From: "Michael S. Tsirkin" <mst@redhat.com>
To: qemu-devel@nongnu.org
Cc: Peter Maydell <peter.maydell@linaro.org>,
Eduardo Habkost <ehabkost@redhat.com>,
Jiang Wang <jiang.wang@bytedance.com>,
qemu-stable@nongnu.org, Stefano Garzarella <sgarzare@redhat.com>
Subject: [PULL 13/57] vhost-vsock: fix migration issue when seqpacket is supported
Date: Tue, 5 Oct 2021 12:01:27 -0400 [thread overview]
Message-ID: <20211005155946.513818-14-mst@redhat.com> (raw)
In-Reply-To: <20211005155946.513818-1-mst@redhat.com>
From: Stefano Garzarella <sgarzare@redhat.com>
Commit 1e08fd0a46 ("vhost-vsock: SOCK_SEQPACKET feature bit support")
enabled the SEQPACKET feature bit.
This commit is released with QEMU 6.1, so if we try to migrate a VM where
the host kernel supports SEQPACKET but machine type version is less than
6.1, we get the following errors:
Features 0x130000002 unsupported. Allowed features: 0x179000000
Failed to load virtio-vhost_vsock:virtio
error while loading state for instance 0x0 of device '0000:00:05.0/virtio-vhost_vsock'
load of migration failed: Operation not permitted
Let's disable the feature bit for machine types < 6.1.
We add a new OnOffAuto property for this, called `seqpacket`.
When it is `auto` (default), QEMU behaves as before, trying to enable the
feature, when it is `on` QEMU will fail if the backend (vhost-vsock
kernel module) doesn't support it.
Fixes: 1e08fd0a46 ("vhost-vsock: SOCK_SEQPACKET feature bit support")
Cc: qemu-stable@nongnu.org
Reported-by: Jiang Wang <jiang.wang@bytedance.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20210921161642.206461-2-sgarzare@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
include/hw/virtio/vhost-vsock.h | 3 +++
hw/core/machine.c | 1 +
hw/virtio/vhost-vsock.c | 19 ++++++++++++++++---
3 files changed, 20 insertions(+), 3 deletions(-)
diff --git a/include/hw/virtio/vhost-vsock.h b/include/hw/virtio/vhost-vsock.h
index 84f4e727c7..3f121a624f 100644
--- a/include/hw/virtio/vhost-vsock.h
+++ b/include/hw/virtio/vhost-vsock.h
@@ -30,6 +30,9 @@ struct VHostVSock {
VHostVSockCommon parent;
VHostVSockConf conf;
+ /* features */
+ OnOffAuto seqpacket;
+
/*< public >*/
};
diff --git a/hw/core/machine.c b/hw/core/machine.c
index 3920a2f2af..74f2a9a984 100644
--- a/hw/core/machine.c
+++ b/hw/core/machine.c
@@ -46,6 +46,7 @@ GlobalProperty hw_compat_6_0[] = {
{ "nvme-ns", "eui64-default", "off"},
{ "e1000", "init-vet", "off" },
{ "e1000e", "init-vet", "off" },
+ { "vhost-vsock-device", "seqpacket", "off" },
};
const size_t hw_compat_6_0_len = G_N_ELEMENTS(hw_compat_6_0);
diff --git a/hw/virtio/vhost-vsock.c b/hw/virtio/vhost-vsock.c
index 1b1a5c70ed..dade0da031 100644
--- a/hw/virtio/vhost-vsock.c
+++ b/hw/virtio/vhost-vsock.c
@@ -114,10 +114,21 @@ static uint64_t vhost_vsock_get_features(VirtIODevice *vdev,
Error **errp)
{
VHostVSockCommon *vvc = VHOST_VSOCK_COMMON(vdev);
+ VHostVSock *vsock = VHOST_VSOCK(vdev);
- virtio_add_feature(&requested_features, VIRTIO_VSOCK_F_SEQPACKET);
- return vhost_get_features(&vvc->vhost_dev, feature_bits,
- requested_features);
+ if (vsock->seqpacket != ON_OFF_AUTO_OFF) {
+ virtio_add_feature(&requested_features, VIRTIO_VSOCK_F_SEQPACKET);
+ }
+
+ requested_features = vhost_get_features(&vvc->vhost_dev, feature_bits,
+ requested_features);
+
+ if (vsock->seqpacket == ON_OFF_AUTO_ON &&
+ !virtio_has_feature(requested_features, VIRTIO_VSOCK_F_SEQPACKET)) {
+ error_setg(errp, "vhost-vsock backend doesn't support seqpacket");
+ }
+
+ return requested_features;
}
static const VMStateDescription vmstate_virtio_vhost_vsock = {
@@ -218,6 +229,8 @@ static void vhost_vsock_device_unrealize(DeviceState *dev)
static Property vhost_vsock_properties[] = {
DEFINE_PROP_UINT64("guest-cid", VHostVSock, conf.guest_cid, 0),
DEFINE_PROP_STRING("vhostfd", VHostVSock, conf.vhostfd),
+ DEFINE_PROP_ON_OFF_AUTO("seqpacket", VHostVSock, seqpacket,
+ ON_OFF_AUTO_AUTO),
DEFINE_PROP_END_OF_LIST(),
};
--
MST
next prev parent reply other threads:[~2021-10-05 16:37 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-05 16:00 [PULL 00/57] pc,pci,virtio: features, fixes Michael S. Tsirkin
2021-10-05 16:00 ` [PULL 01/57] hw/virtio: Acquire RCU read lock in virtqueue_packed_drop_all() Michael S. Tsirkin
2021-10-05 16:00 ` [PULL 02/57] hw/virtio: Have virtqueue_get_avail_bytes() pass caches arg to callees Michael S. Tsirkin
2021-10-05 16:00 ` [PULL 03/57] vhost-vdpa: open device fd in net_init_vhost_vdpa() Michael S. Tsirkin
2021-10-05 16:00 ` [PULL 04/57] vhost-vdpa: classify one time request Michael S. Tsirkin
2021-10-05 16:00 ` [PULL 05/57] vhost-vdpa: prepare for the multiqueue support Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 06/57] vhost-vdpa: let net_vhost_vdpa_init() returns NetClientState * Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 07/57] net: introduce control client Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 08/57] vhost-net: control virtqueue support Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 09/57] virtio-net: use "queue_pairs" instead of "queues" when possible Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 10/57] vhost: record the last virtqueue index for the virtio device Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 11/57] virtio-net: vhost control virtqueue support Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 12/57] vhost-vdpa: multiqueue support Michael S. Tsirkin
