling, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-23 17:05 UTC (permalink / raw)
To: Ian Jackson; +Cc: wei.liu2, xen-devel
On Wed, 2015-11-11 at 15:03 +0000, Ian Jackson wrote:
> > XXX consider combining into a single namespace (i.e. with
> > xengnttab_handle having two open fd's in it on Linux)
>
[...]
> I conclude it would be better to unify them.
Starting to look into this I have some queries on the direction I ought to
take.
Currently the real world implementations are such that a given installation
either provides:
* Neither gnttab nor gntshr (*BSD)
* Only gnttab, not gntshr (mini-os, older Linux versions)
* Both gnttab and ghtshr (newer Linux versions)
In particular gntshr but not gnttab is not found in any real world system.
Should I plan for such systems existing in the API? I think I probably
should.
A second question is then what functionality should be provided for callers
who only want either gnttab or gntshr, but not both. (The canonical example
would be the xc_gntshr_* compat layer in libxenctrl, there may be others).
Previously code could rely on either xc_gnt{tab,shr}_open failing if the
underlying interface was not present, and react accordingly, which might
involve disabling functionality entirely (in such a way as the other calls
to xc_gnt{tab,shr}_* do not need to handle ENOSYS, including by calling
exit(1)).
Now xengnttab_open could succeed because the other interface is available,
which may cause these applications to think they have an interface which
they in reality do not => confusion.
Options I can think of:
* A set of open flags, either FOO_MANDATORY or FOO_ONLY (the latter
allowing us to skip opening the relevant interface).
* Provide an interface to query the actual functionality which a given
handle supports, so apps can react accordingly (e.g. by closing it again
and erroring)
* Force all callers to adjust to handling ENOSYS as part of this
transition
Opinions?
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-23 17:05 ` Ian Campbell
@ 2015-11-24 9:34 ` Ian Campbell
2015-11-25 15:06 ` Ian Campbell
0 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-24 9:34 UTC (permalink / raw)
To: Ian Jackson, Daniel De Graaf; +Cc: wei.liu2, xen-devel
On Mon, 2015-11-23 at 17:05 +0000, Ian Campbell wrote:
Adding Daniel who I should have included the first time, plus a few
more thoughts from me at the end.
> On Wed, 2015-11-11 at 15:03 +0000, Ian Jackson wrote:
> > > XXX consider combining into a single namespace (i.e. with
> > > xengnttab_handle having two open fd's in it on Linux)
> >
> [...]
> > I conclude it would be better to unify them.
>
> Starting to look into this I have some queries on the direction I ought to
> take.
>
> Currently the real world implementations are such that a given installation
> either provides:
> * Neither gnttab nor gntshr (*BSD)
> * Only gnttab, not gntshr (mini-os, older Linux versions)
> * Both gnttab and ghtshr (newer Linux versions)
>
> In particular gntshr but not gnttab is not found in any real world system.
> Should I plan for such systems existing in the API? I think I probably
> should.
>
> A second question is then what functionality should be provided for callers
> who only want either gnttab or gntshr, but not both. (The canonical example
> would be the xc_gntshr_* compat layer in libxenctrl, there may be others).
>
> Previously code could rely on either xc_gnt{tab,shr}_open failing if the
> underlying interface was not present, and react accordingly, which might
> involve disabling functionality entirely (in such a way as the other calls
> to xc_gnt{tab,shr}_* do not need to handle ENOSYS, including by calling
> exit(1)).
>
> Now xengnttab_open could succeed because the other interface is available,
> which may cause these applications to think they have an interface which
> they in reality do not => confusion.
>
> Options I can think of:
>
> * A set of open flags, either FOO_MANDATORY or FOO_ONLY (the latter
> allowing us to skip opening the relevant interface).
> * Provide an interface to query the actual functionality which a given
> handle supports, so apps can react accordingly (e.g. by closing it again
> and erroring)
> * Force all callers to adjust to handling ENOSYS as part of this
> transition
>
> Opinions?
Thinking about this some more overnight, it occurred to me that the
real issue is that gnttab and gntshr are actually two quite difference
APIs. gnttab is all about consuming grant references which are given to
you from elsewhere while gntshr is all about creating grant references
to give to others. They have relatively little in common wrt the
underlying infrastructure, e.g. gnttab is mostly about making GNTTABOP
hypercalls while gntshr mostly interacts with the kernel's grant ref
allocator with no interaction with the hypervisor.
Maybe that and the API issues above constitute an argument for not
combining, I'm really not sure.
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-16 12:30 ` Ian Campbell
@ 2015-11-24 16:32 ` Daniel De Graaf
0 siblings, 0 replies; 111+ messages in thread
From: Daniel De Graaf @ 2015-11-24 16:32 UTC (permalink / raw)
To: Ian Campbell, Ian Jackson; +Cc: wei.liu2, xen-devel
On 16/11/15 07:30, Ian Campbell wrote:
> On Fri, 2015-11-13 at 15:38 -0500, Daniel De Graaf wrote:
>> On 13/11/15 10:02, Ian Campbell wrote:
>>> On Wed, 2015-11-11 at 15:03 +0000, Ian Jackson wrote:
>>>> Ian Campbell writes ("[PATCH XEN v5 07/23] tools: Refactor
>>>> /dev/xen/gnt{dev,shr} wrappers into libxengnttab."):
>>>>> libxengnttab will provide a stable API and ABI for accessing the
>>>>> grant table devices.
