[Xen-devel] [PATCH v5 0/8] Port Xen to Hyper-V

[Xen-devel] [PATCH v5 0/8] Port Xen to Hyper-V

[Wei Liu]

Please see individual patches for changes

Cc: Jan Beulich <jbeulich@suse.com>                                                                                                                                                            
Cc: Andrew Cooper <andrew.cooper3@citrix.com>                                                                                                                                                  
Cc: Wei Liu <wl@xen.org>                                                                                                                                                                       
Cc: Roger Pau Monné <roger.pau@citrix.com>                                                                                                                                                     
Cc: Paul Durrant <paul@xen.org>      

Wei Liu (8):
  x86: introduce CONFIG_GUEST and move code
  x86: add missing headers in hypercall.h
  x86: drop hypervisor_cpuid_base
  x86: introduce hypervisor framework
  x86: rename hypervisor_{alloc,free}_unused_page
  x86: switch xen guest implementation to use hypervisor framework
  x86: be more verbose when running on a hypervisor
  x86: introduce CONFIG_HYPERV and detection code

 xen/arch/x86/Kconfig                          | 15 +++-
 xen/arch/x86/Makefile                         |  2 +-
 xen/arch/x86/guest/Makefile                   |  6 +-
 xen/arch/x86/guest/hyperv/Makefile            |  1 +
 xen/arch/x86/guest/hyperv/hyperv.c            | 54 +++++++++++++
 xen/arch/x86/guest/hypervisor.c               | 78 +++++++++++++++++++
 xen/arch/x86/guest/xen/Makefile               |  4 +
 xen/arch/x86/guest/{ => xen}/hypercall_page.S |  0
 xen/arch/x86/guest/{ => xen}/pvh-boot.c       |  2 +-
 xen/arch/x86/guest/{ => xen}/xen.c            | 60 +++++++-------
 xen/arch/x86/pv/shim.c                        |  6 +-
 xen/arch/x86/setup.c                          |  7 +-
 xen/include/asm-x86/guest.h                   |  2 +
 xen/include/asm-x86/guest/hypercall.h         |  4 +
 xen/include/asm-x86/guest/hyperv.h            | 43 ++++++++++
 xen/include/asm-x86/guest/hypervisor.h        | 62 +++++++++++++++
 xen/include/asm-x86/guest/xen.h               | 23 ++----
 17 files changed, 312 insertions(+), 57 deletions(-)
 create mode 100644 xen/arch/x86/guest/hyperv/Makefile
 create mode 100644 xen/arch/x86/guest/hyperv/hyperv.c
 create mode 100644 xen/arch/x86/guest/hypervisor.c
 create mode 100644 xen/arch/x86/guest/xen/Makefile
 rename xen/arch/x86/guest/{ => xen}/hypercall_page.S (100%)
 rename xen/arch/x86/guest/{ => xen}/pvh-boot.c (99%)
 rename xen/arch/x86/guest/{ => xen}/xen.c (94%)
 create mode 100644 xen/include/asm-x86/guest/hyperv.h
 create mode 100644 xen/include/asm-x86/guest/hypervisor.h

-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

[Xen-devel] [PATCH v5 1/8] x86: introduce CONFIG_GUEST and move code

[Wei Liu]

Xen is able to run as a guest on Xen. We plan to make it able to run
on Hyper-V as well.

Introduce CONFIG_GUEST which is set to true if either running on Xen
or Hyper-V is desired. Restructure code hierarchy for new code to
come.

No functional change intended.

Signed-off-by: Wei Liu <liuwe@microsoft.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
---
 xen/arch/x86/Kconfig                          | 7 +++++--
 xen/arch/x86/Makefile                         | 2 +-
 xen/arch/x86/guest/Makefile                   | 5 +----
 xen/arch/x86/guest/xen/Makefile               | 4 ++++
 xen/arch/x86/guest/{ => xen}/hypercall_page.S | 0
 xen/arch/x86/guest/{ => xen}/pvh-boot.c       | 0
 xen/arch/x86/guest/{ => xen}/xen.c            | 0
 7 files changed, 11 insertions(+), 7 deletions(-)
 create mode 100644 xen/arch/x86/guest/xen/Makefile
 rename xen/arch/x86/guest/{ => xen}/hypercall_page.S (100%)
 rename xen/arch/x86/guest/{ => xen}/pvh-boot.c (100%)
 rename xen/arch/x86/guest/{ => xen}/xen.c (100%)

diff --git a/xen/arch/x86/Kconfig b/xen/arch/x86/Kconfig
index 28b3b4692a..18c8f6fd26 100644
--- a/xen/arch/x86/Kconfig
+++ b/xen/arch/x86/Kconfig
@@ -161,9 +161,12 @@ config XEN_ALIGN_2M
 
 endchoice
 
+config GUEST
+	bool
+
 config XEN_GUEST
-	def_bool n
-	prompt "Xen Guest"
+	bool "Xen Guest"
+	select GUEST
 	---help---
 	  Support for Xen detecting when it is running under Xen.
 
