* [PATCH V7 0/2] Virtio support for toolstack on Arm (Was "IOREQ feature (+ virtio-mmio) on Arm")
@ 2022-04-08 18:21 Oleksandr Tyshchenko
2022-04-08 18:21 ` [PATCH V7 1/2] libxl: Add support for Virtio disk configuration Oleksandr Tyshchenko
2022-04-08 18:21 ` [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm Oleksandr Tyshchenko
0 siblings, 2 replies; 23+ messages in thread
From: Oleksandr Tyshchenko @ 2022-04-08 18:21 UTC (permalink / raw)
To: xen-devel
Cc: Oleksandr Tyshchenko, Wei Liu, Anthony PERARD, George Dunlap,
Nick Rosbrook, Juergen Gross, Stefano Stabellini, Julien Grall,
Bertrand Marquis, Volodymyr Babchuk, Wei Chen, Kaly Xin,
Jiamei Xie, Henry Wang, Alex Bennée
From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
Hello all.
The purpose of this patch series is to add missing virtio-mmio bits to Xen toolstack on Arm.
The Virtio support for toolstack [1] was postponed as the main target was to upstream IOREQ/DM
support on Arm in the first place. Now, we already have IOREQ support in, so we can resume Virtio
enabling work. You can find previous discussions at [2].
Patch series [3] is based on recent "staging" branch
(9120b5737f517fe9d2a3936c38d3a2211630323b livepatch: avoid relocations referencing ignored section symbols)
and tested on Renesas Salvator-X board + H3 ES3.0 SoC (Arm64) with virtio-mmio based virtio-disk backend [4]
running in Dom0 (or Driver domain) and unmodified Linux Guest running on existing virtio-blk driver (frontend).
No issues were observed. Guest domain 'reboot/destroy' use-cases work properly.
Any feedback/help would be highly appreciated.
[1]
https://lore.kernel.org/xen-devel/1610488352-18494-24-git-send-email-olekstysh@gmail.com/
https://lore.kernel.org/xen-devel/1610488352-18494-25-git-send-email-olekstysh@gmail.com/
[2]
https://lists.xenproject.org/archives/html/xen-devel/2021-01/msg02403.html
https://lists.xenproject.org/archives/html/xen-devel/2021-01/msg02536.html
https://lore.kernel.org/xen-devel/1621626361-29076-1-git-send-email-olekstysh@gmail.com/
https://lore.kernel.org/xen-devel/1638982784-14390-1-git-send-email-olekstysh@gmail.com/
[3] https://github.com/otyshchenko1/xen/commits/libxl_virtio2
[4] https://github.com/otyshchenko1/virtio-disk/commits/virtio_grant
Julien Grall (1):
libxl: Introduce basic virtio-mmio support on Arm
Oleksandr Tyshchenko (1):
libxl: Add support for Virtio disk configuration
docs/man/xl-disk-configuration.5.pod.in | 37 +-
tools/golang/xenlight/helpers.gen.go | 6 +
tools/golang/xenlight/types.gen.go | 11 +
tools/include/libxl.h | 6 +
tools/libs/light/libxl_arm.c | 131 +++-
tools/libs/light/libxl_device.c | 57 +-
tools/libs/light/libxl_disk.c | 111 +++-
tools/libs/light/libxl_internal.h | 1 +
tools/libs/light/libxl_types.idl | 10 +
tools/libs/light/libxl_types_internal.idl | 1 +
tools/libs/light/libxl_utils.c | 2 +
tools/libs/util/libxlu_disk_l.c | 952 +++++++++++++++---------------
tools/libs/util/libxlu_disk_l.h | 2 +-
tools/libs/util/libxlu_disk_l.l | 9 +
tools/xl/xl_block.c | 11 +
xen/include/public/arch-arm.h | 7 +
16 files changed, 872 insertions(+), 482 deletions(-)
--
2.7.4
^ permalink raw reply [flat|nested] 23+ messages in thread
* [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-08 18:21 [PATCH V7 0/2] Virtio support for toolstack on Arm (Was "IOREQ feature (+ virtio-mmio) on Arm") Oleksandr Tyshchenko
@ 2022-04-08 18:21 ` Oleksandr Tyshchenko
2022-04-18 21:38 ` Stefano Stabellini
` (2 more replies)
2022-04-08 18:21 ` [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm Oleksandr Tyshchenko
1 sibling, 3 replies; 23+ messages in thread
From: Oleksandr Tyshchenko @ 2022-04-08 18:21 UTC (permalink / raw)
To: xen-devel
Cc: Oleksandr Tyshchenko, Wei Liu, Anthony PERARD, George Dunlap,
Nick Rosbrook, Juergen Gross, Stefano Stabellini, Julien Grall,
Volodymyr Babchuk, Bertrand Marquis
From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
This patch adds basic support for configuring and assisting virtio-mmio
based virtio-disk backend (emualator) which is intended to run out of
Qemu and could be run in any domain.
Although the Virtio block device is quite different from traditional
Xen PV block device (vbd) from the toolstack's point of view:
- as the frontend is virtio-blk which is not a Xenbus driver, nothing
written to Xenstore are fetched by the frontend (the vdev is not
passed to the frontend)
- the ring-ref/event-channel are not used for the backend<->frontend
communication, the proposed IPC for Virtio is IOREQ/DM
it is still a "block device" and ought to be integrated in existing
"disk" handling. So, re-use (and adapt) "disk" parsing/configuration
logic to deal with Virtio devices as well.
For the immediate purpose and an ability to extend that support for
other use-cases in future (Qemu, virtio-pci, etc) perform the following
actions:
- Add new disk backend type (LIBXL_DISK_BACKEND_OTHER) and reflect
that in the configuration
- Introduce new disk protocol field to libxl_device_disk struct
(with LIBXL_DISK_PROTOCOL_XEN and LIBXL_DISK_PROTOCOL_VIRTIO_MMIO
types) and reflect that in the configuration (new "protocol" option
with "xen" protocol being default value)
- Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
An example of domain configuration for Virtio disk:
disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other, protocol=virtio-mmio']
Nothing has changed for default Xen disk configuration.
Please note, this patch is not enough for virtio-disk to work
on Xen (Arm), as for every Virtio device (including disk) we need
to allocate Virtio MMIO params (IRQ and memory region) and pass
them to the backend, also update Guest device-tree. The subsequent
patch will add these missing bits. For the current patch,
the default "irq" and "base" are just written to the Xenstore.
This is not an ideal splitting, but this way we avoid breaking
the bisectability.
Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
---
Changes RFC -> V1:
- no changes
Changes V1 -> V2:
- rebase according to the new location of libxl_virtio_disk.c
Changes V2 -> V3:
- no changes
Changes V3 -> V4:
- rebase according to the new argument for DEFINE_DEVICE_TYPE_STRUCT
Changes V4 -> V5:
- split the changes, change the order of the patches
- update patch description
- don't introduce new "vdisk" configuration option with own parsing logic,
re-use Xen PV block "disk" parsing/configuration logic for the virtio-disk
- introduce "virtio" flag and document it's usage
- add LIBXL_HAVE_DEVICE_DISK_VIRTIO
- update libxlu_disk_l.[ch]
- drop num_disks variable/MAX_VIRTIO_DISKS
- drop Wei's T-b
Changes V5 -> V6:
- rebase on current staging
- use "%"PRIu64 instead of %lu for disk->base in device_disk_add()
- update *.gen.go files
Changes V6 -> V7:
- rebase on current staging
- update *.gen.go files and libxlu_disk_l.[ch] files
- update patch description
- rework significantly to support more flexible configuration
and have more generic basic implementation for being able to extend
that for other use-cases (virtio-pci, qemu, etc).
---
docs/man/xl-disk-configuration.5.pod.in | 37 +-
tools/golang/xenlight/helpers.gen.go | 6 +
tools/golang/xenlight/types.gen.go | 11 +
tools/include/libxl.h | 6 +
tools/libs/light/libxl_device.c | 57 +-
tools/libs/light/libxl_disk.c | 111 +++-
tools/libs/light/libxl_internal.h | 1 +
tools/libs/light/libxl_types.idl | 10 +
tools/libs/light/libxl_types_internal.idl | 1 +
tools/libs/light/libxl_utils.c | 2 +
tools/libs/util/libxlu_disk_l.c | 952 +++++++++++++++---------------
tools/libs/util/libxlu_disk_l.h | 2 +-
tools/libs/util/libxlu_disk_l.l | 9 +
tools/xl/xl_block.c | 11 +
14 files changed, 736 insertions(+), 480 deletions(-)
diff --git a/docs/man/xl-disk-configuration.5.pod.in b/docs/man/xl-disk-configuration.5.pod.in
index 71d0e86..36c851f 100644
--- a/docs/man/xl-disk-configuration.5.pod.in
+++ b/docs/man/xl-disk-configuration.5.pod.in
@@ -232,7 +232,7 @@ Specifies the backend implementation to use
=item Supported values
-phy, qdisk
+phy, qdisk, other
=item Mandatory
@@ -244,11 +244,13 @@ Automatically determine which backend to use.
=back
-This does not affect the guest's view of the device. It controls
-which software implementation of the Xen backend driver is used.
+It controls which software implementation of the backend driver is used.
+Depending on the "protocol" option this may affect the guest's view
+of the device.
Not all backend drivers support all combinations of other options.
-For example, "phy" does not support formats other than "raw".
+For example, "phy" and "other" do not support formats other than "raw" and
+"other" does not support protocols other than "virtio-mmio".
Normally this option should not be specified, in which case libxl will
automatically determine the most suitable backend.
@@ -344,8 +346,35 @@ can be used to disable "hole punching" for file based backends which
were intentionally created non-sparse to avoid fragmentation of the
file.
+=item B<protocol>=I<PROTOCOL>
+
+=over 4
+
+=item Description
+
+Specifies the communication protocol to use for the chosen "backendtype" option
+
+=item Supported values
+
+xen, virtio-mmio
+
+=item Mandatory
+
+No
+
+=item Default value
+
+xen
+
=back
+Besides forcing toolstack to use specific backend implementation, this also
+affects the guest's view of the device. For example, "virtio-mmio" requires
+Virtio frontend driver (virtio-blk) to be used. Please note, the virtual
+device (vdev) is not passed to the guest in that case, but it still must be
+specified for the internal purposes.
+
+=back
=head1 COLO Parameters
diff --git a/tools/golang/xenlight/helpers.gen.go b/tools/golang/xenlight/helpers.gen.go
index b746ff1..3190671 100644
--- a/tools/golang/xenlight/helpers.gen.go
+++ b/tools/golang/xenlight/helpers.gen.go
@@ -1751,6 +1751,9 @@ x.DirectIoSafe = bool(xc.direct_io_safe)
if err := x.DiscardEnable.fromC(&xc.discard_enable);err != nil {
return fmt.Errorf("converting field DiscardEnable: %v", err)
}
+x.Protocol = DiskProtocol(xc.protocol)
+x.Irq = uint32(xc.irq)
+x.Base = uint64(xc.base)
if err := x.ColoEnable.fromC(&xc.colo_enable);err != nil {
return fmt.Errorf("converting field ColoEnable: %v", err)
}
@@ -1788,6 +1791,9 @@ xc.direct_io_safe = C.bool(x.DirectIoSafe)
if err := x.DiscardEnable.toC(&xc.discard_enable); err != nil {
return fmt.Errorf("converting field DiscardEnable: %v", err)
}
+xc.protocol = C.libxl_disk_protocol(x.Protocol)
+xc.irq = C.uint32_t(x.Irq)
+xc.base = C.uint64_t(x.Base)
if err := x.ColoEnable.toC(&xc.colo_enable); err != nil {
return fmt.Errorf("converting field ColoEnable: %v", err)
}
diff --git a/tools/golang/xenlight/types.gen.go b/tools/golang/xenlight/types.gen.go
index b1e84d5..4ff14c2 100644
--- a/tools/golang/xenlight/types.gen.go
+++ b/tools/golang/xenlight/types.gen.go
@@ -99,6 +99,14 @@ DiskBackendUnknown DiskBackend = 0
DiskBackendPhy DiskBackend = 1
DiskBackendTap DiskBackend = 2
DiskBackendQdisk DiskBackend = 3
+DiskBackendOther DiskBackend = 4
+)
+
+type DiskProtocol int
+const(
+DiskProtocolUnknown DiskProtocol = 0
+DiskProtocolXen DiskProtocol = 1
+DiskProtocolVirtioMmio DiskProtocol = 2
)
type NicType int
@@ -643,6 +651,9 @@ Readwrite int
IsCdrom int
DirectIoSafe bool
DiscardEnable Defbool
+Protocol DiskProtocol
+Irq uint32
+Base uint64
ColoEnable Defbool
ColoRestoreEnable Defbool
ColoHost string
diff --git a/tools/include/libxl.h b/tools/include/libxl.h
index 51a9b6c..66dbf63 100644
--- a/tools/include/libxl.h
+++ b/tools/include/libxl.h
@@ -528,6 +528,12 @@
#define LIBXL_HAVE_MAX_GRANT_VERSION 1
/*
+ * LIBXL_HAVE_DEVICE_DISK_PROTOCOL indicates that a 'protocol' field
+ * (of libxl_disk_protocol type) is present in libxl_device_disk.
+ */
+#define LIBXL_HAVE_DEVICE_DISK_PROTOCOL 1
+
+/*
* libxl ABI compatibility
*
* The only guarantee which libxl makes regarding ABI compatibility
diff --git a/tools/libs/light/libxl_device.c b/tools/libs/light/libxl_device.c
index e6025d1..846e8c4 100644
--- a/tools/libs/light/libxl_device.c
+++ b/tools/libs/light/libxl_device.c
@@ -289,9 +289,16 @@ static int disk_try_backend(disk_try_backend_args *a,
libxl_disk_backend backend)
{
libxl__gc *gc = a->gc;
+ libxl_disk_protocol protocol = a->disk->protocol;
/* returns 0 (ie, DISK_BACKEND_UNKNOWN) on failure, or
* backend on success */
+ if ((protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO &&
+ backend != LIBXL_DISK_BACKEND_OTHER) ||
+ (protocol != LIBXL_DISK_PROTOCOL_VIRTIO_MMIO &&
+ backend == LIBXL_DISK_BACKEND_OTHER))
+ goto bad_protocol;
+
switch (backend) {
case LIBXL_DISK_BACKEND_PHY:
if (a->disk->format != LIBXL_DISK_FORMAT_RAW) {
@@ -329,6 +336,29 @@ static int disk_try_backend(disk_try_backend_args *a,
if (a->disk->script) goto bad_script;
return backend;
+ case LIBXL_DISK_BACKEND_OTHER:
+ if (a->disk->format != LIBXL_DISK_FORMAT_RAW)
+ goto bad_format;
+
+ if (a->disk->script)
+ goto bad_script;
+
+ if (libxl_defbool_val(a->disk->colo_enable))
+ goto bad_colo;
+
+ if (a->disk->backend_domid != LIBXL_TOOLSTACK_DOMID) {
+ LOG(DEBUG, "Disk vdev=%s, is using a storage driver domain, "
+ "skipping physical device check", a->disk->vdev);
+ return backend;
+ }
+
+ if (libxl__try_phy_backend(a->stab.st_mode))
+ return backend;
+
+ LOG(DEBUG, "Disk vdev=%s, backend other unsuitable as phys path not a "
+ "block device", a->disk->vdev);
+ return 0;
+
default:
LOG(DEBUG, "Disk vdev=%s, backend %d unknown", a->disk->vdev, backend);
return 0;
@@ -352,6 +382,12 @@ static int disk_try_backend(disk_try_backend_args *a,
LOG(DEBUG, "Disk vdev=%s, backend %s not compatible with colo",
a->disk->vdev, libxl_disk_backend_to_string(backend));
return 0;
+
+ bad_protocol:
+ LOG(DEBUG, "Disk vdev=%s, backend %s not compatible with protocol %s",
+ a->disk->vdev, libxl_disk_backend_to_string(backend),
+ libxl_disk_protocol_to_string(protocol));
+ return 0;
}
int libxl__backendpath_parse_domid(libxl__gc *gc, const char *be_path,
@@ -376,8 +412,9 @@ int libxl__device_disk_set_backend(libxl__gc *gc, libxl_device_disk *disk) {
a.gc = gc;
a.disk = disk;
- LOG(DEBUG, "Disk vdev=%s spec.backend=%s", disk->vdev,
- libxl_disk_backend_to_string(disk->backend));
+ LOG(DEBUG, "Disk vdev=%s spec.backend=%s spec.protocol=%s", disk->vdev,
+ libxl_disk_backend_to_string(disk->backend),
+ libxl_disk_protocol_to_string(disk->protocol));
if (disk->format == LIBXL_DISK_FORMAT_EMPTY) {
if (!disk->is_cdrom) {
@@ -392,7 +429,8 @@ int libxl__device_disk_set_backend(libxl__gc *gc, libxl_device_disk *disk) {
}
memset(&a.stab, 0, sizeof(a.stab));
} else if ((disk->backend == LIBXL_DISK_BACKEND_UNKNOWN ||
- disk->backend == LIBXL_DISK_BACKEND_PHY) &&
+ disk->backend == LIBXL_DISK_BACKEND_PHY ||
+ disk->backend == LIBXL_DISK_BACKEND_OTHER) &&
disk->backend_domid == LIBXL_TOOLSTACK_DOMID &&
!disk->script) {
if (stat(disk->pdev_path, &a.stab)) {
@@ -408,7 +446,8 @@ int libxl__device_disk_set_backend(libxl__gc *gc, libxl_device_disk *disk) {
ok=
disk_try_backend(&a, LIBXL_DISK_BACKEND_PHY) ?:
disk_try_backend(&a, LIBXL_DISK_BACKEND_QDISK) ?:
- disk_try_backend(&a, LIBXL_DISK_BACKEND_TAP);
+ disk_try_backend(&a, LIBXL_DISK_BACKEND_TAP) ?:
+ disk_try_backend(&a, LIBXL_DISK_BACKEND_OTHER);
if (ok)
LOG(DEBUG, "Disk vdev=%s, using backend %s",
disk->vdev,
@@ -441,6 +480,16 @@ char *libxl__device_disk_string_of_backend(libxl_disk_backend backend)
case LIBXL_DISK_BACKEND_QDISK: return "qdisk";
case LIBXL_DISK_BACKEND_TAP: return "phy";
case LIBXL_DISK_BACKEND_PHY: return "phy";
+ case LIBXL_DISK_BACKEND_OTHER: return "other";
+ default: return NULL;
+ }
+}
+
+char *libxl__device_disk_string_of_protocol(libxl_disk_protocol protocol)
+{
+ switch (protocol) {
+ case LIBXL_DISK_PROTOCOL_XEN: return "xen";
+ case LIBXL_DISK_PROTOCOL_VIRTIO_MMIO: return "virtio-mmio";
default: return NULL;
}
}
diff --git a/tools/libs/light/libxl_disk.c b/tools/libs/light/libxl_disk.c
index a5ca778..9bb5098 100644
--- a/tools/libs/light/libxl_disk.c
+++ b/tools/libs/light/libxl_disk.c
@@ -163,6 +163,9 @@ static int libxl__device_disk_setdefault(libxl__gc *gc, uint32_t domid,
rc = libxl__resolve_domid(gc, disk->backend_domname, &disk->backend_domid);
if (rc < 0) return rc;
+ if (disk->protocol == LIBXL_DISK_PROTOCOL_UNKNOWN)
+ disk->protocol = LIBXL_DISK_PROTOCOL_XEN;
+
/* Force Qdisk backend for CDROM devices of guests with a device model. */
if (disk->is_cdrom != 0 &&
libxl__domain_type(gc, domid) == LIBXL_DOMAIN_TYPE_HVM) {
@@ -204,6 +207,9 @@ static int libxl__device_from_disk(libxl__gc *gc, uint32_t domid,
case LIBXL_DISK_BACKEND_QDISK:
device->backend_kind = LIBXL__DEVICE_KIND_QDISK;
break;
+ case LIBXL_DISK_BACKEND_OTHER:
+ device->backend_kind = LIBXL__DEVICE_KIND_VIRTIO_DISK;
+ break;
default:
LOGD(ERROR, domid, "Unrecognized disk backend type: %d",
disk->backend);
@@ -212,7 +218,8 @@ static int libxl__device_from_disk(libxl__gc *gc, uint32_t domid,
device->domid = domid;
device->devid = devid;
- device->kind = LIBXL__DEVICE_KIND_VBD;
+ device->kind = disk->backend == LIBXL_DISK_BACKEND_OTHER ?
+ LIBXL__DEVICE_KIND_VIRTIO_DISK : LIBXL__DEVICE_KIND_VBD;
return 0;
}
@@ -317,6 +324,11 @@ static void device_disk_add(libxl__egc *egc, uint32_t domid,
goto out;
}
+ assert((disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO &&
+ disk->backend == LIBXL_DISK_BACKEND_OTHER) ||
+ (disk->protocol != LIBXL_DISK_PROTOCOL_VIRTIO_MMIO &&
+ disk->backend != LIBXL_DISK_BACKEND_OTHER));
+
switch (disk->backend) {
case LIBXL_DISK_BACKEND_PHY:
dev = disk->pdev_path;
@@ -330,7 +342,14 @@ static void device_disk_add(libxl__egc *egc, uint32_t domid,
assert(device->backend_kind == LIBXL__DEVICE_KIND_VBD);
break;
+ case LIBXL_DISK_BACKEND_OTHER:
+ dev = disk->pdev_path;
+
+ flexarray_append(back, "params");
+ flexarray_append(back, dev);
+ assert(device->backend_kind == LIBXL__DEVICE_KIND_VIRTIO_DISK);
+ break;
case LIBXL_DISK_BACKEND_TAP:
LOG(ERROR, "blktap is not supported");
rc = ERROR_FAIL;
@@ -386,6 +405,12 @@ static void device_disk_add(libxl__egc *egc, uint32_t domid,
flexarray_append_pair(back, "discard-enable",
libxl_defbool_val(disk->discard_enable) ?
"1" : "0");
+ flexarray_append(back, "protocol");
+ flexarray_append(back, libxl__device_disk_string_of_protocol(disk->protocol));
+ if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
+ flexarray_append_pair(back, "base", GCSPRINTF("%"PRIu64, disk->base));
+ flexarray_append_pair(back, "irq", GCSPRINTF("%u", disk->irq));
+ }
flexarray_append(front, "backend-id");
flexarray_append(front, GCSPRINTF("%d", disk->backend_domid));
@@ -532,6 +557,37 @@ static int libxl__disk_from_xenstore(libxl__gc *gc, const char *libxl_path,
}
libxl_string_to_backend(ctx, tmp, &(disk->backend));
+ tmp = libxl__xs_read(gc, XBT_NULL,
+ GCSPRINTF("%s/protocol", libxl_path));
+ if (!tmp) {
+ LOG(ERROR, "Missing xenstore node %s/protocol", libxl_path);
+ goto cleanup;
+ }
+ if (!strcmp(tmp, "xen"))
+ disk->protocol = LIBXL_DISK_PROTOCOL_XEN;
+ else if (!strcmp(tmp, "virtio-mmio"))
+ disk->protocol = LIBXL_DISK_PROTOCOL_VIRTIO_MMIO;
+ else
+ disk->protocol = LIBXL_DISK_PROTOCOL_UNKNOWN;
+
+ if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
+ tmp = libxl__xs_read(gc, XBT_NULL,
+ GCSPRINTF("%s/base", libxl_path));
+ if (!tmp) {
+ LOG(ERROR, "Missing xenstore node %s/base", libxl_path);
+ goto cleanup;
+ }
+ disk->base = strtoul(tmp, NULL, 10);
+
+ tmp = libxl__xs_read(gc, XBT_NULL,
+ GCSPRINTF("%s/irq", libxl_path));
+ if (!tmp) {
+ LOG(ERROR, "Missing xenstore node %s/irq", libxl_path);
+ goto cleanup;
+ }
+ disk->irq = strtoul(tmp, NULL, 10);
+ }
+
disk->vdev = xs_read(ctx->xsh, XBT_NULL,
GCSPRINTF("%s/dev", libxl_path), &len);
if (!disk->vdev) {
@@ -575,6 +631,41 @@ cleanup:
return rc;
}
+static int libxl_device_disk_get_path(libxl__gc *gc, uint32_t domid,
+ char **path)
+{
+ const char *dir;
+ int rc;
+
+ /*
+ * As we don't know exactly what device kind to be used here, guess it
+ * by checking the presence of the corresponding path in Xenstore.
+ * First, try to read path for vbd device (default) and if not exists
+ * read path for virtio_disk device. This will work as long as both Xen PV
+ * and Virtio disk devices are not assigned to the same guest.
+ */
+ *path = GCSPRINTF("%s/device/%s",
+ libxl__xs_libxl_path(gc, domid),
+ libxl__device_kind_to_string(LIBXL__DEVICE_KIND_VBD));
+
+ rc = libxl__xs_read_checked(gc, XBT_NULL, *path, &dir);
+ if (rc)
+ return rc;
+
+ if (dir)
+ return 0;
+
+ *path = GCSPRINTF("%s/device/%s",
+ libxl__xs_libxl_path(gc, domid),
+ libxl__device_kind_to_string(LIBXL__DEVICE_KIND_VIRTIO_DISK));
+
+ rc = libxl__xs_read_checked(gc, XBT_NULL, *path, &dir);
+ if (rc)
+ return rc;
+
+ return 0;
+}
+
int libxl_vdev_to_device_disk(libxl_ctx *ctx, uint32_t domid,
const char *vdev, libxl_device_disk *disk)
{
@@ -588,10 +679,12 @@ int libxl_vdev_to_device_disk(libxl_ctx *ctx, uint32_t domid,
libxl_device_disk_init(disk);
- libxl_path = libxl__domain_device_libxl_path(gc, domid, devid,
- LIBXL__DEVICE_KIND_VBD);
+ rc = libxl_device_disk_get_path(gc, domid, &libxl_path);
+ if (rc)
+ return rc;
- rc = libxl__disk_from_xenstore(gc, libxl_path, devid, disk);
+ rc = libxl__disk_from_xenstore(gc, GCSPRINTF("%s/%d", libxl_path, devid),
+ devid, disk);
GC_FREE;
return rc;
@@ -605,16 +698,19 @@ int libxl_device_disk_getinfo(libxl_ctx *ctx, uint32_t domid,
char *fe_path, *libxl_path;
char *val;
int rc;
+ libxl__device_kind kind;
diskinfo->backend = NULL;
diskinfo->devid = libxl__device_disk_dev_number(disk->vdev, NULL, NULL);
- /* tap devices entries in xenstore are written as vbd devices. */
+ /* tap devices entries in xenstore are written as vbd/virtio_disk devices. */
+ kind = disk->backend == LIBXL_DISK_BACKEND_OTHER ?