[...]
>>>>> +/**
>>>>> + * Memory maps a grant reference from one domain to a local
>>>>> address
>>>>> range.
>>>>> + * Mappings should be unmapped with xengnttab_munmap. If
>>>>> notify_offset
>>>>> or
>>>>> + * notify_port are not -1, this version will attempt to set up an
>>>>> unmap
>>>>> + * notification at the given offset and event channel. When the
>>>>> page
>>>>> is
>>>>> + * unmapped, the byte at the given offset will be zeroed and a
>>>>> wakeup
>>>>> will be
>>>>> + * sent to the given event channel. Logs errors.
>>>>
>>>> What happens if the unmap notification cannot be set up ?
>>>>
>>>> Also "when the page is unmapped" makes it sound like you mean
>>>> xengnttab_munmap but actually I think this is when the grant is
>>>> withdrawn by the grantor ?
>>>>
>>>> If the grant is withdrawn by the grantor, does the page become
>>>> unuseable immediately ? If so, how can anyone ever use this safely ?
>>>
>>> Daniel, could you answer these ones please.
>>
>> This is intended to allow the kernel to send a close-request notification
>> when the application that allocated the grant page exits without calling
>> a proper shutdown (i.e. it crashes, calls _exit, calls execve, etc).
>
> That is the kernel of the grantor or grantee process? It sounds like
> grantor tells grantee (who would then be expected to unmap?)
The kernel providing gntalloc is the one that sends the notification, since
the other domain (gntdev) must take action to unmap the page.
> Who actually does the unmap, the grantee process or their kernel? I suppose
> the process, which is expected to be watching for the notification and is
> required to do the unmap itself. IOW the "munmap notification" is a request
> to please munmap, not a notification that something has been unmapped out
> from beneath the calling process.
Correct. The remote process actually does the unmap; there is no back-channel
available to notify the kernel of the other side directly (and setting one up
would usually be a waste of resources).
> What happens if the unmap notification cannot be set up? Does the call fail
> (and unmap what it has done) or does it succeed?
The only reason for the setup to fail is if an invalid event channel or offset
is specified. It looks like this causes a new mapping to fail (it is freed).
> I think the answer to the other two questions depend on the clarifications
> above, but I think it is the case that nothing is unmapped automatically,
> all that this does is give you a result from evtchn_poll etc with the
> expectation that the caller will call xengnttab_munmap in a controlled way
> by themselves.
>
>> Without this signal, the kernel has no way to request that the mapper of
>> the page release it, and since Xen has no grant revocation mechanism, the
>> page will likely be tied up until the process on the other side is told to
>> release the page through some other method.
>>
>>>>> +/*
>>>>> + * Creates and shares pages with another domain.
>>>>> + *
>>>> ...
>>>>> +void *xengntshr_share_pages(xengntshr_handle *xgs, uint32_t domid,
>>>>> + int count, uint32_t *refs, int
>>>>> writable);
>>>>
>>>> Can this fail ? Can it partially succeed ?
>>>
>>> Daniel?
>>
>> It can fail if you are out of pages to grant (there is a module parameter
>> that can be adjusted via sysfs for the maximum), or in the unlikely case
>> that the kernel itself is out of room in its grant mapping table (or if
>> the syscall itself encounters -ENOMEM).
>
> Does it either completely succeed or undo partial work, or does it return
> partial success somehow?
The limit is checked (and the current count updated) prior to making any
changes.
>>
>>>>> +/*
>>>>> + * Creates and shares a page with another domain, with unmap
>>>>> notification.
>>>>> + *
>>>>> + * @parm xgs a handle to an open grant sharing instance
>>>>> + * @parm domid the domain to share memory with
>>>>> + * @parm refs the grant reference of the pages (output)
>>>>> + * @parm writable true if the other domain can write to the page
>>>>> + * @parm notify_offset The byte offset in the page to use for
>>>>> unmap
>>>>> + * notification; -1 for none.
>>>>> + * @parm notify_port The event channel port to use for unmap
>>>>> notify,
>>>>> or -1
>>>>> + * @return local mapping of the page
>>>>> + */
>>>>> +void *xengntshr_share_page_notify(xengntshr_handle *xgs, uint32_t
>>>>> domid,
>>>>
>>>> What is this `unmap notification' ?
>>>
>>> Daniel?
>>
>> As mentioned above, it is a way for the kernel to request for the other
>> side to unmap a page. It is probably easiest to understand if looking
>> at the libvchan driver: one byte of the page is a "status" byte, and
>> when the side using xengntshr exits or dies, this byte is set to zero
>> and the event channel is notified. When the peer is woken by the notify
>> and sees the state byte set to zero, it removes its own mappings so that
>> the shared pages can be freed.
>
> This is the sending end of the notification requested by the caller of
> xengnttab_map_grant_ref_notify I think?
Yes.
>>>>> +/*
>>>>> + * Unmaps the @count pages starting at @start_address, which were
>>>>> mapped by a
>>>>> + * call to xengntshr_share_*. Never logs.
>>>>
>>>> Linewrap in the comment.
>>>>
>>>>> + */
>>>>> +int xengntshr_munmap(xengntshr_handle *xgs, void *start_address,
>>>>> uint32_t count);
>>>>
>>>> What effect does this have on the peer ?
>>>
>>> Daniel?