diff --git a/xen/arch/x86/Makefile b/xen/arch/x86/Makefile
index a0b2f4ab15..7da5a2631e 100644
--- a/xen/arch/x86/Makefile
+++ b/xen/arch/x86/Makefile
@@ -1,7 +1,7 @@
 subdir-y += acpi
 subdir-y += cpu
 subdir-y += genapic
-subdir-$(CONFIG_XEN_GUEST) += guest
+subdir-$(CONFIG_GUEST) += guest
 subdir-$(CONFIG_HVM) += hvm
 subdir-y += mm
 subdir-$(CONFIG_XENOPROF) += oprofile
diff --git a/xen/arch/x86/guest/Makefile b/xen/arch/x86/guest/Makefile
index 26fb4b1007..6806f04947 100644
--- a/xen/arch/x86/guest/Makefile
+++ b/xen/arch/x86/guest/Makefile
@@ -1,4 +1 @@
-obj-y += hypercall_page.o
-obj-y += xen.o
-
-obj-bin-$(CONFIG_PVH_GUEST) += pvh-boot.init.o
+subdir-$(CONFIG_XEN_GUEST) += xen
diff --git a/xen/arch/x86/guest/xen/Makefile b/xen/arch/x86/guest/xen/Makefile
new file mode 100644
index 0000000000..26fb4b1007
--- /dev/null
+++ b/xen/arch/x86/guest/xen/Makefile
@@ -0,0 +1,4 @@
+obj-y += hypercall_page.o
+obj-y += xen.o
+
+obj-bin-$(CONFIG_PVH_GUEST) += pvh-boot.init.o
diff --git a/xen/arch/x86/guest/hypercall_page.S b/xen/arch/x86/guest/xen/hypercall_page.S
similarity index 100%
rename from xen/arch/x86/guest/hypercall_page.S
rename to xen/arch/x86/guest/xen/hypercall_page.S
diff --git a/xen/arch/x86/guest/pvh-boot.c b/xen/arch/x86/guest/xen/pvh-boot.c
similarity index 100%
rename from xen/arch/x86/guest/pvh-boot.c
rename to xen/arch/x86/guest/xen/pvh-boot.c
diff --git a/xen/arch/x86/guest/xen.c b/xen/arch/x86/guest/xen/xen.c
similarity index 100%
rename from xen/arch/x86/guest/xen.c
rename to xen/arch/x86/guest/xen/xen.c
-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

[Xen-devel] [PATCH v5 2/8] x86: add missing headers in hypercall.h

[Wei Liu]

Include asm_defns.h because ASM_CALL_CONSTRAINT is defined there.

Include xen/lib.h because we need ASSERT_UNREACHABLE.

No functional change.

Signed-off-by: Wei Liu <liuwe@microsoft.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
---
 xen/include/asm-x86/guest/hypercall.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/xen/include/asm-x86/guest/hypercall.h b/xen/include/asm-x86/guest/hypercall.h
index d548816b30..d0d2f5022d 100644
--- a/xen/include/asm-x86/guest/hypercall.h
+++ b/xen/include/asm-x86/guest/hypercall.h
@@ -23,6 +23,8 @@
 
 #include <xen/types.h>
 
+#include <asm/asm_defns.h>
+
 #include <public/xen.h>
 #include <public/sched.h>
 #include <public/hvm/hvm_op.h>
@@ -180,6 +182,8 @@ static inline long xen_hypercall_set_evtchn_upcall_vector(
 
 #else /* CONFIG_XEN_GUEST */
 
+#include <xen/lib.h>
+
 #include <public/sched.h>
 
 static inline void xen_hypercall_console_write(
-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

[Xen-devel] [PATCH v5 3/8] x86: drop hypervisor_cpuid_base

[Wei Liu]

The only user is Xen specific code in PV shim. We can therefore export
the variable directly.

Move __read_mostly to its standard place while at it.

Signed-off-by: Wei Liu <liuwe@microsoft.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
---
 xen/arch/x86/guest/xen/xen.c    | 7 +------
 xen/arch/x86/pv/shim.c          | 2 +-
 xen/include/asm-x86/guest/xen.h | 2 +-
 3 files changed, 3 insertions(+), 8 deletions(-)

diff --git a/xen/arch/x86/guest/xen/xen.c b/xen/arch/x86/guest/xen/xen.c
index a329e7c886..1e37086518 100644
--- a/xen/arch/x86/guest/xen/xen.c
+++ b/xen/arch/x86/guest/xen/xen.c
@@ -37,7 +37,7 @@
 
 bool __read_mostly xen_guest;
 
-static __read_mostly uint32_t xen_cpuid_base;
+uint32_t __read_mostly xen_cpuid_base;
 extern char hypercall_page[];
 static struct rangeset *mem;
 
@@ -297,11 +297,6 @@ int hypervisor_free_unused_page(mfn_t mfn)
     return rangeset_remove_range(mem, mfn_x(mfn), mfn_x(mfn));
 }
 
-uint32_t hypervisor_cpuid_base(void)
-{
-    return xen_cpuid_base;
-}
-
 static void ap_resume(void *unused)
 {
     map_vcpuinfo();
diff --git a/xen/arch/x86/pv/shim.c b/xen/arch/x86/pv/shim.c
index 35bf3945ac..351da970ef 100644
--- a/xen/arch/x86/pv/shim.c
+++ b/xen/arch/x86/pv/shim.c
@@ -995,7 +995,7 @@ domid_t get_initial_domain_id(void)
     if ( !pv_shim )
         return 0;
 
-    cpuid(hypervisor_cpuid_base() + 4, &eax, &ebx, &ecx, &edx);
+    cpuid(xen_cpuid_base + 4, &eax, &ebx, &ecx, &edx);
 
     return (eax & XEN_HVM_CPUID_DOMID_PRESENT) ? ecx : 1;
 }
diff --git a/xen/include/asm-x86/guest/xen.h b/xen/include/asm-x86/guest/xen.h
index 7e04e4a7ab..b015ed1883 100644
--- a/xen/include/asm-x86/guest/xen.h
+++ b/xen/include/asm-x86/guest/xen.h
@@ -30,13 +30,13 @@
 
 extern bool xen_guest;
 extern bool pv_console;
+extern uint32_t xen_cpuid_base;
 
 void probe_hypervisor(void);
 void hypervisor_setup(void);
 void hypervisor_ap_setup(void);
 int hypervisor_alloc_unused_page(mfn_t *mfn);
 int hypervisor_free_unused_page(mfn_t mfn);
-uint32_t hypervisor_cpuid_base(void);
 void hypervisor_resume(void);
 
 DECLARE_PER_CPU(unsigned int, vcpu_id);
-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

[Xen-devel] [PATCH v5 4/8] x86: introduce hypervisor framework

[Wei Liu]

We will soon implement Hyper-V support for Xen. Add a framework for
that.

This requires moving some of the hypervisor_* functions from xen.h to
hypervisor.h.

Signed-off-by: Wei Liu <liuwe@microsoft.com>
---
Changes in v5:
1. Make hypervisor_probe return const char *.