+ LIBXL__DEVICE_KIND_VIRTIO_DISK : LIBXL__DEVICE_KIND_VBD;
fe_path = libxl__domain_device_frontend_path(gc, domid, diskinfo->devid,
- LIBXL__DEVICE_KIND_VBD);
+ kind);
libxl_path = libxl__domain_device_libxl_path(gc, domid, diskinfo->devid,
- LIBXL__DEVICE_KIND_VBD);
+ kind);
diskinfo->backend = xs_read(ctx->xsh, XBT_NULL,
GCSPRINTF("%s/backend", libxl_path), NULL);
if (!diskinfo->backend) {
@@ -1418,6 +1514,7 @@ LIBXL_DEFINE_DEVICE_LIST(disk)
#define libxl__device_disk_update_devid NULL
DEFINE_DEVICE_TYPE_STRUCT(disk, VBD, disks,
+ .get_path = libxl_device_disk_get_path,
.merge = libxl_device_disk_merge,
.dm_needed = libxl_device_disk_dm_needed,
.from_xenstore = (device_from_xenstore_fn_t)libxl__disk_from_xenstore,
diff --git a/tools/libs/light/libxl_internal.h b/tools/libs/light/libxl_internal.h
index a26daec..f4eb75d 100644
--- a/tools/libs/light/libxl_internal.h
+++ b/tools/libs/light/libxl_internal.h
@@ -1493,6 +1493,7 @@ _hidden char * libxl__domain_pvcontrol_read(libxl__gc *gc,
/* from xl_device */
_hidden char *libxl__device_disk_string_of_backend(libxl_disk_backend backend);
+_hidden char *libxl__device_disk_string_of_protocol(libxl_disk_protocol protocol);
_hidden char *libxl__device_disk_string_of_format(libxl_disk_format format);
_hidden const char *libxl__qemu_disk_format_string(libxl_disk_format format);
_hidden int libxl__device_disk_set_backend(libxl__gc*, libxl_device_disk*);
diff --git a/tools/libs/light/libxl_types.idl b/tools/libs/light/libxl_types.idl
index 2a42da2..5d57e3a 100644
--- a/tools/libs/light/libxl_types.idl
+++ b/tools/libs/light/libxl_types.idl
@@ -130,6 +130,13 @@ libxl_disk_backend = Enumeration("disk_backend", [
(1, "PHY"),
(2, "TAP"),
(3, "QDISK"),
+ (4, "OTHER"),
+ ])
+
+libxl_disk_protocol = Enumeration("disk_protocol", [
+ (0, "UNKNOWN"),
+ (1, "XEN"),
+ (2, "VIRTIO_MMIO"),
])
libxl_nic_type = Enumeration("nic_type", [
@@ -704,6 +711,9 @@ libxl_device_disk = Struct("device_disk", [
("is_cdrom", integer),
("direct_io_safe", bool),
("discard_enable", libxl_defbool),
+ ("protocol", libxl_disk_protocol),
+ ("irq", uint32),
+ ("base", uint64),
# Note that the COLO configuration settings should be considered unstable.
# They may change incompatibly in future versions of Xen.
("colo_enable", libxl_defbool),
diff --git a/tools/libs/light/libxl_types_internal.idl b/tools/libs/light/libxl_types_internal.idl
index 3593e21..8f71980 100644
--- a/tools/libs/light/libxl_types_internal.idl
+++ b/tools/libs/light/libxl_types_internal.idl
@@ -32,6 +32,7 @@ libxl__device_kind = Enumeration("device_kind", [
(14, "PVCALLS"),
(15, "VSND"),
(16, "VINPUT"),
+ (17, "VIRTIO_DISK"),
])
libxl__console_backend = Enumeration("console_backend", [
diff --git a/tools/libs/light/libxl_utils.c b/tools/libs/light/libxl_utils.c
index b91c2ca..eb816dd 100644
--- a/tools/libs/light/libxl_utils.c
+++ b/tools/libs/light/libxl_utils.c
@@ -303,6 +303,8 @@ int libxl_string_to_backend(libxl_ctx *ctx, char *s, libxl_disk_backend *backend
*backend = LIBXL_DISK_BACKEND_TAP;
} else if (!strcmp(s, "qdisk")) {
*backend = LIBXL_DISK_BACKEND_QDISK;
+ } else if (!strcmp(s, "other")) {
+ *backend = LIBXL_DISK_BACKEND_OTHER;
} else if (!strcmp(s, "tap")) {
p = strchr(s, ':');
if (!p) {
diff --git a/tools/libs/util/libxlu_disk_l.c b/tools/libs/util/libxlu_disk_l.c
index 32d4b74..af22055 100644
--- a/tools/libs/util/libxlu_disk_l.c
+++ b/tools/libs/util/libxlu_disk_l.c
@@ -549,8 +549,8 @@ static void yynoreturn yy_fatal_error ( const char* msg , yyscan_t yyscanner );
yyg->yy_hold_char = *yy_cp; \
*yy_cp = '\0'; \
yyg->yy_c_buf_p = yy_cp;
-#define YY_NUM_RULES 36
-#define YY_END_OF_BUFFER 37
+#define YY_NUM_RULES 37
+#define YY_END_OF_BUFFER 38
/* This struct is not used in this scanner,
but its presence is necessary. */
struct yy_trans_info
@@ -558,74 +558,75 @@ struct yy_trans_info
flex_int32_t yy_verify;
flex_int32_t yy_nxt;
};
-static const flex_int16_t yy_acclist[575] =
+static const flex_int16_t yy_acclist[589] =
{ 0,
- 35, 35, 37, 33, 34, 36, 8193, 33, 34, 36,
- 16385, 8193, 33, 36,16385, 33, 34, 36, 34, 36,
- 33, 34, 36, 33, 34, 36, 33, 34, 36, 33,
- 34, 36, 33, 34, 36, 33, 34, 36, 33, 34,
- 36, 33, 34, 36, 33, 34, 36, 33, 34, 36,
- 33, 34, 36, 33, 34, 36, 33, 34, 36, 33,
- 34, 36, 33, 34, 36, 33, 34, 36, 35, 36,
- 36, 33, 33, 8193, 33, 8193, 33,16385, 8193, 33,
- 8193, 33, 33, 8224, 33,16416, 33, 33, 33, 33,
- 33, 33, 33, 33, 33, 33, 33, 33, 33, 33,
-
- 33, 33, 33, 33, 33, 33, 33, 33, 33, 35,
- 8193, 33, 8193, 33, 8193, 8224, 33, 8224, 33, 8224,
- 23, 33, 33, 33, 33, 33, 33, 33, 33, 33,
- 33, 33, 33, 33, 33, 33, 33, 33, 33, 33,
- 33, 33, 33, 33, 33, 8224, 33, 8224, 33, 8224,
- 23, 33, 33, 28, 8224, 33,16416, 33, 33, 15,
- 33, 33, 33, 33, 33, 33, 33, 33, 33, 8217,
- 8224, 33,16409,16416, 33, 33, 31, 8224, 33,16416,
- 33, 8216, 8224, 33,16408,16416, 33, 33, 8219, 8224,
- 33,16411,16416, 33, 33, 33, 33, 33, 28, 8224,
-
- 33, 28, 8224, 33, 28, 33, 28, 8224, 33, 3,
- 33, 15, 33, 33, 33, 33, 33, 30, 8224, 33,
- 16416, 33, 33, 33, 8217, 8224, 33, 8217, 8224, 33,
- 8217, 33, 8217, 8224, 33, 33, 31, 8224, 33, 31,
- 8224, 33, 31, 33, 31, 8224, 8216, 8224, 33, 8216,
- 8224, 33, 8216, 33, 8216, 8224, 33, 8219, 8224, 33,
- 8219, 8224, 33, 8219, 33, 8219, 8224, 33, 33, 10,
- 33, 33, 28, 8224, 33, 28, 8224, 33, 28, 8224,
- 28, 33, 28, 33, 3, 33, 33, 33, 33, 33,
- 33, 33, 30, 8224, 33, 30, 8224, 33, 30, 33,
-
- 30, 8224, 33, 33, 29, 8224, 33,16416, 8217, 8224,
- 33, 8217, 8224, 33, 8217, 8224, 8217, 33, 8217, 33,
- 33, 31, 8224, 33, 31, 8224, 33, 31, 8224, 31,
- 33, 31, 8216, 8224, 33, 8216, 8224, 33, 8216, 8224,
- 8216, 33, 8216, 33, 8219, 8224, 33, 8219, 8224, 33,
- 8219, 8224, 8219, 33, 8219, 33, 33, 10, 23, 10,
- 7, 33, 33, 33, 33, 33, 33, 33, 13, 33,
- 30, 8224, 33, 30, 8224, 33, 30, 8224, 30, 33,
- 30, 2, 33, 29, 8224, 33, 29, 8224, 33, 29,
- 33, 29, 8224, 16, 33, 33, 11, 33, 22, 10,
-
- 10, 23, 7, 23, 7, 33, 8, 33, 33, 33,
- 33, 6, 33, 13, 33, 2, 23, 2, 33, 29,
- 8224, 33, 29, 8224, 33, 29, 8224, 29, 33, 29,
- 16, 33, 33, 11, 23, 11, 26, 8224, 33,16416,
- 22, 23, 22, 7, 7, 23, 33, 8, 23, 8,
- 33, 33, 33, 33, 6, 23, 6, 6, 23, 6,
- 23, 33, 2, 2, 23, 33, 33, 11, 11, 23,
- 26, 8224, 33, 26, 8224, 33, 26, 33, 26, 8224,
- 22, 23, 33, 8, 8, 23, 33, 33, 17, 18,
- 6, 6, 23, 6, 6, 33, 33, 14, 33, 26,
-
- 8224, 33, 26, 8224, 33, 26, 8224, 26, 33, 26,
- 33, 33, 33, 17, 23, 17, 18, 23, 18, 6,
- 6, 33, 33, 14, 33, 20, 9, 19, 17, 17,
- 23, 18, 18, 23, 6, 5, 6, 33, 21, 20,
- 23, 20, 9, 23, 9, 19, 23, 19, 4, 6,
- 5, 6, 33, 21, 23, 21, 20, 20, 23, 9,
- 9, 23, 19, 19, 23, 4, 6, 12, 33, 21,
- 21, 23, 12, 33
+ 36, 36, 38, 34, 35, 37, 8193, 34, 35, 37,
+ 16385, 8193, 34, 37,16385, 34, 35, 37, 35, 37,
+ 34, 35, 37, 34, 35, 37, 34, 35, 37, 34,
+ 35, 37, 34, 35, 37, 34, 35, 37, 34, 35,
+ 37, 34, 35, 37, 34, 35, 37, 34, 35, 37,
+ 34, 35, 37, 34, 35, 37, 34, 35, 37, 34,
+ 35, 37, 34, 35, 37, 34, 35, 37, 36, 37,
+ 37, 34, 34, 8193, 34, 8193, 34,16385, 8193, 34,
+ 8193, 34, 34, 8225, 34,16417, 34, 34, 34, 34,
+ 34, 34, 34, 34, 34, 34, 34, 34, 34, 34,
+
+ 34, 34, 34, 34, 34, 34, 34, 34, 34, 34,
+ 36, 8193, 34, 8193, 34, 8193, 8225, 34, 8225, 34,
+ 8225, 24, 34, 34, 34, 34, 34, 34, 34, 34,
+ 34, 34, 34, 34, 34, 34, 34, 34, 34, 34,
+ 34, 34, 34, 34, 34, 34, 34, 8225, 34, 8225,
+ 34, 8225, 24, 34, 34, 29, 8225, 34,16417, 34,
+ 34, 16, 34, 34, 34, 34, 34, 34, 34, 34,
+ 34, 8218, 8225, 34,16410,16417, 34, 34, 32, 8225,
+ 34,16417, 34, 34, 8217, 8225, 34,16409,16417, 34,
+ 34, 8220, 8225, 34,16412,16417, 34, 34, 34, 34,
+
+ 34, 29, 8225, 34, 29, 8225, 34, 29, 34, 29,
+ 8225, 34, 3, 34, 16, 34, 34, 34, 34, 34,
+ 31, 8225, 34,16417, 34, 34, 34, 8218, 8225, 34,
+ 8218, 8225, 34, 8218, 34, 8218, 8225, 34, 34, 32,
+ 8225, 34, 32, 8225, 34, 32, 34, 32, 8225, 34,
+ 8217, 8225, 34, 8217, 8225, 34, 8217, 34, 8217, 8225,
+ 34, 8220, 8225, 34, 8220, 8225, 34, 8220, 34, 8220,
+ 8225, 34, 34, 11, 34, 34, 29, 8225, 34, 29,
+ 8225, 34, 29, 8225, 29, 34, 29, 34, 3, 34,
+ 34, 34, 34, 34, 34, 34, 31, 8225, 34, 31,
+
+ 8225, 34, 31, 34, 31, 8225, 34, 34, 30, 8225,
+ 34,16417, 8218, 8225, 34, 8218, 8225, 34, 8218, 8225,
+ 8218, 34, 8218, 34, 34, 32, 8225, 34, 32, 8225,
+ 34, 32, 8225, 32, 34, 32, 34, 8217, 8225, 34,
+ 8217, 8225, 34, 8217, 8225, 8217, 34, 8217, 34, 8220,
+ 8225, 34, 8220, 8225, 34, 8220, 8225, 8220, 34, 8220,
+ 34, 34, 11, 24, 11, 7, 34, 34, 34, 34,
+ 34, 34, 34, 14, 34, 31, 8225, 34, 31, 8225,
+ 34, 31, 8225, 31, 34, 31, 2, 34, 30, 8225,
+ 34, 30, 8225, 34, 30, 34, 30, 8225, 17, 34,
+
+ 34, 34, 12, 34, 23, 11, 11, 24, 7, 24,
+ 7, 34, 8, 34, 34, 34, 34, 6, 34, 14,
+ 34, 2, 24, 2, 34, 30, 8225, 34, 30, 8225,
+ 34, 30, 8225, 30, 34, 30, 17, 34, 34, 34,
+ 12, 24, 12, 27, 8225, 34,16417, 23, 24, 23,
+ 7, 7, 24, 34, 8, 24, 8, 34, 34, 34,
+ 34, 6, 24, 6, 6, 24, 6, 24, 34, 2,
+ 2, 24, 34, 34, 10, 12, 12, 24, 27, 8225,
+ 34, 27, 8225, 34, 27, 34, 27, 8225, 23, 24,
+ 34, 8, 8, 24, 34, 34, 18, 19, 6, 6,
+
+ 24, 6, 6, 34, 34, 15, 34, 10, 24, 10,
+ 27, 8225, 34, 27, 8225, 34, 27, 8225, 27, 34,
+ 27, 34, 34, 34, 18, 24, 18, 19, 24, 19,
+ 6, 6, 34, 34, 15, 34, 10, 10, 24, 21,
+ 9, 20, 18, 18, 24, 19, 19, 24, 6, 5,
+ 6, 34, 22, 21, 24, 21, 9, 24, 9, 20,
+ 24, 20, 4, 6, 5, 6, 34, 22, 24, 22,
+ 21, 21, 24, 9, 9, 24, 20, 20, 24, 4,
+ 6, 13, 34, 22, 22, 24, 13, 34
} ;
-static const flex_int16_t yy_accept[356] =
+static const flex_int16_t yy_accept[368] =
{ 0,
1, 1, 1, 2, 3, 4, 7, 12, 16, 19,
21, 24, 27, 30, 33, 36, 39, 42, 45, 48,
@@ -633,39 +634,40 @@ static const flex_int16_t yy_accept[356] =
74, 76, 79, 81, 82, 83, 84, 87, 87, 88,
89, 90, 91, 92, 93, 94, 95, 96, 97, 98,
99, 100, 101, 102, 103, 104, 105, 106, 107, 108,
- 109, 110, 111, 113, 115, 116, 118, 120, 121, 122,
+ 109, 110, 111, 112, 114, 116, 117, 119, 121, 122,
123, 124, 125, 126, 127, 128, 129, 130, 131, 132,
133, 134, 135, 136, 137, 138, 139, 140, 141, 142,
- 143, 144, 145, 146, 148, 150, 151, 152, 153, 154,
-
- 158, 159, 160, 162, 163, 164, 165, 166, 167, 168,
- 169, 170, 175, 176, 177, 181, 182, 187, 188, 189,
- 194, 195, 196, 197, 198, 199, 202, 205, 207, 209,
- 210, 212, 214, 215, 216, 217, 218, 222, 223, 224,
- 225, 228, 231, 233, 235, 236, 237, 240, 243, 245,
- 247, 250, 253, 255, 257, 258, 261, 264, 266, 268,
- 269, 270, 271, 272, 273, 276, 279, 281, 283, 284,
- 285, 287, 288, 289, 290, 291, 292, 293, 296, 299,
- 301, 303, 304, 305, 309, 312, 315, 317, 319, 320,
- 321, 322, 325, 328, 330, 332, 333, 336, 339, 341,
-
- 343, 344, 345, 348, 351, 353, 355, 356, 357, 358,
- 360, 361, 362, 363, 364, 365, 366, 367, 368, 369,
- 371, 374, 377, 379, 381, 382, 383, 384, 387, 390,
- 392, 394, 396, 397, 398, 399, 400, 401, 403, 405,
- 406, 407, 408, 409, 410, 411, 412, 413, 414, 416,
- 418, 419, 420, 423, 426, 428, 430, 431, 433, 434,
- 436, 437, 441, 443, 444, 445, 447, 448, 450, 451,
- 452, 453, 454, 455, 457, 458, 460, 462, 463, 464,
- 466, 467, 468, 469, 471, 474, 477, 479, 481, 483,
- 484, 485, 487, 488, 489, 490, 491, 492, 494, 495,
-
- 496, 497, 498, 500, 503, 506, 508, 510, 511, 512,
- 513, 514, 516, 517, 519, 520, 521, 522, 523, 524,
- 526, 527, 528, 529, 530, 532, 533, 535, 536, 538,
- 539, 540, 542, 543, 545, 546, 548, 549, 551, 553,
- 554, 556, 557, 558, 560, 561, 563, 564, 566, 568,
- 570, 571, 573, 575, 575
+ 143, 144, 145, 146, 147, 148, 150, 152, 153, 154,
+
+ 155, 156, 160, 161, 162, 164, 165, 166, 167, 168,
+ 169, 170, 171, 172, 177, 178, 179, 183, 184, 185,
+ 190, 191, 192, 197, 198, 199, 200, 201, 202, 205,
+ 208, 210, 212, 213, 215, 217, 218, 219, 220, 221,
+ 225, 226, 227, 228, 231, 234, 236, 238, 239, 240,
+ 243, 246, 248, 250, 251, 254, 257, 259, 261, 262,
+ 265, 268, 270, 272, 273, 274, 275, 276, 277, 280,
+ 283, 285, 287, 288, 289, 291, 292, 293, 294, 295,
+ 296, 297, 300, 303, 305, 307, 308, 309, 313, 316,
+ 319, 321, 323, 324, 325, 326, 329, 332, 334, 336,
+
+ 337, 338, 341, 344, 346, 348, 349, 350, 353, 356,
+ 358, 360, 361, 362, 363, 365, 366, 367, 368, 369,
+ 370, 371, 372, 373, 374, 376, 379, 382, 384, 386,
+ 387, 388, 389, 392, 395, 397, 399, 401, 402, 403,
+ 404, 405, 406, 407, 409, 411, 412, 413, 414, 415,
+ 416, 417, 418, 419, 420, 422, 424, 425, 426, 429,
+ 432, 434, 436, 437, 439, 440, 441, 443, 444, 448,
+ 450, 451, 452, 454, 455, 457, 458, 459, 460, 461,
+ 462, 464, 465, 467, 469, 470, 471, 473, 474, 475,
+ 476, 477, 479, 482, 485, 487, 489, 491, 492, 493,
+
+ 495, 496, 497, 498, 499, 500, 502, 503, 504, 505,
+ 506, 508, 510, 511, 514, 517, 519, 521, 522, 523,
+ 524, 525, 527, 528, 530, 531, 532, 533, 534, 535,
+ 537, 538, 540, 541, 542, 543, 544, 546, 547, 549,
+ 550, 552, 553, 554, 556, 557, 559, 560, 562, 563,
+ 565, 567, 568, 570, 571, 572, 574, 575, 577, 578,
+ 580, 582, 584, 585, 587, 589, 589
} ;
static const YY_CHAR yy_ec[256] =
@@ -708,216 +710,221 @@ static const YY_CHAR yy_meta[35] =
1, 1, 1, 1
} ;
-static const flex_int16_t yy_base[424] =
+static const flex_int16_t yy_base[438] =
{ 0,
- 0, 0, 901, 900, 902, 897, 33, 36, 905, 905,
- 45, 63, 31, 42, 51, 52, 890, 33, 65, 67,
- 69, 70, 889, 71, 888, 75, 0, 905, 893, 905,
- 91, 94, 0, 0, 103, 886, 112, 0, 89, 98,
- 113, 92, 114, 99, 100, 48, 121, 116, 119, 74,
- 124, 129, 123, 135, 132, 133, 137, 134, 138, 139,
- 141, 0, 155, 0, 0, 164, 0, 0, 849, 142,
- 152, 164, 140, 161, 165, 166, 167, 168, 169, 173,
- 174, 178, 176, 180, 184, 208, 189, 183, 192, 195,
- 215, 191, 193, 223, 0, 0, 905, 208, 204, 236,
-
- 219, 209, 238, 196, 237, 831, 242, 815, 241, 224,
- 243, 261, 244, 259, 277, 266, 286, 250, 288, 298,
- 249, 283, 274, 282, 294, 308, 0, 310, 0, 295,
- 305, 905, 308, 306, 313, 314, 342, 319, 316, 320,
- 331, 0, 349, 0, 342, 344, 356, 0, 358, 0,
- 365, 0, 367, 0, 354, 375, 0, 377, 0, 363,
- 356, 809, 327, 322, 384, 0, 0, 0, 0, 379,
- 905, 382, 384, 386, 390, 372, 392, 403, 0, 410,
- 0, 407, 413, 423, 426, 0, 0, 0, 0, 409,
- 424, 435, 0, 0, 0, 0, 437, 0, 0, 0,
-
- 0, 433, 444, 0, 0, 0, 0, 391, 440, 781,
- 905, 769, 439, 445, 444, 447, 449, 454, 453, 399,
- 464, 0, 0, 0, 0, 757, 465, 476, 0, 478,
- 0, 479, 476, 753, 462, 490, 749, 905, 745, 905,
- 483, 737, 424, 485, 487, 490, 500, 493, 905, 729,
- 905, 502, 518, 0, 0, 0, 0, 905, 498, 721,
- 905, 527, 713, 0, 705, 905, 495, 697, 905, 365,
- 521, 528, 530, 685, 905, 534, 540, 540, 657, 905,
- 537, 542, 650, 905, 553, 0, 557, 0, 0, 551,
- 641, 905, 558, 557, 633, 614, 613, 905, 547, 555,
-
- 563, 565, 569, 584, 0, 0, 0, 0, 583, 570,
- 585, 612, 905, 601, 905, 522, 580, 589, 594, 905,
- 600, 585, 563, 520, 905, 514, 905, 586, 486, 597,
- 480, 441, 905, 416, 905, 345, 905, 334, 905, 601,
- 254, 905, 242, 905, 200, 905, 151, 905, 905, 607,
- 86, 905, 905, 905, 620, 624, 627, 631, 635, 639,
- 643, 647, 651, 655, 659, 663, 667, 671, 675, 679,
- 683, 687, 691, 695, 699, 703, 707, 711, 715, 719,
- 723, 727, 731, 735, 739, 743, 747, 751, 755, 759,
- 763, 767, 771, 775, 779, 783, 787, 791, 795, 799,
-
- 803, 807, 811, 815, 819, 823, 827, 831, 835, 839,
- 843, 847, 851, 855, 859, 863, 867, 871, 875, 879,
- 883, 887, 891
+ 0, 0, 927, 926, 928, 923, 33, 36, 931, 931,
+ 45, 63, 31, 42, 51, 52, 916, 33, 65, 67,
+ 69, 70, 915, 71, 914, 75, 0, 931, 919, 931,
+ 91, 95, 0, 0, 104, 912, 113, 0, 91, 99,
+ 114, 92, 115, 80, 100, 48, 119, 121, 122, 74,
+ 123, 128, 131, 129, 125, 133, 136, 135, 137, 138,
+ 140, 142, 0, 165, 0, 0, 167, 0, 0, 917,
+ 148, 143, 165, 146, 162, 167, 168, 169, 170, 171,
+ 175, 177, 180, 178, 182, 186, 210, 193, 188, 192,
+ 196, 197, 218, 211, 185, 226, 0, 0, 931, 210,
+
+ 222, 253, 195, 223, 232, 229, 231, 910, 236, 909,
+ 237, 230, 249, 267, 253, 261, 279, 224, 242, 288,
+ 275, 291, 301, 268, 284, 298, 285, 301, 313, 0,
+ 315, 0, 302, 310, 931, 314, 311, 313, 319, 346,
+ 323, 320, 324, 353, 0, 355, 0, 332, 348, 362,
+ 0, 364, 0, 325, 366, 0, 374, 0, 353, 376,
+ 0, 383, 0, 362, 364, 914, 331, 379, 391, 0,
+ 0, 0, 0, 378, 931, 384, 386, 389, 393, 399,
+ 395, 411, 0, 415, 0, 402, 411, 424, 427, 0,
+ 0, 0, 0, 410, 425, 436, 0, 0, 0, 0,
+
+ 425, 438, 0, 0, 0, 0, 434, 447, 0, 0,
+ 0, 0, 438, 443, 871, 931, 855, 442, 450, 449,
+ 452, 454, 459, 458, 453, 469, 0, 0, 0, 0,
+ 835, 470, 481, 0, 483, 0, 484, 481, 479, 823,
+ 467, 495, 795, 931, 783, 931, 489, 771, 485, 491,
+ 493, 496, 508, 504, 931, 767, 931, 511, 523, 0,
+ 0, 0, 0, 931, 518, 501, 763, 931, 545, 759,
+ 0, 751, 931, 523, 743, 931, 508, 525, 530, 546,
+ 735, 931, 552, 553, 555, 727, 931, 549, 501, 719,
+ 711, 931, 560, 0, 569, 0, 0, 562, 699, 931,
+
+ 530, 567, 671, 664, 655, 931, 562, 566, 571, 573,
+ 588, 647, 931, 594, 0, 0, 0, 0, 578, 595,
+ 597, 628, 931, 627, 931, 584, 591, 602, 604, 931,
+ 626, 931, 625, 624, 623, 622, 931, 603, 931, 598,
+ 577, 606, 572, 537, 931, 530, 931, 442, 931, 402,
+ 931, 610, 377, 931, 334, 931, 260, 931, 204, 931,
+ 931, 621, 147, 931, 931, 931, 634, 638, 641, 645,
+ 649, 653, 657, 661, 665, 669, 673, 677, 681, 685,
+ 689, 693, 697, 701, 705, 709, 713, 717, 721, 725,
+ 729, 733, 737, 741, 745, 749, 753, 757, 761, 765,
+
+ 769, 773, 777, 781, 785, 789, 793, 797, 801, 805,
+ 809, 813, 817, 821, 825, 829, 833, 837, 841, 845,
+ 849, 853, 857, 861, 865, 869, 873, 877, 881, 885,
+ 889, 893, 897, 901, 905, 909, 913
} ;
-static const flex_int16_t yy_def[424] =
+static const flex_int16_t yy_def[438] =
{ 0,
- 354, 1, 355, 355, 354, 356, 357, 357, 354, 354,
- 358, 358, 12, 12, 12, 12, 12, 12, 12, 12,
- 12, 12, 12, 12, 12, 12, 359, 354, 356, 354,
- 360, 357, 361, 361, 362, 12, 356, 363, 12, 12,
+ 366, 1, 367, 367, 366, 368, 369, 369, 366, 366,
+ 370, 370, 12, 12, 12, 12, 12, 12, 12, 12,
+ 12, 12, 12, 12, 12, 12, 371, 366, 368, 366,
+ 372, 369, 373, 373, 374, 12, 368, 375, 12, 12,
12, 12, 12, 12, 12, 12, 12, 12, 12, 12,
12, 12, 12, 12, 12, 12, 12, 12, 12, 12,
- 12, 359, 360, 361, 361, 364, 365, 365, 354, 12,
+ 12, 12, 371, 372, 373, 373, 376, 377, 377, 366,
12, 12, 12, 12, 12, 12, 12, 12, 12, 12,
- 12, 12, 12, 12, 12, 362, 12, 12, 12, 12,
- 12, 12, 12, 364, 365, 365, 354, 12, 12, 366,
-
- 12, 12, 12, 12, 12, 12, 12, 12, 12, 12,
- 12, 367, 86, 86, 368, 12, 369, 12, 12, 370,
- 12, 12, 12, 12, 12, 371, 372, 366, 372, 12,
- 12, 354, 86, 12, 12, 12, 373, 12, 12, 12,
- 374, 375, 367, 375, 86, 86, 376, 377, 368, 377,
- 378, 379, 369, 379, 12, 380, 381, 370, 381, 12,
- 12, 382, 12, 12, 371, 372, 372, 383, 383, 12,
- 354, 86, 86, 86, 12, 12, 12, 384, 385, 373,
- 385, 12, 12, 386, 374, 375, 375, 387, 387, 86,
- 86, 376, 377, 377, 388, 388, 378, 379, 379, 389,
-
- 389, 12, 380, 381, 381, 390, 390, 12, 12, 391,
- 354, 392, 86, 12, 86, 86, 86, 12, 86, 12,
- 384, 385, 385, 393, 393, 394, 86, 395, 396, 386,
- 396, 86, 86, 397, 12, 398, 391, 354, 399, 354,
- 86, 400, 12, 86, 86, 86, 401, 86, 354, 402,
- 354, 86, 395, 396, 396, 403, 403, 354, 86, 404,
- 354, 405, 406, 406, 399, 354, 86, 407, 354, 12,
- 86, 86, 86, 408, 354, 408, 408, 86, 402, 354,
- 86, 86, 404, 354, 409, 410, 405, 410, 406, 86,
- 407, 354, 12, 86, 411, 412, 408, 354, 408, 408,
-
- 86, 86, 86, 409, 410, 410, 413, 413, 86, 12,
- 86, 414, 354, 415, 354, 408, 408, 86, 86, 354,
- 416, 417, 418, 414, 354, 415, 354, 408, 408, 86,
- 419, 420, 354, 421, 354, 422, 354, 408, 354, 86,
- 423, 354, 420, 354, 421, 354, 422, 354, 354, 86,
- 423, 354, 354, 0, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
-
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354
+ 12, 12, 12, 12, 12, 12, 374, 12, 12, 12,
+ 12, 12, 12, 12, 12, 376, 377, 377, 366, 12,
+
+ 12, 378, 12, 12, 12, 12, 12, 12, 12, 12,
+ 12, 12, 12, 379, 87, 87, 380, 12, 12, 381,
+ 12, 12, 382, 12, 12, 12, 12, 12, 383, 384,
+ 378, 384, 12, 12, 366, 87, 12, 12, 12, 385,
+ 12, 12, 12, 386, 387, 379, 387, 87, 87, 388,
+ 389, 380, 389, 12, 390, 391, 381, 391, 12, 392,
+ 393, 382, 393, 12, 12, 394, 12, 12, 383, 384,
+ 384, 395, 395, 12, 366, 87, 87, 87, 12, 12,
+ 12, 396, 397, 385, 397, 12, 12, 398, 386, 387,
+ 387, 399, 399, 87, 87, 388, 389, 389, 400, 400,
+
+ 12, 390, 391, 391, 401, 401, 12, 392, 393, 393,
+ 402, 402, 12, 12, 403, 366, 404, 87, 12, 87,
+ 87, 87, 12, 87, 12, 396, 397, 397, 405, 405,
+ 406, 87, 407, 408, 398, 408, 87, 87, 12, 409,
+ 12, 410, 403, 366, 411, 366, 87, 412, 12, 87,
+ 87, 87, 413, 87, 366, 414, 366, 87, 407, 408,
+ 408, 415, 415, 366, 87, 12, 416, 366, 417, 418,
+ 418, 411, 366, 87, 419, 366, 12, 87, 87, 87,
+ 420, 366, 420, 420, 87, 414, 366, 87, 87, 421,
+ 416, 366, 422, 423, 417, 423, 418, 87, 419, 366,
+
+ 12, 87, 424, 425, 420, 366, 420, 420, 87, 87,
+ 87, 426, 366, 422, 423, 423, 427, 427, 87, 12,
+ 87, 428, 366, 429, 366, 420, 420, 87, 87, 366,
+ 426, 366, 430, 431, 432, 428, 366, 429, 366, 420,
+ 420, 87, 433, 434, 366, 435, 366, 436, 366, 420,
+ 366, 87, 437, 366, 434, 366, 435, 366, 436, 366,
+ 366, 87, 437, 366, 366, 0, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366
} ;
-static const flex_int16_t yy_nxt[940] =
+static const flex_int16_t yy_nxt[966] =
{ 0,
6, 7, 8, 9, 6, 6, 6, 6, 10, 11,
12, 13, 14, 15, 16, 17, 18, 19, 17, 17,
17, 17, 20, 17, 21, 22, 23, 24, 25, 17,
26, 17, 17, 17, 32, 32, 33, 32, 32, 33,
36, 34, 36, 42, 34, 29, 29, 29, 30, 35,
- 50, 36, 37, 38, 43, 44, 39, 36, 79, 45,
+ 50, 36, 37, 38, 43, 44, 39, 36, 80, 45,
36, 36, 40, 29, 29, 29, 30, 35, 46, 48,
37, 38, 41, 47, 36, 49, 36, 53, 36, 36,
- 36, 56, 58, 36, 36, 55, 82, 60, 51, 342,
- 54, 61, 52, 29, 64, 32, 32, 33, 36, 65,
-
- 70, 36, 34, 29, 29, 29, 30, 36, 36, 36,
- 29, 38, 66, 66, 66, 67, 66, 71, 74, 66,
- 68, 72, 36, 36, 73, 36, 77, 78, 36, 76,
- 36, 53, 36, 36, 75, 85, 80, 83, 36, 86,
- 84, 36, 36, 36, 36, 81, 36, 36, 36, 36,
- 36, 36, 93, 89, 337, 98, 88, 29, 64, 101,
- 90, 36, 91, 65, 92, 87, 29, 95, 89, 99,
- 36, 100, 96, 36, 36, 36, 36, 36, 36, 106,