>>
>> If this removes the (final copy of the) mapping and a notify offset/port
>> is set, that processing happens. Otherwise, the peer cannot tell when
>> this is called.
>
> So this will always success (I think?) but the underlying page cannot be
> freed until the other end unmaps it (whether because of the notification of
> some other reason).
Yes.
> What is the status of the memory at start_address between the call to
> xengntshr_munmap and the otherend doing xengnttab_munmap?
A reference to the page is held in gntalloc's gref_list; the page itself is
not usable by that domain until it can verify (gnttab_query_foreign_access)
that the page is not mapped elsewhere. At that point, the page is released
back to the kernel MM.
> What is the status of the mapping made by the peer via
> xengnttab_map_grant_ref_notify in that same interval?
The gnttab_map_grant_ref_notify is an analogous notification for the gnttab
device, allowing the application using gntalloc to close the channel when
the application that is using it has gone away. The notification in this
case is not needed to clean up the grant mapping (that has already been done
by the gntdev kernel code), but is useful if the sharing application wishes
to clean up (remove the shared pages or perhaps exit) when its peer exits.
> Who is responsible for reclaiming the underlying memory, the kernel or the
> process?
The kernel: gntalloc.c do_cleanup.
> Thanks,
> Ian.
>
--
Daniel De Graaf
National Security Agency
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab.
2015-11-24 9:34 ` Ian Campbell
@ 2015-11-25 15:06 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-25 15:06 UTC (permalink / raw)
To: Ian Jackson, Daniel De Graaf; +Cc: wei.liu2, xen-devel
On Tue, 2015-11-24 at 09:34 +0000, Ian Campbell wrote:
> Thinking about this some more overnight, it occurred to me that the
> real issue is that gnttab and gntshr are actually two quite difference
> APIs. gnttab is all about consuming grant references which are given to
> you from elsewhere while gntshr is all about creating grant references
> to give to others. They have relatively little in common wrt the
> underlying infrastructure, e.g. gnttab is mostly about making GNTTABOP
> hypercalls while gntshr mostly interacts with the kernel's grant ref
> allocator with no interaction with the hypervisor.
>
> Maybe that and the API issues above constitute an argument for not
> combining, I'm really not sure.
Ian and I discussed this (briefly) in real life and based on the above
argumentation we decided that keeping the two APIs separate (but in the
same library) was justified, since there is a reasonable enough air gap
between their functionality (i.e. consuming vs producing grants).
We did wonder a bit about changing the names to make that divide clearer,
but couldn't think of anything especially compelling and decided to stick
with the current shade of yellow.
Ian.
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
2015-11-11 15:13 ` Ian Jackson
2015-11-13 15:21 ` Ian Campbell
@ 2015-11-27 15:46 ` Ian Campbell
1 sibling, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-27 15:46 UTC (permalink / raw)
To: Ian Jackson; +Cc: wei.liu2, xen-devel
On Wed, 2015-11-11 at 15:13 +0000, Ian Jackson wrote:
> > +/*
> > + * Maps a range within one domain to a local address range. Mappings
> > + * should be unmapped with munmap and should follow the same rules as
> > mmap
> > + * regarding page alignment.
> > + *
> > + * prot is as for mmap(2).
> > + *
> > + * Can partially succeed. When a page cannot be mapped, its respective
> > + * field in @err is set to the corresponding errno value.
> > + *
> > + * Returns NULL if no pages can be mapped.
> ...
> > +void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t
> > dom,
> > + int prot, const xen_pfn_t *arr, int *err,
> > + unsigned int num);
>
> If it returns NULL, will all of *err be filled with errno values ?
> Ie is all of err[] always written ?
>
> Does this function ever set errno ?
From my reading of the Linux code it appears to be thus:
If it returns NULL it will have set errno. In that case it appears that err
is invalid.
If it returns non-NULL then all of err has been written to either 0 or a
negative errno val, indicating the state of that particular frame.
errno vals are in the OS space, not the Xen one, for both the global errno
and the err[] array vals.
I think the same is true of FreeBSD from my reading of that code.
In both cases the behaviour is a combined property of the behaviours of the
libxc functions, the privcmd driver and the hypercall on xen side, where
the first two differ a fair bit on different platforms (and call different
hypercalls under the hood, with different behaviour there too)
But the sum _seems_ to add up the same in both cases. Mini-OS is simpler
but looks to do about the same.
IOW I'm not 100% sure of the above but it seems a) like a
sensible/desirable enough behaviour and b) it does look, as far as I can
tell, like the code does this.
> This API invites callers to fail to notice partial failures.
>
> How about permitting err==NULL to mean `on partial failure, unmap
> everything and return NULL setting errno' ?
AFAICT both the FreeBSD and Linux privcmd driver do not handle this case,
so we would be looking at adding functionality to the library to paper over
it.
Looking at the users it seems that those which map batches of pages get
things right, since they naturally tend to be looping over the arrays
anyway.
It's the callers who are mapping a single page which tend to get this
wrong, many of them are pre-existing bugs but a fair few are as a result of
conversions made in this set of series.
In particular xc_map_foreign_pages used to unmap and return NULL with errno
set to the value first non-zero errno from the err array, which I suppose
is what you were thinking of for setting errno in this case?
I'm considering whether to add a second function in the API to map single
pages (since that seems to be the most common usage, and commonly exhibits
this error) vs doing as you suggest, which has the downside of throwing
away all the other errors (which may be more critical than the first).