Changes in v4:
1. Add ASSERT_UNREACHABLE to stubs.
2. Move __read_mostly.
3. Return hops directly.
4. Drop Paul's review tag.
---
 xen/arch/x86/guest/Makefile            |  2 +
 xen/arch/x86/guest/hypervisor.c        | 45 +++++++++++++++++++
 xen/include/asm-x86/guest.h            |  1 +
 xen/include/asm-x86/guest/hypervisor.h | 62 ++++++++++++++++++++++++++
 xen/include/asm-x86/guest/xen.h        | 12 -----
 5 files changed, 110 insertions(+), 12 deletions(-)
 create mode 100644 xen/arch/x86/guest/hypervisor.c
 create mode 100644 xen/include/asm-x86/guest/hypervisor.h

diff --git a/xen/arch/x86/guest/Makefile b/xen/arch/x86/guest/Makefile
index 6806f04947..f63d64bbee 100644
--- a/xen/arch/x86/guest/Makefile
+++ b/xen/arch/x86/guest/Makefile
@@ -1 +1,3 @@
+obj-y += hypervisor.o
+
 subdir-$(CONFIG_XEN_GUEST) += xen
diff --git a/xen/arch/x86/guest/hypervisor.c b/xen/arch/x86/guest/hypervisor.c
new file mode 100644
index 0000000000..c35c610d01
--- /dev/null
+++ b/xen/arch/x86/guest/hypervisor.c
@@ -0,0 +1,45 @@
+/******************************************************************************
+ * arch/x86/guest/hypervisor.c
+ *
+ * Support for detecting and running under a hypervisor.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program 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 General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Copyright (c) 2019 Microsoft.
+ */
+#include <xen/init.h>
+#include <xen/types.h>
+
+#include <asm/cache.h>
+#include <asm/guest/hypervisor.h>
+
+static const struct hypervisor_ops __read_mostly *ops;
+
+const char *__init hypervisor_probe(void)
+{
+    if ( ops )
+        return ops->name;
+
+    return NULL;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/xen/include/asm-x86/guest.h b/xen/include/asm-x86/guest.h
index a38c6b5b3f..8e167165ae 100644
--- a/xen/include/asm-x86/guest.h
+++ b/xen/include/asm-x86/guest.h
@@ -20,6 +20,7 @@
 #define __X86_GUEST_H__
 
 #include <asm/guest/hypercall.h>
+#include <asm/guest/hypervisor.h>
 #include <asm/guest/pvh-boot.h>
 #include <asm/guest/xen.h>
 #include <asm/pv/shim.h>
diff --git a/xen/include/asm-x86/guest/hypervisor.h b/xen/include/asm-x86/guest/hypervisor.h
new file mode 100644
index 0000000000..392f4b90ae
--- /dev/null
+++ b/xen/include/asm-x86/guest/hypervisor.h
@@ -0,0 +1,62 @@
+/******************************************************************************
+ * asm-x86/guest/hypervisor.h
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms and conditions of the GNU General Public
+ * License, version 2, as published by the Free Software Foundation.
+ *
+ * This program 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
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Copyright (c) 2019 Microsoft.
+ */
+
+#ifndef __X86_HYPERVISOR_H__
+#define __X86_HYPERVISOR_H__
+
+struct hypervisor_ops {
+    /* Name of the hypervisor */
+    const char *name;
+    /* Main setup routine */
+    void (*setup)(void);
+    /* AP setup */
+    void (*ap_setup)(void);
+    /* Resume from suspension */
+    void (*resume)(void);
+};
+
+#ifdef CONFIG_GUEST
+
+const char *hypervisor_probe(void);
+void hypervisor_setup(void);
+void hypervisor_ap_setup(void);
+void hypervisor_resume(void);
+
+#else
+
+#include <xen/lib.h>
+#include <xen/types.h>
+
+static inline const char *hypervisor_probe(void) { return NULL; }
+static inline void hypervisor_setup(void) { ASSERT_UNREACHABLE(); }
+static inline void hypervisor_ap_setup(void) { ASSERT_UNREACHABLE(); }
+static inline void hypervisor_resume(void) { ASSERT_UNREACHABLE(); }
+
+#endif  /* CONFIG_GUEST */
+
+#endif /* __X86_HYPERVISOR_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/xen/include/asm-x86/guest/xen.h b/xen/include/asm-x86/guest/xen.h
index b015ed1883..3145f75361 100644
--- a/xen/include/asm-x86/guest/xen.h
+++ b/xen/include/asm-x86/guest/xen.h
@@ -33,11 +33,8 @@ extern bool pv_console;
 extern uint32_t xen_cpuid_base;
 
 void probe_hypervisor(void);
-void hypervisor_setup(void);
-void hypervisor_ap_setup(void);
 int hypervisor_alloc_unused_page(mfn_t *mfn);
 int hypervisor_free_unused_page(mfn_t mfn);
-void hypervisor_resume(void);
 
 DECLARE_PER_CPU(unsigned int, vcpu_id);
 DECLARE_PER_CPU(struct vcpu_info *, vcpu_info);
@@ -49,15 +46,6 @@ DECLARE_PER_CPU(struct vcpu_info *, vcpu_info);
 
 static inline void probe_hypervisor(void) {}
 
-static inline void hypervisor_setup(void)
-{
-    ASSERT_UNREACHABLE();
-}
-static inline void hypervisor_ap_setup(void)
-{
-    ASSERT_UNREACHABLE();
-}
-
 #endif /* CONFIG_XEN_GUEST */
 #endif /* __X86_GUEST_XEN_H__ */
 
-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

[Xen-devel] [PATCH v5 5/8] x86: rename hypervisor_{alloc, free}_unused_page

[Wei Liu]

They are used in Xen code only.

No functional change.