- 105, 85, 36, 36, 102, 36, 107, 36, 103, 36,
- 109, 112, 36, 36, 104, 108, 115, 110, 36, 117,
-
- 36, 36, 36, 335, 36, 36, 122, 111, 29, 29,
- 29, 30, 118, 36, 116, 29, 38, 36, 36, 113,
- 114, 119, 120, 123, 36, 29, 95, 121, 36, 134,
- 131, 96, 130, 36, 125, 124, 126, 126, 66, 127,
- 126, 132, 133, 126, 129, 333, 36, 36, 135, 137,
- 36, 36, 36, 140, 139, 35, 35, 352, 36, 36,
- 85, 141, 141, 66, 142, 141, 160, 145, 141, 144,
- 35, 35, 89, 117, 155, 36, 146, 147, 147, 66,
- 148, 147, 162, 36, 147, 150, 151, 151, 66, 152,
- 151, 36, 36, 151, 154, 120, 161, 36, 156, 156,
-
- 66, 157, 156, 36, 36, 156, 159, 164, 171, 163,
- 29, 166, 29, 168, 36, 36, 167, 170, 169, 35,
- 35, 172, 36, 36, 173, 36, 213, 184, 36, 36,
- 175, 36, 174, 29, 186, 212, 36, 349, 183, 187,
- 177, 176, 178, 178, 66, 179, 178, 182, 348, 178,
- 181, 29, 188, 35, 35, 35, 35, 189, 29, 193,
- 29, 195, 190, 36, 194, 36, 196, 29, 198, 29,
- 200, 191, 36, 199, 36, 201, 219, 29, 204, 29,
- 206, 36, 202, 205, 209, 207, 29, 166, 36, 293,
- 208, 214, 167, 35, 35, 35, 35, 35, 35, 36,
-
- 36, 36, 249, 218, 220, 29, 222, 216, 36, 217,
- 235, 223, 29, 224, 215, 226, 36, 227, 225, 346,
- 35, 35, 36, 228, 228, 66, 229, 228, 29, 186,
- 228, 231, 232, 36, 187, 233, 35, 29, 193, 29,
- 198, 234, 36, 194, 344, 199, 29, 204, 236, 36,
- 35, 241, 205, 242, 36, 35, 35, 270, 35, 35,
- 35, 35, 247, 36, 35, 35, 29, 222, 244, 262,
- 248, 36, 223, 243, 245, 246, 35, 252, 29, 254,
- 29, 256, 258, 342, 255, 259, 257, 35, 35, 339,
- 35, 35, 69, 264, 35, 35, 35, 35, 35, 35,
-
- 267, 35, 35, 275, 35, 35, 35, 35, 271, 35,
- 35, 276, 277, 35, 35, 272, 278, 315, 273, 281,
- 29, 254, 290, 313, 282, 275, 255, 285, 285, 66,
- 286, 285, 35, 35, 285, 288, 295, 298, 296, 35,
- 35, 35, 35, 298, 301, 328, 299, 294, 35, 35,
- 275, 35, 35, 35, 303, 29, 305, 300, 275, 29,
- 307, 306, 35, 35, 302, 308, 337, 36, 35, 35,
- 309, 310, 320, 316, 35, 35, 35, 35, 322, 36,
- 35, 35, 317, 275, 319, 311, 29, 305, 335, 275,
- 318, 321, 306, 323, 35, 35, 35, 35, 330, 329,
-
- 35, 35, 331, 333, 327, 35, 35, 338, 35, 35,
- 353, 340, 35, 35, 350, 325, 275, 315, 35, 35,
- 27, 27, 27, 27, 29, 29, 29, 31, 31, 31,
- 31, 36, 36, 36, 36, 62, 313, 62, 62, 63,
- 63, 63, 63, 65, 269, 65, 65, 35, 35, 35,
- 35, 69, 69, 261, 69, 94, 94, 94, 94, 96,
- 251, 96, 96, 128, 128, 128, 128, 143, 143, 143,
- 143, 149, 149, 149, 149, 153, 153, 153, 153, 158,
- 158, 158, 158, 165, 165, 165, 165, 167, 298, 167,
- 167, 180, 180, 180, 180, 185, 185, 185, 185, 187,
-
- 292, 187, 187, 192, 192, 192, 192, 194, 240, 194,
- 194, 197, 197, 197, 197, 199, 289, 199, 199, 203,
- 203, 203, 203, 205, 284, 205, 205, 210, 210, 210,
- 210, 169, 280, 169, 169, 221, 221, 221, 221, 223,
- 269, 223, 223, 230, 230, 230, 230, 189, 266, 189,
- 189, 196, 211, 196, 196, 201, 261, 201, 201, 207,
- 251, 207, 207, 237, 237, 237, 237, 239, 239, 239,
- 239, 225, 240, 225, 225, 250, 250, 250, 250, 253,
- 253, 253, 253, 255, 238, 255, 255, 260, 260, 260,
- 260, 263, 263, 263, 263, 265, 265, 265, 265, 268,
-
- 268, 268, 268, 274, 274, 274, 274, 279, 279, 279,
- 279, 257, 211, 257, 257, 283, 283, 283, 283, 287,
- 287, 287, 287, 264, 138, 264, 264, 291, 291, 291,
- 291, 297, 297, 297, 297, 304, 304, 304, 304, 306,
- 136, 306, 306, 312, 312, 312, 312, 314, 314, 314,
- 314, 308, 97, 308, 308, 324, 324, 324, 324, 326,
- 326, 326, 326, 332, 332, 332, 332, 334, 334, 334,
- 334, 336, 336, 336, 336, 341, 341, 341, 341, 343,
- 343, 343, 343, 345, 345, 345, 345, 347, 347, 347,
- 347, 351, 351, 351, 351, 36, 30, 59, 57, 36,
-
- 30, 354, 28, 28, 5, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354
+ 36, 57, 59, 36, 36, 55, 83, 61, 51, 36,
+ 54, 62, 52, 29, 65, 56, 32, 32, 33, 66,
+
+ 36, 36, 71, 34, 29, 29, 29, 30, 36, 36,
+ 77, 29, 38, 67, 67, 67, 68, 67, 75, 72,
+ 67, 69, 73, 36, 36, 74, 78, 79, 36, 53,
+ 36, 36, 36, 87, 36, 76, 84, 36, 36, 85,
+ 36, 81, 36, 86, 36, 36, 36, 36, 82, 36,
+ 354, 36, 36, 95, 91, 36, 89, 36, 88, 90,
+ 101, 100, 93, 92, 94, 103, 91, 29, 65, 29,
+ 97, 36, 102, 66, 36, 98, 36, 36, 36, 36,
+ 36, 108, 107, 86, 36, 104, 36, 36, 109, 36,
+ 105, 36, 111, 114, 36, 36, 106, 36, 110, 112,
+
+ 117, 36, 36, 120, 36, 36, 36, 349, 133, 113,
+ 29, 29, 29, 30, 121, 126, 118, 29, 38, 36,
+ 36, 115, 116, 119, 122, 123, 125, 36, 29, 97,
+ 124, 36, 36, 36, 98, 135, 136, 127, 36, 36,
+ 36, 36, 138, 140, 134, 36, 36, 154, 91, 120,
+ 142, 36, 128, 129, 129, 67, 130, 129, 36, 143,
+ 129, 132, 137, 347, 35, 35, 86, 144, 144, 67,
+ 145, 144, 35, 35, 144, 147, 148, 36, 149, 150,
+ 150, 67, 151, 150, 36, 164, 150, 153, 155, 155,
+ 67, 156, 155, 36, 36, 155, 158, 165, 123, 159,
+
+ 36, 160, 160, 67, 161, 160, 166, 36, 160, 163,
+ 36, 36, 167, 175, 168, 29, 170, 29, 172, 36,
+ 36, 171, 36, 173, 174, 35, 35, 176, 36, 36,
+ 177, 188, 36, 36, 36, 179, 201, 345, 178, 217,
+ 36, 180, 187, 35, 35, 181, 182, 182, 67, 183,
+ 182, 186, 194, 182, 185, 29, 190, 29, 192, 35,
+ 35, 191, 36, 193, 29, 197, 29, 199, 29, 203,
+ 198, 36, 200, 36, 204, 195, 29, 205, 29, 209,
+ 364, 207, 206, 218, 210, 29, 211, 36, 36, 213,
+ 219, 212, 214, 29, 170, 35, 35, 35, 35, 171,
+
+ 35, 35, 36, 224, 36, 361, 223, 225, 36, 221,
+ 231, 36, 222, 29, 227, 232, 220, 29, 229, 228,
+ 36, 35, 35, 230, 233, 233, 67, 234, 233, 29,
+ 190, 233, 236, 237, 36, 191, 238, 35, 29, 197,
+ 29, 203, 240, 36, 198, 360, 204, 36, 239, 29,
+ 209, 242, 36, 35, 247, 210, 255, 241, 248, 36,
+ 35, 35, 36, 35, 35, 35, 35, 253, 36, 35,
+ 35, 29, 227, 250, 269, 254, 36, 228, 249, 251,
+ 252, 35, 258, 29, 260, 29, 262, 264, 36, 261,
+ 265, 263, 35, 35, 36, 35, 35, 70, 271, 266,
+
+ 35, 35, 35, 35, 35, 35, 274, 35, 35, 290,
+ 36, 282, 35, 311, 278, 35, 35, 36, 277, 283,
+ 284, 279, 35, 35, 280, 29, 260, 285, 288, 35,
+ 35, 261, 301, 358, 35, 35, 35, 35, 303, 36,
+ 356, 35, 35, 320, 289, 293, 293, 67, 294, 293,
+ 298, 302, 293, 296, 304, 306, 306, 35, 35, 309,
+ 35, 35, 29, 315, 307, 282, 35, 35, 316, 282,
+ 308, 29, 317, 35, 35, 354, 310, 318, 35, 35,
+ 351, 319, 35, 35, 35, 35, 333, 282, 326, 35,
+ 35, 330, 329, 327, 282, 321, 29, 315, 328, 35,
+
+ 35, 282, 316, 334, 36, 335, 325, 340, 35, 35,
+ 341, 342, 343, 35, 35, 35, 35, 35, 35, 350,
+ 352, 35, 35, 362, 365, 323, 349, 347, 345, 313,
+ 339, 337, 35, 35, 27, 27, 27, 27, 29, 29,
+ 29, 31, 31, 31, 31, 36, 36, 36, 36, 63,
+ 332, 63, 63, 64, 64, 64, 64, 66, 282, 66,
+ 66, 35, 35, 35, 35, 70, 70, 325, 70, 96,
+ 96, 96, 96, 98, 323, 98, 98, 131, 131, 131,
+ 131, 146, 146, 146, 146, 152, 152, 152, 152, 157,
+ 157, 157, 157, 162, 162, 162, 162, 169, 169, 169,
+
+ 169, 171, 276, 171, 171, 184, 184, 184, 184, 189,
+ 189, 189, 189, 191, 268, 191, 191, 196, 196, 196,
+ 196, 198, 313, 198, 198, 202, 202, 202, 202, 204,
+ 257, 204, 204, 208, 208, 208, 208, 210, 306, 210,
+ 210, 215, 215, 215, 215, 173, 300, 173, 173, 226,
+ 226, 226, 226, 228, 246, 228, 228, 235, 235, 235,
+ 235, 193, 297, 193, 193, 200, 292, 200, 200, 206,
+ 287, 206, 206, 212, 276, 212, 212, 243, 243, 243,
+ 243, 245, 245, 245, 245, 230, 273, 230, 230, 256,
+ 256, 256, 256, 259, 259, 259, 259, 261, 216, 261,
+
+ 261, 267, 267, 267, 267, 270, 270, 270, 270, 272,
+ 272, 272, 272, 275, 275, 275, 275, 281, 281, 281,
+ 281, 286, 286, 286, 286, 263, 268, 263, 263, 291,
+ 291, 291, 291, 295, 295, 295, 295, 271, 257, 271,
+ 271, 299, 299, 299, 299, 305, 305, 305, 305, 312,
+ 312, 312, 312, 314, 314, 314, 314, 316, 246, 316,
+ 316, 322, 322, 322, 322, 324, 324, 324, 324, 331,
+ 331, 331, 331, 318, 244, 318, 318, 336, 336, 336,
+ 336, 338, 338, 338, 338, 344, 344, 344, 344, 346,
+ 346, 346, 346, 348, 348, 348, 348, 353, 353, 353,
+
+ 353, 355, 355, 355, 355, 357, 357, 357, 357, 359,
+ 359, 359, 359, 363, 363, 363, 363, 216, 141, 139,
+ 99, 36, 30, 60, 58, 36, 30, 366, 28, 28,
+ 5, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366
} ;
-static const flex_int16_t yy_chk[940] =
+static const flex_int16_t yy_chk[966] =
{ 0,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
@@ -927,101 +934,104 @@ static const flex_int16_t yy_chk[940] =
18, 14, 11, 11, 13, 14, 11, 46, 46, 14,
15, 16, 11, 12, 12, 12, 12, 12, 14, 16,
12, 12, 12, 15, 19, 16, 20, 20, 21, 22,
- 24, 22, 24, 50, 26, 21, 50, 26, 19, 351,
- 20, 26, 19, 31, 31, 32, 32, 32, 39, 31,
-
- 39, 42, 32, 35, 35, 35, 35, 40, 44, 45,
- 35, 35, 37, 37, 37, 37, 37, 39, 42, 37,
- 37, 40, 41, 43, 41, 48, 45, 45, 49, 44,
- 47, 47, 53, 51, 43, 53, 48, 51, 52, 54,
- 52, 55, 56, 58, 54, 49, 57, 59, 60, 73,
- 61, 70, 60, 61, 347, 70, 56, 63, 63, 73,
- 58, 71, 59, 63, 59, 55, 66, 66, 57, 71,
- 74, 72, 66, 72, 75, 76, 77, 78, 79, 78,
- 77, 79, 80, 81, 74, 83, 80, 82, 75, 84,
- 82, 85, 88, 85, 76, 81, 87, 83, 87, 89,
-
- 92, 89, 93, 345, 90, 104, 92, 84, 86, 86,
- 86, 86, 90, 99, 88, 86, 86, 98, 102, 86,
- 86, 91, 91, 93, 91, 94, 94, 91, 101, 104,
- 102, 94, 101, 110, 99, 98, 100, 100, 100, 100,
- 100, 103, 103, 100, 100, 343, 105, 103, 105, 107,
- 109, 107, 111, 110, 109, 113, 113, 341, 121, 118,
- 111, 112, 112, 112, 112, 112, 121, 113, 112, 112,
- 114, 114, 116, 116, 118, 116, 114, 115, 115, 115,
- 115, 115, 123, 123, 115, 115, 117, 117, 117, 117,
- 117, 124, 122, 117, 117, 119, 122, 119, 120, 120,
-
- 120, 120, 120, 125, 130, 120, 120, 125, 131, 124,
- 126, 126, 128, 128, 131, 134, 126, 130, 128, 133,
- 133, 133, 135, 136, 133, 139, 164, 140, 138, 140,
- 134, 164, 133, 141, 141, 163, 163, 338, 139, 141,
- 136, 135, 137, 137, 137, 137, 137, 138, 336, 137,
- 137, 143, 143, 145, 145, 146, 146, 143, 147, 147,
- 149, 149, 145, 155, 147, 161, 149, 151, 151, 153,
- 153, 146, 160, 151, 270, 153, 176, 156, 156, 158,
- 158, 176, 155, 156, 161, 158, 165, 165, 170, 270,
- 160, 170, 165, 172, 172, 173, 173, 174, 174, 175,
-
- 208, 177, 220, 175, 177, 178, 178, 173, 220, 174,
- 208, 178, 180, 180, 172, 182, 182, 183, 180, 334,
- 190, 190, 183, 184, 184, 184, 184, 184, 185, 185,
- 184, 184, 190, 243, 185, 191, 191, 192, 192, 197,
- 197, 202, 202, 192, 332, 197, 203, 203, 209, 209,
- 213, 213, 203, 214, 214, 215, 215, 243, 216, 216,
- 217, 217, 218, 218, 219, 219, 221, 221, 215, 235,
- 219, 235, 221, 214, 216, 217, 227, 227, 228, 228,
- 230, 230, 232, 331, 228, 233, 230, 233, 233, 329,
- 232, 232, 236, 236, 241, 241, 244, 244, 245, 245,
-
- 241, 246, 246, 247, 248, 248, 267, 267, 244, 259,
- 259, 247, 247, 252, 252, 245, 248, 326, 246, 252,
- 253, 253, 267, 324, 259, 316, 253, 262, 262, 262,
- 262, 262, 271, 271, 262, 262, 272, 276, 273, 272,
- 272, 273, 273, 277, 278, 316, 276, 271, 281, 281,
- 299, 278, 278, 282, 282, 285, 285, 277, 300, 287,
- 287, 285, 290, 290, 281, 287, 323, 293, 294, 294,
- 290, 293, 303, 299, 301, 301, 302, 302, 310, 310,
- 303, 303, 300, 317, 302, 294, 304, 304, 322, 328,
- 301, 309, 304, 311, 309, 309, 311, 311, 318, 317,
-
- 318, 318, 319, 321, 314, 319, 319, 328, 330, 330,
- 350, 330, 340, 340, 340, 312, 297, 296, 350, 350,
- 355, 355, 355, 355, 356, 356, 356, 357, 357, 357,
- 357, 358, 358, 358, 358, 359, 295, 359, 359, 360,
- 360, 360, 360, 361, 291, 361, 361, 362, 362, 362,
- 362, 363, 363, 283, 363, 364, 364, 364, 364, 365,
- 279, 365, 365, 366, 366, 366, 366, 367, 367, 367,
- 367, 368, 368, 368, 368, 369, 369, 369, 369, 370,
- 370, 370, 370, 371, 371, 371, 371, 372, 274, 372,
- 372, 373, 373, 373, 373, 374, 374, 374, 374, 375,
-
- 268, 375, 375, 376, 376, 376, 376, 377, 265, 377,
- 377, 378, 378, 378, 378, 379, 263, 379, 379, 380,
- 380, 380, 380, 381, 260, 381, 381, 382, 382, 382,
- 382, 383, 250, 383, 383, 384, 384, 384, 384, 385,
- 242, 385, 385, 386, 386, 386, 386, 387, 239, 387,
- 387, 388, 237, 388, 388, 389, 234, 389, 389, 390,
- 226, 390, 390, 391, 391, 391, 391, 392, 392, 392,
- 392, 393, 212, 393, 393, 394, 394, 394, 394, 395,
- 395, 395, 395, 396, 210, 396, 396, 397, 397, 397,
- 397, 398, 398, 398, 398, 399, 399, 399, 399, 400,
-
- 400, 400, 400, 401, 401, 401, 401, 402, 402, 402,
- 402, 403, 162, 403, 403, 404, 404, 404, 404, 405,
- 405, 405, 405, 406, 108, 406, 406, 407, 407, 407,
- 407, 408, 408, 408, 408, 409, 409, 409, 409, 410,
- 106, 410, 410, 411, 411, 411, 411, 412, 412, 412,
- 412, 413, 69, 413, 413, 414, 414, 414, 414, 415,
- 415, 415, 415, 416, 416, 416, 416, 417, 417, 417,
- 417, 418, 418, 418, 418, 419, 419, 419, 419, 420,
- 420, 420, 420, 421, 421, 421, 421, 422, 422, 422,
- 422, 423, 423, 423, 423, 36, 29, 25, 23, 17,
-
- 6, 5, 4, 3, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354, 354,
- 354, 354, 354, 354, 354, 354, 354, 354, 354
+ 24, 22, 24, 50, 26, 21, 50, 26, 19, 44,
+ 20, 26, 19, 31, 31, 21, 32, 32, 32, 31,
+
+ 39, 42, 39, 32, 35, 35, 35, 35, 40, 45,
+ 44, 35, 35, 37, 37, 37, 37, 37, 42, 39,
+ 37, 37, 40, 41, 43, 41, 45, 45, 47, 47,
+ 48, 49, 51, 54, 55, 43, 51, 52, 54, 52,
+ 53, 48, 56, 53, 58, 57, 59, 60, 49, 61,
+ 363, 62, 72, 61, 62, 74, 56, 71, 55, 57,
+ 72, 71, 60, 59, 60, 74, 58, 64, 64, 67,
+ 67, 75, 73, 64, 73, 67, 76, 77, 78, 79,
+ 80, 79, 78, 80, 81, 75, 82, 84, 81, 83,
+ 76, 85, 83, 86, 95, 86, 77, 89, 82, 84,
+
+ 88, 90, 88, 91, 103, 91, 92, 359, 103, 85,
+ 87, 87, 87, 87, 92, 95, 89, 87, 87, 100,
+ 94, 87, 87, 90, 93, 93, 94, 93, 96, 96,
+ 93, 101, 104, 118, 96, 105, 105, 100, 106, 112,
+ 107, 105, 107, 109, 104, 109, 111, 118, 119, 119,
+ 111, 119, 101, 102, 102, 102, 102, 102, 113, 112,
+ 102, 102, 106, 357, 115, 115, 113, 114, 114, 114,
+ 114, 114, 116, 116, 114, 114, 115, 124, 116, 117,
+ 117, 117, 117, 117, 121, 124, 117, 117, 120, 120,
+ 120, 120, 120, 125, 127, 120, 120, 125, 122, 121,
+
+ 122, 123, 123, 123, 123, 123, 126, 126, 123, 123,
+ 128, 133, 127, 134, 128, 129, 129, 131, 131, 134,
+ 137, 129, 138, 131, 133, 136, 136, 136, 139, 142,
+ 136, 143, 141, 143, 154, 137, 154, 355, 136, 167,
+ 167, 138, 142, 148, 148, 139, 140, 140, 140, 140,
+ 140, 141, 148, 140, 140, 144, 144, 146, 146, 149,
+ 149, 144, 159, 146, 150, 150, 152, 152, 155, 155,
+ 150, 164, 152, 165, 155, 149, 157, 157, 160, 160,
+ 353, 159, 157, 168, 160, 162, 162, 174, 168, 164,
+ 174, 162, 165, 169, 169, 176, 176, 177, 177, 169,
+
+ 178, 178, 179, 180, 181, 350, 179, 181, 180, 177,
+ 186, 186, 178, 182, 182, 187, 176, 184, 184, 182,
+ 187, 194, 194, 184, 188, 188, 188, 188, 188, 189,
+ 189, 188, 188, 194, 201, 189, 195, 195, 196, 196,
+ 202, 202, 207, 207, 196, 348, 202, 213, 201, 208,
+ 208, 214, 214, 218, 218, 208, 225, 213, 219, 219,
+ 220, 220, 225, 221, 221, 222, 222, 223, 223, 224,
+ 224, 226, 226, 220, 241, 224, 241, 226, 219, 221,
+ 222, 232, 232, 233, 233, 235, 235, 237, 239, 233,
+ 238, 235, 238, 238, 249, 237, 237, 242, 242, 239,
+
+ 247, 247, 250, 250, 251, 251, 247, 252, 252, 266,
+ 266, 253, 289, 289, 250, 254, 254, 277, 249, 253,
+ 253, 251, 258, 258, 252, 259, 259, 254, 258, 265,
+ 265, 259, 277, 346, 274, 274, 278, 278, 279, 301,
+ 344, 279, 279, 301, 265, 269, 269, 269, 269, 269,
+ 274, 278, 269, 269, 280, 283, 284, 280, 280, 285,
+ 288, 288, 293, 293, 283, 307, 285, 285, 293, 308,
+ 284, 295, 295, 298, 298, 343, 288, 295, 302, 302,
+ 341, 298, 309, 309, 310, 310, 319, 326, 307, 319,
+ 319, 311, 310, 308, 327, 302, 314, 314, 309, 311,
+
+ 311, 340, 314, 320, 320, 321, 338, 326, 321, 321,
+ 327, 328, 329, 328, 328, 329, 329, 342, 342, 340,
+ 342, 352, 352, 352, 362, 336, 335, 334, 333, 331,
+ 324, 322, 362, 362, 367, 367, 367, 367, 368, 368,
+ 368, 369, 369, 369, 369, 370, 370, 370, 370, 371,
+ 312, 371, 371, 372, 372, 372, 372, 373, 305, 373,
+ 373, 374, 374, 374, 374, 375, 375, 304, 375, 376,
+ 376, 376, 376, 377, 303, 377, 377, 378, 378, 378,
+ 378, 379, 379, 379, 379, 380, 380, 380, 380, 381,
+ 381, 381, 381, 382, 382, 382, 382, 383, 383, 383,
+
+ 383, 384, 299, 384, 384, 385, 385, 385, 385, 386,
+ 386, 386, 386, 387, 291, 387, 387, 388, 388, 388,
+ 388, 389, 290, 389, 389, 390, 390, 390, 390, 391,
+ 286, 391, 391, 392, 392, 392, 392, 393, 281, 393,
+ 393, 394, 394, 394, 394, 395, 275, 395, 395, 396,
+ 396, 396, 396, 397, 272, 397, 397, 398, 398, 398,
+ 398, 399, 270, 399, 399, 400, 267, 400, 400, 401,
+ 256, 401, 401, 402, 248, 402, 402, 403, 403, 403,
+ 403, 404, 404, 404, 404, 405, 245, 405, 405, 406,
+ 406, 406, 406, 407, 407, 407, 407, 408, 243, 408,
+
+ 408, 409, 409, 409, 409, 410, 410, 410, 410, 411,
+ 411, 411, 411, 412, 412, 412, 412, 413, 413, 413,
+ 413, 414, 414, 414, 414, 415, 240, 415, 415, 416,
+ 416, 416, 416, 417, 417, 417, 417, 418, 231, 418,
+ 418, 419, 419, 419, 419, 420, 420, 420, 420, 421,
+ 421, 421, 421, 422, 422, 422, 422, 423, 217, 423,
+ 423, 424, 424, 424, 424, 425, 425, 425, 425, 426,
+ 426, 426, 426, 427, 215, 427, 427, 428, 428, 428,
+ 428, 429, 429, 429, 429, 430, 430, 430, 430, 431,
+ 431, 431, 431, 432, 432, 432, 432, 433, 433, 433,
+
+ 433, 434, 434, 434, 434, 435, 435, 435, 435, 436,
+ 436, 436, 436, 437, 437, 437, 437, 166, 110, 108,
+ 70, 36, 29, 25, 23, 17, 6, 5, 4, 3,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366, 366, 366, 366, 366, 366,
+ 366, 366, 366, 366, 366
} ;
#define YY_TRAILING_MASK 0x2000
@@ -1160,9 +1170,17 @@ static void setbackendtype(DiskParseContext *dpc, const char *str) {
if ( !strcmp(str,"phy")) DSET(dpc,backend,BACKEND,str,PHY);
else if (!strcmp(str,"tap")) DSET(dpc,backend,BACKEND,str,TAP);
else if (!strcmp(str,"qdisk")) DSET(dpc,backend,BACKEND,str,QDISK);
+ else if (!strcmp(str,"other")) DSET(dpc,backend,BACKEND,str,OTHER);
else xlu__disk_err(dpc,str,"unknown value for backendtype");
}
+/* Sets ->protocol from the string. IDL should provide something for this. */
+static void setprotocol(DiskParseContext *dpc, const char *str) {
+ if (!strcmp(str,"xen")) DSET(dpc,protocol,PROTOCOL,str,XEN);
+ else if (!strcmp(str,"virtio-mmio")) DSET(dpc,protocol,PROTOCOL,str,VIRTIO_MMIO);
+ else xlu__disk_err(dpc,str,"unknown value for protocol");
+}
+
/* Sets ->colo-port from the string. COLO need this. */
static void setcoloport(DiskParseContext *dpc, const char *str) {
int port = atoi(str);
@@ -1199,9 +1217,9 @@ static int vdev_and_devtype(DiskParseContext *dpc, char *str) {
#undef DPC /* needs to be defined differently the actual lexer */
#define DPC ((DiskParseContext*)yyextra)
-#line 1202 "libxlu_disk_l.c"
+#line 1220 "libxlu_disk_l.c"
-#line 1204 "libxlu_disk_l.c"
+#line 1222 "libxlu_disk_l.c"
#define INITIAL 0
#define LEXERR 1
@@ -1477,13 +1495,13 @@ YY_DECL
}
{
-#line 177 "libxlu_disk_l.l"
+#line 185 "libxlu_disk_l.l"
-#line 180 "libxlu_disk_l.l"
+#line 188 "libxlu_disk_l.l"
/*----- the scanner rules which do the parsing -----*/
-#line 1486 "libxlu_disk_l.c"
+#line 1504 "libxlu_disk_l.c"
while ( /*CONSTCOND*/1 ) /* loops until end-of-file is reached */
{
@@ -1515,14 +1533,14 @@ yy_match:
while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state )
{
yy_current_state = (int) yy_def[yy_current_state];
- if ( yy_current_state >= 355 )
+ if ( yy_current_state >= 367 )
yy_c = yy_meta[yy_c];
}
yy_current_state = yy_nxt[yy_base[yy_current_state] + yy_c];
*yyg->yy_state_ptr++ = yy_current_state;
++yy_cp;
}
- while ( yy_current_state != 354 );
+ while ( yy_current_state != 366 );
yy_find_action:
yy_current_state = *--yyg->yy_state_ptr;
@@ -1572,152 +1590,158 @@ do_action: /* This label is used only to access EOF actions. */
case 1:
/* rule 1 can match eol */
YY_RULE_SETUP
-#line 182 "libxlu_disk_l.l"
+#line 190 "libxlu_disk_l.l"
{ /* ignore whitespace before parameters */ }
YY_BREAK
/* ordinary parameters setting enums or strings */
case 2:
/* rule 2 can match eol */
YY_RULE_SETUP
-#line 186 "libxlu_disk_l.l"
+#line 194 "libxlu_disk_l.l"
{ STRIP(','); setformat(DPC, FROMEQUALS); }
YY_BREAK
case 3:
YY_RULE_SETUP
-#line 188 "libxlu_disk_l.l"
+#line 196 "libxlu_disk_l.l"
{ DPC->disk->is_cdrom = 1; }
YY_BREAK
case 4:
YY_RULE_SETUP
-#line 189 "libxlu_disk_l.l"
+#line 197 "libxlu_disk_l.l"
{ DPC->disk->is_cdrom = 1; }
YY_BREAK
case 5:
YY_RULE_SETUP
-#line 190 "libxlu_disk_l.l"
+#line 198 "libxlu_disk_l.l"
{ DPC->disk->is_cdrom = 0; }
YY_BREAK
case 6:
/* rule 6 can match eol */
YY_RULE_SETUP
-#line 191 "libxlu_disk_l.l"
+#line 199 "libxlu_disk_l.l"
{ xlu__disk_err(DPC,yytext,"unknown value for type"); }
YY_BREAK
case 7:
/* rule 7 can match eol */
YY_RULE_SETUP
-#line 193 "libxlu_disk_l.l"
+#line 201 "libxlu_disk_l.l"
{ STRIP(','); setaccess(DPC, FROMEQUALS); }
YY_BREAK
case 8:
/* rule 8 can match eol */
YY_RULE_SETUP
-#line 194 "libxlu_disk_l.l"
+#line 202 "libxlu_disk_l.l"
{ STRIP(','); SAVESTRING("backend", backend_domname, FROMEQUALS); }
YY_BREAK
case 9:
/* rule 9 can match eol */
YY_RULE_SETUP
-#line 195 "libxlu_disk_l.l"
+#line 203 "libxlu_disk_l.l"
{ STRIP(','); setbackendtype(DPC,FROMEQUALS); }
YY_BREAK
case 10:
/* rule 10 can match eol */
YY_RULE_SETUP
-#line 197 "libxlu_disk_l.l"
-{ STRIP(','); SAVESTRING("vdev", vdev, FROMEQUALS); }
+#line 204 "libxlu_disk_l.l"
+{ STRIP(','); setprotocol(DPC,FROMEQUALS); }
YY_BREAK
case 11:
/* rule 11 can match eol */
YY_RULE_SETUP
-#line 198 "libxlu_disk_l.l"
-{ STRIP(','); SAVESTRING("script", script, FROMEQUALS); }
+#line 206 "libxlu_disk_l.l"
+{ STRIP(','); SAVESTRING("vdev", vdev, FROMEQUALS); }
YY_BREAK
case 12:
+/* rule 12 can match eol */
YY_RULE_SETUP
-#line 199 "libxlu_disk_l.l"
-{ DPC->disk->direct_io_safe = 1; }
+#line 207 "libxlu_disk_l.l"
+{ STRIP(','); SAVESTRING("script", script, FROMEQUALS); }
YY_BREAK
case 13:
YY_RULE_SETUP
-#line 200 "libxlu_disk_l.l"
-{ libxl_defbool_set(&DPC->disk->discard_enable, true); }
+#line 208 "libxlu_disk_l.l"
+{ DPC->disk->direct_io_safe = 1; }
YY_BREAK
case 14:
YY_RULE_SETUP
-#line 201 "libxlu_disk_l.l"
-{ libxl_defbool_set(&DPC->disk->discard_enable, false); }
+#line 209 "libxlu_disk_l.l"
+{ libxl_defbool_set(&DPC->disk->discard_enable, true); }
YY_BREAK
-/* Note that the COLO configuration settings should be considered unstable.