I did briefly consider allow err == NULL iff nr == 1, but that seems worse
than both the other options.
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap.
2015-11-11 15:14 ` Ian Jackson
@ 2015-11-27 15:57 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-27 15:57 UTC (permalink / raw)
To: Ian Jackson; +Cc: wei.liu2, xen-devel
On Wed, 2015-11-11 at 15:14 +0000, Ian Jackson wrote:
> Ian Campbell writes ("[PATCH XEN v5 14/23] tools: foreignmemory: provide
> xenforeignmemory_unmap."):
> > And require it be used instead of direct munmap.
> >
> > This will allow e.g. Valgrind hooks to help track incorrect use of
> > foreign mappings.
> >
> > Switch all uses of xenforeignmemory_map to use
> > xenforeignmemory_unmap, not that foreign mappings via the libxc compat
> > xc_map_foreign_* interface will not take advantage of this and will
> > need converting.
> ...
> > +int xenforeignmemory_unmap(xenforeignmemory_handle *fmem,
> > + void *addr, unsigned int num);
> > +
>
> What kind of errors might this return ?
Same as munmap(2), i.e. only EINVAL according to http://pubs.opengroup.org/
onlinepubs/9699919799/functions/munmap.html
Ian,
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
2015-11-13 15:27 ` Ian Jackson
@ 2015-11-27 17:18 ` Ian Campbell
2015-11-27 17:48 ` Andrew Cooper
0 siblings, 1 reply; 111+ messages in thread
From: Ian Campbell @ 2015-11-27 17:18 UTC (permalink / raw)
To: Ian Jackson; +Cc: Andrew Cooper, Paul Durrant, wei.liu2, xen-devel
On Fri, 2015-11-13 at 15:27 +0000, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH XEN v5 13/23] tools: Refactor foreign
> memory mapping into libxenforeignmemory"):
> > On Wed, 2015-11-11 at 15:13 +0000, Ian Jackson wrote:
> > Are you suggesting:
> >
> > void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
> > int prot, const xen_pfn_t *arr,
> > int err_out[num],
> > unsigned int num);
> >
> > Is that (a var sized array based on another param) really allowed?
>
> Yes, if the array size precedes the array.
Sadly:
for i in include/xenforeignmemory.h; do \
gcc -x c -ansi -Wall -Werror -pedantic -I/local/scratch/ianc/devel/libxenctrl-split/xen.git/tools/libs/foreignmemory/../../../tools/include \
-S -o /dev/null $i || exit 1; \
echo $i; \
done >headers.chk.new
include/xenforeignmemory.h:142:28: error: ISO C90 forbids variable length array ‘arr’ [-Werror=vla]
const xen_pfn_t *arr[pages], int err[pages]);
^
include/xenforeignmemory.h:142:28: error: ISO C90 forbids variable length array ‘err’ [-Werror=vla]
cc1: all warnings being treated as errors
/local/scratch/ianc/devel/libxenctrl-split/xen.git/tools/libs/foreignmemory/../../../tools/Rules.mk:200: recipe for target 'headers.chk' failed
It's -ansi (AKA -std=c90) together with -pedantic which does it. -std=c99
is happy with it (including with -pedantic).
The original motivation for this check was [0]:
As part of the tidyup, we should choose a particular C standard (89,
probably) and ensure that the API/ABI complies with `gcc -std=c$VER
-pedantic`. This will help to provide a consistent API on other
platforms (I seem to recall an effort to port libvchan to windows.)
It's not clear to me how much the "89 probably" was influenced by the "on
other platforms ... windows" bit (maybe Andy or Paul has an opinion).
C99 was 16 years ago now, I'm struggling to think of a reason not to move
the baseline for tools stuff at least to that.
https://en.wikipedia.org/wiki/Visual_C%2B%2B might be one such reason I
suppose, although I'm not convinced a libvchan port to Windows, even if not
entirely hypothetical, would be using any of xen.git/tools/libs/* rather
than the equivalent frameworks provided by the Windows PV drivers.
Ian.
[0] http://article.gmane.org/gmane.comp.emulators.xen.devel/247265
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
2015-11-27 17:18 ` Ian Campbell
@ 2015-11-27 17:48 ` Andrew Cooper
2015-11-29 9:54 ` Paul Durrant
[not found] ` <9AAE0902D5BC7E449B7C8E4E778ABCD02F69D32F@AMSPEX01CL01.citrite.net>
0 siblings, 2 replies; 111+ messages in thread
From: Andrew Cooper @ 2015-11-27 17:48 UTC (permalink / raw)
To: Ian Campbell, Ian Jackson; +Cc: Paul Durrant, wei.liu2, xen-devel
On 27/11/15 17:18, Ian Campbell wrote:
> On Fri, 2015-11-13 at 15:27 +0000, Ian Jackson wrote:
>> Ian Campbell writes ("Re: [PATCH XEN v5 13/23] tools: Refactor foreign
>> memory mapping into libxenforeignmemory"):
>>> On Wed, 2015-11-11 at 15:13 +0000, Ian Jackson wrote:
>>> Are you suggesting:
>>>
>>> void *xenforeignmemory_map(xenforeignmemory_handle *fmem, uint32_t dom,
>>> int prot, const xen_pfn_t *arr,
>>> int err_out[num],
>>> unsigned int num);
>>>
>>> Is that (a var sized array based on another param) really allowed?