Signed-off-by: Wei Liu <liuwe@microsoft.com>
Reviewed-by: Paul Durrant <pdurrant@amazon.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
---
 xen/arch/x86/guest/xen/xen.c    | 6 +++---
 xen/arch/x86/pv/shim.c          | 4 ++--
 xen/include/asm-x86/guest/xen.h | 4 ++--
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/xen/arch/x86/guest/xen/xen.c b/xen/arch/x86/guest/xen/xen.c
index 1e37086518..0f5b5267c5 100644
--- a/xen/arch/x86/guest/xen/xen.c
+++ b/xen/arch/x86/guest/xen/xen.c
@@ -93,7 +93,7 @@ static void map_shared_info(void)
     unsigned int i;
     unsigned long rc;
 
-    if ( hypervisor_alloc_unused_page(&mfn) )
+    if ( xg_alloc_unused_page(&mfn) )
         panic("unable to reserve shared info memory page\n");
 
     xatp.gpfn = mfn_x(mfn);
@@ -280,7 +280,7 @@ void hypervisor_ap_setup(void)
     init_evtchn();
 }
 
-int hypervisor_alloc_unused_page(mfn_t *mfn)
+int xg_alloc_unused_page(mfn_t *mfn)
 {
     unsigned long m;
     int rc;
@@ -292,7 +292,7 @@ int hypervisor_alloc_unused_page(mfn_t *mfn)
     return rc;
 }
 
-int hypervisor_free_unused_page(mfn_t mfn)
+int xg_free_unused_page(mfn_t mfn)
 {
     return rangeset_remove_range(mem, mfn_x(mfn), mfn_x(mfn));
 }
diff --git a/xen/arch/x86/pv/shim.c b/xen/arch/x86/pv/shim.c
index 351da970ef..7a898fdbe5 100644
--- a/xen/arch/x86/pv/shim.c
+++ b/xen/arch/x86/pv/shim.c
@@ -742,7 +742,7 @@ static long pv_shim_grant_table_op(unsigned int cmd,
                 };
                 mfn_t mfn;
 
-                rc = hypervisor_alloc_unused_page(&mfn);
+                rc = xg_alloc_unused_page(&mfn);
                 if ( rc )
                 {
                     gprintk(XENLOG_ERR,
@@ -754,7 +754,7 @@ static long pv_shim_grant_table_op(unsigned int cmd,
                 rc = xen_hypercall_memory_op(XENMEM_add_to_physmap, &xatp);
                 if ( rc )
                 {
-                    hypervisor_free_unused_page(mfn);
+                    xg_free_unused_page(mfn);
                     break;
                 }
 
diff --git a/xen/include/asm-x86/guest/xen.h b/xen/include/asm-x86/guest/xen.h
index 3145f75361..01dc3ee6f6 100644
--- a/xen/include/asm-x86/guest/xen.h
+++ b/xen/include/asm-x86/guest/xen.h
@@ -33,8 +33,8 @@ extern bool pv_console;
 extern uint32_t xen_cpuid_base;
 
 void probe_hypervisor(void);
-int hypervisor_alloc_unused_page(mfn_t *mfn);
-int hypervisor_free_unused_page(mfn_t mfn);
+int xg_alloc_unused_page(mfn_t *mfn);
+int xg_free_unused_page(mfn_t mfn);
 
 DECLARE_PER_CPU(unsigned int, vcpu_id);
 DECLARE_PER_CPU(struct vcpu_info *, vcpu_info);
-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

[Xen-devel] [PATCH v5 6/8] x86: switch xen guest implementation to use hypervisor framework

[Wei Liu]

Signed-off-by: Wei Liu <liuwe@microsoft.com>
---
Changes in v5:
1. xen_probe -> xg_probe
2. Drop xen_ prefix for private functions
---
 xen/arch/x86/guest/hypervisor.c   | 27 +++++++++++++++++-
 xen/arch/x86/guest/xen/pvh-boot.c |  2 +-
 xen/arch/x86/guest/xen/xen.c      | 47 ++++++++++++++++++-------------
 xen/arch/x86/setup.c              |  2 +-
 xen/include/asm-x86/guest/xen.h   |  5 ++--
 5 files changed, 59 insertions(+), 24 deletions(-)

diff --git a/xen/arch/x86/guest/hypervisor.c b/xen/arch/x86/guest/hypervisor.c
index c35c610d01..6cd86d0c02 100644
--- a/xen/arch/x86/guest/hypervisor.c
+++ b/xen/arch/x86/guest/hypervisor.c
@@ -22,18 +22,43 @@
 #include <xen/types.h>
 
 #include <asm/cache.h>
-#include <asm/guest/hypervisor.h>
+#include <asm/guest.h>
 
 static const struct hypervisor_ops __read_mostly *ops;
 
 const char *__init hypervisor_probe(void)
 {
+    if ( !cpu_has_hypervisor )
+        return NULL;
+
+    if ( ops )
+        return ops->name;
+
+    ops = xg_probe();
     if ( ops )
         return ops->name;
 
     return NULL;
 }
 
+void __init hypervisor_setup(void)
+{
+    if ( ops && ops->setup )
+        ops->setup();
+}
+
+void hypervisor_ap_setup(void)
+{
+    if ( ops && ops->ap_setup )
+        ops->ap_setup();
+}
+
+void hypervisor_resume(void)
+{
+    if ( ops && ops->resume )
+        ops->resume();
+}
+
 /*
  * Local variables:
  * mode: C
diff --git a/xen/arch/x86/guest/xen/pvh-boot.c b/xen/arch/x86/guest/xen/pvh-boot.c
index ca8e156f7d..498625eae0 100644
--- a/xen/arch/x86/guest/xen/pvh-boot.c
+++ b/xen/arch/x86/guest/xen/pvh-boot.c
@@ -103,7 +103,7 @@ void __init pvh_init(multiboot_info_t **mbi, module_t **mod)
 {
     convert_pvh_info(mbi, mod);
 
-    probe_hypervisor();
+    hypervisor_probe();
     ASSERT(xen_guest);
 
     get_memory_map();
diff --git a/xen/arch/x86/guest/xen/xen.c b/xen/arch/x86/guest/xen/xen.c
index 0f5b5267c5..10daa043e4 100644
--- a/xen/arch/x86/guest/xen/xen.c
+++ b/xen/arch/x86/guest/xen/xen.c
@@ -67,22 +67,6 @@ static void __init find_xen_leaves(void)
     }
 }
 