- * They may change incompatibly in future versions of Xen. */
case 15:
YY_RULE_SETUP
-#line 204 "libxlu_disk_l.l"
-{ libxl_defbool_set(&DPC->disk->colo_enable, true); }
+#line 210 "libxlu_disk_l.l"
+{ libxl_defbool_set(&DPC->disk->discard_enable, false); }
YY_BREAK
+/* Note that the COLO configuration settings should be considered unstable.
+ * They may change incompatibly in future versions of Xen. */
case 16:
YY_RULE_SETUP
-#line 205 "libxlu_disk_l.l"
-{ libxl_defbool_set(&DPC->disk->colo_enable, false); }
+#line 213 "libxlu_disk_l.l"
+{ libxl_defbool_set(&DPC->disk->colo_enable, true); }
YY_BREAK
case 17:
-/* rule 17 can match eol */
YY_RULE_SETUP
-#line 206 "libxlu_disk_l.l"
-{ STRIP(','); SAVESTRING("colo-host", colo_host, FROMEQUALS); }
+#line 214 "libxlu_disk_l.l"
+{ libxl_defbool_set(&DPC->disk->colo_enable, false); }
YY_BREAK
case 18:
/* rule 18 can match eol */
YY_RULE_SETUP
-#line 207 "libxlu_disk_l.l"
-{ STRIP(','); setcoloport(DPC, FROMEQUALS); }
+#line 215 "libxlu_disk_l.l"
+{ STRIP(','); SAVESTRING("colo-host", colo_host, FROMEQUALS); }
YY_BREAK
case 19:
/* rule 19 can match eol */
YY_RULE_SETUP
-#line 208 "libxlu_disk_l.l"
-{ STRIP(','); SAVESTRING("colo-export", colo_export, FROMEQUALS); }
+#line 216 "libxlu_disk_l.l"
+{ STRIP(','); setcoloport(DPC, FROMEQUALS); }
YY_BREAK
case 20:
/* rule 20 can match eol */
YY_RULE_SETUP
-#line 209 "libxlu_disk_l.l"
-{ STRIP(','); SAVESTRING("active-disk", active_disk, FROMEQUALS); }
+#line 217 "libxlu_disk_l.l"
+{ STRIP(','); SAVESTRING("colo-export", colo_export, FROMEQUALS); }
YY_BREAK
case 21:
/* rule 21 can match eol */
YY_RULE_SETUP
-#line 210 "libxlu_disk_l.l"
+#line 218 "libxlu_disk_l.l"
+{ STRIP(','); SAVESTRING("active-disk", active_disk, FROMEQUALS); }
+ YY_BREAK
+case 22:
+/* rule 22 can match eol */
+YY_RULE_SETUP
+#line 219 "libxlu_disk_l.l"
{ STRIP(','); SAVESTRING("hidden-disk", hidden_disk, FROMEQUALS); }
YY_BREAK
/* the target magic parameter, eats the rest of the string */
-case 22:
+case 23:
YY_RULE_SETUP
-#line 214 "libxlu_disk_l.l"
+#line 223 "libxlu_disk_l.l"
{ STRIP(','); SAVESTRING("target", pdev_path, FROMEQUALS); }
YY_BREAK
/* unknown parameters */
-case 23:
-/* rule 23 can match eol */
+case 24:
+/* rule 24 can match eol */
YY_RULE_SETUP
-#line 218 "libxlu_disk_l.l"
+#line 227 "libxlu_disk_l.l"
{ xlu__disk_err(DPC,yytext,"unknown parameter"); }
YY_BREAK
/* deprecated prefixes */
/* the "/.*" in these patterns ensures that they count as if they
* matched the whole string, so these patterns take precedence */
-case 24:
+case 25:
YY_RULE_SETUP
-#line 225 "libxlu_disk_l.l"
+#line 234 "libxlu_disk_l.l"
{
STRIP(':');
DPC->had_depr_prefix=1; DEPRECATE("use `[format=]...,'");
setformat(DPC, yytext);
}
YY_BREAK
-case 25:
+case 26:
YY_RULE_SETUP
-#line 231 "libxlu_disk_l.l"
+#line 240 "libxlu_disk_l.l"
{
char *newscript;
STRIP(':');
@@ -1731,65 +1755,65 @@ YY_RULE_SETUP
free(newscript);
}
YY_BREAK
-case 26:
+case 27:
*yy_cp = yyg->yy_hold_char; /* undo effects of setting up yytext */
yyg->yy_c_buf_p = yy_cp = yy_bp + 8;
YY_DO_BEFORE_ACTION; /* set up yytext again */
YY_RULE_SETUP
-#line 244 "libxlu_disk_l.l"
+#line 253 "libxlu_disk_l.l"
{ DPC->had_depr_prefix=1; DEPRECATE(0); }
YY_BREAK
-case 27:
+case 28:
YY_RULE_SETUP
-#line 245 "libxlu_disk_l.l"
+#line 254 "libxlu_disk_l.l"
{ DPC->had_depr_prefix=1; DEPRECATE(0); }
YY_BREAK
-case 28:
+case 29:
*yy_cp = yyg->yy_hold_char; /* undo effects of setting up yytext */
yyg->yy_c_buf_p = yy_cp = yy_bp + 4;
YY_DO_BEFORE_ACTION; /* set up yytext again */
YY_RULE_SETUP
-#line 246 "libxlu_disk_l.l"
+#line 255 "libxlu_disk_l.l"
{ DPC->had_depr_prefix=1; DEPRECATE(0); }
YY_BREAK
-case 29:
+case 30:
*yy_cp = yyg->yy_hold_char; /* undo effects of setting up yytext */
yyg->yy_c_buf_p = yy_cp = yy_bp + 6;
YY_DO_BEFORE_ACTION; /* set up yytext again */
YY_RULE_SETUP
-#line 247 "libxlu_disk_l.l"
+#line 256 "libxlu_disk_l.l"
{ DPC->had_depr_prefix=1; DEPRECATE(0); }
YY_BREAK
-case 30:
+case 31:
*yy_cp = yyg->yy_hold_char; /* undo effects of setting up yytext */
yyg->yy_c_buf_p = yy_cp = yy_bp + 5;
YY_DO_BEFORE_ACTION; /* set up yytext again */
YY_RULE_SETUP
-#line 248 "libxlu_disk_l.l"
+#line 257 "libxlu_disk_l.l"
{ DPC->had_depr_prefix=1; DEPRECATE(0); }
YY_BREAK
-case 31:
+case 32:
*yy_cp = yyg->yy_hold_char; /* undo effects of setting up yytext */
yyg->yy_c_buf_p = yy_cp = yy_bp + 4;
YY_DO_BEFORE_ACTION; /* set up yytext again */
YY_RULE_SETUP
-#line 249 "libxlu_disk_l.l"
+#line 258 "libxlu_disk_l.l"
{ DPC->had_depr_prefix=1; DEPRECATE(0); }
YY_BREAK
-case 32:
-/* rule 32 can match eol */
+case 33:
+/* rule 33 can match eol */
YY_RULE_SETUP
-#line 251 "libxlu_disk_l.l"
+#line 260 "libxlu_disk_l.l"
{
xlu__disk_err(DPC,yytext,"unknown deprecated disk prefix");
return 0;
}
YY_BREAK
/* positional parameters */
-case 33:
-/* rule 33 can match eol */
+case 34:
+/* rule 34 can match eol */
YY_RULE_SETUP
-#line 258 "libxlu_disk_l.l"
+#line 267 "libxlu_disk_l.l"
{
STRIP(',');
@@ -1816,27 +1840,27 @@ YY_RULE_SETUP
}
}
YY_BREAK
-case 34:
+case 35:
YY_RULE_SETUP
-#line 284 "libxlu_disk_l.l"
+#line 293 "libxlu_disk_l.l"
{
BEGIN(LEXERR);
yymore();
}
YY_BREAK
-case 35:
+case 36:
YY_RULE_SETUP
-#line 288 "libxlu_disk_l.l"
+#line 297 "libxlu_disk_l.l"
{
xlu__disk_err(DPC,yytext,"bad disk syntax"); return 0;
}
YY_BREAK
-case 36:
+case 37:
YY_RULE_SETUP
-#line 291 "libxlu_disk_l.l"
+#line 300 "libxlu_disk_l.l"
YY_FATAL_ERROR( "flex scanner jammed" );
YY_BREAK
-#line 1839 "libxlu_disk_l.c"
+#line 1863 "libxlu_disk_l.c"
case YY_STATE_EOF(INITIAL):
case YY_STATE_EOF(LEXERR):
yyterminate();
@@ -2104,7 +2128,7 @@ static int yy_get_next_buffer (yyscan_t yyscanner)
while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state )
{
yy_current_state = (int) yy_def[yy_current_state];
- if ( yy_current_state >= 355 )
+ if ( yy_current_state >= 367 )
yy_c = yy_meta[yy_c];
}
yy_current_state = yy_nxt[yy_base[yy_current_state] + yy_c];
@@ -2128,11 +2152,11 @@ static int yy_get_next_buffer (yyscan_t yyscanner)
while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state )
{
yy_current_state = (int) yy_def[yy_current_state];
- if ( yy_current_state >= 355 )
+ if ( yy_current_state >= 367 )
yy_c = yy_meta[yy_c];
}
yy_current_state = yy_nxt[yy_base[yy_current_state] + yy_c];
- yy_is_jam = (yy_current_state == 354);
+ yy_is_jam = (yy_current_state == 366);
if ( ! yy_is_jam )
*yyg->yy_state_ptr++ = yy_current_state;
@@ -2941,4 +2965,4 @@ void yyfree (void * ptr , yyscan_t yyscanner)
#define YYTABLES_NAME "yytables"
-#line 291 "libxlu_disk_l.l"
+#line 300 "libxlu_disk_l.l"
diff --git a/tools/libs/util/libxlu_disk_l.h b/tools/libs/util/libxlu_disk_l.h
index 6abeecf..509aad6 100644
--- a/tools/libs/util/libxlu_disk_l.h
+++ b/tools/libs/util/libxlu_disk_l.h
@@ -694,7 +694,7 @@ extern int yylex (yyscan_t yyscanner);
#undef yyTABLES_NAME
#endif
-#line 291 "libxlu_disk_l.l"
+#line 300 "libxlu_disk_l.l"
#line 699 "libxlu_disk_l.h"
#undef xlu__disk_yyIN_HEADER
diff --git a/tools/libs/util/libxlu_disk_l.l b/tools/libs/util/libxlu_disk_l.l
index 3bd639a..bfd90ae 100644
--- a/tools/libs/util/libxlu_disk_l.l
+++ b/tools/libs/util/libxlu_disk_l.l
@@ -122,9 +122,17 @@ static void setbackendtype(DiskParseContext *dpc, const char *str) {
if ( !strcmp(str,"phy")) DSET(dpc,backend,BACKEND,str,PHY);
else if (!strcmp(str,"tap")) DSET(dpc,backend,BACKEND,str,TAP);
else if (!strcmp(str,"qdisk")) DSET(dpc,backend,BACKEND,str,QDISK);
+ else if (!strcmp(str,"other")) DSET(dpc,backend,BACKEND,str,OTHER);
else xlu__disk_err(dpc,str,"unknown value for backendtype");
}
+/* Sets ->protocol from the string. IDL should provide something for this. */
+static void setprotocol(DiskParseContext *dpc, const char *str) {
+ if (!strcmp(str,"xen")) DSET(dpc,protocol,PROTOCOL,str,XEN);
+ else if (!strcmp(str,"virtio-mmio")) DSET(dpc,protocol,PROTOCOL,str,VIRTIO_MMIO);
+ else xlu__disk_err(dpc,str,"unknown value for protocol");
+}
+
/* Sets ->colo-port from the string. COLO need this. */
static void setcoloport(DiskParseContext *dpc, const char *str) {
int port = atoi(str);
@@ -192,6 +200,7 @@ devtype=[^,]*,? { xlu__disk_err(DPC,yytext,"unknown value for type"); }
access=[^,]*,? { STRIP(','); setaccess(DPC, FROMEQUALS); }
backend=[^,]*,? { STRIP(','); SAVESTRING("backend", backend_domname, FROMEQUALS); }
backendtype=[^,]*,? { STRIP(','); setbackendtype(DPC,FROMEQUALS); }
+protocol=[^,]*,? { STRIP(','); setprotocol(DPC,FROMEQUALS); }
vdev=[^,]*,? { STRIP(','); SAVESTRING("vdev", vdev, FROMEQUALS); }
script=[^,]*,? { STRIP(','); SAVESTRING("script", script, FROMEQUALS); }
diff --git a/tools/xl/xl_block.c b/tools/xl/xl_block.c
index 70eed43..7990035 100644
--- a/tools/xl/xl_block.c
+++ b/tools/xl/xl_block.c
@@ -50,6 +50,11 @@ int main_blockattach(int argc, char **argv)
return 0;
}
+ if (disk.protocol != LIBXL_DISK_PROTOCOL_XEN) {
+ fprintf(stderr, "block-attach is only supported for protocol xen\n");
+ return 1;
+ }
+
if (libxl_device_disk_add(ctx, fe_domid, &disk, 0)) {
fprintf(stderr, "libxl_device_disk_add failed.\n");
return 1;
@@ -119,6 +124,12 @@ int main_blockdetach(int argc, char **argv)
fprintf(stderr, "Error: Device %s not connected.\n", argv[optind+1]);
return 1;
}
+
+ if (disk.protocol != LIBXL_DISK_PROTOCOL_XEN) {
+ fprintf(stderr, "block-detach is only supported for protocol xen\n");
+ return 1;
+ }
+
rc = !force ? libxl_device_disk_safe_remove(ctx, domid, &disk, 0) :
libxl_device_disk_destroy(ctx, domid, &disk, 0);
if (rc) {
--
2.7.4
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-08 18:21 [PATCH V7 0/2] Virtio support for toolstack on Arm (Was "IOREQ feature (+ virtio-mmio) on Arm") Oleksandr Tyshchenko
2022-04-08 18:21 ` [PATCH V7 1/2] libxl: Add support for Virtio disk configuration Oleksandr Tyshchenko
@ 2022-04-08 18:21 ` Oleksandr Tyshchenko
2022-04-18 21:41 ` Stefano Stabellini
` (2 more replies)
1 sibling, 3 replies; 23+ messages in thread
From: Oleksandr Tyshchenko @ 2022-04-08 18:21 UTC (permalink / raw)
To: xen-devel
Cc: Julien Grall, Wei Liu, Anthony PERARD, Juergen Gross,
Stefano Stabellini, Julien Grall, Bertrand Marquis,
Volodymyr Babchuk, Jiamei Xie, Henry Wang, Oleksandr Tyshchenko
From: Julien Grall <julien.grall@arm.com>
This patch introduces helpers to allocate Virtio MMIO params
(IRQ and memory region) and create specific device node in
the Guest device-tree with allocated params. In order to deal
with multiple Virtio devices, reserve corresponding ranges.
For now, we reserve 1MB for memory regions and 10 SPIs.
As these helpers should be used for every Virtio device attached
to the Guest, call them for Virtio disk(s).
Please note, with statically allocated Virtio IRQs there is
a risk of a clash with a physical IRQs of passthrough devices.
For the first version, it's fine, but we should consider allocating
the Virtio IRQs automatically. Thankfully, we know in advance which
IRQs will be used for passthrough to be able to choose non-clashed
ones.
Signed-off-by: Julien Grall <julien.grall@arm.com>
Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
Tested-by: Jiamei Xie <Jiamei.xie@arm.com>
Reviewed-by: Henry Wang <Henry.Wang@arm.com>
---
@Jiamei, @Henry I decided to leave your T-b and R-b tags with the minor
change I made, are you still happy with that?
s/if (disk->virtio)/if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
Please note, this is a split/cleanup/hardening of Julien's PoC:
"Add support for Guest IO forwarding to a device emulator"
Changes RFC -> V1:
- was squashed with:
"[RFC PATCH V1 09/12] libxl: Handle virtio-mmio irq in more correct way"
"[RFC PATCH V1 11/12] libxl: Insert "dma-coherent" property into virtio-mmio device node"
"[RFC PATCH V1 12/12] libxl: Fix duplicate memory node in DT"
- move VirtIO MMIO #define-s to xen/include/public/arch-arm.h
Changes V1 -> V2:
- update the author of a patch
Changes V2 -> V3:
- no changes
Changes V3 -> V4:
- no changes
Changes V4 -> V5:
- split the changes, change the order of the patches
- drop an extra "virtio" configuration option
- update patch description
- use CONTAINER_OF instead of own implementation
- reserve ranges for Virtio MMIO params and put them
in correct location
- create helpers to allocate Virtio MMIO params, add
corresponding sanity-сhecks
- add comment why MMIO size 0x200 is chosen
- update debug print
- drop Wei's T-b
Changes V5 -> V6:
- rebase on current staging
Changes V6 -> V7:
- rebase on current staging
- add T-b and R-b tags
- update according to the recent changes to
"libxl: Add support for Virtio disk configuration"
---
tools/libs/light/libxl_arm.c | 131 +++++++++++++++++++++++++++++++++++++++++-
xen/include/public/arch-arm.h | 7 +++
2 files changed, 136 insertions(+), 2 deletions(-)
diff --git a/tools/libs/light/libxl_arm.c b/tools/libs/light/libxl_arm.c
index eef1de0..8132a47 100644
--- a/tools/libs/light/libxl_arm.c
+++ b/tools/libs/light/libxl_arm.c
@@ -8,6 +8,56 @@
#include <assert.h>
#include <xen/device_tree_defs.h>
+/*
+ * There is no clear requirements for the total size of Virtio MMIO region.
+ * The size of control registers is 0x100 and device-specific configuration
+ * registers starts at the offset 0x100, however it's size depends on the device
+ * and the driver. Pick the biggest known size at the moment to cover most
+ * of the devices (also consider allowing the user to configure the size via
+ * config file for the one not conforming with the proposed value).
+ */
+#define VIRTIO_MMIO_DEV_SIZE xen_mk_ullong(0x200)
+
+static uint64_t virtio_mmio_base;
+static uint32_t virtio_mmio_irq;
+
+static void init_virtio_mmio_params(void)
+{
+ virtio_mmio_base = GUEST_VIRTIO_MMIO_BASE;
+ virtio_mmio_irq = GUEST_VIRTIO_MMIO_SPI_FIRST;
+}
+
+static uint64_t alloc_virtio_mmio_base(libxl__gc *gc)
+{
+ uint64_t base = virtio_mmio_base;
+
+ /* Make sure we have enough reserved resources */
+ if ((virtio_mmio_base + VIRTIO_MMIO_DEV_SIZE >
+ GUEST_VIRTIO_MMIO_BASE + GUEST_VIRTIO_MMIO_SIZE)) {
+ LOG(ERROR, "Ran out of reserved range for Virtio MMIO BASE 0x%"PRIx64"\n",
+ virtio_mmio_base);
+ return 0;
+ }
+ virtio_mmio_base += VIRTIO_MMIO_DEV_SIZE;
+
+ return base;
+}
+
+static uint32_t alloc_virtio_mmio_irq(libxl__gc *gc)
+{
+ uint32_t irq = virtio_mmio_irq;
+
+ /* Make sure we have enough reserved resources */
+ if (virtio_mmio_irq > GUEST_VIRTIO_MMIO_SPI_LAST) {
+ LOG(ERROR, "Ran out of reserved range for Virtio MMIO IRQ %u\n",
+ virtio_mmio_irq);
+ return 0;
+ }
+ virtio_mmio_irq++;
+
+ return irq;
+}
+
static const char *gicv_to_string(libxl_gic_version gic_version)
{
switch (gic_version) {
@@ -26,8 +76,8 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
{
uint32_t nr_spis = 0;
unsigned int i;
- uint32_t vuart_irq;
- bool vuart_enabled = false;
+ uint32_t vuart_irq, virtio_irq = 0;
+ bool vuart_enabled = false, virtio_enabled = false;
/*
* If pl011 vuart is enabled then increment the nr_spis to allow allocation
@@ -39,6 +89,35 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
vuart_enabled = true;
}
+ /*
+ * Virtio MMIO params are non-unique across the whole system and must be
+ * initialized for every new guest.
+ */
+ init_virtio_mmio_params();
+ for (i = 0; i < d_config->num_disks; i++) {
+ libxl_device_disk *disk = &d_config->disks[i];
+
+ if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
+ disk->base = alloc_virtio_mmio_base(gc);
+ if (!disk->base)
+ return ERROR_FAIL;
+
+ disk->irq = alloc_virtio_mmio_irq(gc);
+ if (!disk->irq)
+ return ERROR_FAIL;
+
+ if (virtio_irq < disk->irq)
+ virtio_irq = disk->irq;
+ virtio_enabled = true;
+
+ LOG(DEBUG, "Allocate Virtio MMIO params for Vdev %s: IRQ %u BASE 0x%"PRIx64,
+ disk->vdev, disk->irq, disk->base);
+ }
+ }
+
+ if (virtio_enabled)
+ nr_spis += (virtio_irq - 32) + 1;
+
for (i = 0; i < d_config->b_info.num_irqs; i++) {
uint32_t irq = d_config->b_info.irqs[i];
uint32_t spi;
@@ -58,6 +137,13 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
return ERROR_FAIL;
}
+ /* The same check as for vpl011 */
+ if (virtio_enabled &&
+ (irq >= GUEST_VIRTIO_MMIO_SPI_FIRST && irq <= virtio_irq)) {
+ LOG(ERROR, "Physical IRQ %u conflicting with Virtio MMIO IRQ range\n", irq);
+ return ERROR_FAIL;
+ }
+
if (irq < 32)
continue;
@@ -787,6 +873,39 @@ static int make_vpci_node(libxl__gc *gc, void *fdt,
return 0;
}
+
+static int make_virtio_mmio_node(libxl__gc *gc, void *fdt,
+ uint64_t base, uint32_t irq)
+{
+ int res;
+ gic_interrupt intr;
+ /* Placeholder for virtio@ + a 64-bit number + \0 */
+ char buf[24];
+
+ snprintf(buf, sizeof(buf), "virtio@%"PRIx64, base);
+ res = fdt_begin_node(fdt, buf);
+ if (res) return res;
+
+ res = fdt_property_compat(gc, fdt, 1, "virtio,mmio");
+ if (res) return res;
+
+ res = fdt_property_regs(gc, fdt, GUEST_ROOT_ADDRESS_CELLS, GUEST_ROOT_SIZE_CELLS,
+ 1, base, VIRTIO_MMIO_DEV_SIZE);
+ if (res) return res;
+
+ set_interrupt(intr, irq, 0xf, DT_IRQ_TYPE_EDGE_RISING);
+ res = fdt_property_interrupts(gc, fdt, &intr, 1);
+ if (res) return res;
+
+ res = fdt_property(fdt, "dma-coherent", NULL, 0);
+ if (res) return res;
+
+ res = fdt_end_node(fdt);
+ if (res) return res;
+
+ return 0;
+}
+
static const struct arch_info *get_arch_info(libxl__gc *gc,
const struct xc_dom_image *dom)
{
@@ -988,6 +1107,7 @@ static int libxl__prepare_dtb(libxl__gc *gc, libxl_domain_config *d_config,
size_t fdt_size = 0;
int pfdt_size = 0;
libxl_domain_build_info *const info = &d_config->b_info;
+ unsigned int i;
const libxl_version_info *vers;
const struct arch_info *ainfo;
@@ -1094,6 +1214,13 @@ next_resize:
if (d_config->num_pcidevs)
FDT( make_vpci_node(gc, fdt, ainfo, dom) );
+ for (i = 0; i < d_config->num_disks; i++) {
+ libxl_device_disk *disk = &d_config->disks[i];
+
+ if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
+ FDT( make_virtio_mmio_node(gc, fdt, disk->base, disk->irq) );
+ }
+
if (pfdt)
FDT( copy_partial_fdt(gc, fdt, pfdt) );
diff --git a/xen/include/public/arch-arm.h b/xen/include/public/arch-arm.h
index ab05fe1..c8b6058 100644
--- a/xen/include/public/arch-arm.h
+++ b/xen/include/public/arch-arm.h
@@ -407,6 +407,10 @@ typedef uint64_t xen_callback_t;
/* Physical Address Space */
+/* Virtio MMIO mappings */
+#define GUEST_VIRTIO_MMIO_BASE xen_mk_ullong(0x02000000)
+#define GUEST_VIRTIO_MMIO_SIZE xen_mk_ullong(0x00100000)
+
/*
* vGIC mappings: Only one set of mapping is used by the guest.
* Therefore they can overlap.
@@ -493,6 +497,9 @@ typedef uint64_t xen_callback_t;
#define GUEST_VPL011_SPI 32
+#define GUEST_VIRTIO_MMIO_SPI_FIRST 33
+#define GUEST_VIRTIO_MMIO_SPI_LAST 43
+
/* PSCI functions */
#define PSCI_cpu_suspend 0
#define PSCI_cpu_off 1
--
2.7.4
^ permalink raw reply related [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-08 18:21 ` [PATCH V7 1/2] libxl: Add support for Virtio disk configuration Oleksandr Tyshchenko
@ 2022-04-18 21:38 ` Stefano Stabellini
2022-04-19 17:15 ` Oleksandr
2022-04-22 9:41 ` Roger Pau Monné
2022-04-25 7:43 ` Juergen Gross
2 siblings, 1 reply; 23+ messages in thread
From: Stefano Stabellini @ 2022-04-18 21:38 UTC (permalink / raw)
To: Oleksandr Tyshchenko
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Juergen Gross, Stefano Stabellini,
Julien Grall, Volodymyr Babchuk, Bertrand Marquis
> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>
> This patch adds basic support for configuring and assisting virtio-mmio
> based virtio-disk backend (emualator) which is intended to run out of
^ emulator
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-08 18:21 ` [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm Oleksandr Tyshchenko
@ 2022-04-18 21:41 ` Stefano Stabellini
2022-04-18 23:03 ` Julien Grall
2022-04-19 16:07 ` Oleksandr
2022-04-20 5:20 ` Henry Wang
2022-04-22 2:42 ` Jiamei Xie
2 siblings, 2 replies; 23+ messages in thread
From: Stefano Stabellini @ 2022-04-18 21:41 UTC (permalink / raw)
To: Oleksandr Tyshchenko
Cc: xen-devel, Julien Grall, Wei Liu, Anthony PERARD, Juergen Gross,
Stefano Stabellini, Julien Grall, Bertrand Marquis,
Volodymyr Babchuk, Jiamei Xie, Henry Wang, Oleksandr Tyshchenko
[-- Attachment #1: Type: text/plain, Size: 11917 bytes --]
On Fri, 8 Apr 2022, Oleksandr Tyshchenko wrote:
> From: Julien Grall <julien.grall@arm.com>
>
> This patch introduces helpers to allocate Virtio MMIO params
> (IRQ and memory region) and create specific device node in
> the Guest device-tree with allocated params. In order to deal
> with multiple Virtio devices, reserve corresponding ranges.
> For now, we reserve 1MB for memory regions and 10 SPIs.
>
> As these helpers should be used for every Virtio device attached
> to the Guest, call them for Virtio disk(s).
>
> Please note, with statically allocated Virtio IRQs there is
> a risk of a clash with a physical IRQs of passthrough devices.
> For the first version, it's fine, but we should consider allocating
> the Virtio IRQs automatically. Thankfully, we know in advance which
> IRQs will be used for passthrough to be able to choose non-clashed
> ones.
>
> Signed-off-by: Julien Grall <julien.grall@arm.com>
> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> Tested-by: Jiamei Xie <Jiamei.xie@arm.com>
> Reviewed-by: Henry Wang <Henry.Wang@arm.com>
I realize that we are at v7 and I haven't reviewed this before, so I'll
limit my comments. I'll clarify the ones that I think are more important
from the one that are less important.
> ---
> @Jiamei, @Henry I decided to leave your T-b and R-b tags with the minor
> change I made, are you still happy with that?
>
> s/if (disk->virtio)/if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
>
> Please note, this is a split/cleanup/hardening of Julien's PoC:
> "Add support for Guest IO forwarding to a device emulator"
>
> Changes RFC -> V1:
> - was squashed with:
> "[RFC PATCH V1 09/12] libxl: Handle virtio-mmio irq in more correct way"
> "[RFC PATCH V1 11/12] libxl: Insert "dma-coherent" property into virtio-mmio device node"
> "[RFC PATCH V1 12/12] libxl: Fix duplicate memory node in DT"
> - move VirtIO MMIO #define-s to xen/include/public/arch-arm.h
>
> Changes V1 -> V2:
> - update the author of a patch
>
> Changes V2 -> V3:
> - no changes
>
> Changes V3 -> V4:
> - no changes
>
> Changes V4 -> V5:
> - split the changes, change the order of the patches
> - drop an extra "virtio" configuration option
> - update patch description
> - use CONTAINER_OF instead of own implementation
> - reserve ranges for Virtio MMIO params and put them
> in correct location
> - create helpers to allocate Virtio MMIO params, add
> corresponding sanity-сhecks
> - add comment why MMIO size 0x200 is chosen
> - update debug print
> - drop Wei's T-b
>
> Changes V5 -> V6:
> - rebase on current staging
>
> Changes V6 -> V7:
> - rebase on current staging
> - add T-b and R-b tags
> - update according to the recent changes to
> "libxl: Add support for Virtio disk configuration"
> ---
> tools/libs/light/libxl_arm.c | 131 +++++++++++++++++++++++++++++++++++++++++-
> xen/include/public/arch-arm.h | 7 +++
> 2 files changed, 136 insertions(+), 2 deletions(-)
>
> diff --git a/tools/libs/light/libxl_arm.c b/tools/libs/light/libxl_arm.c
> index eef1de0..8132a47 100644
> --- a/tools/libs/light/libxl_arm.c
> +++ b/tools/libs/light/libxl_arm.c
> @@ -8,6 +8,56 @@
> #include <assert.h>
> #include <xen/device_tree_defs.h>
>
> +/*
> + * There is no clear requirements for the total size of Virtio MMIO region.
> + * The size of control registers is 0x100 and device-specific configuration
> + * registers starts at the offset 0x100, however it's size depends on the device
> + * and the driver. Pick the biggest known size at the moment to cover most
> + * of the devices (also consider allowing the user to configure the size via
> + * config file for the one not conforming with the proposed value).
> + */
> +#define VIRTIO_MMIO_DEV_SIZE xen_mk_ullong(0x200)
Actually, I don't think we need to make this generic. We only support
virtio-disk now and I think it is fine if this was called
VIRTIO_DISK_MMIO_DEV_SIZE and the size was exactly the size needed by
virtio-disk. When/if we support another virtio protocol we can add set
the appropriate size of that one as well.
At the moment as you can see below:
*/
for (i = 0; i < d_config->num_disks; i++) {
libxl_device_disk *disk = &d_config->disks[i];
if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
it is all very virtio-disk specific, so we might as well exploit it for
the better :-)
Given that we are at v7 of this series, I don't think this is very
important, so I think it is OK if we keep the code as is.
> +static uint64_t virtio_mmio_base;
> +static uint32_t virtio_mmio_irq;
No need for these two variables to be global in this file, they could be
local variables in libxl__arch_domain_prepare_config.
I think that this is a change we should make if possible before
committing these patches.