>> Yes, if the array size precedes the array.
> Sadly:
>
> for i in include/xenforeignmemory.h; do \
> gcc -x c -ansi -Wall -Werror -pedantic -I/local/scratch/ianc/devel/libxenctrl-split/xen.git/tools/libs/foreignmemory/../../../tools/include \
> -S -o /dev/null $i || exit 1; \
> echo $i; \
> done >headers.chk.new
> include/xenforeignmemory.h:142:28: error: ISO C90 forbids variable length array ‘arr’ [-Werror=vla]
> const xen_pfn_t *arr[pages], int err[pages]);
> ^
> include/xenforeignmemory.h:142:28: error: ISO C90 forbids variable length array ‘err’ [-Werror=vla]
> cc1: all warnings being treated as errors
> /local/scratch/ianc/devel/libxenctrl-split/xen.git/tools/libs/foreignmemory/../../../tools/Rules.mk:200: recipe for target 'headers.chk' failed
>
> It's -ansi (AKA -std=c90) together with -pedantic which does it. -std=c99
> is happy with it (including with -pedantic).
>
> The original motivation for this check was [0]:
> As part of the tidyup, we should choose a particular C standard (89,
> probably) and ensure that the API/ABI complies with `gcc -std=c$VER
> -pedantic`. This will help to provide a consistent API on other
> platforms (I seem to recall an effort to port libvchan to windows.)
>
> It's not clear to me how much the "89 probably" was influenced by the "on
> other platforms ... windows" bit (maybe Andy or Paul has an opinion).
I was under the impression that MSVC still doesn't have C99 support.
According to wikipedia, VS2013 gained some C99, and 2015 "achieved a
level of the C99 standard compliance in the library that is similar to
other C compilers".
According to MSDN about Visual C++:
C99 Conformance: Visual Studio 2015 fully implements the C99 Standard
Library, with the exception of any library features that depend on
compiler features not yet supported by the Visual C++ compiler (for
example, <tgmath.h> is not implemented).
which is wonderfully circular.
According to
https://msdn.microsoft.com/en-us/library/zb1574zs%28v=vs.140%29.aspx
which appears to be the documentation, VLAs are still not supported,
even in VS2015.
>
> C99 was 16 years ago now, I'm struggling to think of a reason not to move
> the baseline for tools stuff at least to that.
>
> https://en.wikipedia.org/wiki/Visual_C%2B%2B might be one such reason I
> suppose, although I'm not convinced a libvchan port to Windows, even if not
> entirely hypothetical, would be using any of xen.git/tools/libs/* rather
> than the equivalent frameworks provided by the Windows PV drivers.
It would be nice to at least be able to use the same header files, for
ease of porting userspace software.
In this case, VLAs are just being used as an aid for the compiler to
spot errors. It doesn't change the API/ABI, and could be #ifdef'd
around, if we care both for using C99 in general, and Windows support.
~Andrew
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
2015-11-27 17:48 ` Andrew Cooper
@ 2015-11-29 9:54 ` Paul Durrant
[not found] ` <9AAE0902D5BC7E449B7C8E4E778ABCD02F69D32F@AMSPEX01CL01.citrite.net>
1 sibling, 0 replies; 111+ messages in thread
From: Paul Durrant @ 2015-11-29 9:54 UTC (permalink / raw)
To: Andrew Cooper, Ian Campbell, Ian Jackson; +Cc: win-pv-devel, Wei Liu, xen-devel
> -----Original Message-----
[snip]
> > C99 was 16 years ago now, I'm struggling to think of a reason not to move
> > the baseline for tools stuff at least to that.
> >
> > https://en.wikipedia.org/wiki/Visual_C%2B%2B might be one such reason I
> > suppose, although I'm not convinced a libvchan port to Windows, even if
> not
> > entirely hypothetical, would be using any of xen.git/tools/libs/* rather
> > than the equivalent frameworks provided by the Windows PV drivers.
>
> It would be nice to at least be able to use the same header files, for
> ease of porting userspace software.
>
It's possible that libvchan on Windows will make use of the tools/libs headers. As Andy says it would ease porting client software.
> In this case, VLAs are just being used as an aid for the compiler to
> spot errors. It doesn't change the API/ABI, and could be #ifdef'd
> around, if we care both for using C99 in general, and Windows support.
>
We still compile with VS2012 in Citrix and Xen Project uses VS2013 so we can't rely on C99. A #ifdef here does seem like the best solution.
Paul
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
[not found] ` <9AAE0902D5BC7E449B7C8E4E778ABCD02F69D32F@AMSPEX01CL01.citrite.net>
@ 2015-11-30 9:51 ` Ian Campbell
[not found] ` <1448877118.15768.11.camel@citrix.com>
1 sibling, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-11-30 9:51 UTC (permalink / raw)
To: Paul Durrant, Andrew Cooper, Ian Jackson; +Cc: win-pv-devel, Wei Liu, xen-devel
On Sun, 2015-11-29 at 09:54 +0000, Paul Durrant wrote:
> > -----Original Message-----
> [snip]
> > > C99 was 16 years ago now, I'm struggling to think of a reason not to
> > > move
> > > the baseline for tools stuff at least to that.
> > >
> > > https://en.wikipedia.org/wiki/Visual_C%2B%2B might be one such reason
> > > I
> > > suppose, although I'm not convinced a libvchan port to Windows, even
> > > if
> > not
> > > entirely hypothetical, would be using any of xen.git/tools/libs/*
> > > rather
> > > than the equivalent frameworks provided by the Windows PV drivers.