-void __init probe_hypervisor(void)
-{
-    if ( xen_guest || !cpu_has_hypervisor )
-        return;
-
-    find_xen_leaves();
-
-    if ( !xen_cpuid_base )
-        return;
-
-    /* Fill the hypercall page. */
-    wrmsrl(cpuid_ebx(xen_cpuid_base + 2), __pa(hypercall_page));
-
-    xen_guest = true;
-}
-
 static void map_shared_info(void)
 {
     mfn_t mfn;
@@ -245,7 +229,7 @@ static void init_evtchn(void)
     }
 }
 
-void __init hypervisor_setup(void)
+static void __init setup(void)
 {
     init_memmap();
 
@@ -273,7 +257,7 @@ void __init hypervisor_setup(void)
     init_evtchn();
 }
 
-void hypervisor_ap_setup(void)
+static void ap_setup(void)
 {
     set_vcpu_id();
     map_vcpuinfo();
@@ -303,7 +287,7 @@ static void ap_resume(void *unused)
     init_evtchn();
 }
 
-void hypervisor_resume(void)
+static void resume(void)
 {
     /* Reset shared info page. */
     map_shared_info();
@@ -326,6 +310,31 @@ void hypervisor_resume(void)
         pv_console_init();
 }
 
+static const struct hypervisor_ops xg_ops = {
+    .name = "Xen",
+    .setup = setup,
+    .ap_setup = ap_setup,
+    .resume = resume,
+};
+
+const struct hypervisor_ops *__init xg_probe(void)
+{
+    if ( xen_guest )
+        return &xg_ops;
+
+    find_xen_leaves();
+
+    if ( !xen_cpuid_base )
+        return NULL;
+
+    /* Fill the hypercall page. */
+    wrmsrl(cpuid_ebx(xen_cpuid_base + 2), __pa(hypercall_page));
+
+    xen_guest = true;
+
+    return &xg_ops;
+}
+
 /*
  * Local variables:
  * mode: C
diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c
index 7d27f36053..a6b354c29f 100644
--- a/xen/arch/x86/setup.c
+++ b/xen/arch/x86/setup.c
@@ -763,7 +763,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
      * allocing any xenheap structures wanted in lower memory. */
     kexec_early_calculations();
 
-    probe_hypervisor();
+    hypervisor_probe();
 
     parse_video_info();
 
diff --git a/xen/include/asm-x86/guest/xen.h b/xen/include/asm-x86/guest/xen.h
index 01dc3ee6f6..2042a9a0c2 100644
--- a/xen/include/asm-x86/guest/xen.h
+++ b/xen/include/asm-x86/guest/xen.h
@@ -23,6 +23,7 @@
 
 #include <asm/e820.h>
 #include <asm/fixmap.h>
+#include <asm/guest/hypervisor.h>
 
 #define XEN_shared_info ((struct shared_info *)fix_to_virt(FIX_XEN_SHARED_INFO))
 
@@ -32,7 +33,7 @@ extern bool xen_guest;
 extern bool pv_console;
 extern uint32_t xen_cpuid_base;
 
-void probe_hypervisor(void);
+const struct hypervisor_ops *xg_probe(void);
 int xg_alloc_unused_page(mfn_t *mfn);
 int xg_free_unused_page(mfn_t mfn);
 
@@ -44,7 +45,7 @@ DECLARE_PER_CPU(struct vcpu_info *, vcpu_info);
 #define xen_guest 0
 #define pv_console 0
 
-static inline void probe_hypervisor(void) {}
+static inline const struct hypervisor_ops *xg_probe(void) { return NULL; }
 
 #endif /* CONFIG_XEN_GUEST */
 #endif /* __X86_GUEST_XEN_H__ */
-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

[Xen-devel] [PATCH v5 7/8] x86: be more verbose when running on a hypervisor

[Wei Liu]

Also replace reference to xen_guest.

Signed-off-by: Wei Liu <liuwe@microsoft.com>
---
Changes in v5:
1. Cache and use hypervisor name instead
---
 xen/arch/x86/setup.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c
index a6b354c29f..fc049eaac8 100644
--- a/xen/arch/x86/setup.c
+++ b/xen/arch/x86/setup.c
@@ -700,6 +700,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
         .max_grant_frames = -1,
         .max_maptrack_frames = -1,
     };
+    const char *hypervisor_name;
 
     /* Critical region without IDT or TSS.  Any fault is deadly! */
 
@@ -763,7 +764,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
      * allocing any xenheap structures wanted in lower memory. */
     kexec_early_calculations();
 
-    hypervisor_probe();
+    hypervisor_name = hypervisor_probe();
 
     parse_video_info();
 
@@ -788,6 +789,8 @@ void __init noreturn __start_xen(unsigned long mbi_p)
     printk("Command line: %s\n", cmdline);
 
     printk("Xen image load base address: %#lx\n", xen_phys_start);
+    if ( hypervisor_name )
+        printk("Running on %s\n", hypervisor_name);
 
 #ifdef CONFIG_VIDEO
     printk("Video information:\n");
@@ -1569,7 +1572,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
             max_cpus = nr_cpu_ids;
     }
 
-    if ( xen_guest )
+    if ( hypervisor_name )
         hypervisor_setup();
 
     /* Low mappings were only needed for some BIOS table parsing. */
-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

[Xen-devel] [PATCH v5 8/8] x86: introduce CONFIG_HYPERV and detection code

[Wei Liu]

We use the same code structure as we did for Xen.

As starters, detect Hyper-V in probe routine. More complex
functionalities will be added later.