> +static void init_virtio_mmio_params(void)
> +{
> + virtio_mmio_base = GUEST_VIRTIO_MMIO_BASE;
> + virtio_mmio_irq = GUEST_VIRTIO_MMIO_SPI_FIRST;
> +}
> +
> +static uint64_t alloc_virtio_mmio_base(libxl__gc *gc)
> +{
> + uint64_t base = virtio_mmio_base;
If virtio_mmio_base is local in libxl__arch_domain_prepare_config, then
it could be passed here as parameter (as pointer).
If we make the dev MMIO size (GUEST_VIRTIO_MMIO_SIZE) specific to
virtio-disk, then we could pass it also as argument to this function:
static uint64_t alloc_virtio_mmio_base(libxl__gc *gc,
uint64 *virtio_mmio_base,
uint64_t dev_mmio_size)
> + /* Make sure we have enough reserved resources */
> + if ((virtio_mmio_base + VIRTIO_MMIO_DEV_SIZE >
> + GUEST_VIRTIO_MMIO_BASE + GUEST_VIRTIO_MMIO_SIZE)) {
> + LOG(ERROR, "Ran out of reserved range for Virtio MMIO BASE 0x%"PRIx64"\n",
> + virtio_mmio_base);
> + return 0;
> + }
> + virtio_mmio_base += VIRTIO_MMIO_DEV_SIZE;
> +
> + return base;
> +}
> +
> +static uint32_t alloc_virtio_mmio_irq(libxl__gc *gc)
> +{
> + uint32_t irq = virtio_mmio_irq;
> +
> + /* Make sure we have enough reserved resources */
> + if (virtio_mmio_irq > GUEST_VIRTIO_MMIO_SPI_LAST) {
> + LOG(ERROR, "Ran out of reserved range for Virtio MMIO IRQ %u\n",
> + virtio_mmio_irq);
> + return 0;
> + }
> + virtio_mmio_irq++;
> +
> + return irq;
> +}
> +
> static const char *gicv_to_string(libxl_gic_version gic_version)
> {
> switch (gic_version) {
> @@ -26,8 +76,8 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> {
> uint32_t nr_spis = 0;
> unsigned int i;
> - uint32_t vuart_irq;
> - bool vuart_enabled = false;
> + uint32_t vuart_irq, virtio_irq = 0;
> + bool vuart_enabled = false, virtio_enabled = false;
>
> /*
> * If pl011 vuart is enabled then increment the nr_spis to allow allocation
> @@ -39,6 +89,35 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> vuart_enabled = true;
> }
>
> + /*
> + * Virtio MMIO params are non-unique across the whole system and must be
> + * initialized for every new guest.
> + */
> + init_virtio_mmio_params();
No need to initialize them in init_virtio_mmio_params, you could just
do:
uint64_t virtio_mmio_base = GUEST_VIRTIO_MMIO_BASE;
uint32_t virtio_mmio_irq = GUEST_VIRTIO_MMIO_SPI_FIRST;
at the top of libxl__arch_domain_prepare_config
> + for (i = 0; i < d_config->num_disks; i++) {
> + libxl_device_disk *disk = &d_config->disks[i];
> +
> + if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
> + disk->base = alloc_virtio_mmio_base(gc);
> + if (!disk->base)
> + return ERROR_FAIL;
> +
> + disk->irq = alloc_virtio_mmio_irq(gc);
> + if (!disk->irq)
> + return ERROR_FAIL;
> +
> + if (virtio_irq < disk->irq)
> + virtio_irq = disk->irq;
> + virtio_enabled = true;
> +
> + LOG(DEBUG, "Allocate Virtio MMIO params for Vdev %s: IRQ %u BASE 0x%"PRIx64,
> + disk->vdev, disk->irq, disk->base);
> + }
> + }
> +
> + if (virtio_enabled)
> + nr_spis += (virtio_irq - 32) + 1;
> +
> for (i = 0; i < d_config->b_info.num_irqs; i++) {
> uint32_t irq = d_config->b_info.irqs[i];
> uint32_t spi;
> @@ -58,6 +137,13 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> return ERROR_FAIL;
> }
>
> + /* The same check as for vpl011 */
> + if (virtio_enabled &&
> + (irq >= GUEST_VIRTIO_MMIO_SPI_FIRST && irq <= virtio_irq)) {
NIT: alignment:
if (virtio_enabled &&
(irq >= GUEST_VIRTIO_MMIO_SPI_FIRST && irq <= virtio_irq)) {
> + LOG(ERROR, "Physical IRQ %u conflicting with Virtio MMIO IRQ range\n", irq);
> + return ERROR_FAIL;
> + }
> +
> if (irq < 32)
> continue;
>
> @@ -787,6 +873,39 @@ static int make_vpci_node(libxl__gc *gc, void *fdt,
> return 0;
> }
>
> +
> +static int make_virtio_mmio_node(libxl__gc *gc, void *fdt,
> + uint64_t base, uint32_t irq)
> +{
> + int res;
> + gic_interrupt intr;
> + /* Placeholder for virtio@ + a 64-bit number + \0 */
> + char buf[24];
> +
> + snprintf(buf, sizeof(buf), "virtio@%"PRIx64, base);
> + res = fdt_begin_node(fdt, buf);
> + if (res) return res;
> +
> + res = fdt_property_compat(gc, fdt, 1, "virtio,mmio");
> + if (res) return res;
> +
> + res = fdt_property_regs(gc, fdt, GUEST_ROOT_ADDRESS_CELLS, GUEST_ROOT_SIZE_CELLS,
> + 1, base, VIRTIO_MMIO_DEV_SIZE);
> + if (res) return res;
> +
> + set_interrupt(intr, irq, 0xf, DT_IRQ_TYPE_EDGE_RISING);
> + res = fdt_property_interrupts(gc, fdt, &intr, 1);
> + if (res) return res;
> +
> + res = fdt_property(fdt, "dma-coherent", NULL, 0);
> + if (res) return res;
> +
> + res = fdt_end_node(fdt);
> + if (res) return res;
> +
> + return 0;
> +}
> +
> static const struct arch_info *get_arch_info(libxl__gc *gc,
> const struct xc_dom_image *dom)
> {
> @@ -988,6 +1107,7 @@ static int libxl__prepare_dtb(libxl__gc *gc, libxl_domain_config *d_config,
> size_t fdt_size = 0;
> int pfdt_size = 0;
> libxl_domain_build_info *const info = &d_config->b_info;
> + unsigned int i;
>
> const libxl_version_info *vers;
> const struct arch_info *ainfo;
> @@ -1094,6 +1214,13 @@ next_resize:
> if (d_config->num_pcidevs)
> FDT( make_vpci_node(gc, fdt, ainfo, dom) );
>
> + for (i = 0; i < d_config->num_disks; i++) {
> + libxl_device_disk *disk = &d_config->disks[i];
> +
> + if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
> + FDT( make_virtio_mmio_node(gc, fdt, disk->base, disk->irq) );
> + }
> +
> if (pfdt)
> FDT( copy_partial_fdt(gc, fdt, pfdt) );
>
> diff --git a/xen/include/public/arch-arm.h b/xen/include/public/arch-arm.h
> index ab05fe1..c8b6058 100644
> --- a/xen/include/public/arch-arm.h
> +++ b/xen/include/public/arch-arm.h
> @@ -407,6 +407,10 @@ typedef uint64_t xen_callback_t;
>
> /* Physical Address Space */
>
> +/* Virtio MMIO mappings */
> +#define GUEST_VIRTIO_MMIO_BASE xen_mk_ullong(0x02000000)
> +#define GUEST_VIRTIO_MMIO_SIZE xen_mk_ullong(0x00100000)
> +
> /*
> * vGIC mappings: Only one set of mapping is used by the guest.
> * Therefore they can overlap.
> @@ -493,6 +497,9 @@ typedef uint64_t xen_callback_t;
>
> #define GUEST_VPL011_SPI 32
>
> +#define GUEST_VIRTIO_MMIO_SPI_FIRST 33
> +#define GUEST_VIRTIO_MMIO_SPI_LAST 43
> +
> /* PSCI functions */
> #define PSCI_cpu_suspend 0
> #define PSCI_cpu_off 1
> --
> 2.7.4
>
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-18 21:41 ` Stefano Stabellini
@ 2022-04-18 23:03 ` Julien Grall
2022-04-19 16:07 ` Oleksandr
1 sibling, 0 replies; 23+ messages in thread
From: Julien Grall @ 2022-04-18 23:03 UTC (permalink / raw)
To: Stefano Stabellini
Cc: Oleksandr Tyshchenko, xen-devel, Julien Grall, Wei Liu,
Anthony PERARD, Juergen Gross, Bertrand Marquis,
Volodymyr Babchuk, Jiamei Xie, Henry Wang, Oleksandr Tyshchenko
[-- Attachment #1: Type: text/plain, Size: 493 bytes --]
Hi,
Sorry for the formatting issues.
On Mon, 18 Apr 2022, 21:41 Stefano Stabellini, <sstabellini@kernel.org>
wrote:
> > +static uint64_t virtio_mmio_base;
> > +static uint32_t virtio_mmio_irq;
>
> No need for these two variables to be global in this file, they could be
> local variables in libxl__arch_domain_prepare_config.
It is more than "no need". Libxl can be used by a daemon to handle multiple
domains. So we can't use static variable to keep track of local information.
Cheers,
[-- Attachment #2: Type: text/html, Size: 1038 bytes --]
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-18 21:41 ` Stefano Stabellini
2022-04-18 23:03 ` Julien Grall
@ 2022-04-19 16:07 ` Oleksandr
2022-04-20 0:13 ` Stefano Stabellini
1 sibling, 1 reply; 23+ messages in thread
From: Oleksandr @ 2022-04-19 16:07 UTC (permalink / raw)
To: Stefano Stabellini
Cc: xen-devel, Julien Grall, Wei Liu, Anthony PERARD, Juergen Gross,
Julien Grall, Bertrand Marquis, Volodymyr Babchuk, Jiamei Xie,
Henry Wang, Oleksandr Tyshchenko
On 19.04.22 00:41, Stefano Stabellini wrote:
Hello Stefano
> On Fri, 8 Apr 2022, Oleksandr Tyshchenko wrote:
>> From: Julien Grall <julien.grall@arm.com>
>>
>> This patch introduces helpers to allocate Virtio MMIO params
>> (IRQ and memory region) and create specific device node in
>> the Guest device-tree with allocated params. In order to deal
>> with multiple Virtio devices, reserve corresponding ranges.
>> For now, we reserve 1MB for memory regions and 10 SPIs.
>>
>> As these helpers should be used for every Virtio device attached
>> to the Guest, call them for Virtio disk(s).
>>
>> Please note, with statically allocated Virtio IRQs there is
>> a risk of a clash with a physical IRQs of passthrough devices.
>> For the first version, it's fine, but we should consider allocating
>> the Virtio IRQs automatically. Thankfully, we know in advance which
>> IRQs will be used for passthrough to be able to choose non-clashed
>> ones.
>>
>> Signed-off-by: Julien Grall <julien.grall@arm.com>
>> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>> Tested-by: Jiamei Xie <Jiamei.xie@arm.com>
>> Reviewed-by: Henry Wang <Henry.Wang@arm.com>
> I realize that we are at v7 and I haven't reviewed this before, so I'll
> limit my comments. I'll clarify the ones that I think are more important
> from the one that are less important.
thank you
>
>
>> ---
>> @Jiamei, @Henry I decided to leave your T-b and R-b tags with the minor
>> change I made, are you still happy with that?
>>
>> s/if (disk->virtio)/if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
>>
>> Please note, this is a split/cleanup/hardening of Julien's PoC:
>> "Add support for Guest IO forwarding to a device emulator"
>>
>> Changes RFC -> V1:
>> - was squashed with:
>> "[RFC PATCH V1 09/12] libxl: Handle virtio-mmio irq in more correct way"
>> "[RFC PATCH V1 11/12] libxl: Insert "dma-coherent" property into virtio-mmio device node"
>> "[RFC PATCH V1 12/12] libxl: Fix duplicate memory node in DT"
>> - move VirtIO MMIO #define-s to xen/include/public/arch-arm.h
>>
>> Changes V1 -> V2:
>> - update the author of a patch
>>
>> Changes V2 -> V3:
>> - no changes
>>
>> Changes V3 -> V4:
>> - no changes
>>
>> Changes V4 -> V5:
>> - split the changes, change the order of the patches
>> - drop an extra "virtio" configuration option
>> - update patch description
>> - use CONTAINER_OF instead of own implementation
>> - reserve ranges for Virtio MMIO params and put them
>> in correct location
>> - create helpers to allocate Virtio MMIO params, add
>> corresponding sanity-сhecks
>> - add comment why MMIO size 0x200 is chosen
>> - update debug print
>> - drop Wei's T-b
>>
>> Changes V5 -> V6:
>> - rebase on current staging
>>
>> Changes V6 -> V7:
>> - rebase on current staging
>> - add T-b and R-b tags
>> - update according to the recent changes to
>> "libxl: Add support for Virtio disk configuration"
>> ---
>> tools/libs/light/libxl_arm.c | 131 +++++++++++++++++++++++++++++++++++++++++-
>> xen/include/public/arch-arm.h | 7 +++
>> 2 files changed, 136 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/libs/light/libxl_arm.c b/tools/libs/light/libxl_arm.c
>> index eef1de0..8132a47 100644
>> --- a/tools/libs/light/libxl_arm.c
>> +++ b/tools/libs/light/libxl_arm.c
>> @@ -8,6 +8,56 @@
>> #include <assert.h>
>> #include <xen/device_tree_defs.h>
>>
>> +/*
>> + * There is no clear requirements for the total size of Virtio MMIO region.
>> + * The size of control registers is 0x100 and device-specific configuration
>> + * registers starts at the offset 0x100, however it's size depends on the device
>> + * and the driver. Pick the biggest known size at the moment to cover most
>> + * of the devices (also consider allowing the user to configure the size via
>> + * config file for the one not conforming with the proposed value).
>> + */
>> +#define VIRTIO_MMIO_DEV_SIZE xen_mk_ullong(0x200)
> Actually, I don't think we need to make this generic. We only support
> virtio-disk now and I think it is fine if this was called
> VIRTIO_DISK_MMIO_DEV_SIZE and the size was exactly the size needed by
> virtio-disk. When/if we support another virtio protocol we can add set
> the appropriate size of that one as well.
I would prefer if we keep this generic, although we are going to support
virtio-blk as the first virtio-mmio device, there is nothing disk
specific in that value. The same value (0x200) is used
by SW which also generates virtio-mmio device tree nodes like we do in
Xen libxl, I am speaking about kvmtool, Qemu here.
>
> At the moment as you can see below:
>
> */
> for (i = 0; i < d_config->num_disks; i++) {
> libxl_device_disk *disk = &d_config->disks[i];
>
> if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
>
> it is all very virtio-disk specific, so we might as well exploit it for
> the better :-)
>
> Given that we are at v7 of this series, I don't think this is very
> important, so I think it is OK if we keep the code as is.
ok
>
>
>> +static uint64_t virtio_mmio_base;
>> +static uint32_t virtio_mmio_irq;
> No need for these two variables to be global in this file, they could be
> local variables in libxl__arch_domain_prepare_config.
>
> I think that this is a change we should make if possible before
> committing these patches.
I agree with that, will do
>
>
>
>> +static void init_virtio_mmio_params(void)
>> +{
>> + virtio_mmio_base = GUEST_VIRTIO_MMIO_BASE;
>> + virtio_mmio_irq = GUEST_VIRTIO_MMIO_SPI_FIRST;
>> +}
>> +
>> +static uint64_t alloc_virtio_mmio_base(libxl__gc *gc)
>> +{
>> + uint64_t base = virtio_mmio_base;
> If virtio_mmio_base is local in libxl__arch_domain_prepare_config, then
> it could be passed here as parameter (as pointer).
yes, good point, will do
>
> If we make the dev MMIO size (GUEST_VIRTIO_MMIO_SIZE) specific to
> virtio-disk, then we could pass it also as argument to this function:
>
> static uint64_t alloc_virtio_mmio_base(libxl__gc *gc,
> uint64 *virtio_mmio_base,
> uint64_t dev_mmio_size)
Please see my comment regarding VIRTIO_MMIO_DEV_SIZE above
>
>
>> + /* Make sure we have enough reserved resources */
>> + if ((virtio_mmio_base + VIRTIO_MMIO_DEV_SIZE >
>> + GUEST_VIRTIO_MMIO_BASE + GUEST_VIRTIO_MMIO_SIZE)) {
>> + LOG(ERROR, "Ran out of reserved range for Virtio MMIO BASE 0x%"PRIx64"\n",
>> + virtio_mmio_base);
>> + return 0;
>> + }
>> + virtio_mmio_base += VIRTIO_MMIO_DEV_SIZE;
>> +
>> + return base;
>> +}
>> +
>> +static uint32_t alloc_virtio_mmio_irq(libxl__gc *gc)
>> +{
>> + uint32_t irq = virtio_mmio_irq;
>> +
>> + /* Make sure we have enough reserved resources */
>> + if (virtio_mmio_irq > GUEST_VIRTIO_MMIO_SPI_LAST) {
>> + LOG(ERROR, "Ran out of reserved range for Virtio MMIO IRQ %u\n",
>> + virtio_mmio_irq);
>> + return 0;
>> + }
>> + virtio_mmio_irq++;
>> +
>> + return irq;
>> +}
>> +
>> static const char *gicv_to_string(libxl_gic_version gic_version)
>> {
>> switch (gic_version) {
>> @@ -26,8 +76,8 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
>> {
>> uint32_t nr_spis = 0;
>> unsigned int i;
>> - uint32_t vuart_irq;
>> - bool vuart_enabled = false;
>> + uint32_t vuart_irq, virtio_irq = 0;
>> + bool vuart_enabled = false, virtio_enabled = false;
>>
>> /*
>> * If pl011 vuart is enabled then increment the nr_spis to allow allocation
>> @@ -39,6 +89,35 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
>> vuart_enabled = true;
>> }
>>
>> + /*
>> + * Virtio MMIO params are non-unique across the whole system and must be
>> + * initialized for every new guest.
>> + */
>> + init_virtio_mmio_params();
> No need to initialize them in init_virtio_mmio_params, you could just
> do:
>
> uint64_t virtio_mmio_base = GUEST_VIRTIO_MMIO_BASE;
> uint32_t virtio_mmio_irq = GUEST_VIRTIO_MMIO_SPI_FIRST;
>
> at the top of libxl__arch_domain_prepare_config
yes, will do
>
>
>> + for (i = 0; i < d_config->num_disks; i++) {
>> + libxl_device_disk *disk = &d_config->disks[i];
>> +
>> + if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
>> + disk->base = alloc_virtio_mmio_base(gc);
>> + if (!disk->base)
>> + return ERROR_FAIL;
>> +
>> + disk->irq = alloc_virtio_mmio_irq(gc);
>> + if (!disk->irq)
>> + return ERROR_FAIL;
>> +
>> + if (virtio_irq < disk->irq)
>> + virtio_irq = disk->irq;
>> + virtio_enabled = true;
>> +
>> + LOG(DEBUG, "Allocate Virtio MMIO params for Vdev %s: IRQ %u BASE 0x%"PRIx64,
>> + disk->vdev, disk->irq, disk->base);
>> + }
>> + }
>> +
>> + if (virtio_enabled)
>> + nr_spis += (virtio_irq - 32) + 1;
>> +
>> for (i = 0; i < d_config->b_info.num_irqs; i++) {
>> uint32_t irq = d_config->b_info.irqs[i];
>> uint32_t spi;
>> @@ -58,6 +137,13 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
>> return ERROR_FAIL;
>> }
>>
>> + /* The same check as for vpl011 */
>> + if (virtio_enabled &&
>> + (irq >= GUEST_VIRTIO_MMIO_SPI_FIRST && irq <= virtio_irq)) {
> NIT: alignment:
>
> if (virtio_enabled &&
> (irq >= GUEST_VIRTIO_MMIO_SPI_FIRST && irq <= virtio_irq)) {
ok
>
>
>
>> + LOG(ERROR, "Physical IRQ %u conflicting with Virtio MMIO IRQ range\n", irq);
>> + return ERROR_FAIL;
>> + }
>> +
>> if (irq < 32)
>> continue;
>>
>> @@ -787,6 +873,39 @@ static int make_vpci_node(libxl__gc *gc, void *fdt,
>> return 0;
>> }
>>
>> +
>> +static int make_virtio_mmio_node(libxl__gc *gc, void *fdt,
>> + uint64_t base, uint32_t irq)
>> +{
>> + int res;
>> + gic_interrupt intr;
>> + /* Placeholder for virtio@ + a 64-bit number + \0 */
>> + char buf[24];
>> +
>> + snprintf(buf, sizeof(buf), "virtio@%"PRIx64, base);
>> + res = fdt_begin_node(fdt, buf);
>> + if (res) return res;
>> +
>> + res = fdt_property_compat(gc, fdt, 1, "virtio,mmio");
>> + if (res) return res;
>> +
>> + res = fdt_property_regs(gc, fdt, GUEST_ROOT_ADDRESS_CELLS, GUEST_ROOT_SIZE_CELLS,
>> + 1, base, VIRTIO_MMIO_DEV_SIZE);
>> + if (res) return res;
>> +
>> + set_interrupt(intr, irq, 0xf, DT_IRQ_TYPE_EDGE_RISING);
>> + res = fdt_property_interrupts(gc, fdt, &intr, 1);
>> + if (res) return res;
>> +
>> + res = fdt_property(fdt, "dma-coherent", NULL, 0);
>> + if (res) return res;
>> +
>> + res = fdt_end_node(fdt);
>> + if (res) return res;
>> +
>> + return 0;
>> +}
>> +
>> static const struct arch_info *get_arch_info(libxl__gc *gc,
>> const struct xc_dom_image *dom)
>> {
>> @@ -988,6 +1107,7 @@ static int libxl__prepare_dtb(libxl__gc *gc, libxl_domain_config *d_config,
>> size_t fdt_size = 0;
>> int pfdt_size = 0;
>> libxl_domain_build_info *const info = &d_config->b_info;
>> + unsigned int i;
>>
>> const libxl_version_info *vers;
>> const struct arch_info *ainfo;
>> @@ -1094,6 +1214,13 @@ next_resize:
>> if (d_config->num_pcidevs)
>> FDT( make_vpci_node(gc, fdt, ainfo, dom) );
>>
>> + for (i = 0; i < d_config->num_disks; i++) {
>> + libxl_device_disk *disk = &d_config->disks[i];
>> +
>> + if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
>> + FDT( make_virtio_mmio_node(gc, fdt, disk->base, disk->irq) );
>> + }
>> +
>> if (pfdt)
>> FDT( copy_partial_fdt(gc, fdt, pfdt) );
>>
>> diff --git a/xen/include/public/arch-arm.h b/xen/include/public/arch-arm.h
>> index ab05fe1..c8b6058 100644
>> --- a/xen/include/public/arch-arm.h
>> +++ b/xen/include/public/arch-arm.h
>> @@ -407,6 +407,10 @@ typedef uint64_t xen_callback_t;
>>
>> /* Physical Address Space */
>>
>> +/* Virtio MMIO mappings */
>> +#define GUEST_VIRTIO_MMIO_BASE xen_mk_ullong(0x02000000)
>> +#define GUEST_VIRTIO_MMIO_SIZE xen_mk_ullong(0x00100000)
>> +
>> /*
>> * vGIC mappings: Only one set of mapping is used by the guest.
>> * Therefore they can overlap.
>> @@ -493,6 +497,9 @@ typedef uint64_t xen_callback_t;
>>
>> #define GUEST_VPL011_SPI 32
>>
>> +#define GUEST_VIRTIO_MMIO_SPI_FIRST 33
>> +#define GUEST_VIRTIO_MMIO_SPI_LAST 43
>> +
>> /* PSCI functions */
>> #define PSCI_cpu_suspend 0
>> #define PSCI_cpu_off 1
>> --
>> 2.7.4
--
Regards,
Oleksandr Tyshchenko
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-18 21:38 ` Stefano Stabellini
@ 2022-04-19 17:15 ` Oleksandr
0 siblings, 0 replies; 23+ messages in thread
From: Oleksandr @ 2022-04-19 17:15 UTC (permalink / raw)
To: Stefano Stabellini
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Juergen Gross, Julien Grall,
Volodymyr Babchuk, Bertrand Marquis
On 19.04.22 00:38, Stefano Stabellini wrote:
Hello Stefano
>> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>>
>> This patch adds basic support for configuring and assisting virtio-mmio
>> based virtio-disk backend (emualator) which is intended to run out of
> ^ emulator
ok)
--
Regards,
Oleksandr Tyshchenko
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-19 16:07 ` Oleksandr
@ 2022-04-20 0:13 ` Stefano Stabellini
0 siblings, 0 replies; 23+ messages in thread
From: Stefano Stabellini @ 2022-04-20 0:13 UTC (permalink / raw)
To: Oleksandr
Cc: Stefano Stabellini, xen-devel, Julien Grall, Wei Liu,
Anthony PERARD, Juergen Gross, Julien Grall, Bertrand Marquis,
Volodymyr Babchuk, Jiamei Xie, Henry Wang, Oleksandr Tyshchenko
[-- Attachment #1: Type: text/plain, Size: 5086 bytes --]
On Tue, 19 Apr 2022, Oleksandr wrote:
> On 19.04.22 00:41, Stefano Stabellini wrote:
> Hello Stefano
>
> > On Fri, 8 Apr 2022, Oleksandr Tyshchenko wrote:
> > > From: Julien Grall <julien.grall@arm.com>
> > >
> > > This patch introduces helpers to allocate Virtio MMIO params
> > > (IRQ and memory region) and create specific device node in
> > > the Guest device-tree with allocated params. In order to deal
> > > with multiple Virtio devices, reserve corresponding ranges.
> > > For now, we reserve 1MB for memory regions and 10 SPIs.
> > >
> > > As these helpers should be used for every Virtio device attached
> > > to the Guest, call them for Virtio disk(s).
> > >
> > > Please note, with statically allocated Virtio IRQs there is
> > > a risk of a clash with a physical IRQs of passthrough devices.
> > > For the first version, it's fine, but we should consider allocating
> > > the Virtio IRQs automatically. Thankfully, we know in advance which
> > > IRQs will be used for passthrough to be able to choose non-clashed
> > > ones.
> > >
> > > Signed-off-by: Julien Grall <julien.grall@arm.com>
> > > Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> > > Tested-by: Jiamei Xie <Jiamei.xie@arm.com>
> > > Reviewed-by: Henry Wang <Henry.Wang@arm.com>
> > I realize that we are at v7 and I haven't reviewed this before, so I'll
> > limit my comments. I'll clarify the ones that I think are more important
> > from the one that are less important.
>
> thank you
>
>
> >
> >
> > > ---
> > > @Jiamei, @Henry I decided to leave your T-b and R-b tags with the minor
> > > change I made, are you still happy with that?
> > >
> > > s/if (disk->virtio)/if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
> > >
> > > Please note, this is a split/cleanup/hardening of Julien's PoC:
> > > "Add support for Guest IO forwarding to a device emulator"
> > >
> > > Changes RFC -> V1:
> > > - was squashed with:
> > > "[RFC PATCH V1 09/12] libxl: Handle virtio-mmio irq in more correct
> > > way"
> > > "[RFC PATCH V1 11/12] libxl: Insert "dma-coherent" property into
> > > virtio-mmio device node"
> > > "[RFC PATCH V1 12/12] libxl: Fix duplicate memory node in DT"
> > > - move VirtIO MMIO #define-s to xen/include/public/arch-arm.h
> > >
> > > Changes V1 -> V2:
> > > - update the author of a patch
> > >
> > > Changes V2 -> V3:
> > > - no changes
> > >
> > > Changes V3 -> V4:
> > > - no changes
> > >
> > > Changes V4 -> V5:
> > > - split the changes, change the order of the patches
> > > - drop an extra "virtio" configuration option
> > > - update patch description
> > > - use CONTAINER_OF instead of own implementation
> > > - reserve ranges for Virtio MMIO params and put them
> > > in correct location
> > > - create helpers to allocate Virtio MMIO params, add
> > > corresponding sanity-сhecks
> > > - add comment why MMIO size 0x200 is chosen
> > > - update debug print
> > > - drop Wei's T-b
> > >
> > > Changes V5 -> V6:
> > > - rebase on current staging
> > >
> > > Changes V6 -> V7:
> > > - rebase on current staging
> > > - add T-b and R-b tags
> > > - update according to the recent changes to
> > > "libxl: Add support for Virtio disk configuration"
> > > ---
> > > tools/libs/light/libxl_arm.c | 131
> > > +++++++++++++++++++++++++++++++++++++++++-
> > > xen/include/public/arch-arm.h | 7 +++
> > > 2 files changed, 136 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/tools/libs/light/libxl_arm.c b/tools/libs/light/libxl_arm.c
> > > index eef1de0..8132a47 100644
> > > --- a/tools/libs/light/libxl_arm.c
> > > +++ b/tools/libs/light/libxl_arm.c
> > > @@ -8,6 +8,56 @@
> > > #include <assert.h>
> > > #include <xen/device_tree_defs.h>
> > > +/*
> > > + * There is no clear requirements for the total size of Virtio MMIO
> > > region.
> > > + * The size of control registers is 0x100 and device-specific
> > > configuration
> > > + * registers starts at the offset 0x100, however it's size depends on the
> > > device
> > > + * and the driver. Pick the biggest known size at the moment to cover
> > > most
> > > + * of the devices (also consider allowing the user to configure the size
> > > via
> > > + * config file for the one not conforming with the proposed value).
> > > + */
> > > +#define VIRTIO_MMIO_DEV_SIZE xen_mk_ullong(0x200)
> > Actually, I don't think we need to make this generic. We only support
> > virtio-disk now and I think it is fine if this was called
> > VIRTIO_DISK_MMIO_DEV_SIZE and the size was exactly the size needed by
> > virtio-disk. When/if we support another virtio protocol we can add set
> > the appropriate size of that one as well.
>
> I would prefer if we keep this generic, although we are going to support
> virtio-blk as the first virtio-mmio device, there is nothing disk specific in
> that value. The same value (0x200) is used
> by SW which also generates virtio-mmio device tree nodes like we do in Xen
> libxl, I am speaking about kvmtool, Qemu here.
OK
^ permalink raw reply [flat|nested] 23+ messages in thread
* RE: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-08 18:21 ` [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm Oleksandr Tyshchenko
2022-04-18 21:41 ` Stefano Stabellini
@ 2022-04-20 5:20 ` Henry Wang
2022-04-20 9:07 ` Oleksandr
2022-04-22 2:42 ` Jiamei Xie
2 siblings, 1 reply; 23+ messages in thread
From: Henry Wang @ 2022-04-20 5:20 UTC (permalink / raw)
To: Oleksandr Tyshchenko, xen-devel
Cc: Julien Grall, Wei Liu, Anthony PERARD, Juergen Gross,
Stefano Stabellini, Julien Grall, Bertrand Marquis,
Volodymyr Babchuk, Jiamei Xie, Oleksandr Tyshchenko
Hi Oleksandr,
> -----Original Message-----
> From: Oleksandr Tyshchenko <olekstysh@gmail.com>
> Subject: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
>
> From: Julien Grall <julien.grall@arm.com>
>
> This patch introduces helpers to allocate Virtio MMIO params
> (IRQ and memory region) and create specific device node in
> the Guest device-tree with allocated params. In order to deal
> with multiple Virtio devices, reserve corresponding ranges.