> >
> > It would be nice to at least be able to use the same header files, for
> > ease of porting userspace software.
> >
>
> It's possible that libvchan on Windows will make use of the tools/libs
> headers. As Andy says it would ease porting client software.
>
> > In this case, VLAs are just being used as an aid for the compiler to
> > spot errors. It doesn't change the API/ABI, and could be #ifdef'd
> > around, if we care both for using C99 in general, and Windows support.
> >
>
> We still compile with VS2012 in Citrix and Xen Project uses VS2013 so we
> can't rely on C99. A #ifdef here does seem like the best solution.
Would you expect new projects (i.e. stuff based on tools/libs) to continue
to have requirements to build with those older versions? (Although with
Andy's link saying even VS2015 doesn't do VLAs maybe it is a bit moot).
I don't think using VLA here is worth an ifdef, but I think it is worth
reordering the arguments such that once we end up with a new enough
compiler baseline (in $donkeys years) we can switch without changing the
ABI (I think that's the case).
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory
[not found] ` <1448877118.15768.11.camel@citrix.com>
@ 2015-12-01 10:16 ` Paul Durrant
0 siblings, 0 replies; 111+ messages in thread
From: Paul Durrant @ 2015-12-01 10:16 UTC (permalink / raw)
To: Ian Campbell, Andrew Cooper, Ian Jackson; +Cc: win-pv-devel, Wei Liu, xen-devel
> -----Original Message-----
> From: Ian Campbell [mailto:ian.campbell@citrix.com]
> Sent: 30 November 2015 09:52
> To: Paul Durrant; Andrew Cooper; Ian Jackson
> Cc: Wei Liu; xen-devel@lists.xen.org; win-pv-devel@lists.xenproject.org
> Subject: Re: [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping
> into libxenforeignmemory
>
> On Sun, 2015-11-29 at 09:54 +0000, Paul Durrant wrote:
> > > -----Original Message-----
> > [snip]
> > > > C99 was 16 years ago now, I'm struggling to think of a reason not to
> > > > move
> > > > the baseline for tools stuff at least to that.
> > > >
> > > > https://en.wikipedia.org/wiki/Visual_C%2B%2B might be one such
> reason
> > > > I
> > > > suppose, although I'm not convinced a libvchan port to Windows, even
> > > > if
> > > not
> > > > entirely hypothetical, would be using any of xen.git/tools/libs/*
> > > > rather
> > > > than the equivalent frameworks provided by the Windows PV drivers.
> > >
> > > It would be nice to at least be able to use the same header files, for
> > > ease of porting userspace software.
> > >
> >
> > It's possible that libvchan on Windows will make use of the tools/libs
> > headers. As Andy says it would ease porting client software.
> >
> > > In this case, VLAs are just being used as an aid for the compiler to
> > > spot errors. It doesn't change the API/ABI, and could be #ifdef'd
> > > around, if we care both for using C99 in general, and Windows support.
> > >
> >
> > We still compile with VS2012 in Citrix and Xen Project uses VS2013 so we
> > can't rely on C99. A #ifdef here does seem like the best solution.
>
> Would you expect new projects (i.e. stuff based on tools/libs) to continue
> to have requirements to build with those older versions? (Although with
> Andy's link saying even VS2015 doesn't do VLAs maybe it is a bit moot).
>
It would certainly be preferable to have something that did build under a sufficiently old C standard to cover as many environments as is reasonable, and I think any supported MSVC build environment should be included in that set (although I have no problem using #ifdefs to achieve that).
> I don't think using VLA here is worth an ifdef, but I think it is worth
> reordering the arguments such that once we end up with a new enough
> compiler baseline (in $donkeys years) we can switch without changing the
> ABI (I think that's the case).
>
IIUC correctly the use of VLAs does not change the calling convention in any way, right? It would still mean a pointer to err_out is passed and the use of a VLA is so that bounds checking can be performed?
I certainly don't think we should get ourselves into a situation where additions to compilers or C standards do cause changes in the ABI (and we already have to be careful with Windows 32-bit since it defaults to pascal calling convention).
Paul
> Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
* Re: [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall.
2015-11-13 15:49 ` Andrew Cooper
@ 2015-12-02 14:55 ` Ian Campbell
0 siblings, 0 replies; 111+ messages in thread
From: Ian Campbell @ 2015-12-02 14:55 UTC (permalink / raw)
To: Andrew Cooper, ian.jackson, wei.liu2, xen-devel
On Fri, 2015-11-13 at 15:49 +0000, Andrew Cooper wrote:
> On 09/11/15 12:00, Ian Campbell wrote:
> > diff --git a/tools/libs/call/linux.c b/tools/libs/call/linux.c
> > new file mode 100644
> > index 0000000..906ca7e
> > --- /dev/null
> > +++ b/tools/libs/call/linux.c
> > @@ -0,0 +1,132 @@
> > +/*
> > + * This library is free software; you can redistribute it and/or
> > + * modify it under the terms of the GNU Lesser General Public
> > + * License as published by the Free Software Foundation;
> > + * version 2.1 of the License.
> > + *
> > + * This library is distributed in the hope that it will be useful,
> > + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> > + * Lesser General Public License for more details.
> > + *
> > + * You should have received a copy of the GNU Lesser General Public
> > + * License along with this library; If not, see <http://www.gnu.org/li
> > censes/>.
> > + *
> > + * Split out from xc_linus_osdep.c:
> > + *
> > + * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
> > + */
> > +
> > +#include <errno.h>
> > +#include <fcntl.h>
> > +#include <unistd.h>
> > +
> > +#include <sys/mman.h>
> > +#include <sys/ioctl.h>
> > +
> > +#include "private.h"
> > +
> > +int osdep_xencall_open(xencall_handle *xcall)
> > +{
> > + int flags, saved_errno;
> > + int fd = open("/proc/xen/privcmd", O_RDWR);
>
> I know this is a straight copy, but these new libraries really should
> start with the PVOps interface of /dev/xen/privcmd before falling back
> to the older classic-linux interfaces under /proc
Right, I'll pick this up for (almost, modulo resolving a conflict or two)
free when I rebase over Doug's patch to do this, which I'll be doing before
reposting.
> Ideally, once all these library improvements are done, there should be
> no need to mount xenfs in a Linux PVops domain.
>
> > +
> > + if ( fd == -1 )
> > + {
> > + PERROR("Could not obtain handle on privileged command
> > interface");
> > + return -1;
> > + }
> > +
> > + /* Although we return the file handle as the 'xc handle' the API
> > + does not specify / guarentee that this integer is in fact
> > + a file handle. Thus we must take responsiblity to ensure
> > + it doesn't propagate (ie leak) outside the process */
>
> Again, I know this is a straight copy, but it is racy. The open() call
> needs O_CLOEXEC.
>
> There might be some issues here to do with older versions of libc.
Or indeed super old versions of Linux (O_CLOEXEC came with 2.6.23 according
to man).
I'm tempted to hedge and use O_CLOEXEC with /dev/xen/privcmd, and leave the
legacy /proc/xen/privcmd path doing it this way, after all we've managed to
get this far...
> > + /* Do not copy the VMA to child process on fork. Avoid the page
> > being COW
> > + on hypercall. */
> > + rc = madvise(p, npages * PAGE_SIZE, MADV_DONTFORK);
>
> This also seems racy, although it is unclear from the manpages how to
> avoid the race.
pthread_atfork() might be one approach, and libxencall already uses
pthreads (for locking the cache), so it wouldn't be a problem to add more
uses, I don't think.
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply [flat|nested] 111+ messages in thread
end of thread, other threads:[~2015-12-02 14:55 UTC | newest]
Thread overview: 111+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-11-09 11:59 [Qemu-devel] [Minios-devel] [PATCH v4 0/<VARIOUS>] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 00/23] " Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 01/23] tools/Rules.mk: Properly handle libraries with recursive dependencies Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 02/23] tools: Refactor "xentoollog" into its own library Ian Campbell
2015-11-13 15:05 ` Andrew Cooper
2015-11-13 15:35 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 03/23] tools/libxc: Remove osdep indirection for xc_evtchn Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 04/23] tools: Refactor /dev/xen/evtchn wrappers into libxenevtchn Ian Campbell
2015-11-11 14:50 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 05/23] tools: Arrange to check public headers for ANSI compatiblity Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 06/23] tools/libxc: Remove osdep indirection for xc_gnt{shr, tab} Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 07/23] tools: Refactor /dev/xen/gnt{dev, shr} wrappers into libxengnttab Ian Campbell
2015-11-11 14:34 ` Wei Liu
2015-11-11 15:03 ` Ian Jackson
2015-11-13 15:02 ` Ian Campbell
2015-11-13 20:38 ` Daniel De Graaf
2015-11-16 12:30 ` Ian Campbell
2015-11-24 16:32 ` Daniel De Graaf
2015-11-23 17:05 ` Ian Campbell
2015-11-24 9:34 ` Ian Campbell
2015-11-25 15:06 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 08/23] tools/libxc: Remove osdep indirection for privcmd Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 09/23] tools: Refactor hypercall calling wrappers into libxencall Ian Campbell
2015-11-11 15:08 ` Ian Jackson
2015-11-11 15:31 ` Wei Liu
2015-11-13 15:17 ` Ian Campbell
2015-11-13 15:16 ` Ian Campbell
2015-11-13 16:56 ` Ian Campbell
2015-11-13 16:57 ` Ian Campbell
2015-11-13 17:16 ` Roger Pau Monné
2015-11-13 17:23 ` Ian Campbell
2015-11-13 15:35 ` Andrew Cooper
2015-11-13 15:49 ` Andrew Cooper
2015-12-02 14:55 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 10/23] tools/libxc: drop xc_map_foreign_bulk_compat wrappers Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 11/23] tools: Remove xc_map_foreign_batch Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 12/23] tools: Implement xc_map_foreign_range(s) in terms of common helper Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 13/23] tools: Refactor foreign memory mapping into libxenforeignmemory Ian Campbell
2015-11-11 15:13 ` Ian Jackson
2015-11-13 15:21 ` Ian Campbell
2015-11-13 15:27 ` Ian Jackson
2015-11-27 17:18 ` Ian Campbell