Signed-off-by: Wei Liu <liuwe@microsoft.com>
Reviewed-by: Paul Durrant <pdurrant@amazon.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
---
 xen/arch/x86/Kconfig               |  8 +++++
 xen/arch/x86/guest/Makefile        |  1 +
 xen/arch/x86/guest/hyperv/Makefile |  1 +
 xen/arch/x86/guest/hyperv/hyperv.c | 54 ++++++++++++++++++++++++++++++
 xen/arch/x86/guest/hypervisor.c    |  8 +++++
 xen/include/asm-x86/guest.h        |  1 +
 xen/include/asm-x86/guest/hyperv.h | 43 ++++++++++++++++++++++++
 7 files changed, 116 insertions(+)
 create mode 100644 xen/arch/x86/guest/hyperv/Makefile
 create mode 100644 xen/arch/x86/guest/hyperv/hyperv.c
 create mode 100644 xen/include/asm-x86/guest/hyperv.h

diff --git a/xen/arch/x86/Kconfig b/xen/arch/x86/Kconfig
index 18c8f6fd26..0a02b6ee3f 100644
--- a/xen/arch/x86/Kconfig
+++ b/xen/arch/x86/Kconfig
@@ -164,6 +164,14 @@ endchoice
 config GUEST
 	bool
 
+config HYPERV_GUEST
+	bool "Hyper-V Guest"
+	select GUEST
+	---help---
+	  Support for Xen detecting when it is running under Hyper-V.
+
+	  If unsure, say N.
+
 config XEN_GUEST
 	bool "Xen Guest"
 	select GUEST
diff --git a/xen/arch/x86/guest/Makefile b/xen/arch/x86/guest/Makefile
index f63d64bbee..f164196772 100644
--- a/xen/arch/x86/guest/Makefile
+++ b/xen/arch/x86/guest/Makefile
@@ -1,3 +1,4 @@
 obj-y += hypervisor.o
 
+subdir-$(CONFIG_HYPERV_GUEST) += hyperv
 subdir-$(CONFIG_XEN_GUEST) += xen
diff --git a/xen/arch/x86/guest/hyperv/Makefile b/xen/arch/x86/guest/hyperv/Makefile
new file mode 100644
index 0000000000..68170109a9
--- /dev/null
+++ b/xen/arch/x86/guest/hyperv/Makefile
@@ -0,0 +1 @@
+obj-y += hyperv.o
diff --git a/xen/arch/x86/guest/hyperv/hyperv.c b/xen/arch/x86/guest/hyperv/hyperv.c
new file mode 100644
index 0000000000..916e08ff89
--- /dev/null
+++ b/xen/arch/x86/guest/hyperv/hyperv.c
@@ -0,0 +1,54 @@
+/******************************************************************************
+ * arch/x86/guest/hyperv/hyperv.c
+ *
+ * Support for detecting and running under Hyper-V.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program 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 General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Copyright (c) 2019 Microsoft.
+ */
+#include <xen/init.h>
+
+#include <asm/guest.h>
+
+static const struct hypervisor_ops hyperv_ops = {
+    .name = "Hyper-V",
+};
+
+const struct hypervisor_ops * __init hyperv_probe(void)
+{
+    uint32_t eax, ebx, ecx, edx;
+
+    cpuid(0x40000000, &eax, &ebx, &ecx, &edx);
+    if ( !((ebx == 0x7263694d) &&  /* "Micr" */
+           (ecx == 0x666f736f) &&  /* "osof" */
+           (edx == 0x76482074)) )  /* "t Hv" */
+        return NULL;
+
+    cpuid(0x40000001, &eax, &ebx, &ecx, &edx);
+    if ( eax != 0x31237648 )    /* Hv#1 */
+        return NULL;
+
+    return &hyperv_ops;
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/xen/arch/x86/guest/hypervisor.c b/xen/arch/x86/guest/hypervisor.c
index 6cd86d0c02..65c25115e0 100644
--- a/xen/arch/x86/guest/hypervisor.c
+++ b/xen/arch/x86/guest/hypervisor.c
@@ -38,6 +38,14 @@ const char *__init hypervisor_probe(void)
     if ( ops )
         return ops->name;
 
+    /*
+     * Detection of Hyper-V must come after Xen to avoid false positive due
+     * to viridian support
+     */
+    ops = hyperv_probe();
+    if ( ops )
+        return ops->name;
+
     return NULL;
 }
 
diff --git a/xen/include/asm-x86/guest.h b/xen/include/asm-x86/guest.h
index 8e167165ae..94448606d4 100644
--- a/xen/include/asm-x86/guest.h
+++ b/xen/include/asm-x86/guest.h
@@ -20,6 +20,7 @@
 #define __X86_GUEST_H__
 
 #include <asm/guest/hypercall.h>
+#include <asm/guest/hyperv.h>
 #include <asm/guest/hypervisor.h>
 #include <asm/guest/pvh-boot.h>
 #include <asm/guest/xen.h>
diff --git a/xen/include/asm-x86/guest/hyperv.h b/xen/include/asm-x86/guest/hyperv.h
new file mode 100644
index 0000000000..3f88b94c77
--- /dev/null
+++ b/xen/include/asm-x86/guest/hyperv.h
@@ -0,0 +1,43 @@
+/******************************************************************************
+ * asm-x86/guest/hyperv.h
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms and conditions of the GNU General Public
+ * License, version 2, as published by the Free Software Foundation.
+ *
+ * This program 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
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Copyright (c) 2019 Microsoft.
+ */
+
+#ifndef __X86_GUEST_HYPERV_H__
+#define __X86_GUEST_HYPERV_H__
+
+#ifdef CONFIG_HYPERV_GUEST
+
+#include <asm/guest/hypervisor.h>
+
+const struct hypervisor_ops *hyperv_probe(void);
+
+#else
+
+static inline const struct hypervisor_ops *hyperv_probe(void) { return NULL; }
+
+#endif /* CONFIG_HYPERV_GUEST */
+#endif /* __X86_GUEST_HYPERV_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
2.20.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 4/8] x86: introduce hypervisor framework

[Jan Beulich]

On 30.11.2019 12:57, Wei Liu wrote:
> We will soon implement Hyper-V support for Xen. Add a framework for
> that.
> 
> This requires moving some of the hypervisor_* functions from xen.h to
> hypervisor.h.
> 
> Signed-off-by: Wei Liu <liuwe@microsoft.com>

Acked-by: Jan Beulich <jbeulich@suse.com>
with one more adjustment (sorry for noticing only now):