> For now, we reserve 1MB for memory regions and 10 SPIs.
>
> As these helpers should be used for every Virtio device attached
> to the Guest, call them for Virtio disk(s).
>
> Please note, with statically allocated Virtio IRQs there is
> a risk of a clash with a physical IRQs of passthrough devices.
> For the first version, it's fine, but we should consider allocating
> the Virtio IRQs automatically. Thankfully, we know in advance which
> IRQs will be used for passthrough to be able to choose non-clashed
> ones.
>
> Signed-off-by: Julien Grall <julien.grall@arm.com>
> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> Tested-by: Jiamei Xie <Jiamei.xie@arm.com>
> Reviewed-by: Henry Wang <Henry.Wang@arm.com>
> ---
> @Jiamei, @Henry I decided to leave your T-b and R-b tags with the minor
> change I made, are you still happy with that?
Sorry for the late response, just checked the code and yes I am happy with
keeping my Reviewed-by. Thanks for your effort in rebasing the patch!
Kind regards,
Henry
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-20 5:20 ` Henry Wang
@ 2022-04-20 9:07 ` Oleksandr
0 siblings, 0 replies; 23+ messages in thread
From: Oleksandr @ 2022-04-20 9:07 UTC (permalink / raw)
To: Henry Wang
Cc: xen-devel, Julien Grall, Wei Liu, Anthony PERARD, Juergen Gross,
Stefano Stabellini, Julien Grall, Bertrand Marquis,
Volodymyr Babchuk, Jiamei Xie, Oleksandr Tyshchenko
On 20.04.22 08:20, Henry Wang wrote:
> Hi Oleksandr,
Hi Henry
>
>> -----Original Message-----
>> From: Oleksandr Tyshchenko <olekstysh@gmail.com>
>> Subject: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
>>
>> From: Julien Grall <julien.grall@arm.com>
>>
>> This patch introduces helpers to allocate Virtio MMIO params
>> (IRQ and memory region) and create specific device node in
>> the Guest device-tree with allocated params. In order to deal
>> with multiple Virtio devices, reserve corresponding ranges.
>> For now, we reserve 1MB for memory regions and 10 SPIs.
>>
>> As these helpers should be used for every Virtio device attached
>> to the Guest, call them for Virtio disk(s).
>>
>> Please note, with statically allocated Virtio IRQs there is
>> a risk of a clash with a physical IRQs of passthrough devices.
>> For the first version, it's fine, but we should consider allocating
>> the Virtio IRQs automatically. Thankfully, we know in advance which
>> IRQs will be used for passthrough to be able to choose non-clashed
>> ones.
>>
>> Signed-off-by: Julien Grall <julien.grall@arm.com>
>> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>> Tested-by: Jiamei Xie <Jiamei.xie@arm.com>
>> Reviewed-by: Henry Wang <Henry.Wang@arm.com>
>> ---
>> @Jiamei, @Henry I decided to leave your T-b and R-b tags with the minor
>> change I made, are you still happy with that?
> Sorry for the late response, just checked the code and yes I am happy with
> keeping my Reviewed-by. Thanks for your effort in rebasing the patch!
Thank you for confirming! Note that I will have to drop tags for the
next version due to non-minor changes I am currently making to address
review comments.
>
> Kind regards,
> Henry
--
Regards,
Oleksandr Tyshchenko
^ permalink raw reply [flat|nested] 23+ messages in thread
* RE: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-08 18:21 ` [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm Oleksandr Tyshchenko
2022-04-18 21:41 ` Stefano Stabellini
2022-04-20 5:20 ` Henry Wang
@ 2022-04-22 2:42 ` Jiamei Xie
2022-04-22 10:34 ` Oleksandr Tyshchenko
2 siblings, 1 reply; 23+ messages in thread
From: Jiamei Xie @ 2022-04-22 2:42 UTC (permalink / raw)
To: Oleksandr Tyshchenko, xen-devel
Cc: Julien Grall, Wei Liu, Anthony PERARD, Juergen Gross,
Stefano Stabellini, Julien Grall, Bertrand Marquis,
Volodymyr Babchuk, Henry Wang, Oleksandr Tyshchenko
Hi Oleksandr,
I am happy to keep my T-b tag. I have tested this latest patch series and it works.
Regards,
Jiamei Xie
> -----Original Message-----
> From: Xen-devel <xen-devel-bounces@lists.xenproject.org> On Behalf Of
> Oleksandr Tyshchenko
> Sent: 2022年4月9日 2:21
> To: xen-devel@lists.xenproject.org
> Cc: Julien Grall <Julien.Grall@arm.com>; Wei Liu <wl@xen.org>; Anthony
> PERARD <anthony.perard@citrix.com>; Juergen Gross <jgross@suse.com>;
> Stefano Stabellini <sstabellini@kernel.org>; Julien Grall <julien@xen.org>;
> Bertrand Marquis <Bertrand.Marquis@arm.com>; Volodymyr Babchuk
> <Volodymyr_Babchuk@epam.com>; Jiamei Xie <Jiamei.Xie@arm.com>;
> Henry Wang <Henry.Wang@arm.com>; Oleksandr Tyshchenko
> <oleksandr_tyshchenko@epam.com>
> Subject: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
>
> From: Julien Grall <julien.grall@arm.com>
>
> This patch introduces helpers to allocate Virtio MMIO params
> (IRQ and memory region) and create specific device node in
> the Guest device-tree with allocated params. In order to deal
> with multiple Virtio devices, reserve corresponding ranges.
> For now, we reserve 1MB for memory regions and 10 SPIs.
>
> As these helpers should be used for every Virtio device attached
> to the Guest, call them for Virtio disk(s).
>
> Please note, with statically allocated Virtio IRQs there is
> a risk of a clash with a physical IRQs of passthrough devices.
> For the first version, it's fine, but we should consider allocating
> the Virtio IRQs automatically. Thankfully, we know in advance which
> IRQs will be used for passthrough to be able to choose non-clashed
> ones.
>
> Signed-off-by: Julien Grall <julien.grall@arm.com>
> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> Tested-by: Jiamei Xie <Jiamei.xie@arm.com>
> Reviewed-by: Henry Wang <Henry.Wang@arm.com>
> ---
> @Jiamei, @Henry I decided to leave your T-b and R-b tags with the minor
> change I made, are you still happy with that?
>
> s/if (disk->virtio)/if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
>
> Please note, this is a split/cleanup/hardening of Julien's PoC:
> "Add support for Guest IO forwarding to a device emulator"
>
> Changes RFC -> V1:
> - was squashed with:
> "[RFC PATCH V1 09/12] libxl: Handle virtio-mmio irq in more correct way"
> "[RFC PATCH V1 11/12] libxl: Insert "dma-coherent" property into virtio-
> mmio device node"
> "[RFC PATCH V1 12/12] libxl: Fix duplicate memory node in DT"
> - move VirtIO MMIO #define-s to xen/include/public/arch-arm.h
>
> Changes V1 -> V2:
> - update the author of a patch
>
> Changes V2 -> V3:
> - no changes
>
> Changes V3 -> V4:
> - no changes
>
> Changes V4 -> V5:
> - split the changes, change the order of the patches
> - drop an extra "virtio" configuration option
> - update patch description
> - use CONTAINER_OF instead of own implementation
> - reserve ranges for Virtio MMIO params and put them
> in correct location
> - create helpers to allocate Virtio MMIO params, add
> corresponding sanity-сhecks
> - add comment why MMIO size 0x200 is chosen
> - update debug print
> - drop Wei's T-b
>
> Changes V5 -> V6:
> - rebase on current staging
>
> Changes V6 -> V7:
> - rebase on current staging
> - add T-b and R-b tags
> - update according to the recent changes to
> "libxl: Add support for Virtio disk configuration"
> ---
> tools/libs/light/libxl_arm.c | 131
> +++++++++++++++++++++++++++++++++++++++++-
> xen/include/public/arch-arm.h | 7 +++
> 2 files changed, 136 insertions(+), 2 deletions(-)
>
> diff --git a/tools/libs/light/libxl_arm.c b/tools/libs/light/libxl_arm.c
> index eef1de0..8132a47 100644
> --- a/tools/libs/light/libxl_arm.c
> +++ b/tools/libs/light/libxl_arm.c
> @@ -8,6 +8,56 @@
> #include <assert.h>
> #include <xen/device_tree_defs.h>
>
> +/*
> + * There is no clear requirements for the total size of Virtio MMIO region.
> + * The size of control registers is 0x100 and device-specific configuration
> + * registers starts at the offset 0x100, however it's size depends on the
> device
> + * and the driver. Pick the biggest known size at the moment to cover most
> + * of the devices (also consider allowing the user to configure the size via
> + * config file for the one not conforming with the proposed value).
> + */
> +#define VIRTIO_MMIO_DEV_SIZE xen_mk_ullong(0x200)
> +
> +static uint64_t virtio_mmio_base;
> +static uint32_t virtio_mmio_irq;
> +
> +static void init_virtio_mmio_params(void)
> +{
> + virtio_mmio_base = GUEST_VIRTIO_MMIO_BASE;
> + virtio_mmio_irq = GUEST_VIRTIO_MMIO_SPI_FIRST;
> +}
> +
> +static uint64_t alloc_virtio_mmio_base(libxl__gc *gc)
> +{
> + uint64_t base = virtio_mmio_base;
> +
> + /* Make sure we have enough reserved resources */
> + if ((virtio_mmio_base + VIRTIO_MMIO_DEV_SIZE >
> + GUEST_VIRTIO_MMIO_BASE + GUEST_VIRTIO_MMIO_SIZE)) {
> + LOG(ERROR, "Ran out of reserved range for Virtio MMIO BASE
> 0x%"PRIx64"\n",
> + virtio_mmio_base);
> + return 0;
> + }
> + virtio_mmio_base += VIRTIO_MMIO_DEV_SIZE;
> +
> + return base;
> +}
> +
> +static uint32_t alloc_virtio_mmio_irq(libxl__gc *gc)
> +{
> + uint32_t irq = virtio_mmio_irq;
> +
> + /* Make sure we have enough reserved resources */
> + if (virtio_mmio_irq > GUEST_VIRTIO_MMIO_SPI_LAST) {
> + LOG(ERROR, "Ran out of reserved range for Virtio MMIO IRQ %u\n",
> + virtio_mmio_irq);
> + return 0;
> + }
> + virtio_mmio_irq++;
> +
> + return irq;
> +}
> +
> static const char *gicv_to_string(libxl_gic_version gic_version)
> {
> switch (gic_version) {
> @@ -26,8 +76,8 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> {
> uint32_t nr_spis = 0;
> unsigned int i;
> - uint32_t vuart_irq;
> - bool vuart_enabled = false;
> + uint32_t vuart_irq, virtio_irq = 0;
> + bool vuart_enabled = false, virtio_enabled = false;
>
> /*
> * If pl011 vuart is enabled then increment the nr_spis to allow allocation
> @@ -39,6 +89,35 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> vuart_enabled = true;
> }
>
> + /*
> + * Virtio MMIO params are non-unique across the whole system and
> must be
> + * initialized for every new guest.
> + */
> + init_virtio_mmio_params();
> + for (i = 0; i < d_config->num_disks; i++) {
> + libxl_device_disk *disk = &d_config->disks[i];
> +
> + if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
> + disk->base = alloc_virtio_mmio_base(gc);
> + if (!disk->base)
> + return ERROR_FAIL;
> +
> + disk->irq = alloc_virtio_mmio_irq(gc);
> + if (!disk->irq)
> + return ERROR_FAIL;
> +
> + if (virtio_irq < disk->irq)
> + virtio_irq = disk->irq;
> + virtio_enabled = true;
> +
> + LOG(DEBUG, "Allocate Virtio MMIO params for Vdev %s: IRQ %u
> BASE 0x%"PRIx64,
> + disk->vdev, disk->irq, disk->base);
> + }
> + }
> +
> + if (virtio_enabled)
> + nr_spis += (virtio_irq - 32) + 1;
> +
> for (i = 0; i < d_config->b_info.num_irqs; i++) {
> uint32_t irq = d_config->b_info.irqs[i];
> uint32_t spi;
> @@ -58,6 +137,13 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> return ERROR_FAIL;
> }
>
> + /* The same check as for vpl011 */
> + if (virtio_enabled &&
> + (irq >= GUEST_VIRTIO_MMIO_SPI_FIRST && irq <= virtio_irq)) {
> + LOG(ERROR, "Physical IRQ %u conflicting with Virtio MMIO IRQ
> range\n", irq);
> + return ERROR_FAIL;
> + }
> +
> if (irq < 32)
> continue;
>
> @@ -787,6 +873,39 @@ static int make_vpci_node(libxl__gc *gc, void *fdt,
> return 0;
> }
>
> +
> +static int make_virtio_mmio_node(libxl__gc *gc, void *fdt,
> + uint64_t base, uint32_t irq)
> +{
> + int res;
> + gic_interrupt intr;
> + /* Placeholder for virtio@ + a 64-bit number + \0 */
> + char buf[24];
> +
> + snprintf(buf, sizeof(buf), "virtio@%"PRIx64, base);
> + res = fdt_begin_node(fdt, buf);
> + if (res) return res;
> +
> + res = fdt_property_compat(gc, fdt, 1, "virtio,mmio");
> + if (res) return res;
> +
> + res = fdt_property_regs(gc, fdt, GUEST_ROOT_ADDRESS_CELLS,
> GUEST_ROOT_SIZE_CELLS,
> + 1, base, VIRTIO_MMIO_DEV_SIZE);
> + if (res) return res;
> +
> + set_interrupt(intr, irq, 0xf, DT_IRQ_TYPE_EDGE_RISING);
> + res = fdt_property_interrupts(gc, fdt, &intr, 1);
> + if (res) return res;
> +
> + res = fdt_property(fdt, "dma-coherent", NULL, 0);
> + if (res) return res;
> +
> + res = fdt_end_node(fdt);
> + if (res) return res;
> +
> + return 0;
> +}
> +
> static const struct arch_info *get_arch_info(libxl__gc *gc,
> const struct xc_dom_image *dom)
> {
> @@ -988,6 +1107,7 @@ static int libxl__prepare_dtb(libxl__gc *gc,
> libxl_domain_config *d_config,
> size_t fdt_size = 0;
> int pfdt_size = 0;
> libxl_domain_build_info *const info = &d_config->b_info;
> + unsigned int i;
>
> const libxl_version_info *vers;
> const struct arch_info *ainfo;
> @@ -1094,6 +1214,13 @@ next_resize:
> if (d_config->num_pcidevs)
> FDT( make_vpci_node(gc, fdt, ainfo, dom) );
>
> + for (i = 0; i < d_config->num_disks; i++) {
> + libxl_device_disk *disk = &d_config->disks[i];
> +
> + if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
> + FDT( make_virtio_mmio_node(gc, fdt, disk->base, disk->irq) );
> + }
> +
> if (pfdt)
> FDT( copy_partial_fdt(gc, fdt, pfdt) );
>
> diff --git a/xen/include/public/arch-arm.h b/xen/include/public/arch-arm.h
> index ab05fe1..c8b6058 100644
> --- a/xen/include/public/arch-arm.h
> +++ b/xen/include/public/arch-arm.h
> @@ -407,6 +407,10 @@ typedef uint64_t xen_callback_t;
>
> /* Physical Address Space */
>
> +/* Virtio MMIO mappings */
> +#define GUEST_VIRTIO_MMIO_BASE xen_mk_ullong(0x02000000)
> +#define GUEST_VIRTIO_MMIO_SIZE xen_mk_ullong(0x00100000)
> +
> /*
> * vGIC mappings: Only one set of mapping is used by the guest.
> * Therefore they can overlap.
> @@ -493,6 +497,9 @@ typedef uint64_t xen_callback_t;
>
> #define GUEST_VPL011_SPI 32
>
> +#define GUEST_VIRTIO_MMIO_SPI_FIRST 33
> +#define GUEST_VIRTIO_MMIO_SPI_LAST 43
> +
> /* PSCI functions */
> #define PSCI_cpu_suspend 0
> #define PSCI_cpu_off 1
> --
> 2.7.4
>
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-08 18:21 ` [PATCH V7 1/2] libxl: Add support for Virtio disk configuration Oleksandr Tyshchenko
2022-04-18 21:38 ` Stefano Stabellini
@ 2022-04-22 9:41 ` Roger Pau Monné
2022-04-23 7:39 ` Oleksandr
2022-04-25 7:43 ` Juergen Gross
2 siblings, 1 reply; 23+ messages in thread
From: Roger Pau Monné @ 2022-04-22 9:41 UTC (permalink / raw)
To: Oleksandr Tyshchenko
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Juergen Gross, Stefano Stabellini,
Julien Grall, Volodymyr Babchuk, Bertrand Marquis
On Fri, Apr 08, 2022 at 09:21:04PM +0300, Oleksandr Tyshchenko wrote:
> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>
> This patch adds basic support for configuring and assisting virtio-mmio
> based virtio-disk backend (emualator) which is intended to run out of
> Qemu and could be run in any domain.
> Although the Virtio block device is quite different from traditional
> Xen PV block device (vbd) from the toolstack's point of view:
> - as the frontend is virtio-blk which is not a Xenbus driver, nothing
> written to Xenstore are fetched by the frontend (the vdev is not
> passed to the frontend)
> - the ring-ref/event-channel are not used for the backend<->frontend
> communication, the proposed IPC for Virtio is IOREQ/DM
> it is still a "block device" and ought to be integrated in existing
> "disk" handling. So, re-use (and adapt) "disk" parsing/configuration
> logic to deal with Virtio devices as well.
>
> For the immediate purpose and an ability to extend that support for
> other use-cases in future (Qemu, virtio-pci, etc) perform the following
> actions:
> - Add new disk backend type (LIBXL_DISK_BACKEND_OTHER) and reflect
> that in the configuration
> - Introduce new disk protocol field to libxl_device_disk struct
> (with LIBXL_DISK_PROTOCOL_XEN and LIBXL_DISK_PROTOCOL_VIRTIO_MMIO
> types) and reflect that in the configuration (new "protocol" option
> with "xen" protocol being default value)
> - Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
> one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
>
> An example of domain configuration for Virtio disk:
> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other, protocol=virtio-mmio']
>
> Nothing has changed for default Xen disk configuration.
>
> Please note, this patch is not enough for virtio-disk to work
> on Xen (Arm), as for every Virtio device (including disk) we need
> to allocate Virtio MMIO params (IRQ and memory region) and pass
> them to the backend, also update Guest device-tree. The subsequent
> patch will add these missing bits. For the current patch,
> the default "irq" and "base" are just written to the Xenstore.
> This is not an ideal splitting, but this way we avoid breaking
> the bisectability.
>
> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> ---
> Changes RFC -> V1:
> - no changes
>
> Changes V1 -> V2:
> - rebase according to the new location of libxl_virtio_disk.c
>
> Changes V2 -> V3:
> - no changes
>
> Changes V3 -> V4:
> - rebase according to the new argument for DEFINE_DEVICE_TYPE_STRUCT
>
> Changes V4 -> V5:
> - split the changes, change the order of the patches
> - update patch description
> - don't introduce new "vdisk" configuration option with own parsing logic,
> re-use Xen PV block "disk" parsing/configuration logic for the virtio-disk
> - introduce "virtio" flag and document it's usage
> - add LIBXL_HAVE_DEVICE_DISK_VIRTIO
> - update libxlu_disk_l.[ch]
> - drop num_disks variable/MAX_VIRTIO_DISKS
> - drop Wei's T-b
>
> Changes V5 -> V6:
> - rebase on current staging
> - use "%"PRIu64 instead of %lu for disk->base in device_disk_add()
> - update *.gen.go files
>
> Changes V6 -> V7:
> - rebase on current staging
> - update *.gen.go files and libxlu_disk_l.[ch] files
> - update patch description
> - rework significantly to support more flexible configuration
> and have more generic basic implementation for being able to extend
> that for other use-cases (virtio-pci, qemu, etc).
> ---
> docs/man/xl-disk-configuration.5.pod.in | 37 +-
> tools/golang/xenlight/helpers.gen.go | 6 +
> tools/golang/xenlight/types.gen.go | 11 +
> tools/include/libxl.h | 6 +
> tools/libs/light/libxl_device.c | 57 +-
> tools/libs/light/libxl_disk.c | 111 +++-
> tools/libs/light/libxl_internal.h | 1 +
> tools/libs/light/libxl_types.idl | 10 +
> tools/libs/light/libxl_types_internal.idl | 1 +
> tools/libs/light/libxl_utils.c | 2 +
> tools/libs/util/libxlu_disk_l.c | 952 +++++++++++++++---------------
> tools/libs/util/libxlu_disk_l.h | 2 +-
> tools/libs/util/libxlu_disk_l.l | 9 +
> tools/xl/xl_block.c | 11 +
> 14 files changed, 736 insertions(+), 480 deletions(-)
>
> diff --git a/docs/man/xl-disk-configuration.5.pod.in b/docs/man/xl-disk-configuration.5.pod.in
> index 71d0e86..36c851f 100644
> --- a/docs/man/xl-disk-configuration.5.pod.in
> +++ b/docs/man/xl-disk-configuration.5.pod.in
> @@ -232,7 +232,7 @@ Specifies the backend implementation to use
>
> =item Supported values
>
> -phy, qdisk
> +phy, qdisk, other
>
> =item Mandatory
>
> @@ -244,11 +244,13 @@ Automatically determine which backend to use.
>
> =back
>
> -This does not affect the guest's view of the device. It controls
> -which software implementation of the Xen backend driver is used.
> +It controls which software implementation of the backend driver is used.
> +Depending on the "protocol" option this may affect the guest's view
> +of the device.
>
> Not all backend drivers support all combinations of other options.
> -For example, "phy" does not support formats other than "raw".
> +For example, "phy" and "other" do not support formats other than "raw" and
> +"other" does not support protocols other than "virtio-mmio".
> Normally this option should not be specified, in which case libxl will
> automatically determine the most suitable backend.
>
> @@ -344,8 +346,35 @@ can be used to disable "hole punching" for file based backends which
> were intentionally created non-sparse to avoid fragmentation of the
> file.
>
> +=item B<protocol>=I<PROTOCOL>
> +
> +=over 4
> +
> +=item Description
> +
> +Specifies the communication protocol to use for the chosen "backendtype" option
> +
> +=item Supported values
> +
> +xen, virtio-mmio
From a user PoV, I think it would be better to just select xen or
virtio here, but not the underlying configuration mechanism used to
expose the devices to the guest.
We would likely need to add a different option to select mmio or pci
then, but that should be set by default based on architecture/guest
type. For example on x86 it should default to pci, while on Arm I
guess it will depend on whether the guest has PCI or not?
In any case, I think we should offer an option that's selecting
between xen or virtio protocol, and the way to expose the
configuration of the device shouldn't need to be explicitly selected
by the user.
Thanks, Roger.
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
2022-04-22 2:42 ` Jiamei Xie
@ 2022-04-22 10:34 ` Oleksandr Tyshchenko
0 siblings, 0 replies; 23+ messages in thread
From: Oleksandr Tyshchenko @ 2022-04-22 10:34 UTC (permalink / raw)
To: Jiamei Xie
Cc: xen-devel, Julien Grall, Wei Liu, Anthony PERARD, Juergen Gross,
Stefano Stabellini, Julien Grall, Bertrand Marquis,
Volodymyr Babchuk, Henry Wang, Oleksandr Tyshchenko
[-- Attachment #1: Type: text/plain, Size: 11960 bytes --]
On Fri, Apr 22, 2022 at 5:43 AM Jiamei Xie <Jiamei.Xie@arm.com> wrote:
> Hi Oleksandr,
>
Hi Jiamei
[Sorry for the possible format issues]
>
> I am happy to keep my T-b tag. I have tested this latest patch series
> and it works.
Thank you for the testing and confirmation!
>
>
> Regards,
> Jiamei Xie
>
> > -----Original Message-----
> > From: Xen-devel <xen-devel-bounces@lists.xenproject.org> On Behalf Of
> > Oleksandr Tyshchenko
> > Sent: 2022年4月9日 2:21
> > To: xen-devel@lists.xenproject.org
> > Cc: Julien Grall <Julien.Grall@arm.com>; Wei Liu <wl@xen.org>; Anthony
> > PERARD <anthony.perard@citrix.com>; Juergen Gross <jgross@suse.com>;
> > Stefano Stabellini <sstabellini@kernel.org>; Julien Grall <
> julien@xen.org>;
> > Bertrand Marquis <Bertrand.Marquis@arm.com>; Volodymyr Babchuk
> > <Volodymyr_Babchuk@epam.com>; Jiamei Xie <Jiamei.Xie@arm.com>;
> > Henry Wang <Henry.Wang@arm.com>; Oleksandr Tyshchenko
> > <oleksandr_tyshchenko@epam.com>
> > Subject: [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm
> >
> > From: Julien Grall <julien.grall@arm.com>
> >
> > This patch introduces helpers to allocate Virtio MMIO params
> > (IRQ and memory region) and create specific device node in
> > the Guest device-tree with allocated params. In order to deal
> > with multiple Virtio devices, reserve corresponding ranges.
> > For now, we reserve 1MB for memory regions and 10 SPIs.
> >
> > As these helpers should be used for every Virtio device attached
> > to the Guest, call them for Virtio disk(s).
> >
> > Please note, with statically allocated Virtio IRQs there is
> > a risk of a clash with a physical IRQs of passthrough devices.
> > For the first version, it's fine, but we should consider allocating
> > the Virtio IRQs automatically. Thankfully, we know in advance which
> > IRQs will be used for passthrough to be able to choose non-clashed
> > ones.
> >
> > Signed-off-by: Julien Grall <julien.grall@arm.com>
> > Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> > Tested-by: Jiamei Xie <Jiamei.xie@arm.com>
> > Reviewed-by: Henry Wang <Henry.Wang@arm.com>
> > ---
> > @Jiamei, @Henry I decided to leave your T-b and R-b tags with the minor
> > change I made, are you still happy with that?
> >
> > s/if (disk->virtio)/if (disk->protocol ==
> LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
> >
> > Please note, this is a split/cleanup/hardening of Julien's PoC:
> > "Add support for Guest IO forwarding to a device emulator"
> >
> > Changes RFC -> V1:
> > - was squashed with:
> > "[RFC PATCH V1 09/12] libxl: Handle virtio-mmio irq in more correct
> way"
> > "[RFC PATCH V1 11/12] libxl: Insert "dma-coherent" property into
> virtio-
> > mmio device node"
> > "[RFC PATCH V1 12/12] libxl: Fix duplicate memory node in DT"
> > - move VirtIO MMIO #define-s to xen/include/public/arch-arm.h
> >
> > Changes V1 -> V2:
> > - update the author of a patch
> >
> > Changes V2 -> V3:
> > - no changes
> >
> > Changes V3 -> V4:
> > - no changes
> >
> > Changes V4 -> V5:
> > - split the changes, change the order of the patches
> > - drop an extra "virtio" configuration option
> > - update patch description
> > - use CONTAINER_OF instead of own implementation
> > - reserve ranges for Virtio MMIO params and put them
> > in correct location
> > - create helpers to allocate Virtio MMIO params, add
> > corresponding sanity-сhecks
> > - add comment why MMIO size 0x200 is chosen
> > - update debug print
> > - drop Wei's T-b
> >
> > Changes V5 -> V6:
> > - rebase on current staging
> >
> > Changes V6 -> V7:
> > - rebase on current staging
> > - add T-b and R-b tags
> > - update according to the recent changes to
> > "libxl: Add support for Virtio disk configuration"
> > ---
> > tools/libs/light/libxl_arm.c | 131
> > +++++++++++++++++++++++++++++++++++++++++-
> > xen/include/public/arch-arm.h | 7 +++
> > 2 files changed, 136 insertions(+), 2 deletions(-)
> >
> > diff --git a/tools/libs/light/libxl_arm.c b/tools/libs/light/libxl_arm.c
> > index eef1de0..8132a47 100644
> > --- a/tools/libs/light/libxl_arm.c
> > +++ b/tools/libs/light/libxl_arm.c
> > @@ -8,6 +8,56 @@
> > #include <assert.h>
> > #include <xen/device_tree_defs.h>
> >
> > +/*
> > + * There is no clear requirements for the total size of Virtio MMIO
> region.
> > + * The size of control registers is 0x100 and device-specific
> configuration
> > + * registers starts at the offset 0x100, however it's size depends on
> the
> > device
> > + * and the driver. Pick the biggest known size at the moment to cover
> most
> > + * of the devices (also consider allowing the user to configure the
> size via
> > + * config file for the one not conforming with the proposed value).
> > + */
> > +#define VIRTIO_MMIO_DEV_SIZE xen_mk_ullong(0x200)
> > +
> > +static uint64_t virtio_mmio_base;
> > +static uint32_t virtio_mmio_irq;
> > +
> > +static void init_virtio_mmio_params(void)
> > +{
> > + virtio_mmio_base = GUEST_VIRTIO_MMIO_BASE;
> > + virtio_mmio_irq = GUEST_VIRTIO_MMIO_SPI_FIRST;
> > +}
> > +
> > +static uint64_t alloc_virtio_mmio_base(libxl__gc *gc)
> > +{
> > + uint64_t base = virtio_mmio_base;
> > +
> > + /* Make sure we have enough reserved resources */
> > + if ((virtio_mmio_base + VIRTIO_MMIO_DEV_SIZE >
> > + GUEST_VIRTIO_MMIO_BASE + GUEST_VIRTIO_MMIO_SIZE)) {
> > + LOG(ERROR, "Ran out of reserved range for Virtio MMIO BASE
> > 0x%"PRIx64"\n",
> > + virtio_mmio_base);
> > + return 0;
> > + }
> > + virtio_mmio_base += VIRTIO_MMIO_DEV_SIZE;
> > +
> > + return base;
> > +}
> > +
> > +static uint32_t alloc_virtio_mmio_irq(libxl__gc *gc)
> > +{
> > + uint32_t irq = virtio_mmio_irq;
> > +
> > + /* Make sure we have enough reserved resources */
> > + if (virtio_mmio_irq > GUEST_VIRTIO_MMIO_SPI_LAST) {
> > + LOG(ERROR, "Ran out of reserved range for Virtio MMIO IRQ %u\n",
> > + virtio_mmio_irq);
> > + return 0;
> > + }
> > + virtio_mmio_irq++;
> > +
> > + return irq;
> > +}
> > +
> > static const char *gicv_to_string(libxl_gic_version gic_version)
> > {
> > switch (gic_version) {
> > @@ -26,8 +76,8 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> > {
> > uint32_t nr_spis = 0;
> > unsigned int i;
> > - uint32_t vuart_irq;
> > - bool vuart_enabled = false;
> > + uint32_t vuart_irq, virtio_irq = 0;
> > + bool vuart_enabled = false, virtio_enabled = false;
> >
> > /*
> > * If pl011 vuart is enabled then increment the nr_spis to allow
> allocation
> > @@ -39,6 +89,35 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> > vuart_enabled = true;
> > }
> >
> > + /*
> > + * Virtio MMIO params are non-unique across the whole system and
> > must be
> > + * initialized for every new guest.
> > + */
> > + init_virtio_mmio_params();
> > + for (i = 0; i < d_config->num_disks; i++) {
> > + libxl_device_disk *disk = &d_config->disks[i];
> > +
> > + if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO) {
> > + disk->base = alloc_virtio_mmio_base(gc);
> > + if (!disk->base)
> > + return ERROR_FAIL;
> > +
> > + disk->irq = alloc_virtio_mmio_irq(gc);
> > + if (!disk->irq)
> > + return ERROR_FAIL;
> > +
> > + if (virtio_irq < disk->irq)
> > + virtio_irq = disk->irq;
> > + virtio_enabled = true;
> > +
> > + LOG(DEBUG, "Allocate Virtio MMIO params for Vdev %s: IRQ %u
> > BASE 0x%"PRIx64,
> > + disk->vdev, disk->irq, disk->base);
> > + }
> > + }
> > +
> > + if (virtio_enabled)
> > + nr_spis += (virtio_irq - 32) + 1;
> > +
> > for (i = 0; i < d_config->b_info.num_irqs; i++) {
> > uint32_t irq = d_config->b_info.irqs[i];
> > uint32_t spi;
> > @@ -58,6 +137,13 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc,
> > return ERROR_FAIL;
> > }
> >
> > + /* The same check as for vpl011 */
> > + if (virtio_enabled &&
> > + (irq >= GUEST_VIRTIO_MMIO_SPI_FIRST && irq <= virtio_irq)) {
> > + LOG(ERROR, "Physical IRQ %u conflicting with Virtio MMIO IRQ
> > range\n", irq);
> > + return ERROR_FAIL;
> > + }
> > +
> > if (irq < 32)
> > continue;
> >
> > @@ -787,6 +873,39 @@ static int make_vpci_node(libxl__gc *gc, void *fdt,
> > return 0;
> > }
> >
> > +
> > +static int make_virtio_mmio_node(libxl__gc *gc, void *fdt,
> > + uint64_t base, uint32_t irq)
> > +{
> > + int res;
> > + gic_interrupt intr;
> > + /* Placeholder for virtio@ + a 64-bit number + \0 */
> > + char buf[24];
> > +
> > + snprintf(buf, sizeof(buf), "virtio@%"PRIx64, base);
> > + res = fdt_begin_node(fdt, buf);
> > + if (res) return res;
> > +
> > + res = fdt_property_compat(gc, fdt, 1, "virtio,mmio");
> > + if (res) return res;
> > +
> > + res = fdt_property_regs(gc, fdt, GUEST_ROOT_ADDRESS_CELLS,
> > GUEST_ROOT_SIZE_CELLS,
> > + 1, base, VIRTIO_MMIO_DEV_SIZE);
> > + if (res) return res;
> > +
> > + set_interrupt(intr, irq, 0xf, DT_IRQ_TYPE_EDGE_RISING);
> > + res = fdt_property_interrupts(gc, fdt, &intr, 1);
> > + if (res) return res;
> > +
> > + res = fdt_property(fdt, "dma-coherent", NULL, 0);
> > + if (res) return res;
> > +
> > + res = fdt_end_node(fdt);
> > + if (res) return res;
> > +
> > + return 0;
> > +}
> > +
> > static const struct arch_info *get_arch_info(libxl__gc *gc,
> > const struct xc_dom_image
> *dom)
> > {
> > @@ -988,6 +1107,7 @@ static int libxl__prepare_dtb(libxl__gc *gc,
> > libxl_domain_config *d_config,
> > size_t fdt_size = 0;
> > int pfdt_size = 0;
> > libxl_domain_build_info *const info = &d_config->b_info;
> > + unsigned int i;
> >
> > const libxl_version_info *vers;
> > const struct arch_info *ainfo;
> > @@ -1094,6 +1214,13 @@ next_resize:
> > if (d_config->num_pcidevs)
> > FDT( make_vpci_node(gc, fdt, ainfo, dom) );
> >
> > + for (i = 0; i < d_config->num_disks; i++) {
> > + libxl_device_disk *disk = &d_config->disks[i];
> > +
> > + if (disk->protocol == LIBXL_DISK_PROTOCOL_VIRTIO_MMIO)
> > + FDT( make_virtio_mmio_node(gc, fdt, disk->base,
> disk->irq) );
> > + }
> > +
> > if (pfdt)
> > FDT( copy_partial_fdt(gc, fdt, pfdt) );
> >
> > diff --git a/xen/include/public/arch-arm.h
> b/xen/include/public/arch-arm.h
> > index ab05fe1..c8b6058 100644
> > --- a/xen/include/public/arch-arm.h
> > +++ b/xen/include/public/arch-arm.h
> > @@ -407,6 +407,10 @@ typedef uint64_t xen_callback_t;
> >
> > /* Physical Address Space */
> >
> > +/* Virtio MMIO mappings */
> > +#define GUEST_VIRTIO_MMIO_BASE xen_mk_ullong(0x02000000)
> > +#define GUEST_VIRTIO_MMIO_SIZE xen_mk_ullong(0x00100000)
> > +
> > /*
> > * vGIC mappings: Only one set of mapping is used by the guest.
> > * Therefore they can overlap.
> > @@ -493,6 +497,9 @@ typedef uint64_t xen_callback_t;
> >
> > #define GUEST_VPL011_SPI 32
> >
> > +#define GUEST_VIRTIO_MMIO_SPI_FIRST 33
> > +#define GUEST_VIRTIO_MMIO_SPI_LAST 43
> > +
> > /* PSCI functions */
> > #define PSCI_cpu_suspend 0
> > #define PSCI_cpu_off 1
> > --
> > 2.7.4
> >
>
>
--
Regards,
Oleksandr Tyshchenko
[-- Attachment #2: Type: text/html, Size: 16651 bytes --]
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-22 9:41 ` Roger Pau Monné
@ 2022-04-23 7:39 ` Oleksandr
2022-04-25 13:12 ` Roger Pau Monné
0 siblings, 1 reply; 23+ messages in thread
From: Oleksandr @ 2022-04-23 7:39 UTC (permalink / raw)
To: Roger Pau Monné
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Juergen Gross, Stefano Stabellini,
Julien Grall, Volodymyr Babchuk, Bertrand Marquis
On 22.04.22 12:41, Roger Pau Monné wrote:
Hello Roger
> On Fri, Apr 08, 2022 at 09:21:04PM +0300, Oleksandr Tyshchenko wrote:
>> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>>
>> This patch adds basic support for configuring and assisting virtio-mmio
>> based virtio-disk backend (emualator) which is intended to run out of
>> Qemu and could be run in any domain.
>> Although the Virtio block device is quite different from traditional
>> Xen PV block device (vbd) from the toolstack's point of view:
>> - as the frontend is virtio-blk which is not a Xenbus driver, nothing
>> written to Xenstore are fetched by the frontend (the vdev is not
>> passed to the frontend)
>> - the ring-ref/event-channel are not used for the backend<->frontend
>> communication, the proposed IPC for Virtio is IOREQ/DM
>> it is still a "block device" and ought to be integrated in existing
>> "disk" handling. So, re-use (and adapt) "disk" parsing/configuration
>> logic to deal with Virtio devices as well.
>>
>> For the immediate purpose and an ability to extend that support for
>> other use-cases in future (Qemu, virtio-pci, etc) perform the following
>> actions:
>> - Add new disk backend type (LIBXL_DISK_BACKEND_OTHER) and reflect
>> that in the configuration
>> - Introduce new disk protocol field to libxl_device_disk struct
>> (with LIBXL_DISK_PROTOCOL_XEN and LIBXL_DISK_PROTOCOL_VIRTIO_MMIO
>> types) and reflect that in the configuration (new "protocol" option
>> with "xen" protocol being default value)
>> - Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
>> one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
>>
>> An example of domain configuration for Virtio disk:
>> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other, protocol=virtio-mmio']
>>
>> Nothing has changed for default Xen disk configuration.
>>
>> Please note, this patch is not enough for virtio-disk to work
>> on Xen (Arm), as for every Virtio device (including disk) we need
>> to allocate Virtio MMIO params (IRQ and memory region) and pass
>> them to the backend, also update Guest device-tree. The subsequent
>> patch will add these missing bits. For the current patch,
>> the default "irq" and "base" are just written to the Xenstore.
>> This is not an ideal splitting, but this way we avoid breaking
>> the bisectability.
>>
>> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>> ---
>> Changes RFC -> V1:
>> - no changes
>>
>> Changes V1 -> V2:
>> - rebase according to the new location of libxl_virtio_disk.c
>>
>> Changes V2 -> V3:
>> - no changes
>>
>> Changes V3 -> V4:
>> - rebase according to the new argument for DEFINE_DEVICE_TYPE_STRUCT
>>
>> Changes V4 -> V5:
>> - split the changes, change the order of the patches
>> - update patch description
>> - don't introduce new "vdisk" configuration option with own parsing logic,
>> re-use Xen PV block "disk" parsing/configuration logic for the virtio-disk
>> - introduce "virtio" flag and document it's usage
>> - add LIBXL_HAVE_DEVICE_DISK_VIRTIO
>> - update libxlu_disk_l.[ch]
>> - drop num_disks variable/MAX_VIRTIO_DISKS
>> - drop Wei's T-b
>>
>> Changes V5 -> V6:
>> - rebase on current staging
>> - use "%"PRIu64 instead of %lu for disk->base in device_disk_add()
>> - update *.gen.go files
>>
>> Changes V6 -> V7:
>> - rebase on current staging
>> - update *.gen.go files and libxlu_disk_l.[ch] files
>> - update patch description
>> - rework significantly to support more flexible configuration
>> and have more generic basic implementation for being able to extend
>> that for other use-cases (virtio-pci, qemu, etc).
>> ---
>> docs/man/xl-disk-configuration.5.pod.in | 37 +-
>> tools/golang/xenlight/helpers.gen.go | 6 +
>> tools/golang/xenlight/types.gen.go | 11 +
>> tools/include/libxl.h | 6 +
>> tools/libs/light/libxl_device.c | 57 +-
>> tools/libs/light/libxl_disk.c | 111 +++-
>> tools/libs/light/libxl_internal.h | 1 +
>> tools/libs/light/libxl_types.idl | 10 +
>> tools/libs/light/libxl_types_internal.idl | 1 +
>> tools/libs/light/libxl_utils.c | 2 +
>> tools/libs/util/libxlu_disk_l.c | 952 +++++++++++++++---------------
>> tools/libs/util/libxlu_disk_l.h | 2 +-
>> tools/libs/util/libxlu_disk_l.l | 9 +
>> tools/xl/xl_block.c | 11 +
>> 14 files changed, 736 insertions(+), 480 deletions(-)
>>
>> diff --git a/docs/man/xl-disk-configuration.5.pod.in b/docs/man/xl-disk-configuration.5.pod.in
>> index 71d0e86..36c851f 100644
>> --- a/docs/man/xl-disk-configuration.5.pod.in
>> +++ b/docs/man/xl-disk-configuration.5.pod.in
>> @@ -232,7 +232,7 @@ Specifies the backend implementation to use
>>
>> =item Supported values
>>
>> -phy, qdisk
>> +phy, qdisk, other
>>
>> =item Mandatory
>>
>> @@ -244,11 +244,13 @@ Automatically determine which backend to use.
>>
>> =back
>>
>> -This does not affect the guest's view of the device. It controls
>> -which software implementation of the Xen backend driver is used.
>> +It controls which software implementation of the backend driver is used.
>> +Depending on the "protocol" option this may affect the guest's view
>> +of the device.
>>
>> Not all backend drivers support all combinations of other options.
>> -For example, "phy" does not support formats other than "raw".
>> +For example, "phy" and "other" do not support formats other than "raw" and
>> +"other" does not support protocols other than "virtio-mmio".
>> Normally this option should not be specified, in which case libxl will
>> automatically determine the most suitable backend.
>>
>> @@ -344,8 +346,35 @@ can be used to disable "hole punching" for file based backends which
>> were intentionally created non-sparse to avoid fragmentation of the
>> file.
>>
>> +=item B<protocol>=I<PROTOCOL>
>> +
>> +=over 4
>> +
>> +=item Description
>> +
>> +Specifies the communication protocol to use for the chosen "backendtype" option
>> +
>> +=item Supported values
>> +
>> +xen, virtio-mmio
> From a user PoV, I think it would be better to just select xen or
> virtio here, but not the underlying configuration mechanism used to
> expose the devices to the guest.
I got your point.
>
> We would likely need to add a different option to select mmio or pci
> then, but that should be set by default based on architecture/guest
> type. For example on x86 it should default to pci, while on Arm I
> guess it will depend on whether the guest has PCI or not?
>
> In any case, I think we should offer an option that's selecting
> between xen or virtio protocol, and the way to expose the
> configuration of the device shouldn't need to be explicitly selected
> by the user.
ok, for now I will use "xen and virtio" values for the "protocol"
option, then internally toolstack will assume that "virtio" really means
"virtio-mmio".
When there is a need to expand that support to "virtio-pci", we will see
how to deal with it from the configuration PoV, probably like you
suggested above by adding another option (e.g. "transport") with default
values based on the architecture/guest type.
>
> Thanks, Roger.
--
Regards,
Oleksandr Tyshchenko
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-08 18:21 ` [PATCH V7 1/2] libxl: Add support for Virtio disk configuration Oleksandr Tyshchenko
2022-04-18 21:38 ` Stefano Stabellini
2022-04-22 9:41 ` Roger Pau Monné
@ 2022-04-25 7:43 ` Juergen Gross
2022-04-25 12:02 ` Oleksandr
2 siblings, 1 reply; 23+ messages in thread
From: Juergen Gross @ 2022-04-25 7:43 UTC (permalink / raw)
To: Oleksandr Tyshchenko, xen-devel
Cc: Oleksandr Tyshchenko, Wei Liu, Anthony PERARD, George Dunlap,
Nick Rosbrook, Stefano Stabellini, Julien Grall,
Volodymyr Babchuk, Bertrand Marquis
[-- Attachment #1.1.1: Type: text/plain, Size: 4251 bytes --]
On 08.04.22 20:21, Oleksandr Tyshchenko wrote:
> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>
> This patch adds basic support for configuring and assisting virtio-mmio
> based virtio-disk backend (emualator) which is intended to run out of
> Qemu and could be run in any domain.
> Although the Virtio block device is quite different from traditional
> Xen PV block device (vbd) from the toolstack's point of view:
> - as the frontend is virtio-blk which is not a Xenbus driver, nothing
> written to Xenstore are fetched by the frontend (the vdev is not
> passed to the frontend)
I thought about the future support on x86.
There we don't have a device tree (and I don't want to introduce it),
so the only ways to specify the backend domain id would be to:
- add some information to ACPI tables
- use boot parameters
- use Xenstore
Thinking further of hotplugging virtio devices, Xenstore seems to be the
only real suitable alternative. Using virtio mechanisms doesn't seem
appropriate, as such information should be retrieved in "platform
specific" ways (see e.g. specifying an "endpoint" in the virtio IOMMU
device [1], [2]). I think the Xenstore information for that purpose
could be rather minimal and it should be device-type agnostic. Having
just a directory with endpoints and associated backend domids would
probably be enough (not needed in this series, of course).
This doesn't preclude the device tree variant you are using, as this
would be required for dom0less systems anyway.
OTOH I'd like you to modify the commit message to make it more clear
that in future frontend data might be written to Xenstore in order to
support other use cases.
> - the ring-ref/event-channel are not used for the backend<->frontend
> communication, the proposed IPC for Virtio is IOREQ/DM
> it is still a "block device" and ought to be integrated in existing
> "disk" handling. So, re-use (and adapt) "disk" parsing/configuration
> logic to deal with Virtio devices as well.
>
> For the immediate purpose and an ability to extend that support for
> other use-cases in future (Qemu, virtio-pci, etc) perform the following
> actions:
> - Add new disk backend type (LIBXL_DISK_BACKEND_OTHER) and reflect
> that in the configuration
> - Introduce new disk protocol field to libxl_device_disk struct
> (with LIBXL_DISK_PROTOCOL_XEN and LIBXL_DISK_PROTOCOL_VIRTIO_MMIO
> types) and reflect that in the configuration (new "protocol" option
> with "xen" protocol being default value)
And with the hotplug option in mind I start to feel unueasy with naming
the new Xenstore node "protocol", as the frontend disk nodes for "normal"
disks already have a "protocol" entry specifying 64- or 32-bit protocol.
Maybe we should really name it "transport" instead?
> - Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
> one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
>
> An example of domain configuration for Virtio disk:
> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other, protocol=virtio-mmio']
With Roger's feedback this would then be "transport=virtio", the "mmio"
part should then be something like "adapter=mmio" (in contrast to
"adapter=pci"), and "adapter" only needed in case of a device tree and
PCI being available.
>
> Nothing has changed for default Xen disk configuration.
>
> Please note, this patch is not enough for virtio-disk to work
> on Xen (Arm), as for every Virtio device (including disk) we need
> to allocate Virtio MMIO params (IRQ and memory region) and pass
> them to the backend, also update Guest device-tree. The subsequent
> patch will add these missing bits. For the current patch,
> the default "irq" and "base" are just written to the Xenstore.
> This is not an ideal splitting, but this way we avoid breaking
> the bisectability.
>
> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
I'm fine with the overall approach and couldn't spot any real issues
in the code.
Juergen
[1]: https://github.com/oasis-tcs/virtio-spec/blob/master/virtio-iommu.tex
[2]: https://medium.com/@michael2012zhao_67085/virtio-iommu-789369049443
[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 3149 bytes --]
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 495 bytes --]
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-25 7:43 ` Juergen Gross
@ 2022-04-25 12:02 ` Oleksandr
2022-04-25 12:09 ` Juergen Gross
0 siblings, 1 reply; 23+ messages in thread
From: Oleksandr @ 2022-04-25 12:02 UTC (permalink / raw)
To: Juergen Gross
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Stefano Stabellini, Julien Grall,
Volodymyr Babchuk, Bertrand Marquis
On 25.04.22 10:43, Juergen Gross wrote:
Hello Juergen
Thank you for the feedback.
> On 08.04.22 20:21, Oleksandr Tyshchenko wrote:
>> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>>
>> This patch adds basic support for configuring and assisting virtio-mmio
>> based virtio-disk backend (emualator) which is intended to run out of
>> Qemu and could be run in any domain.
>> Although the Virtio block device is quite different from traditional
>> Xen PV block device (vbd) from the toolstack's point of view:
>> - as the frontend is virtio-blk which is not a Xenbus driver, nothing
>> written to Xenstore are fetched by the frontend (the vdev is not
>> passed to the frontend)
>
> I thought about the future support on x86.
>
> There we don't have a device tree (and I don't want to introduce it),
> so the only ways to specify the backend domain id would be to:
>
> - add some information to ACPI tables
> - use boot parameters
> - use Xenstore
I understand that, and agree
> Thinking further of hotplugging virtio devices, Xenstore seems to be the
> only real suitable alternative. Using virtio mechanisms doesn't seem
> appropriate, as such information should be retrieved in "platform
> specific" ways (see e.g. specifying an "endpoint" in the virtio IOMMU
> device [1], [2]). I think the Xenstore information for that purpose
> could be rather minimal and it should be device-type agnostic. Having
> just a directory with endpoints and associated backend domids would
> probably be enough (not needed in this series, of course).
Just to make it clear, we are speaking about the possible ways to
communicate backend domid for another series [1], so about the x86's
alternative of device-tree bindings "xen,dev-domid" [2].
I was thinking we could avoid using Xenstore at the guest side for that
purpose, but I didn't think about hotplug...
I assume, all Xenstore bits wouldn't go the outside Xen grant
DMA-mapping layer (grant-dma-ops.c)?
>
> This doesn't preclude the device tree variant you are using, as this
> would be required for dom0less systems anyway.
good
>
> OTOH I'd like you to modify the commit message to make it more clear
> that in future frontend data might be written to Xenstore in order to
> support other use cases.
ok, will do
>
>> - the ring-ref/event-channel are not used for the backend<->frontend
>> communication, the proposed IPC for Virtio is IOREQ/DM
>> it is still a "block device" and ought to be integrated in existing
>> "disk" handling. So, re-use (and adapt) "disk" parsing/configuration
>> logic to deal with Virtio devices as well.
>>
>> For the immediate purpose and an ability to extend that support for
>> other use-cases in future (Qemu, virtio-pci, etc) perform the following
>> actions:
>> - Add new disk backend type (LIBXL_DISK_BACKEND_OTHER) and reflect
>> that in the configuration
>> - Introduce new disk protocol field to libxl_device_disk struct
>> (with LIBXL_DISK_PROTOCOL_XEN and LIBXL_DISK_PROTOCOL_VIRTIO_MMIO
>> types) and reflect that in the configuration (new "protocol" option
>> with "xen" protocol being default value)
>
> And with the hotplug option in mind I start to feel unueasy with naming
> the new Xenstore node "protocol", as the frontend disk nodes for "normal"
> disks already have a "protocol" entry specifying 64- or 32-bit protocol.
I noticed the "protocol" node at the frontend side for traditional Xen
PV block device which handles yet another purpose, but I didn't think
much about it
since the new "protocol" node in only for the backend's use. If we start
thinking of frontend's Xenstore nodes, then yes, will clash...
>
>
> Maybe we should really name it "transport" instead?
... For me the "transport" name is associated with virtio transports:
mmio, pci, ccw. But, I would be ok with that name. Another possible name
could be "specification".
>
>
>> - Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
>> one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
>>
>> An example of domain configuration for Virtio disk:
>> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other,
>> protocol=virtio-mmio']
>
> With Roger's feedback this would then be "transport=virtio", the "mmio"
> part should then be something like "adapter=mmio" (in contrast to
> "adapter=pci"), and "adapter" only needed in case of a device tree and
> PCI being available.
ok, will rename. Can we add "adapter" (or whenever the name would be)
option later, when there is a real need? For now, I mean within the
current series which adds only virtio-mmio bits on Arm, we can assume
that "transport=virtio" implies using virtio-mmio.
BTW, if we named the main option "specification", the secondary option
"transport" would good fit from my PoV.
For example:
disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other,
specification=virtio, transport=mmio']
But, I am perfectly fine with "transport/adapter".
>
>>
>> Nothing has changed for default Xen disk configuration.
>>
>> Please note, this patch is not enough for virtio-disk to work
>> on Xen (Arm), as for every Virtio device (including disk) we need
>> to allocate Virtio MMIO params (IRQ and memory region) and pass
>> them to the backend, also update Guest device-tree. The subsequent
>> patch will add these missing bits. For the current patch,
>> the default "irq" and "base" are just written to the Xenstore.
>> This is not an ideal splitting, but this way we avoid breaking
>> the bisectability.
>>
>> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>
> I'm fine with the overall approach and couldn't spot any real issues
> in the code.
Thank you.
[1]
https://lore.kernel.org/xen-devel/1650646263-22047-1-git-send-email-olekstysh@gmail.com/
[2]
https://lore.kernel.org/xen-devel/1650646263-22047-5-git-send-email-olekstysh@gmail.com/
https://lore.kernel.org/xen-devel/1649964960-24864-1-git-send-email-olekstysh@gmail.com/
>
>
>
> Juergen
>
> [1]:
> https://github.com/oasis-tcs/virtio-spec/blob/master/virtio-iommu.tex
> [2]: https://medium.com/@michael2012zhao_67085/virtio-iommu-789369049443
--
Regards,
Oleksandr Tyshchenko
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-25 12:02 ` Oleksandr
@ 2022-04-25 12:09 ` Juergen Gross
2022-04-25 12:59 ` Oleksandr
0 siblings, 1 reply; 23+ messages in thread
From: Juergen Gross @ 2022-04-25 12:09 UTC (permalink / raw)
To: Oleksandr
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Stefano Stabellini, Julien Grall,
Volodymyr Babchuk, Bertrand Marquis
[-- Attachment #1.1.1: Type: text/plain, Size: 4333 bytes --]
On 25.04.22 14:02, Oleksandr wrote:
>
> On 25.04.22 10:43, Juergen Gross wrote:
>
>
> Hello Juergen
>
>
> Thank you for the feedback.
>
>> On 08.04.22 20:21, Oleksandr Tyshchenko wrote:
>>> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>>>
>>> This patch adds basic support for configuring and assisting virtio-mmio
>>> based virtio-disk backend (emualator) which is intended to run out of
>>> Qemu and could be run in any domain.
>>> Although the Virtio block device is quite different from traditional
>>> Xen PV block device (vbd) from the toolstack's point of view:
>>> - as the frontend is virtio-blk which is not a Xenbus driver, nothing
>>> written to Xenstore are fetched by the frontend (the vdev is not
>>> passed to the frontend)
>>
>> I thought about the future support on x86.
>>
>> There we don't have a device tree (and I don't want to introduce it),
>> so the only ways to specify the backend domain id would be to:
>>
>> - add some information to ACPI tables
>> - use boot parameters
>> - use Xenstore
>
> I understand that, and agree
>
>
>> Thinking further of hotplugging virtio devices, Xenstore seems to be the
>> only real suitable alternative. Using virtio mechanisms doesn't seem
>> appropriate, as such information should be retrieved in "platform
>> specific" ways (see e.g. specifying an "endpoint" in the virtio IOMMU
>> device [1], [2]). I think the Xenstore information for that purpose
>> could be rather minimal and it should be device-type agnostic. Having
>> just a directory with endpoints and associated backend domids would
>> probably be enough (not needed in this series, of course).
>
> Just to make it clear, we are speaking about the possible ways to communicate
> backend domid for another series [1], so about the x86's alternative of
> device-tree bindings "xen,dev-domid" [2].
> I was thinking we could avoid using Xenstore at the guest side for that purpose,
> but I didn't think about hotplug...
> I assume, all Xenstore bits wouldn't go the outside Xen grant DMA-mapping layer
> (grant-dma-ops.c)?
I think it would be another driver under drivers/xen/ without the need to
touch any other frontend related file or Xen-related architecture specific
code.
>> And with the hotplug option in mind I start to feel unueasy with naming
>> the new Xenstore node "protocol", as the frontend disk nodes for "normal"
>> disks already have a "protocol" entry specifying 64- or 32-bit protocol.
>
>
> I noticed the "protocol" node at the frontend side for traditional Xen PV block
> device which handles yet another purpose, but I didn't think much about it
> since the new "protocol" node in only for the backend's use. If we start
> thinking of frontend's Xenstore nodes, then yes, will clash...
>
>>
>>
>> Maybe we should really name it "transport" instead?
>
> ... For me the "transport" name is associated with virtio transports: mmio, pci,
> ccw. But, I would be ok with that name. Another possible name could be
> "specification".
Yeah, looking at the virtio spec this makes sense.
So I would be fine with "specification".
>>> - Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
>>> one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
>>>
>>> An example of domain configuration for Virtio disk:
>>> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other, protocol=virtio-mmio']
>>
>> With Roger's feedback this would then be "transport=virtio", the "mmio"
>> part should then be something like "adapter=mmio" (in contrast to
>> "adapter=pci"), and "adapter" only needed in case of a device tree and
>> PCI being available.
>
> ok, will rename. Can we add "adapter" (or whenever the name would be) option
> later, when there is a real need? For now, I mean within the current series
> which adds only virtio-mmio bits on Arm, we can assume that "transport=virtio"
> implies using virtio-mmio.
Yes, we should add it only when needed.
>
> BTW, if we named the main option "specification", the secondary option
> "transport" would good fit from my PoV.
> For example:
> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other, specification=virtio,
> transport=mmio']
Fine with me.
Juergen
[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 3149 bytes --]
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 495 bytes --]
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-25 12:09 ` Juergen Gross
@ 2022-04-25 12:59 ` Oleksandr
0 siblings, 0 replies; 23+ messages in thread
From: Oleksandr @ 2022-04-25 12:59 UTC (permalink / raw)
To: Juergen Gross
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Stefano Stabellini, Julien Grall,
Volodymyr Babchuk, Bertrand Marquis
On 25.04.22 15:09, Juergen Gross wrote:
Hello Juergen
> On 25.04.22 14:02, Oleksandr wrote:
>>
>> On 25.04.22 10:43, Juergen Gross wrote:
>>
>>
>> Hello Juergen
>>
>>
>> Thank you for the feedback.
>>
>>> On 08.04.22 20:21, Oleksandr Tyshchenko wrote:
>>>> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>>>>
>>>> This patch adds basic support for configuring and assisting
>>>> virtio-mmio
>>>> based virtio-disk backend (emualator) which is intended to run out of
>>>> Qemu and could be run in any domain.
>>>> Although the Virtio block device is quite different from traditional
>>>> Xen PV block device (vbd) from the toolstack's point of view:
>>>> - as the frontend is virtio-blk which is not a Xenbus driver,
>>>> nothing
>>>> written to Xenstore are fetched by the frontend (the vdev is not
>>>> passed to the frontend)
>>>
>>> I thought about the future support on x86.
>>>
>>> There we don't have a device tree (and I don't want to introduce it),
>>> so the only ways to specify the backend domain id would be to:
>>>
>>> - add some information to ACPI tables
>>> - use boot parameters
>>> - use Xenstore
>>
>> I understand that, and agree
>>
>>
>>> Thinking further of hotplugging virtio devices, Xenstore seems to be
>>> the
>>> only real suitable alternative. Using virtio mechanisms doesn't seem
>>> appropriate, as such information should be retrieved in "platform
>>> specific" ways (see e.g. specifying an "endpoint" in the virtio IOMMU
>>> device [1], [2]). I think the Xenstore information for that purpose
>>> could be rather minimal and it should be device-type agnostic. Having
>>> just a directory with endpoints and associated backend domids would
>>> probably be enough (not needed in this series, of course).
>>
>> Just to make it clear, we are speaking about the possible ways to
>> communicate backend domid for another series [1], so about the x86's
>> alternative of device-tree bindings "xen,dev-domid" [2].
>> I was thinking we could avoid using Xenstore at the guest side for
>> that purpose, but I didn't think about hotplug...
>> I assume, all Xenstore bits wouldn't go the outside Xen grant
>> DMA-mapping layer (grant-dma-ops.c)?
>
> I think it would be another driver under drivers/xen/ without the need to
> touch any other frontend related file or Xen-related architecture
> specific
> code.
ok
>
>
>>> And with the hotplug option in mind I start to feel unueasy with naming
>>> the new Xenstore node "protocol", as the frontend disk nodes for
>>> "normal"
>>> disks already have a "protocol" entry specifying 64- or 32-bit
>>> protocol.
>>
>>
>> I noticed the "protocol" node at the frontend side for traditional
>> Xen PV block device which handles yet another purpose, but I didn't
>> think much about it
>> since the new "protocol" node in only for the backend's use. If we
>> start thinking of frontend's Xenstore nodes, then yes, will clash...
>>
>>>
>>>
>>> Maybe we should really name it "transport" instead?
>>
>> ... For me the "transport" name is associated with virtio transports:
>> mmio, pci, ccw. But, I would be ok with that name. Another possible
>> name could be "specification".
>
> Yeah, looking at the virtio spec this makes sense.
>
> So I would be fine with "specification".
ok, thank you for the confirmation.
>
>>>> - Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
>>>> one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
>>>>
>>>> An example of domain configuration for Virtio disk:
>>>> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other,
>>>> protocol=virtio-mmio']
>>>
>>> With Roger's feedback this would then be "transport=virtio", the "mmio"
>>> part should then be something like "adapter=mmio" (in contrast to
>>> "adapter=pci"), and "adapter" only needed in case of a device tree and
>>> PCI being available.
>>
>> ok, will rename. Can we add "adapter" (or whenever the name would be)
>> option later, when there is a real need? For now, I mean within the
>> current series which adds only virtio-mmio bits on Arm, we can assume
>> that "transport=virtio" implies using virtio-mmio.
>
> Yes, we should add it only when needed.
ok, good
>
>>
>> BTW, if we named the main option "specification", the secondary
>> option "transport" would good fit from my PoV.
>> For example:
>> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other,
>> specification=virtio, transport=mmio']
>
> Fine with me.
perfect
>
>
> Juergen
--
Regards,
Oleksandr Tyshchenko
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-23 7:39 ` Oleksandr
@ 2022-04-25 13:12 ` Roger Pau Monné
2022-04-27 8:20 ` Oleksandr
0 siblings, 1 reply; 23+ messages in thread
From: Roger Pau Monné @ 2022-04-25 13:12 UTC (permalink / raw)
To: Oleksandr
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Juergen Gross, Stefano Stabellini,
Julien Grall, Volodymyr Babchuk, Bertrand Marquis
On Sat, Apr 23, 2022 at 10:39:14AM +0300, Oleksandr wrote:
>
> On 22.04.22 12:41, Roger Pau Monné wrote:
>
>
> Hello Roger
>
> > On Fri, Apr 08, 2022 at 09:21:04PM +0300, Oleksandr Tyshchenko wrote:
> > > From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> > >
> > > This patch adds basic support for configuring and assisting virtio-mmio
> > > based virtio-disk backend (emualator) which is intended to run out of
> > > Qemu and could be run in any domain.
> > > Although the Virtio block device is quite different from traditional
> > > Xen PV block device (vbd) from the toolstack's point of view:
> > > - as the frontend is virtio-blk which is not a Xenbus driver, nothing
> > > written to Xenstore are fetched by the frontend (the vdev is not
> > > passed to the frontend)
> > > - the ring-ref/event-channel are not used for the backend<->frontend
> > > communication, the proposed IPC for Virtio is IOREQ/DM
> > > it is still a "block device" and ought to be integrated in existing
> > > "disk" handling. So, re-use (and adapt) "disk" parsing/configuration
> > > logic to deal with Virtio devices as well.
> > >
> > > For the immediate purpose and an ability to extend that support for
> > > other use-cases in future (Qemu, virtio-pci, etc) perform the following
> > > actions:
> > > - Add new disk backend type (LIBXL_DISK_BACKEND_OTHER) and reflect
> > > that in the configuration
> > > - Introduce new disk protocol field to libxl_device_disk struct
> > > (with LIBXL_DISK_PROTOCOL_XEN and LIBXL_DISK_PROTOCOL_VIRTIO_MMIO
> > > types) and reflect that in the configuration (new "protocol" option
> > > with "xen" protocol being default value)
> > > - Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
> > > one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
> > >
> > > An example of domain configuration for Virtio disk:
> > > disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other, protocol=virtio-mmio']
> > >
> > > Nothing has changed for default Xen disk configuration.
> > >
> > > Please note, this patch is not enough for virtio-disk to work
> > > on Xen (Arm), as for every Virtio device (including disk) we need
> > > to allocate Virtio MMIO params (IRQ and memory region) and pass
> > > them to the backend, also update Guest device-tree. The subsequent
> > > patch will add these missing bits. For the current patch,
> > > the default "irq" and "base" are just written to the Xenstore.
> > > This is not an ideal splitting, but this way we avoid breaking
> > > the bisectability.
> > >
> > > Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> > > ---
> > > Changes RFC -> V1:
> > > - no changes
> > >
> > > Changes V1 -> V2:
> > > - rebase according to the new location of libxl_virtio_disk.c
> > >
> > > Changes V2 -> V3:
> > > - no changes
> > >
> > > Changes V3 -> V4:
> > > - rebase according to the new argument for DEFINE_DEVICE_TYPE_STRUCT
> > >
> > > Changes V4 -> V5:
> > > - split the changes, change the order of the patches
> > > - update patch description
> > > - don't introduce new "vdisk" configuration option with own parsing logic,
> > > re-use Xen PV block "disk" parsing/configuration logic for the virtio-disk
> > > - introduce "virtio" flag and document it's usage
> > > - add LIBXL_HAVE_DEVICE_DISK_VIRTIO
> > > - update libxlu_disk_l.[ch]
> > > - drop num_disks variable/MAX_VIRTIO_DISKS
> > > - drop Wei's T-b
> > >
> > > Changes V5 -> V6:
> > > - rebase on current staging
> > > - use "%"PRIu64 instead of %lu for disk->base in device_disk_add()
> > > - update *.gen.go files
> > >
> > > Changes V6 -> V7:
> > > - rebase on current staging
> > > - update *.gen.go files and libxlu_disk_l.[ch] files
> > > - update patch description
> > > - rework significantly to support more flexible configuration
> > > and have more generic basic implementation for being able to extend
> > > that for other use-cases (virtio-pci, qemu, etc).
> > > ---
> > > docs/man/xl-disk-configuration.5.pod.in | 37 +-
> > > tools/golang/xenlight/helpers.gen.go | 6 +
> > > tools/golang/xenlight/types.gen.go | 11 +
> > > tools/include/libxl.h | 6 +
> > > tools/libs/light/libxl_device.c | 57 +-
> > > tools/libs/light/libxl_disk.c | 111 +++-
> > > tools/libs/light/libxl_internal.h | 1 +
> > > tools/libs/light/libxl_types.idl | 10 +
> > > tools/libs/light/libxl_types_internal.idl | 1 +
> > > tools/libs/light/libxl_utils.c | 2 +
> > > tools/libs/util/libxlu_disk_l.c | 952 +++++++++++++++---------------
> > > tools/libs/util/libxlu_disk_l.h | 2 +-
> > > tools/libs/util/libxlu_disk_l.l | 9 +
> > > tools/xl/xl_block.c | 11 +
> > > 14 files changed, 736 insertions(+), 480 deletions(-)
> > >
> > > diff --git a/docs/man/xl-disk-configuration.5.pod.in b/docs/man/xl-disk-configuration.5.pod.in
> > > index 71d0e86..36c851f 100644
> > > --- a/docs/man/xl-disk-configuration.5.pod.in
> > > +++ b/docs/man/xl-disk-configuration.5.pod.in
> > > @@ -232,7 +232,7 @@ Specifies the backend implementation to use
> > > =item Supported values
> > > -phy, qdisk
> > > +phy, qdisk, other
> > > =item Mandatory
> > > @@ -244,11 +244,13 @@ Automatically determine which backend to use.
> > > =back
> > > -This does not affect the guest's view of the device. It controls
> > > -which software implementation of the Xen backend driver is used.
> > > +It controls which software implementation of the backend driver is used.
> > > +Depending on the "protocol" option this may affect the guest's view
> > > +of the device.
> > > Not all backend drivers support all combinations of other options.
> > > -For example, "phy" does not support formats other than "raw".
> > > +For example, "phy" and "other" do not support formats other than "raw" and
> > > +"other" does not support protocols other than "virtio-mmio".
> > > Normally this option should not be specified, in which case libxl will
> > > automatically determine the most suitable backend.
> > > @@ -344,8 +346,35 @@ can be used to disable "hole punching" for file based backends which
> > > were intentionally created non-sparse to avoid fragmentation of the
> > > file.
> > > +=item B<protocol>=I<PROTOCOL>
> > > +
> > > +=over 4
> > > +
> > > +=item Description
> > > +
> > > +Specifies the communication protocol to use for the chosen "backendtype" option
> > > +
> > > +=item Supported values
> > > +
> > > +xen, virtio-mmio
> > From a user PoV, I think it would be better to just select xen or
> > virtio here, but not the underlying configuration mechanism used to
> > expose the devices to the guest.
>
> I got your point.
>
>
>
> >
> > We would likely need to add a different option to select mmio or pci
> > then, but that should be set by default based on architecture/guest
> > type. For example on x86 it should default to pci, while on Arm I
> > guess it will depend on whether the guest has PCI or not?
> >
> > In any case, I think we should offer an option that's selecting
> > between xen or virtio protocol, and the way to expose the
> > configuration of the device shouldn't need to be explicitly selected
> > by the user.
>
>
> ok, for now I will use "xen and virtio" values for the "protocol" option,
> then internally toolstack will assume that "virtio" really means
> "virtio-mmio".
>
> When there is a need to expand that support to "virtio-pci", we will see how
> to deal with it from the configuration PoV, probably like you suggested
> above by adding another option (e.g. "transport") with default values based
> on the architecture/guest type.
I think this likely also wants to be a separate field in libxl_device_disk,
which could be left empty and libxl will attempt to set a default.
For example have the following in libxl_types.idl:
libxl_device_protocol = Enumeration("device_protocol", [
(0, "UNKNOWN"),
(1, "XEN"),
(2, "VIRTIO"),
])
libxl_device_configuration = Enumeration("device_configuration", [
(0, "UNKNOWN"),
(1, "XENBUS"),
(2, "MMIO"),
])
libxl_device_disk = Struct("device_disk", [
("protocol", libxl_device_protocol),
("configuration", libxl_device_configuration),
])
I don't like libxl_device_configuration much, I think is too generic,
but I can't think of anything better. Maybe others can provide better
names.
Thanks, Roger.
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-25 13:12 ` Roger Pau Monné
@ 2022-04-27 8:20 ` Oleksandr
2022-04-27 8:27 ` Roger Pau Monné
0 siblings, 1 reply; 23+ messages in thread
From: Oleksandr @ 2022-04-27 8:20 UTC (permalink / raw)
To: Roger Pau Monné
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Juergen Gross, Stefano Stabellini,
Julien Grall, Volodymyr Babchuk, Bertrand Marquis
On 25.04.22 16:12, Roger Pau Monné wrote:
Hello Roger
> On Sat, Apr 23, 2022 at 10:39:14AM +0300, Oleksandr wrote:
>> On 22.04.22 12:41, Roger Pau Monné wrote:
>>
>>
>> Hello Roger
>>
>>> On Fri, Apr 08, 2022 at 09:21:04PM +0300, Oleksandr Tyshchenko wrote:
>>>> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>>>>
>>>> This patch adds basic support for configuring and assisting virtio-mmio
>>>> based virtio-disk backend (emualator) which is intended to run out of
>>>> Qemu and could be run in any domain.
>>>> Although the Virtio block device is quite different from traditional
>>>> Xen PV block device (vbd) from the toolstack's point of view:
>>>> - as the frontend is virtio-blk which is not a Xenbus driver, nothing
>>>> written to Xenstore are fetched by the frontend (the vdev is not
>>>> passed to the frontend)
>>>> - the ring-ref/event-channel are not used for the backend<->frontend
>>>> communication, the proposed IPC for Virtio is IOREQ/DM
>>>> it is still a "block device" and ought to be integrated in existing
>>>> "disk" handling. So, re-use (and adapt) "disk" parsing/configuration
>>>> logic to deal with Virtio devices as well.
>>>>
>>>> For the immediate purpose and an ability to extend that support for
>>>> other use-cases in future (Qemu, virtio-pci, etc) perform the following
>>>> actions:
>>>> - Add new disk backend type (LIBXL_DISK_BACKEND_OTHER) and reflect
>>>> that in the configuration
>>>> - Introduce new disk protocol field to libxl_device_disk struct
>>>> (with LIBXL_DISK_PROTOCOL_XEN and LIBXL_DISK_PROTOCOL_VIRTIO_MMIO
>>>> types) and reflect that in the configuration (new "protocol" option
>>>> with "xen" protocol being default value)
>>>> - Add new device kind (LIBXL__DEVICE_KIND_VIRTIO_DISK) as current
>>>> one (LIBXL__DEVICE_KIND_VBD) doesn't fit into Virtio disk model
>>>>
>>>> An example of domain configuration for Virtio disk:
>>>> disk = [ 'phy:/dev/mmcblk0p3, xvda1, backendtype=other, protocol=virtio-mmio']
>>>>
>>>> Nothing has changed for default Xen disk configuration.
>>>>
>>>> Please note, this patch is not enough for virtio-disk to work
>>>> on Xen (Arm), as for every Virtio device (including disk) we need
>>>> to allocate Virtio MMIO params (IRQ and memory region) and pass
>>>> them to the backend, also update Guest device-tree. The subsequent
>>>> patch will add these missing bits. For the current patch,
>>>> the default "irq" and "base" are just written to the Xenstore.
>>>> This is not an ideal splitting, but this way we avoid breaking
>>>> the bisectability.
>>>>
>>>> Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>>>> ---
>>>> Changes RFC -> V1:
>>>> - no changes
>>>>
>>>> Changes V1 -> V2:
>>>> - rebase according to the new location of libxl_virtio_disk.c
>>>>
>>>> Changes V2 -> V3:
>>>> - no changes
>>>>
>>>> Changes V3 -> V4:
>>>> - rebase according to the new argument for DEFINE_DEVICE_TYPE_STRUCT
>>>>
>>>> Changes V4 -> V5:
>>>> - split the changes, change the order of the patches
>>>> - update patch description
>>>> - don't introduce new "vdisk" configuration option with own parsing logic,
>>>> re-use Xen PV block "disk" parsing/configuration logic for the virtio-disk
>>>> - introduce "virtio" flag and document it's usage
>>>> - add LIBXL_HAVE_DEVICE_DISK_VIRTIO
>>>> - update libxlu_disk_l.[ch]
>>>> - drop num_disks variable/MAX_VIRTIO_DISKS
>>>> - drop Wei's T-b
>>>>
>>>> Changes V5 -> V6:
>>>> - rebase on current staging
>>>> - use "%"PRIu64 instead of %lu for disk->base in device_disk_add()
>>>> - update *.gen.go files
>>>>
>>>> Changes V6 -> V7:
>>>> - rebase on current staging
>>>> - update *.gen.go files and libxlu_disk_l.[ch] files
>>>> - update patch description
>>>> - rework significantly to support more flexible configuration
>>>> and have more generic basic implementation for being able to extend
>>>> that for other use-cases (virtio-pci, qemu, etc).
>>>> ---
>>>> docs/man/xl-disk-configuration.5.pod.in | 37 +-
>>>> tools/golang/xenlight/helpers.gen.go | 6 +
>>>> tools/golang/xenlight/types.gen.go | 11 +
>>>> tools/include/libxl.h | 6 +
>>>> tools/libs/light/libxl_device.c | 57 +-
>>>> tools/libs/light/libxl_disk.c | 111 +++-
>>>> tools/libs/light/libxl_internal.h | 1 +
>>>> tools/libs/light/libxl_types.idl | 10 +
>>>> tools/libs/light/libxl_types_internal.idl | 1 +
>>>> tools/libs/light/libxl_utils.c | 2 +
>>>> tools/libs/util/libxlu_disk_l.c | 952 +++++++++++++++---------------
>>>> tools/libs/util/libxlu_disk_l.h | 2 +-
>>>> tools/libs/util/libxlu_disk_l.l | 9 +
>>>> tools/xl/xl_block.c | 11 +
>>>> 14 files changed, 736 insertions(+), 480 deletions(-)
>>>>
>>>> diff --git a/docs/man/xl-disk-configuration.5.pod.in b/docs/man/xl-disk-configuration.5.pod.in
>>>> index 71d0e86..36c851f 100644
>>>> --- a/docs/man/xl-disk-configuration.5.pod.in
>>>> +++ b/docs/man/xl-disk-configuration.5.pod.in
>>>> @@ -232,7 +232,7 @@ Specifies the backend implementation to use
>>>> =item Supported values
>>>> -phy, qdisk
>>>> +phy, qdisk, other
>>>> =item Mandatory
>>>> @@ -244,11 +244,13 @@ Automatically determine which backend to use.
>>>> =back
>>>> -This does not affect the guest's view of the device. It controls
>>>> -which software implementation of the Xen backend driver is used.
>>>> +It controls which software implementation of the backend driver is used.
>>>> +Depending on the "protocol" option this may affect the guest's view
>>>> +of the device.
>>>> Not all backend drivers support all combinations of other options.
>>>> -For example, "phy" does not support formats other than "raw".
>>>> +For example, "phy" and "other" do not support formats other than "raw" and
>>>> +"other" does not support protocols other than "virtio-mmio".
>>>> Normally this option should not be specified, in which case libxl will
>>>> automatically determine the most suitable backend.
>>>> @@ -344,8 +346,35 @@ can be used to disable "hole punching" for file based backends which
>>>> were intentionally created non-sparse to avoid fragmentation of the
>>>> file.
>>>> +=item B<protocol>=I<PROTOCOL>
>>>> +
>>>> +=over 4
>>>> +
>>>> +=item Description
>>>> +
>>>> +Specifies the communication protocol to use for the chosen "backendtype" option
>>>> +
>>>> +=item Supported values
>>>> +
>>>> +xen, virtio-mmio
>>> From a user PoV, I think it would be better to just select xen or
>>> virtio here, but not the underlying configuration mechanism used to
>>> expose the devices to the guest.
>> I got your point.
>>
>>
>>
>>> We would likely need to add a different option to select mmio or pci
>>> then, but that should be set by default based on architecture/guest
>>> type. For example on x86 it should default to pci, while on Arm I
>>> guess it will depend on whether the guest has PCI or not?
>>>
>>> In any case, I think we should offer an option that's selecting
>>> between xen or virtio protocol, and the way to expose the
>>> configuration of the device shouldn't need to be explicitly selected
>>> by the user.
>>
>> ok, for now I will use "xen and virtio" values for the "protocol" option,
>> then internally toolstack will assume that "virtio" really means
>> "virtio-mmio".
>>
>> When there is a need to expand that support to "virtio-pci", we will see how
>> to deal with it from the configuration PoV, probably like you suggested
>> above by adding another option (e.g. "transport") with default values based
>> on the architecture/guest type.
> I think this likely also wants to be a separate field in libxl_device_disk,
> which could be left empty and libxl will attempt to set a default.
> For example have the following in libxl_types.idl:
>
> libxl_device_protocol = Enumeration("device_protocol", [
> (0, "UNKNOWN"),
> (1, "XEN"),
> (2, "VIRTIO"),
> ])
>
> libxl_device_configuration = Enumeration("device_configuration", [
> (0, "UNKNOWN"),
> (1, "XENBUS"),
> (2, "MMIO"),
> ])
>
> libxl_device_disk = Struct("device_disk", [
> ("protocol", libxl_device_protocol),
> ("configuration", libxl_device_configuration),
> ])
>
> I don't like libxl_device_configuration much, I think is too generic,
> but I can't think of anything better. Maybe others can provide better
> names.
Here [1] we had a discussion regarding user configuration options.
"protocol" cannot be used as it will lead to the confusion (at least
with Xen PV block device which already has "protocol" frontend's entry
in Xenstore).
Preliminary, we had agreed on the following names:
1. specification: xen or virtio
2. transport: mmio or pci
Please tell me, are you OK with them?
[1]
https://lore.kernel.org/xen-devel/e657458d-d33e-a340-d9fe-152ec97eefec@suse.com/
>
> Thanks, Roger.
--
Regards,
Oleksandr Tyshchenko
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-27 8:20 ` Oleksandr
@ 2022-04-27 8:27 ` Roger Pau Monné
2022-04-27 8:36 ` Juergen Gross
0 siblings, 1 reply; 23+ messages in thread
From: Roger Pau Monné @ 2022-04-27 8:27 UTC (permalink / raw)
To: Oleksandr
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Juergen Gross, Stefano Stabellini,
Julien Grall, Volodymyr Babchuk, Bertrand Marquis
On Wed, Apr 27, 2022 at 11:20:04AM +0300, Oleksandr wrote:
>
> On 25.04.22 16:12, Roger Pau Monné wrote:
>
>
> Hello Roger
>
>
> > On Sat, Apr 23, 2022 at 10:39:14AM +0300, Oleksandr wrote:
> > > On 22.04.22 12:41, Roger Pau Monné wrote:
> > >
> > >
> > > Hello Roger
> > >
> > > > On Fri, Apr 08, 2022 at 09:21:04PM +0300, Oleksandr Tyshchenko wrote:
> > > > > From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> > > > > --- a/docs/man/xl-disk-configuration.5.pod.in
> > > > > +++ b/docs/man/xl-disk-configuration.5.pod.in
> > > > > @@ -232,7 +232,7 @@ Specifies the backend implementation to use
> > > > > =item Supported values
> > > > > -phy, qdisk
> > > > > +phy, qdisk, other
> > > > > =item Mandatory
> > > > > @@ -244,11 +244,13 @@ Automatically determine which backend to use.
> > > > > =back
> > > > > -This does not affect the guest's view of the device. It controls
> > > > > -which software implementation of the Xen backend driver is used.
> > > > > +It controls which software implementation of the backend driver is used.
> > > > > +Depending on the "protocol" option this may affect the guest's view
> > > > > +of the device.
> > > > > Not all backend drivers support all combinations of other options.
> > > > > -For example, "phy" does not support formats other than "raw".
> > > > > +For example, "phy" and "other" do not support formats other than "raw" and
> > > > > +"other" does not support protocols other than "virtio-mmio".
> > > > > Normally this option should not be specified, in which case libxl will
> > > > > automatically determine the most suitable backend.
> > > > > @@ -344,8 +346,35 @@ can be used to disable "hole punching" for file based backends which
> > > > > were intentionally created non-sparse to avoid fragmentation of the
> > > > > file.
> > > > > +=item B<protocol>=I<PROTOCOL>
> > > > > +
> > > > > +=over 4
> > > > > +
> > > > > +=item Description
> > > > > +
> > > > > +Specifies the communication protocol to use for the chosen "backendtype" option
> > > > > +
> > > > > +=item Supported values
> > > > > +
> > > > > +xen, virtio-mmio
> > > > From a user PoV, I think it would be better to just select xen or
> > > > virtio here, but not the underlying configuration mechanism used to
> > > > expose the devices to the guest.
> > > I got your point.
> > >
> > >
> > >
> > > > We would likely need to add a different option to select mmio or pci
> > > > then, but that should be set by default based on architecture/guest
> > > > type. For example on x86 it should default to pci, while on Arm I
> > > > guess it will depend on whether the guest has PCI or not?
> > > >
> > > > In any case, I think we should offer an option that's selecting
> > > > between xen or virtio protocol, and the way to expose the
> > > > configuration of the device shouldn't need to be explicitly selected
> > > > by the user.
> > >
> > > ok, for now I will use "xen and virtio" values for the "protocol" option,
> > > then internally toolstack will assume that "virtio" really means
> > > "virtio-mmio".
> > >
> > > When there is a need to expand that support to "virtio-pci", we will see how
> > > to deal with it from the configuration PoV, probably like you suggested
> > > above by adding another option (e.g. "transport") with default values based
> > > on the architecture/guest type.
> > I think this likely also wants to be a separate field in libxl_device_disk,
> > which could be left empty and libxl will attempt to set a default.
> > For example have the following in libxl_types.idl:
> >
> > libxl_device_protocol = Enumeration("device_protocol", [
> > (0, "UNKNOWN"),
> > (1, "XEN"),
> > (2, "VIRTIO"),
> > ])
> >
> > libxl_device_configuration = Enumeration("device_configuration", [
> > (0, "UNKNOWN"),
> > (1, "XENBUS"),
> > (2, "MMIO"),
> > ])
> >
> > libxl_device_disk = Struct("device_disk", [
> > ("protocol", libxl_device_protocol),
> > ("configuration", libxl_device_configuration),
> > ])
> >
> > I don't like libxl_device_configuration much, I think is too generic,
> > but I can't think of anything better. Maybe others can provide better
> > names.
>
>
> Here [1] we had a discussion regarding user configuration options.
> "protocol" cannot be used as it will lead to the confusion (at least with
> Xen PV block device which already has "protocol" frontend's entry in
> Xenstore).
>
> Preliminary, we had agreed on the following names:
> 1. specification: xen or virtio
> 2. transport: mmio or pci
>
> Please tell me, are you OK with them?
Yes, that's fine. My main point is that libxl_device_disk should
contain both fields, so a 3rd party toolstack can set 'specification =
virtio' and let Xen decide the transport to use.
I'm dubious whether we want to have a xenbus transport for xen
specification to use, but I guess it's fine to say that for
specification == xen transport is ignored.
Thanks, Roger.
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH V7 1/2] libxl: Add support for Virtio disk configuration
2022-04-27 8:27 ` Roger Pau Monné
@ 2022-04-27 8:36 ` Juergen Gross
0 siblings, 0 replies; 23+ messages in thread
From: Juergen Gross @ 2022-04-27 8:36 UTC (permalink / raw)
To: Roger Pau Monné, Oleksandr
Cc: xen-devel, Oleksandr Tyshchenko, Wei Liu, Anthony PERARD,
George Dunlap, Nick Rosbrook, Stefano Stabellini, Julien Grall,
Volodymyr Babchuk, Bertrand Marquis
[-- Attachment #1.1.1: Type: text/plain, Size: 5129 bytes --]
On 27.04.22 10:27, Roger Pau Monné wrote:
> On Wed, Apr 27, 2022 at 11:20:04AM +0300, Oleksandr wrote:
>>
>> On 25.04.22 16:12, Roger Pau Monné wrote:
>>
>>
>> Hello Roger
>>
>>
>>> On Sat, Apr 23, 2022 at 10:39:14AM +0300, Oleksandr wrote:
>>>> On 22.04.22 12:41, Roger Pau Monné wrote:
>>>>
>>>>
>>>> Hello Roger
>>>>
>>>>> On Fri, Apr 08, 2022 at 09:21:04PM +0300, Oleksandr Tyshchenko wrote:
>>>>>> From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
>>>>>> --- a/docs/man/xl-disk-configuration.5.pod.in
>>>>>> +++ b/docs/man/xl-disk-configuration.5.pod.in
>>>>>> @@ -232,7 +232,7 @@ Specifies the backend implementation to use
>>>>>> =item Supported values
>>>>>> -phy, qdisk
>>>>>> +phy, qdisk, other
>>>>>> =item Mandatory
>>>>>> @@ -244,11 +244,13 @@ Automatically determine which backend to use.
>>>>>> =back
>>>>>> -This does not affect the guest's view of the device. It controls
>>>>>> -which software implementation of the Xen backend driver is used.
>>>>>> +It controls which software implementation of the backend driver is used.
>>>>>> +Depending on the "protocol" option this may affect the guest's view
>>>>>> +of the device.
>>>>>> Not all backend drivers support all combinations of other options.
>>>>>> -For example, "phy" does not support formats other than "raw".
>>>>>> +For example, "phy" and "other" do not support formats other than "raw" and
>>>>>> +"other" does not support protocols other than "virtio-mmio".
>>>>>> Normally this option should not be specified, in which case libxl will
>>>>>> automatically determine the most suitable backend.
>>>>>> @@ -344,8 +346,35 @@ can be used to disable "hole punching" for file based backends which
>>>>>> were intentionally created non-sparse to avoid fragmentation of the
>>>>>> file.
>>>>>> +=item B<protocol>=I<PROTOCOL>
>>>>>> +
>>>>>> +=over 4
>>>>>> +
>>>>>> +=item Description
>>>>>> +
>>>>>> +Specifies the communication protocol to use for the chosen "backendtype" option
>>>>>> +
>>>>>> +=item Supported values
>>>>>> +
>>>>>> +xen, virtio-mmio
>>>>> From a user PoV, I think it would be better to just select xen or
>>>>> virtio here, but not the underlying configuration mechanism used to
>>>>> expose the devices to the guest.
>>>> I got your point.
>>>>
>>>>
>>>>
>>>>> We would likely need to add a different option to select mmio or pci
>>>>> then, but that should be set by default based on architecture/guest
>>>>> type. For example on x86 it should default to pci, while on Arm I
>>>>> guess it will depend on whether the guest has PCI or not?
>>>>>
>>>>> In any case, I think we should offer an option that's selecting
>>>>> between xen or virtio protocol, and the way to expose the
>>>>> configuration of the device shouldn't need to be explicitly selected
>>>>> by the user.
>>>>
>>>> ok, for now I will use "xen and virtio" values for the "protocol" option,
>>>> then internally toolstack will assume that "virtio" really means
>>>> "virtio-mmio".
>>>>
>>>> When there is a need to expand that support to "virtio-pci", we will see how
>>>> to deal with it from the configuration PoV, probably like you suggested
>>>> above by adding another option (e.g. "transport") with default values based
>>>> on the architecture/guest type.
>>> I think this likely also wants to be a separate field in libxl_device_disk,
>>> which could be left empty and libxl will attempt to set a default.
>>> For example have the following in libxl_types.idl:
>>>
>>> libxl_device_protocol = Enumeration("device_protocol", [
>>> (0, "UNKNOWN"),
>>> (1, "XEN"),
>>> (2, "VIRTIO"),
>>> ])
>>>
>>> libxl_device_configuration = Enumeration("device_configuration", [
>>> (0, "UNKNOWN"),
>>> (1, "XENBUS"),
>>> (2, "MMIO"),
>>> ])
>>>
>>> libxl_device_disk = Struct("device_disk", [
>>> ("protocol", libxl_device_protocol),
>>> ("configuration", libxl_device_configuration),
>>> ])
>>>
>>> I don't like libxl_device_configuration much, I think is too generic,
>>> but I can't think of anything better. Maybe others can provide better
>>> names.
>>
>>
>> Here [1] we had a discussion regarding user configuration options.
>> "protocol" cannot be used as it will lead to the confusion (at least with
>> Xen PV block device which already has "protocol" frontend's entry in
>> Xenstore).
>>
>> Preliminary, we had agreed on the following names:
>> 1. specification: xen or virtio
>> 2. transport: mmio or pci
>>
>> Please tell me, are you OK with them?
>
> Yes, that's fine. My main point is that libxl_device_disk should
> contain both fields, so a 3rd party toolstack can set 'specification =
> virtio' and let Xen decide the transport to use.
>
> I'm dubious whether we want to have a xenbus transport for xen
> specification to use, but I guess it's fine to say that for
> specification == xen transport is ignored.
At least for now, I guess. It might be a thing in case someone decides
to add argo support.
Juergen
[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 3149 bytes --]
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 495 bytes --]
^ permalink raw reply [flat|nested] 23+ messages in thread
end of thread, other threads:[~2022-04-27 8:36 UTC | newest]
Thread overview: 23+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-04-08 18:21 [PATCH V7 0/2] Virtio support for toolstack on Arm (Was "IOREQ feature (+ virtio-mmio) on Arm") Oleksandr Tyshchenko
2022-04-08 18:21 ` [PATCH V7 1/2] libxl: Add support for Virtio disk configuration Oleksandr Tyshchenko
2022-04-18 21:38 ` Stefano Stabellini
2022-04-19 17:15 ` Oleksandr
2022-04-22 9:41 ` Roger Pau Monné
2022-04-23 7:39 ` Oleksandr
2022-04-25 13:12 ` Roger Pau Monné
2022-04-27 8:20 ` Oleksandr
2022-04-27 8:27 ` Roger Pau Monné
2022-04-27 8:36 ` Juergen Gross
2022-04-25 7:43 ` Juergen Gross
2022-04-25 12:02 ` Oleksandr
2022-04-25 12:09 ` Juergen Gross
2022-04-25 12:59 ` Oleksandr
2022-04-08 18:21 ` [PATCH V7 2/2] libxl: Introduce basic virtio-mmio support on Arm Oleksandr Tyshchenko
2022-04-18 21:41 ` Stefano Stabellini
2022-04-18 23:03 ` Julien Grall
2022-04-19 16:07 ` Oleksandr
2022-04-20 0:13 ` Stefano Stabellini
2022-04-20 5:20 ` Henry Wang
2022-04-20 9:07 ` Oleksandr
2022-04-22 2:42 ` Jiamei Xie
2022-04-22 10:34 ` Oleksandr Tyshchenko
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.