search for: virtqueue_is_broken

Hi, this patch-set tries to solve various hang situations when virtio devices (network or block) are hot-unplugged from a KVM guest. On System z there exists no handshake mechanism between host and guest when a device is hot-unplugged. The device is removed and no further I/O is possible. The guest is notified about the hard removal with a CRW machine check. As per architecture, the host must

Hi, this patch-set tries to solve various hang situations when virtio devices (network or block) are hot-unplugged from a KVM guest. On System z there exists no handshake mechanism between host and guest when a device is hot-unplugged. The device is removed and no further I/O is possible. The guest is notified about the hard removal with a CRW machine check. As per architecture, the host must

...struct vring_virtqueue *vq = to_vvq(_vq); + const struct vring_virtqueue *vq = to_vvq(_vq); return vq->packed_ring ? vq->packed.vring.num : vq->split.vring.num; } @@ -2819,9 +2819,9 @@ void __virtqueue_unbreak(struct virtqueue *_vq) } EXPORT_SYMBOL_GPL(__virtqueue_unbreak); -bool virtqueue_is_broken(struct virtqueue *_vq) +bool virtqueue_is_broken(const struct virtqueue *_vq) { - struct vring_virtqueue *vq = to_vvq(_vq); + const struct vring_virtqueue *vq = to_vvq(_vq); return READ_ONCE(vq->broken); } @@ -2868,9 +2868,9 @@ void __virtio_unbreak_device(struct virtio_device *dev) } EX...

...epending on the result of the host notification operation. If the host kick failed the current virtqueue is now flagged as 'broken'. Patches [3,4] add code to verify host kicks by testing the return value of virtqueue_kick() in order to avoid potential loops. Patch [5] adds a new function virtqueue_is_broken(). This function should be used to verify the state of a virtqueue when a previous virtqueue_get_buf() returned a NULL pointer. Patch [6,7,8,9] add virtqueue_is_broken() calls to handle potential errors when a virtqueue_bet_buf() doesn't deliver any more buffers. Heinz Graalfs (9): virtio_...

...epending on the result of the host notification operation. If the host kick failed the current virtqueue is now flagged as 'broken'. Patches [3,4] add code to verify host kicks by testing the return value of virtqueue_kick() in order to avoid potential loops. Patch [5] adds a new function virtqueue_is_broken(). This function should be used to verify the state of a virtqueue when a previous virtqueue_get_buf() returned a NULL pointer. Patch [6,7,8,9] add virtqueue_is_broken() calls to handle potential errors when a virtqueue_bet_buf() doesn't deliver any more buffers. Heinz Graalfs (9): virtio_...

This patch adds 2 new functions: virtqueue_set_broken(): to be called when a virtqueue kick operation fails. virtqueue_is_broken(): can be called to query the virtqueue state after a host was kicked. Signed-off-by: Heinz Graalfs <graalfs at linux.vnet.ibm.com> --- drivers/virtio/virtio_ring.c | 16 ++++++++++++++++ include/linux/virtio.h | 4 ++++ 2 files changed, 20 insertions(+) diff --git a/drivers/virtio/...

...4,10 @@ static bool virtnet_send_command(struct virtnet_info *vi, u8 class, u8 cmd, > > * into the hypervisor, so the request should be handled immediately. > > */ > > while (!virtqueue_get_buf(vi->cvq, &tmp) && > > - !virtqueue_is_broken(vi->cvq)) > > + !virtqueue_is_broken(vi->cvq)) { > > + cond_resched(); > > cpu_relax(); > > + } > > > > return vi->ctrl->status == VIRTIO_NET_OK; > > } > > -- > > 2.39.3 &...

...bool virtnet_send_command(struct virtnet_info *vi, u8 class, u8 cmd, >>> * into the hypervisor, so the request should be handled immediately. >>> */ >>> while (!virtqueue_get_buf(vi->cvq, &tmp) && >>> - !virtqueue_is_broken(vi->cvq)) >>> + !virtqueue_is_broken(vi->cvq)) { >>> + cond_resched(); >>> cpu_relax(); >>> + } >>> >>> return vi->ctrl->status == VIRTIO_NET_OK; >>> } >>...

...ruct virtnet_info *vi, u8 class, u8 cmd, > > > > * into the hypervisor, so the request should be handled immediately. > > > > */ > > > > while (!virtqueue_get_buf(vi->cvq, &tmp) && > > > > - !virtqueue_is_broken(vi->cvq)) > > > > + !virtqueue_is_broken(vi->cvq)) { > > > > + cond_resched(); > > > > cpu_relax(); > > > > + } > > > > > > > > return vi->ctrl->status =...

This patch series performs a clean up of the code in virtio_ring and virtio_pci, modifying it to conform with the Linux kernel coding style guidance [1]. The modifications ensure the code easy to read and understand. This small series does few short cleanups in the code. Patch-1 Remove unnecessary num zero check, which performs in power_of_2. Patch-2 Avoid using inline for small functions.

...vq command which turns out to have several side effects: 1) infinite poll for buggy devices 2) bad interaction with scheduler So this series tries to use sleep + timeout instead of busy polling. Please review. Thanks Changes since RFC: - switch to use BAD_RING in virtio_break_device() - check virtqueue_is_broken() after being woken up - use more_used() instead of virtqueue_get_buf() to allow caller to get buffers afterwards - break the virtio-net device when timeout - get buffer manually since the virtio core check more_used() instead Jason Wang (4): virtio-net: convert rx mode setting to use workqueu...