> --- /dev/null
> +++ b/xen/arch/x86/guest/hypervisor.c
> @@ -0,0 +1,45 @@
> +/******************************************************************************
> + * arch/x86/guest/hypervisor.c
> + *
> + * Support for detecting and running under a hypervisor.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program 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 General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; If not, see <http://www.gnu.org/licenses/>.
> + *
> + * Copyright (c) 2019 Microsoft.
> + */
> +#include <xen/init.h>
> +#include <xen/types.h>
> +
> +#include <asm/cache.h>
> +#include <asm/guest/hypervisor.h>
> +
> +static const struct hypervisor_ops __read_mostly *ops;

The __read_mostly is misplaced - it's an attribute of the variable,
not its type, and hence belongs after the * . It just so happens
that the compiler is (still) relatively relaxed in what it accepts,
but I think at least the gcc manual has a warning towards future
more strict behavior.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 6/8] x86: switch xen guest implementation to use hypervisor framework

[Jan Beulich]

On 30.11.2019 12:57, Wei Liu wrote:
> Signed-off-by: Wei Liu <liuwe@microsoft.com>

Acked-by: Jan Beulich <jbeulich@suse.com>
again with one more remark:

> @@ -326,6 +310,31 @@ void hypervisor_resume(void)
>          pv_console_init();
>  }
>  
> +static const struct hypervisor_ops xg_ops = {

Along with other static variable not having an xg_ prefix,
this one could lose its one, too. But I'm not going to make
this a requirement.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 7/8] x86: be more verbose when running on a hypervisor

[Jan Beulich]

On 30.11.2019 12:57, Wei Liu wrote:
> Also replace reference to xen_guest.
> 
> Signed-off-by: Wei Liu <liuwe@microsoft.com>

Acked-by: Jan Beulich <jbeulich@suse.com>

However, ...

> --- a/xen/arch/x86/setup.c
> +++ b/xen/arch/x86/setup.c
> @@ -700,6 +700,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
>          .max_grant_frames = -1,
>          .max_maptrack_frames = -1,
>      };
> +    const char *hypervisor_name;
>  
>      /* Critical region without IDT or TSS.  Any fault is deadly! */
>  
> @@ -763,7 +764,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
>       * allocing any xenheap structures wanted in lower memory. */
>      kexec_early_calculations();
>  
> -    hypervisor_probe();
> +    hypervisor_name = hypervisor_probe();

... you no longer calling this function multiple time, why does
patch 4 still put in a respective guard?

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 4/8] x86: introduce hypervisor framework

[Wei Liu]

On Tue, Dec 03, 2019 at 03:49:33PM +0100, Jan Beulich wrote:
> On 30.11.2019 12:57, Wei Liu wrote:
> > We will soon implement Hyper-V support for Xen. Add a framework for
> > that.
> > 
> > This requires moving some of the hypervisor_* functions from xen.h to
> > hypervisor.h.
> > 
> > Signed-off-by: Wei Liu <liuwe@microsoft.com>
> 
> Acked-by: Jan Beulich <jbeulich@suse.com>
[...]
> > +#include <asm/guest/hypervisor.h>
> > +
> > +static const struct hypervisor_ops __read_mostly *ops;
> 
> The __read_mostly is misplaced - it's an attribute of the variable,
> not its type, and hence belongs after the * . It just so happens
> that the compiler is (still) relatively relaxed in what it accepts,
> but I think at least the gcc manual has a warning towards future
> more strict behavior.

Fixed.

Wei.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 6/8] x86: switch xen guest implementation to use hypervisor framework

[Wei Liu]

On Tue, Dec 03, 2019 at 03:52:47PM +0100, Jan Beulich wrote:
> On 30.11.2019 12:57, Wei Liu wrote:
> > Signed-off-by: Wei Liu <liuwe@microsoft.com>
> 
> Acked-by: Jan Beulich <jbeulich@suse.com>
> again with one more remark:
> 
> > @@ -326,6 +310,31 @@ void hypervisor_resume(void)
> >          pv_console_init();
> >  }
> >  
> > +static const struct hypervisor_ops xg_ops = {
> 
> Along with other static variable not having an xg_ prefix,
> this one could lose its one, too. But I'm not going to make
> this a requirement.

I will drop the xg_ prefix locally per your comment here.

Wei.

> 
> Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 7/8] x86: be more verbose when running on a hypervisor

[Wei Liu]

On Tue, Dec 03, 2019 at 03:54:35PM +0100, Jan Beulich wrote:
> On 30.11.2019 12:57, Wei Liu wrote:
> > Also replace reference to xen_guest.
> > 
> > Signed-off-by: Wei Liu <liuwe@microsoft.com>
> 
> Acked-by: Jan Beulich <jbeulich@suse.com>

Thanks.

> 
> However, ...
> 
> > --- a/xen/arch/x86/setup.c
> > +++ b/xen/arch/x86/setup.c
> > @@ -700,6 +700,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
> >          .max_grant_frames = -1,
> >          .max_maptrack_frames = -1,
> >      };
> > +    const char *hypervisor_name;
> >  
> >      /* Critical region without IDT or TSS.  Any fault is deadly! */
> >  
> > @@ -763,7 +764,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
> >       * allocing any xenheap structures wanted in lower memory. */
> >      kexec_early_calculations();
> >  
> > -    hypervisor_probe();
> > +    hypervisor_name = hypervisor_probe();
> 
> ... you no longer calling this function multiple time, why does
> patch 4 still put in a respective guard?

Remnant from previous iterations.

I can submit a follow-up patch to drop that -- do really want to
invalidate all the reviews and acks I got so far.

Wei.

> 
> Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 8/8] x86: introduce CONFIG_HYPERV and detection code

[Wei Liu]

On Sat, Nov 30, 2019 at 11:57:37AM +0000, Wei Liu wrote:
[...]
> + */
> +#include <xen/init.h>
> +
> +#include <asm/guest.h>
> +
> +static const struct hypervisor_ops hyperv_ops = {

Since xg_ops has lost its xg_ prefix, I also take the liberty to drop
the hyperv_ prefix here to make things more consistent.

