* [PATCH v2 0/2] vhost: VDUSE startup workaround and fix @ 2025-09-16 9:35 Maxime Coquelin 2025-09-16 9:35 ` [PATCH v2 1/2] vhost: add VDUSE virtqueue ready state polling workaround Maxime Coquelin ` (2 more replies) 0 siblings, 3 replies; 4+ messages in thread From: Maxime Coquelin @ 2025-09-16 9:35 UTC (permalink / raw) To: dev, david.marchand, chenbox; +Cc: Maxime Coquelin This series is a follow-up of the workaround to ensure queues are in a ready state before starting the device. It also adds a patch to ensure data passed to the kernel are properly initialized. Changes in v2: ============== - Improve comments (David) - Initialize vq_info properly (David) - Add David's R-by Maxime Coquelin (2): vhost: add VDUSE virtqueue ready state polling workaround vhost: fix uninitialized VQ info in VDUSE vring setup lib/vhost/vduse.c | 63 ++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 60 insertions(+), 3 deletions(-) -- 2.51.0 ^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH v2 1/2] vhost: add VDUSE virtqueue ready state polling workaround 2025-09-16 9:35 [PATCH v2 0/2] vhost: VDUSE startup workaround and fix Maxime Coquelin @ 2025-09-16 9:35 ` Maxime Coquelin 2025-09-16 9:35 ` [PATCH v2 2/2] vhost: fix uninitialized VQ info in VDUSE vring setup Maxime Coquelin 2025-09-16 13:09 ` [PATCH v2 0/2] vhost: VDUSE startup workaround and fix Patrick Robb 2 siblings, 0 replies; 4+ messages in thread From: Maxime Coquelin @ 2025-09-16 9:35 UTC (permalink / raw) To: dev, david.marchand, chenbox; +Cc: Maxime Coquelin, stable Add workaround to poll virtqueue ready states before starting device when VIRTIO_DEVICE_STATUS_DRIVER_OK is set in vduse_events_handler(). For each virtqueue, poll using VDUSE_VQ_GET_INFO ioctl to check vq_info->ready state with configurable retry limit. This addresses timing issues where device start was attempted before all virtqueues were properly initialized and ready. A notification mechanism will be introduced in the next version of the VDUSE uAPI. When it lands, we would only apply this workaround when the kernel does not support it. Fixes: a9120db8b98b ("vhost: add VDUSE device startup") Cc: stable@dpdk.org Reviewed-by: David Marchand <david.marchand@redhat.com> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com> --- lib/vhost/vduse.c | 61 +++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 59 insertions(+), 2 deletions(-) diff --git a/lib/vhost/vduse.c b/lib/vhost/vduse.c index 9de7f04a4f..422c4ab8f3 100644 --- a/lib/vhost/vduse.c +++ b/lib/vhost/vduse.c @@ -272,6 +272,55 @@ vduse_vring_cleanup(struct virtio_net *dev, unsigned int index) vq->last_avail_idx = 0; } +/* + * Tests show that virtqueues get ready at the first retry at worst, + * but let's be on the safe side and allow more retries. + */ +#define VDUSE_VQ_READY_POLL_MAX_RETRIES 100 + +static int +vduse_wait_for_virtqueues_ready(struct virtio_net *dev) +{ + unsigned int i; + int ret; + + for (i = 0; i < dev->nr_vring; i++) { + int retry_count = 0; + + while (retry_count < VDUSE_VQ_READY_POLL_MAX_RETRIES) { + struct vduse_vq_info vq_info = { 0 }; + + vq_info.index = i; + ret = ioctl(dev->vduse_dev_fd, VDUSE_VQ_GET_INFO, &vq_info); + if (ret) { + VHOST_CONFIG_LOG(dev->ifname, ERR, + "Failed to get VQ %u info while polling ready state: %s", + i, strerror(errno)); + return -1; + } + + if (vq_info.ready) { + VHOST_CONFIG_LOG(dev->ifname, DEBUG, + "VQ %u is ready after %u retries", i, retry_count); + break; + } + + retry_count++; + usleep(1000); + } + + if (retry_count >= VDUSE_VQ_READY_POLL_MAX_RETRIES) { + VHOST_CONFIG_LOG(dev->ifname, ERR, + "VQ %u ready state polling timeout after %u retries", + i, VDUSE_VQ_READY_POLL_MAX_RETRIES); + return -1; + } + } + + VHOST_CONFIG_LOG(dev->ifname, INFO, "All virtqueues are ready after polling"); + return 0; +} + static void vduse_device_start(struct virtio_net *dev, bool reconnect) { @@ -414,10 +463,18 @@ vduse_events_handler(int fd, void *arg, int *close __rte_unused) } if ((old_status ^ dev->status) & VIRTIO_DEVICE_STATUS_DRIVER_OK) { - if (dev->status & VIRTIO_DEVICE_STATUS_DRIVER_OK) + if (dev->status & VIRTIO_DEVICE_STATUS_DRIVER_OK) { + /* Poll virtqueues ready states before starting device */ + ret = vduse_wait_for_virtqueues_ready(dev); + if (ret < 0) { + VHOST_CONFIG_LOG(dev->ifname, ERR, + "Failed to wait for virtqueues ready, aborting device start"); + return; + } vduse_device_start(dev, false); - else + } else { vduse_device_stop(dev); + } } VHOST_CONFIG_LOG(dev->ifname, INFO, "Request %s (%u) handled successfully", -- 2.51.0 ^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH v2 2/2] vhost: fix uninitialized VQ info in VDUSE vring setup 2025-09-16 9:35 [PATCH v2 0/2] vhost: VDUSE startup workaround and fix Maxime Coquelin 2025-09-16 9:35 ` [PATCH v2 1/2] vhost: add VDUSE virtqueue ready state polling workaround Maxime Coquelin @ 2025-09-16 9:35 ` Maxime Coquelin 2025-09-16 13:09 ` [PATCH v2 0/2] vhost: VDUSE startup workaround and fix Patrick Robb 2 siblings, 0 replies; 4+ messages in thread From: Maxime Coquelin @ 2025-09-16 9:35 UTC (permalink / raw) To: dev, david.marchand, chenbox; +Cc: Maxime Coquelin, stable Initialize vduse_vq_info structure to zero to avoid using uninitialized memory when setting up VDUSE virtual rings. This ensures all fields start with known values. Fixes: a9120db8b98b ("vhost: add VDUSE device startup") Cc: stable@dpdk.org Reported-by: David Marchand <david.marchand@redhat.com> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com> --- lib/vhost/vduse.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/vhost/vduse.c b/lib/vhost/vduse.c index 422c4ab8f3..6f286c2963 100644 --- a/lib/vhost/vduse.c +++ b/lib/vhost/vduse.c @@ -141,7 +141,7 @@ vduse_vring_setup(struct virtio_net *dev, unsigned int index, bool reconnect) { struct vhost_virtqueue *vq = dev->virtqueue[index]; struct vhost_vring_addr *ra = &vq->ring_addrs; - struct vduse_vq_info vq_info; + struct vduse_vq_info vq_info = { 0 }; struct vduse_vq_eventfd vq_efd; int ret; -- 2.51.0 ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v2 0/2] vhost: VDUSE startup workaround and fix 2025-09-16 9:35 [PATCH v2 0/2] vhost: VDUSE startup workaround and fix Maxime Coquelin 2025-09-16 9:35 ` [PATCH v2 1/2] vhost: add VDUSE virtqueue ready state polling workaround Maxime Coquelin 2025-09-16 9:35 ` [PATCH v2 2/2] vhost: fix uninitialized VQ info in VDUSE vring setup Maxime Coquelin @ 2025-09-16 13:09 ` Patrick Robb 2 siblings, 0 replies; 4+ messages in thread From: Patrick Robb @ 2025-09-16 13:09 UTC (permalink / raw) To: Maxime Coquelin; +Cc: dev [-- Attachment #1: Type: text/plain, Size: 781 bytes --] Putting in a retest at the Community Lab. On Tue, Sep 16, 2025 at 5:36 AM Maxime Coquelin <maxime.coquelin@redhat.com> wrote: > This series is a follow-up of the workaround to ensure queues are > in a ready state before starting the device. > > It also adds a patch to ensure data passed to the kernel are properly > initialized. > > Changes in v2: > ============== > - Improve comments (David) > - Initialize vq_info properly (David) > - Add David's R-by > > Maxime Coquelin (2): > vhost: add VDUSE virtqueue ready state polling workaround > vhost: fix uninitialized VQ info in VDUSE vring setup > > lib/vhost/vduse.c | 63 ++++++++++++++++++++++++++++++++++++++++++++--- > 1 file changed, 60 insertions(+), 3 deletions(-) > > -- > 2.51.0 > > [-- Attachment #2: Type: text/html, Size: 1140 bytes --] ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2025-09-16 13:16 UTC | newest] Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2025-09-16 9:35 [PATCH v2 0/2] vhost: VDUSE startup workaround and fix Maxime Coquelin 2025-09-16 9:35 ` [PATCH v2 1/2] vhost: add VDUSE virtqueue ready state polling workaround Maxime Coquelin 2025-09-16 9:35 ` [PATCH v2 2/2] vhost: fix uninitialized VQ info in VDUSE vring setup Maxime Coquelin 2025-09-16 13:09 ` [PATCH v2 0/2] vhost: VDUSE startup workaround and fix Patrick Robb
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).