2015-11-27 17:48 ` Andrew Cooper
2015-11-29 9:54 ` Paul Durrant
[not found] ` <9AAE0902D5BC7E449B7C8E4E778ABCD02F69D32F@AMSPEX01CL01.citrite.net>
2015-11-30 9:51 ` Ian Campbell
[not found] ` <1448877118.15768.11.camel@citrix.com>
2015-12-01 10:16 ` Paul Durrant
2015-11-27 15:46 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 14/23] tools: foreignmemory: provide xenforeignmemory_unmap Ian Campbell
2015-11-11 15:14 ` Ian Jackson
2015-11-27 15:57 ` Ian Campbell
2015-11-13 16:16 ` Andrew Cooper
2015-11-13 16:17 ` Andrew Cooper
2015-11-09 12:00 ` [PATCH XEN v5 15/23] foreignmemory: use size_t for size arguments Ian Campbell
2015-11-11 15:15 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 16/23] tools/libs/evtchn: Review and update doc comments Ian Campbell
2015-11-11 15:15 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 17/23] tools/libs: Clean up hard tabs Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 18/23] tools/libs/gnttab: Review and update doc comments Ian Campbell
2015-11-11 15:17 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 19/23] tools/libs/call: Update some log messages to not refer to xc Ian Campbell
2015-11-13 16:20 ` Andrew Cooper
2015-11-23 13:10 ` Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 20/23] tools/libs/call: Avoid xc_memalign in netbsd and solaris backends Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 21/23] tools/libs/foreignmemory: Mention restrictions on fork in docs Ian Campbell
2015-11-11 15:18 ` Ian Jackson
2015-11-09 12:00 ` [PATCH XEN v5 22/23] tools: Update CFLAGS for qemu-xen to allow it to use new libraries Ian Campbell
2015-11-09 12:00 ` [PATCH XEN v5 23/23] HACK: Update Config.mk to pull all the right bits from my xenbits trees Ian Campbell
2015-11-11 15:20 ` [PATCH XEN v5 00/23] Begin to disentangle libxenctrl and provide some stable libraries Ian Jackson
2015-11-09 12:01 ` [Qemu-devel] [PATCH QEMU-XEN v5 0/9] " Ian Campbell
2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN v5 1/9] xen_console: correctly cleanup primary console on teardown Ian Campbell
2015-11-09 12:01 ` [Qemu-devel] " Ian Campbell
2015-11-09 12:01 ` [Qemu-devel] [PATCH QEMU-XEN v5 2/9] xen: Switch to libxenevtchn interface for compat shims Ian Campbell
2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [Qemu-devel] [PATCH QEMU-XEN v5 3/9] xen: Switch to libxengnttab " Ian Campbell
2015-11-09 12:01 ` Ian Campbell
2015-11-10 14:02 ` Stefano Stabellini
2015-11-10 14:02 ` [Qemu-devel] " Stefano Stabellini
2015-11-09 12:01 ` [Qemu-devel] [PATCH QEMU-XEN v5 4/9] xen: Switch uses of xc_map_foreign_range into xc_map_foreign_bulk Ian Campbell
2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [Qemu-devel] [PATCH QEMU-XEN v5 5/9] xen: Switch uses of xc_map_foreign_pages " Ian Campbell
2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [Qemu-devel] [PATCH QEMU-XEN v5 6/9] xen: Switch uses of xc_map_foreign_bulk to use libxenforeignmemory API Ian Campbell
2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN v5 7/9] xen: Use stable library interfaces when they are available Ian Campbell
2015-11-09 12:01 ` [Qemu-devel] " Ian Campbell
2015-11-10 14:00 ` Stefano Stabellini
2015-11-10 14:00 ` Stefano Stabellini
2015-11-09 12:01 ` [Qemu-devel] [PATCH QEMU-XEN v5 8/9] xen: domainbuild: reopen libxenctrl interface after forking for domain watcher Ian Campbell
2015-11-09 12:01 ` Ian Campbell
2015-11-09 12:01 ` [Qemu-devel] [PATCH QEMU-XEN v5 9/9] xen: make it possible to build without the Xen PV domain builder Ian Campbell
2015-11-09 12:01 ` Ian Campbell
2015-11-10 12:48 ` Stefano Stabellini
2015-11-10 12:48 ` [Qemu-devel] " Stefano Stabellini
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 1/5] qemu-xen-traditional: Use xentoollog as a separate library Ian Campbell
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 2/5] qemu-xen-traditional: Use libxenevtchn Ian Campbell
2015-11-11 15:19 ` Ian Jackson
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 3/5] qemu-xen-traditional: Use libxengnttab Ian Campbell
2015-11-11 15:20 ` Ian Jackson
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 4/5] qemu-xen-traditional: Add libxencall to rpath-link Ian Campbell
2015-11-11 15:20 ` Ian Jackson
2015-11-09 12:01 ` [PATCH QEMU-XEN-TRADITIONAL v5 5/5] qemu-xen-traditional: Add libxenforeignmemory " Ian Campbell
2015-11-11 15:21 ` Ian Jackson
2015-11-11 15:23 ` [PATCH QEMU-XEN-TRADITIONAL v5 0/5] Begin to disentangle libxenctrl and provide some stable libraries Ian Jackson
2015-11-09 12:01 ` [PATCH MINI-OS " Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 1/5] mini-os: Include libxentoollog with libxc Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 2/5] mini-os: Include libxenevtchn " Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 3/5] mini-os: Include libxengnttab " Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 4/5] mini-os: Include libxencall " Ian Campbell
2015-11-09 12:01 ` [PATCH MINI-OS v5 5/5] mini-os: Include libxenforeignmemory " Ian Campbell
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.