2021-10-05 16:01 ` Michael S. Tsirkin [this message]
2021-10-05 16:01 ` [PULL 14/57] vhost-vsock: handle common features in vhost-vsock-common Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 15/57] acpi: add helper routines to initialize ACPI tables Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 16/57] acpi: build_rsdt: use acpi_table_begin()/acpi_table_end() instead of build_header() Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 17/57] acpi: build_xsdt: " Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 18/57] acpi: build_slit: " Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 19/57] acpi: build_fadt: " Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 20/57] acpi: build_tpm2: " Michael S. Tsirkin
2021-10-05 16:01 ` [PULL 21/57] acpi: acpi_build_hest: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 22/57] acpi: build_mcfg: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 23/57] acpi: build_hmat: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 24/57] acpi: nvdimm_build_nfit: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 25/57] acpi: nvdimm_build_ssdt: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 26/57] acpi: vmgenid_build_acpi: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 27/57] acpi: x86: build_dsdt: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 28/57] acpi: build_hpet: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 29/57] acpi: build_tpm_tcpa: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 30/57] acpi: arm/x86: build_srat: " Michael S. Tsirkin
2021-10-05 16:02 ` [PULL 31/57] acpi: use build_append_int_noprefix() API to compose SRAT table Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 32/57] acpi: build_dmar_q35: use acpi_table_begin()/acpi_table_end() instead of build_header() Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 33/57] acpi: build_waet: " Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 34/57] acpi: build_amd_iommu: " Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 35/57] acpi: madt: arm/x86: " Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 36/57] acpi: x86: remove dead code Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 37/57] acpi: x86: set enabled when composing _MAT entries Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 38/57] acpi: x86: madt: use build_append_int_noprefix() API to compose MADT table Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 39/57] acpi: arm/virt: " Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 40/57] acpi: build_dsdt_microvm: use acpi_table_begin()/acpi_table_end() instead of build_header() Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 41/57] acpi: arm: virt: build_dsdt: " Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 42/57] acpi: arm: virt: build_iort: " Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 43/57] acpi: arm/virt: convert build_iort() to endian agnostic build_append_FOO() API Michael S. Tsirkin
2021-10-05 16:03 ` [PULL 44/57] acpi: arm/virt: build_spcr: fix invalid cast Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 45/57] acpi: arm/virt: build_spcr: use acpi_table_begin()/acpi_table_end() instead of build_header() Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 46/57] acpi: arm/virt: build_gtdt: " Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 47/57] acpi: build_facs: use build_append_int_noprefix() API to compose table Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 48/57] acpi: remove no longer used build_header() Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 49/57] acpi: AcpiGenericAddress no longer used to map/access fields of MMIO, drop packed attribute Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 50/57] bios-tables-test: allow changes in DSDT ACPI tables for q35 Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 51/57] hw/i386/acpi: fix conflicting IO address range for acpi pci hotplug in q35 Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 52/57] bios-tables-test: Update ACPI DSDT table golden blobs for q35 Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 53/57] virtio-balloon: Fix page-poison subsection name Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 54/57] nvdimm: release the correct device list Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 55/57] hw/i386/amd_iommu: Rename amdviPCI TypeInfo Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 56/57] hw/i386/amd_iommu: Rename SysBus specific functions as amdvi_sysbus_X() Michael S. Tsirkin
2021-10-05 16:04 ` [PULL 57/57] hw/i386/amd_iommu: Add description/category to TYPE_AMD_IOMMU_PCI Michael S. Tsirkin
2021-10-05 17:21 ` [PULL 00/57] pc,pci,virtio: features, fixes Richard Henderson
2021-10-05 21:32 ` Michael S. Tsirkin
2021-10-05 23:36 ` Richard Henderson
2021-10-11 3:32 ` Jason Wang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20211005155946.513818-14-mst@redhat.com \
--to=mst@redhat.com \
--cc=ehabkost@redhat.com \
--cc=jiang.wang@bytedance.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-stable@nongnu.org \
--cc=sgarzare@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).