...rtnet_info *vi, u8 class, u8 cmd, >>>>> * into the hypervisor, so the request should be handled immediately. >>>>> */ >>>>> while (!virtqueue_get_buf(vi->cvq, &tmp) && >>>>> - !virtqueue_is_broken(vi->cvq)) >>>>> + !virtqueue_is_broken(vi->cvq)) { >>>>> + cond_resched(); >>>>> cpu_relax(); >>>>> + } >>>>> >>>>> return vi->ctrl->s...

...t; > > > > > * into the hypervisor, so the request should be handled immediately. > > > > > > */ > > > > > > while (!virtqueue_get_buf(vi->cvq, &tmp) && > > > > > > - !virtqueue_is_broken(vi->cvq)) > > > > > > + !virtqueue_is_broken(vi->cvq)) { > > > > > > + cond_resched(); > > > > > > cpu_relax(); > > > > > > + } > > > > > > > >...

...struct vring_virtqueue *vq = to_vvq(_vq); + const struct vring_virtqueue *vq = to_vvq(_vq); return vq->packed_ring ? vq->packed.vring.num : vq->split.vring.num; } @@ -2819,9 +2819,9 @@ void __virtqueue_unbreak(struct virtqueue *_vq) } EXPORT_SYMBOL_GPL(__virtqueue_unbreak); -bool virtqueue_is_broken(struct virtqueue *_vq) +bool virtqueue_is_broken(const struct virtqueue *_vq) { - struct vring_virtqueue *vq = to_vvq(_vq); + const struct vring_virtqueue *vq = to_vvq(_vq); return READ_ONCE(vq->broken); } @@ -2868,9 +2868,9 @@ void __virtio_unbreak_device(struct virtio_device *dev) } EX...

...--------------------------------------- > 3 files changed, 61 insertions(+), 53 deletions(-) > > diff --git a/include/linux/virtio.h b/include/linux/virtio.h > index 3c19bd3..8df7ba8 100644 > --- a/include/linux/virtio.h > +++ b/include/linux/virtio.h > @@ -78,6 +78,7 @@ bool virtqueue_is_broken(struct virtqueue *vq); > /** > * virtio_device - representation of a device using virtio > * @index: unique position on the virtio bus > + * @failed: saved value for CONFIG_S_FAILED bit (for restore) Have you considered s/failed/saved_failed/ ? > * @dev: underlying device. &...

...--------------------------------------- > 3 files changed, 61 insertions(+), 53 deletions(-) > > diff --git a/include/linux/virtio.h b/include/linux/virtio.h > index 3c19bd3..8df7ba8 100644 > --- a/include/linux/virtio.h > +++ b/include/linux/virtio.h > @@ -78,6 +78,7 @@ bool virtqueue_is_broken(struct virtqueue *vq); > /** > * virtio_device - representation of a device using virtio > * @index: unique position on the virtio bus > + * @failed: saved value for CONFIG_S_FAILED bit (for restore) Have you considered s/failed/saved_failed/ ? > * @dev: underlying device. &...

...(!virtqueue_indirect_desc_table_add(vq, desc, *num)) { > virtqueue_kick(vq); > - wait_event(vb->acked, virtqueue_get_buf(vq, &len)); > - vb->balloon_page_chunk.chunk_num = 0; > + if (busy_wait) > + while (!virtqueue_get_buf(vq, &len) && > + !virtqueue_is_broken(vq)) > + cpu_relax(); > + else > + wait_event(vb->acked, virtqueue_get_buf(vq, &len)); This is something I didn't previously notice. As you always keep a single buffer in flight, you do not really need indirect at all. Just add all descriptors in the ring directly, then...

...(!virtqueue_indirect_desc_table_add(vq, desc, *num)) { > virtqueue_kick(vq); > - wait_event(vb->acked, virtqueue_get_buf(vq, &len)); > - vb->balloon_page_chunk.chunk_num = 0; > + if (busy_wait) > + while (!virtqueue_get_buf(vq, &len) && > + !virtqueue_is_broken(vq)) > + cpu_relax(); > + else > + wait_event(vb->acked, virtqueue_get_buf(vq, &len)); This is something I didn't previously notice. As you always keep a single buffer in flight, you do not really need indirect at all. Just add all descriptors in the ring directly, then...

...Please review. Thanks Changes since V2: - Don't use interrupt but cond_resched() Changes since V1: - use RTNL to synchronize rx mode worker - use completion for simplicity - don't try to harden CVQ command Changes since RFC: - switch to use BAD_RING in virtio_break_device() - check virtqueue_is_broken() after being woken up - use more_used() instead of virtqueue_get_buf() to allow caller to get buffers afterwards - break the virtio-net device when timeout - get buffer manually since the virtio core check more_used() instead Jason Wang (2): virtio-net: convert rx mode setting to use work...

...vers/virtio/virtio.c | 57 +++++++++++++++++++++++++++++++++++++++++-------- 2 files changed, 54 insertions(+), 9 deletions(-) diff --git a/include/linux/virtio.h b/include/linux/virtio.h index 8df7ba8..5636b11 100644 --- a/include/linux/virtio.h +++ b/include/linux/virtio.h @@ -79,6 +79,9 @@ bool virtqueue_is_broken(struct virtqueue *vq); * virtio_device - representation of a device using virtio * @index: unique position on the virtio bus * @failed: saved value for CONFIG_S_FAILED bit (for restore) + * @config_enabled: configuration change reporting enabled + * @config_changed: configuration change repor...