Wei.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 7/8] x86: be more verbose when running on a hypervisor

[Jan Beulich]

On 03.12.2019 17:37, Wei Liu wrote:
> On Tue, Dec 03, 2019 at 03:54:35PM +0100, Jan Beulich wrote:
>> On 30.11.2019 12:57, Wei Liu wrote:
>>> Also replace reference to xen_guest.
>>>
>>> Signed-off-by: Wei Liu <liuwe@microsoft.com>
>>
>> Acked-by: Jan Beulich <jbeulich@suse.com>
> 
> Thanks.
> 
>>
>> However, ...
>>
>>> --- a/xen/arch/x86/setup.c
>>> +++ b/xen/arch/x86/setup.c
>>> @@ -700,6 +700,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
>>>          .max_grant_frames = -1,
>>>          .max_maptrack_frames = -1,
>>>      };
>>> +    const char *hypervisor_name;
>>>  
>>>      /* Critical region without IDT or TSS.  Any fault is deadly! */
>>>  
>>> @@ -763,7 +764,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
>>>       * allocing any xenheap structures wanted in lower memory. */
>>>      kexec_early_calculations();
>>>  
>>> -    hypervisor_probe();
>>> +    hypervisor_name = hypervisor_probe();
>>
>> ... you no longer calling this function multiple time, why does
>> patch 4 still put in a respective guard?
> 
> Remnant from previous iterations.
> 
> I can submit a follow-up patch to drop that -- do really want to
> invalidate all the reviews and acks I got so far.

According to my records patch 4 had no acks except mine, which you
could keep with this change (in fact I was thinking of making it
dependent upon the dropping of this leftover). Subsequent patches
may only need re-basing, which doesn't imply dropping of any acks.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 7/8] x86: be more verbose when running on a hypervisor

[Wei Liu]

On Tue, Dec 03, 2019 at 05:58:28PM +0100, Jan Beulich wrote:
> On 03.12.2019 17:37, Wei Liu wrote:
> > On Tue, Dec 03, 2019 at 03:54:35PM +0100, Jan Beulich wrote:
> >> On 30.11.2019 12:57, Wei Liu wrote:
> >>> Also replace reference to xen_guest.
> >>>
> >>> Signed-off-by: Wei Liu <liuwe@microsoft.com>
> >>
> >> Acked-by: Jan Beulich <jbeulich@suse.com>
> > 
> > Thanks.
> > 
> >>
> >> However, ...
> >>
> >>> --- a/xen/arch/x86/setup.c
> >>> +++ b/xen/arch/x86/setup.c
> >>> @@ -700,6 +700,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
> >>>          .max_grant_frames = -1,
> >>>          .max_maptrack_frames = -1,
> >>>      };
> >>> +    const char *hypervisor_name;
> >>>  
> >>>      /* Critical region without IDT or TSS.  Any fault is deadly! */
> >>>  
> >>> @@ -763,7 +764,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
> >>>       * allocing any xenheap structures wanted in lower memory. */
> >>>      kexec_early_calculations();
> >>>  
> >>> -    hypervisor_probe();
> >>> +    hypervisor_name = hypervisor_probe();
> >>
> >> ... you no longer calling this function multiple time, why does
> >> patch 4 still put in a respective guard?
> > 
> > Remnant from previous iterations.
> > 
> > I can submit a follow-up patch to drop that -- do really want to
> > invalidate all the reviews and acks I got so far.
> 
> According to my records patch 4 had no acks except mine, which you
> could keep with this change (in fact I was thinking of making it
> dependent upon the dropping of this leftover). Subsequent patches
> may only need re-basing, which doesn't imply dropping of any acks.

OK. In that case, I will drop it locally. If that causes any substantial
changes, I will post another version; otherwise I will just keep all the
tags and push this series soon-ish.

How does that sound to you?

Wei.

> 
> Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH v5 7/8] x86: be more verbose when running on a hypervisor

[Wei Liu]

On Tue, Dec 03, 2019 at 05:09:43PM +0000, Wei Liu wrote:
> On Tue, Dec 03, 2019 at 05:58:28PM +0100, Jan Beulich wrote:
> > On 03.12.2019 17:37, Wei Liu wrote:
> > > On Tue, Dec 03, 2019 at 03:54:35PM +0100, Jan Beulich wrote:
> > >> On 30.11.2019 12:57, Wei Liu wrote:
> > >>> Also replace reference to xen_guest.
> > >>>
> > >>> Signed-off-by: Wei Liu <liuwe@microsoft.com>
> > >>
> > >> Acked-by: Jan Beulich <jbeulich@suse.com>
> > > 
> > > Thanks.
> > > 
> > >>
> > >> However, ...
> > >>
> > >>> --- a/xen/arch/x86/setup.c
> > >>> +++ b/xen/arch/x86/setup.c
> > >>> @@ -700,6 +700,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
> > >>>          .max_grant_frames = -1,
> > >>>          .max_maptrack_frames = -1,
> > >>>      };
> > >>> +    const char *hypervisor_name;
> > >>>  
> > >>>      /* Critical region without IDT or TSS.  Any fault is deadly! */
> > >>>  
> > >>> @@ -763,7 +764,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
> > >>>       * allocing any xenheap structures wanted in lower memory. */
> > >>>      kexec_early_calculations();
> > >>>  
> > >>> -    hypervisor_probe();
> > >>> +    hypervisor_name = hypervisor_probe();
> > >>
> > >> ... you no longer calling this function multiple time, why does
> > >> patch 4 still put in a respective guard?
> > > 
> > > Remnant from previous iterations.
> > > 
> > > I can submit a follow-up patch to drop that -- do really want to
> > > invalidate all the reviews and acks I got so far.
> > 
> > According to my records patch 4 had no acks except mine, which you
> > could keep with this change (in fact I was thinking of making it
> > dependent upon the dropping of this leftover). Subsequent patches
> > may only need re-basing, which doesn't imply dropping of any acks.
> 
> OK. In that case, I will drop it locally. If that causes any substantial
> changes, I will post another version; otherwise I will just keep all the
> tags and push this series soon-ish.
> 
> How does that sound to you?

And it turns out it is indeed trivial. Dropping that hunk in patch 4
only requires a minor fixup to patch 6.

Wei.

> 
> Wei.
> 
> > 
> > Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel