virtio-pci: disable vring processing when bus-mastering is disabled
RFC PATCH 1/1 virtio-pci: disable vring processing when bus-mastering is disabled
Currently the SLOF firmware for pseries guests will disable/re-enable a PCI device multiple times via IO/MEM/MASTER bits of PCI_COMMAND register after the initial probe/feature negotiation, as it tends to work with a single device at a time at various stages like probing and running block/network bootloaders without doing a full reset in-between.
In QEMU, when PCI_COMMAND_MASTER is disabled we disable the corresponding IOMMU memory region, so DMA accesses (including to vring fields like idx/flags) will no longer undergo the necessary translation. Normally we wouldn't expect this to happen since it would be misbehavior on the driver side to continue driving DMA requests.
However, in the case of pseries, with iommu_platform=on, we trigger the following sequence when tearing down the virtio-blk dataplane ioeventfd in response to the guest unsetting PCI_COMMAND_MASTER:
#2 0x0000555555922651 in virtqueue_map_desc (vdev=vdev@entry=0x555556dbcfb0, p_num_sg=p_num_sg@entry=0x7fffe657e1a8, addr=addr@entry=0x7fffe657e240, iov=iov@entry=0x7fffe6580240, max_num_sg=max_num_sg@entry=1024, is_write=is_write@entry=false, pa=0, sz=0)
at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:757
#3 0x0000555555922a89 in virtqueue_pop (vq=vq@entry=0x555556dc8660, sz=sz@entry=184)
at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:950
#4 0x00005555558d3eca in virtio_blk_get_request (vq=0x555556dc8660, s=0x555556dbcfb0)
at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:255
#5 0x00005555558d3eca in virtio_blk_handle_vq (s=0x555556dbcfb0, vq=0x555556dc8660)
at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:776
#6 0x000055555591dd66 in virtio_queue_notify_aio_vq (vq=vq@entry=0x555556dc8660)
at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1550
#7 0x000055555591ecef in virtio_queue_notify_aio_vq (vq=0x555556dc8660)
at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1546
#8 0x000055555591ecef in virtio_queue_host_notifier_aio_poll (opaque=0x555556dc86c8)
at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:2527
#9 0x0000555555d02164 in run_poll_handlers_once (ctx=ctx@entry=0x55555688bfc0, timeout=timeout@entry=0x7fffe65844a8)
at /home/mdroth/w/qemu.git/util/aio-posix.c:520
#10 0x0000555555d02d1b in try_poll_mode (timeout=0x7fffe65844a8, ctx=0x55555688bfc0)
at /home/mdroth/w/qemu.git/util/aio-posix.c:607
#11 0x0000555555d02d1b in aio_poll (ctx=ctx@entry=0x55555688bfc0, blocking=blocking@entry=true)
at /home/mdroth/w/qemu.git/util/aio-posix.c:639
#12 0x0000555555d0004d in aio_wait_bh_oneshot (ctx=0x55555688bfc0, cb=cb@entry=0x5555558d5130 <virtio_blk_data_plane_stop_bh>, opaque=opaque@entry=0x555556de86f0)
at /home/mdroth/w/qemu.git/util/aio-wait.c:71
#13 0x00005555558d59bf in virtio_blk_data_plane_stop (vdev=<optimized out>)
at /home/mdroth/w/qemu.git/hw/block/dataplane/virtio-blk.c:288
#14 0x0000555555b906a1 in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38)
at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:245
#15 0x0000555555b90dbb in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38)
at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:237
#16 0x0000555555b92a8e in virtio_pci_stop_ioeventfd (proxy=0x555556db4e40)
at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:292
#17 0x0000555555b92a8e in virtio_write_config (pci_dev=0x555556db4e40, address=<optimized out>, val=1048832, len=<optimized out>)
at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:613I.e. the calling code is only scheduling a one-shot BH for virtio_blk_data_plane_stop_bh, but somehow we end up trying to process an additional virtqueue entry before we get there. This is likely due to the following check in virtio_queue_host_notifier_aio_poll:
static bool virtio_queue_host_notifier_aio_poll(void *opaque)
{
EventNotifier *n = opaque;
VirtQueue *vq = container_of(n, VirtQueue, host_notifier);
bool progress; if (!vq->vring.desc || virtio_queue_empty(vq)) {
return false;
}
progress = virtio_queue_notify_aio_vq(vq);namely the call to virtio_queue_empty(). In this case, since no new requests have actually been issued, shadow_avail_idx == last_avail_idx, so we actually try to access the vring via vring_avail_idx() to get the latest non-shadowed idx:
int virtio_queue_empty(VirtQueue *vq)
{
bool empty;
... if (vq->shadow_avail_idx != vq->last_avail_idx) {
return 0;
}
rcu_read_lock();
empty = vring_avail_idx(vq) == vq->last_avail_idx;
rcu_read_unlock();
return empty;but since the IOMMU region has been disabled we get a bogus value (0 usually), which causes virtio_queue_empty() to falsely report that there are entries to be processed, which causes errors such as:
"virtio: zero sized buffers are not allowed"
or
"virtio-blk missing headers"
and puts the device in an error state.
This patch works around the issue by introducing virtio_set_disabled(), which piggy-backs off the vdev->broken flag we already use to bypass checks like virtio_queue_empty(), and sets/unsets it in response to enabling/disabling bus-mastering.
NOTES:
- It's possible we could also work around this in SLOF by doing a full reset instead of relying on PCI_COMMAND to enable/disable, but in any case the QEMU behavior seems wrong.
- This leaves some other oddities in play, like the fact that DRIVER_OK also gets unset in response to bus-mastering being disabled, but not restored (however the device seems to continue working)
- Similarly, we disable the host notifier via virtio_bus_stop_ioeventfd(), which seems to move the handling out of virtio-blk dataplane and back into the main IO thread, and it ends up staying there till a reset (but otherwise continues working normally)
--- hw/virtio/virtio-pci.c | 12 ++++++++---- hw/virtio/virtio.c | 7 ++++++- include/hw/virtio/virtio.h | 1 + 3 files changed, 15 insertions(+), 5 deletions(-)
hw/virtio/virtio-pci.c +8 -4
@@ -608,10 +608,14 @@ static void virtio_write_config(PCIDevice *pci_dev, uint32_t address,pcie_cap_flr_write_config (pci_dev, address, val, len); } -if (range_covers_byte (address, len,PCI_COMMAND ) && - !(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { -virtio_pci_stop_ioeventfd (proxy); -virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); +if (range_covers_byte (address, len,PCI_COMMAND )) { +if (!(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { +virtio_set_disabled (vdev, true); +virtio_pci_stop_ioeventfd (proxy); +virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); + }else { +virtio_set_disabled (vdev, false); + } } if (proxy->config_cap &&
hw/virtio/virtio.c +6 -1
@@ -2575,6 +2575,11 @@ void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) vdev->bus_name =g_strdup (bus_name); } +void virtio_set_disabled (VirtIODevice *vdev,bool disable) +{ + vdev->broken = disable; +} + void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...) {va_list ap;
@@ -2588,7 +2593,7 @@ void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...)virtio_notify_config (vdev); } - vdev->broken = true; +virtio_set_disabled (vdev, true); }static void virtio_memory_listener_commit (MemoryListener *listener)
include/hw/virtio/virtio.h +1 -0
@@ -168,6 +168,7 @@ void virtio_init(VirtIODevice *vdev, const char *name, uint16_t device_id, size_t config_size);void virtio_cleanup (VirtIODevice *vdev); +void virtio_set_disabled (VirtIODevice *vdev,bool disable); void virtio_error(VirtIODevice *vdev, const char *fmt, ...) GCC_FMT_ATTR(2, 3);/* Set the child bus name. */
Signature · 1 line
2.17.1
Quoted 157 lines · Michael Roth
Currently the SLOF firmware for pseries guests will disable/re-enable a PCI device multiple times via IO/MEM/MASTER bits of
PCI_COMMANDregister after the initial probe/feature negotiation, as it tends to work with a single device at a time at various stages like probing and running block/network bootloaders without doing a full reset in-between.In QEMU, when
PCI_COMMAND_MASTERis disabled we disable the corresponding IOMMU memory region, so DMA accesses (including to vring fields like idx/flags) will no longer undergo the necessary translation. Normally we wouldn't expect this to happen since it would be misbehavior on the driver side to continue driving DMA requests.However, in the case of pseries, with iommu_platform=on, we trigger the following sequence when tearing down the virtio-blk dataplane ioeventfd in response to the guest unsetting
PCI_COMMAND_MASTER:#2
0x0000555555922651invirtqueue_map_desc(vdev=vdev@entry=0x555556dbcfb0, p_num_sg=p_num_sg@entry=0x7fffe657e1a8, addr=addr@entry=0x7fffe657e240, iov=iov@entry=0x7fffe6580240, max_num_sg=max_num_sg@entry=1024, is_write=is_write@entry=false, pa=0, sz=0)at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:757 #3 0x0000555555922a89 in virtqueue_pop (vq=vq@entry=0x555556dc8660, sz=sz@entry=184) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:950 #4 0x00005555558d3eca in virtio_blk_get_request (vq=0x555556dc8660, s=0x555556dbcfb0) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:255 #5 0x00005555558d3eca in virtio_blk_handle_vq (s=0x555556dbcfb0, vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:776 #6 0x000055555591dd66 in virtio_queue_notify_aio_vq (vq=vq@entry=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1550 #7 0x000055555591ecef in virtio_queue_notify_aio_vq (vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1546 #8 0x000055555591ecef in virtio_queue_host_notifier_aio_poll (opaque=0x555556dc86c8) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:2527 #9 0x0000555555d02164 in run_poll_handlers_once (ctx=ctx@entry=0x55555688bfc0, timeout=timeout@entry=0x7fffe65844a8) at /home/mdroth/w/qemu.git/util/aio-posix.c:520 #10 0x0000555555d02d1b in try_poll_mode (timeout=0x7fffe65844a8, ctx=0x55555688bfc0) at /home/mdroth/w/qemu.git/util/aio-posix.c:607 #11 0x0000555555d02d1b in aio_poll (ctx=ctx@entry=0x55555688bfc0, blocking=blocking@entry=true) at /home/mdroth/w/qemu.git/util/aio-posix.c:639 #12 0x0000555555d0004d in aio_wait_bh_oneshot (ctx=0x55555688bfc0, cb=cb@entry=0x5555558d5130 <virtio_blk_data_plane_stop_bh>, opaque=opaque@entry=0x555556de86f0) at /home/mdroth/w/qemu.git/util/aio-wait.c:71 #13 0x00005555558d59bf in virtio_blk_data_plane_stop (vdev=<optimized out>) at /home/mdroth/w/qemu.git/hw/block/dataplane/virtio-blk.c:288 #14 0x0000555555b906a1 in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:245 #15 0x0000555555b90dbb in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:237 #16 0x0000555555b92a8e in virtio_pci_stop_ioeventfd (proxy=0x555556db4e40) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:292 #17 0x0000555555b92a8e in virtio_write_config (pci_dev=0x555556db4e40, address=<optimized out>, val=1048832, len=<optimized out>) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:613I.e. the calling code is only scheduling a one-shot BH for
virtio_blk_data_plane_stop_bh, but somehow we end up trying to process an additional virtqueue entry before we get there. This is likely due to the following check invirtio_queue_host_notifier_aio_poll:static bool virtio_queue_host_notifier_aio_poll(void *opaque) { EventNotifier *n = opaque; VirtQueue *vq = container_of(n, VirtQueue, host_notifier); bool progress;if (!vq->vring.desc || virtio_queue_empty(vq)) { return false; } progress = virtio_queue_notify_aio_vq(vq);namely the call to
virtio_queue_empty(). In this case, since no new requests have actually been issued,shadow_avail_idx==last_avail_idx, so we actually try to access the vring viavring_avail_idx()to get the latest non-shadowed idx:int virtio_queue_empty(VirtQueue *vq) { bool empty; ...if (vq->shadow_avail_idx != vq->last_avail_idx) { return 0; } rcu_read_lock(); empty = vring_avail_idx(vq) == vq->last_avail_idx; rcu_read_unlock(); return empty;but since the IOMMU region has been disabled we get a bogus value (0 usually), which causes
virtio_queue_empty()to falsely report that there are entries to be processed, which causes errors such as:"virtio: zero sized buffers are not allowed"
or
"virtio-blk missing headers"
and puts the device in an error state.
This patch works around the issue by introducing
virtio_set_disabled(), which piggy-backs off thevdev->brokenflag we already use to bypass checks likevirtio_queue_empty(), and sets/unsets it in response to enabling/disabling bus-mastering.NOTES:
- It's possible we could also work around this in SLOF by doing a full reset instead of relying on
PCI_COMMANDto enable/disable, but in any case the QEMU behavior seems wrong.
- This leaves some other oddities in play, like the fact thatDRIVER_OKalso gets unset in response to bus-mastering being disabled, but not restored (however the device seems to continue working)
- Similarly, we disable the host notifier viavirtio_bus_stop_ioeventfd(), which seems to move the handling out of virtio-blk dataplane and back into the main IO thread, and it ends up staying there till a reset (but otherwise continues working normally)Cc David Gibson <[email protected]>,Cc Alexey Kardashevskiy <[email protected]>Cc "Michael S. Tsirkin" <[email protected]>Signed-off-by Michael Roth <[email protected]>--- hw/virtio/virtio-pci.c | 12 ++++++++---- hw/virtio/virtio.c | 7 ++++++- include/hw/virtio/virtio.h | 1 + 3 files changed, 15 insertions(+), 5 deletions(-)hw/virtio/virtio-pci.c +8
-4@@ -608,10 +608,14 @@ static void virtio_write_config(PCIDevice *pci_dev, uint32_t address,pcie_cap_flr_write_config (pci_dev, address, val, len); } -if (range_covers_byte (address, len,PCI_COMMAND ) && - !(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { -virtio_pci_stop_ioeventfd (proxy); -virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); +if (range_covers_byte (address, len,PCI_COMMAND )) { +if (!(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { +virtio_set_disabled (vdev, true); +virtio_pci_stop_ioeventfd (proxy); +virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); + }else { +virtio_set_disabled (vdev, false); + } } if (proxy->config_cap &&hw/virtio/virtio.c +0
-0@@ -2575,6 +2575,11 @@ void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) vdev->bus_name =g_strdup (bus_name);
hw/virtio/virtio.c void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) · line 2585
Quoted 10 lines
hw/virtio/virtio.c +5
-0} +void virtio_set_disabled (VirtIODevice *vdev,bool disable) +{ + vdev->broken = disable; +} + void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...) {va_list ap;
Hmm. I think just clear and immediate set of bus master while device was not doing any DMA actually should be fine and should not require a reset. Now it's true that right now guests reset first thing which will clear the broken flag, but I'd say it's wrong to require this specific order.
I think the easiest thing is to add a "disabled" flag.
Quoted 14 lines
hw/virtio/virtio.c +1
-1@@ -2588,7 +2593,7 @@ void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...)virtio_notify_config (vdev); } - vdev->broken = true; +virtio_set_disabled (vdev, true); }static void virtio_memory_listener_commit (MemoryListener *listener)include/hw/virtio/virtio.h +0
-0
include/hw/virtio/virtio.h void virtio_init(VirtIODevice *vdev, const char *name, · line 174
Quoted 7 lines
include/hw/virtio/virtio.h +1
-0@@ -168,6 +168,7 @@ void virtio_init(VirtIODevice *vdev, const char *name, uint16_t device_id, size_t config_size);void virtio_cleanup (VirtIODevice *vdev); +void virtio_set_disabled (VirtIODevice *vdev,bool disable); void virtio_error(VirtIODevice *vdev, const char *fmt, ...) GCC_FMT_ATTR(2, 3);/* Set the child bus name. */
One open question here is cross version migration. What exactly happens if we migrate to an old qemu? from an old qemu?
Quoted 2 lines
Signature · 1 line
2.17.1
Quoting Michael S. Tsirkin (2019-11-13 04:09:02)
Quoted 160 lines · Michael Roth
Quoted 159 lines
Currently the SLOF firmware for pseries guests will disable/re-enable a PCI device multiple times via IO/MEM/MASTER bits of
PCI_COMMANDregister after the initial probe/feature negotiation, as it tends to work with a single device at a time at various stages like probing and running block/network bootloaders without doing a full reset in-between.In QEMU, when
PCI_COMMAND_MASTERis disabled we disable the corresponding IOMMU memory region, so DMA accesses (including to vring fields like idx/flags) will no longer undergo the necessary translation. Normally we wouldn't expect this to happen since it would be misbehavior on the driver side to continue driving DMA requests.However, in the case of pseries, with iommu_platform=on, we trigger the following sequence when tearing down the virtio-blk dataplane ioeventfd in response to the guest unsetting
PCI_COMMAND_MASTER:#2
0x0000555555922651invirtqueue_map_desc(vdev=vdev@entry=0x555556dbcfb0, p_num_sg=p_num_sg@entry=0x7fffe657e1a8, addr=addr@entry=0x7fffe657e240, iov=iov@entry=0x7fffe6580240, max_num_sg=max_num_sg@entry=1024, is_write=is_write@entry=false, pa=0, sz=0)at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:757 #3 0x0000555555922a89 in virtqueue_pop (vq=vq@entry=0x555556dc8660, sz=sz@entry=184) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:950 #4 0x00005555558d3eca in virtio_blk_get_request (vq=0x555556dc8660, s=0x555556dbcfb0) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:255 #5 0x00005555558d3eca in virtio_blk_handle_vq (s=0x555556dbcfb0, vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:776 #6 0x000055555591dd66 in virtio_queue_notify_aio_vq (vq=vq@entry=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1550 #7 0x000055555591ecef in virtio_queue_notify_aio_vq (vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1546 #8 0x000055555591ecef in virtio_queue_host_notifier_aio_poll (opaque=0x555556dc86c8) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:2527 #9 0x0000555555d02164 in run_poll_handlers_once (ctx=ctx@entry=0x55555688bfc0, timeout=timeout@entry=0x7fffe65844a8) at /home/mdroth/w/qemu.git/util/aio-posix.c:520 #10 0x0000555555d02d1b in try_poll_mode (timeout=0x7fffe65844a8, ctx=0x55555688bfc0) at /home/mdroth/w/qemu.git/util/aio-posix.c:607 #11 0x0000555555d02d1b in aio_poll (ctx=ctx@entry=0x55555688bfc0, blocking=blocking@entry=true) at /home/mdroth/w/qemu.git/util/aio-posix.c:639 #12 0x0000555555d0004d in aio_wait_bh_oneshot (ctx=0x55555688bfc0, cb=cb@entry=0x5555558d5130 <virtio_blk_data_plane_stop_bh>, opaque=opaque@entry=0x555556de86f0) at /home/mdroth/w/qemu.git/util/aio-wait.c:71 #13 0x00005555558d59bf in virtio_blk_data_plane_stop (vdev=<optimized out>) at /home/mdroth/w/qemu.git/hw/block/dataplane/virtio-blk.c:288 #14 0x0000555555b906a1 in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:245 #15 0x0000555555b90dbb in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:237 #16 0x0000555555b92a8e in virtio_pci_stop_ioeventfd (proxy=0x555556db4e40) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:292 #17 0x0000555555b92a8e in virtio_write_config (pci_dev=0x555556db4e40, address=<optimized out>, val=1048832, len=<optimized out>) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:613I.e. the calling code is only scheduling a one-shot BH for
virtio_blk_data_plane_stop_bh, but somehow we end up trying to process an additional virtqueue entry before we get there. This is likely due to the following check invirtio_queue_host_notifier_aio_poll:static bool virtio_queue_host_notifier_aio_poll(void *opaque) { EventNotifier *n = opaque; VirtQueue *vq = container_of(n, VirtQueue, host_notifier); bool progress;if (!vq->vring.desc || virtio_queue_empty(vq)) { return false; } progress = virtio_queue_notify_aio_vq(vq);namely the call to
virtio_queue_empty(). In this case, since no new requests have actually been issued,shadow_avail_idx==last_avail_idx, so we actually try to access the vring viavring_avail_idx()to get the latest non-shadowed idx:int virtio_queue_empty(VirtQueue *vq) { bool empty; ...if (vq->shadow_avail_idx != vq->last_avail_idx) { return 0; } rcu_read_lock(); empty = vring_avail_idx(vq) == vq->last_avail_idx; rcu_read_unlock(); return empty;but since the IOMMU region has been disabled we get a bogus value (0 usually), which causes
virtio_queue_empty()to falsely report that there are entries to be processed, which causes errors such as:"virtio: zero sized buffers are not allowed"
or
"virtio-blk missing headers"
and puts the device in an error state.
This patch works around the issue by introducing
virtio_set_disabled(), which piggy-backs off thevdev->brokenflag we already use to bypass checks likevirtio_queue_empty(), and sets/unsets it in response to enabling/disabling bus-mastering.NOTES:
- It's possible we could also work around this in SLOF by doing a full reset instead of relying on
PCI_COMMANDto enable/disable, but in any case the QEMU behavior seems wrong.
- This leaves some other oddities in play, like the fact thatDRIVER_OKalso gets unset in response to bus-mastering being disabled, but not restored (however the device seems to continue working)
- Similarly, we disable the host notifier viavirtio_bus_stop_ioeventfd(), which seems to move the handling out of virtio-blk dataplane and back into the main IO thread, and it ends up staying there till a reset (but otherwise continues working normally)Cc David Gibson <[email protected]>,Cc Alexey Kardashevskiy <[email protected]>Cc "Michael S. Tsirkin" <[email protected]>Signed-off-by Michael Roth <[email protected]>--- hw/virtio/virtio-pci.c | 12 ++++++++---- hw/virtio/virtio.c | 7 ++++++- include/hw/virtio/virtio.h | 1 + 3 files changed, 15 insertions(+), 5 deletions(-)hw/virtio/virtio-pci.c +8
-4@@ -608,10 +608,14 @@ static void virtio_write_config(PCIDevice *pci_dev, uint32_t address,pcie_cap_flr_write_config (pci_dev, address, val, len); } -if (range_covers_byte (address, len,PCI_COMMAND ) && - !(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { -virtio_pci_stop_ioeventfd (proxy); -virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); +if (range_covers_byte (address, len,PCI_COMMAND )) { +if (!(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { +virtio_set_disabled (vdev, true); +virtio_pci_stop_ioeventfd (proxy); +virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); + }else { +virtio_set_disabled (vdev, false); + } } if (proxy->config_cap &&hw/virtio/virtio.c +0
-0@@ -2575,6 +2575,11 @@ void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) vdev->bus_name =g_strdup (bus_name); }
Quoted 15 lines
hw/virtio/virtio.c void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) · line 2585
Quoted 8 lines
hw/virtio/virtio.c +5
-0+void virtio_set_disabled (VirtIODevice *vdev,bool disable) +{ + vdev->broken = disable; +} + void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...) {va_list ap;Hmm. I think just clear and immediate set of bus master while device was not doing any DMA actually should be fine and should not require a reset. Now it's true that right now guests reset first thing which will clear the broken flag, but I'd say it's wrong to require this specific order.
I think the easiest thing is to add a "disabled" flag.
Agreed. Some comments on that below.
Quoted 22 lines
Quoted 22 lines
hw/virtio/virtio.c +1
-1@@ -2588,7 +2593,7 @@ void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...)virtio_notify_config (vdev); } - vdev->broken = true; +virtio_set_disabled (vdev, true); }static void virtio_memory_listener_commit (MemoryListener *listener)include/hw/virtio/virtio.h +1
-0@@ -168,6 +168,7 @@ void virtio_init(VirtIODevice *vdev, const char *name, uint16_t device_id, size_t config_size);void virtio_cleanup (VirtIODevice *vdev); +void virtio_set_disabled (VirtIODevice *vdev,bool disable); void virtio_error(VirtIODevice *vdev, const char *fmt, ...) GCC_FMT_ATTR(2, 3);
Quoted 6 lines
include/hw/virtio/virtio.h void virtio_init(VirtIODevice *vdev, const char *name, · line 174
Quoted 1 line
include/hw/virtio/virtio.h +0
-0/* Set the child bus name. */ One open question here is cross version migration. What exactly happens if we migrate to an old qemu? from an old qemu?
Currently vdev->broken is already migrated via a subsection if it is set.
So, with the current proposed patch I would expect the following:
old -> new:
dev->broken set due to proposed virtio_set_disabled():
- N/A
dev->broken set due to other/existing reasons:
- value is migrated, migration succeeds
- device continues not working until a reset, or bus-mastering enabled/re-enabled. (which is good, but also makes me realize that a device that was put into broken state for reasons other than virtio_set_disabled() should not get 'unbroken' simply because bus-master bit was cycled. a separate flag is probably needed)
- PASS
dev->broken not set:
- not migrated
- PASS
new -> old:
dev->broken set due to proposed virtio_set_disabled():
- migration succeeds to any QEMU that already has handling for dev->broken.
- migration fails for any QEMU that doesn't have handling for dev->broken as it does now, but will a higher chance of triggering
- device won't work until it is reset. for most guests we will probably get a reset before the device is used again anyway. for the SLOF case the device will stay broken after bus-mastering re-enabled, but that's the case for existing QEMU now anyway
- PASS, but with increased chance of migration breakage for QEMU's that don't have handling for dev->broken.
dev->broken set due to other/existing reasons:
- migration succeeds to any QEMU that already has handling for dev->broken. device will require a reset as it does now.
- migration fails for any QEMU that doesn't have handling for dev->broken as it does now
- PASS
dev->broken not set:
- not migrated
- PASS
With a new dev->disabled flag, which we'd likely send using a subsection like with dev->broken, I would expect the following:
old -> new:
dev->disabled set:
- N/A
dev->disabled not set:
- not migrated
- if source disables BM just before migration we will likely go into an error state that either sets dev->broken or puts the device in some other possibly bad state. either case would require subsequent reset, just as they would without migration
- PASS
new -> old:
dev->disabled set:
- migration stream fails
- FAIL, but as expected, and doesn't seem common currently outside of SLOF early-boot else we'd probably have more reports of breakage from vring access while BM isn't set.
dev->broken not set:
- not migrated
- PASS
So re-using 'broken' is slightly better from a migration standpoint, but as noted above it is probably wrong to unset 'broken' just because BM bit gets cycled, so a new 'disabled' flag is probably needed.
We could probably get by with just adding a check for dev->disabled in virtio_queue_empty(), or even earlier in virtio_queue_host_notifier_aio_poll(), but it seems more proper to also add it in most of the same places we currently check for dev->broken.
That seems somewhat redundant though, so I think maybe the best approach is to:
- replace most dev->broken checks with checks for dev->disabled
- set dev->disabled whenever dev->broken gets set
- add a check in virtio_set_disabled() that only allows us to re-enable if dev->broken hasn't also been set.
I'll work on a follow-up using that approach if it seems reasonable to you.
Thanks!
Quoted 4 lines
Quoted 3 lines
Signature · 2 lines
-- 2.17.1
Quoted 282 lines · Michael Roth
Quoting Michael S. Tsirkin (2019-11-13 04:09:02)
Quoted 160 lines · Michael Roth
Quoted 159 lines
Currently the SLOF firmware for pseries guests will disable/re-enable a PCI device multiple times via IO/MEM/MASTER bits of
PCI_COMMANDregister after the initial probe/feature negotiation, as it tends to work with a single device at a time at various stages like probing and running block/network bootloaders without doing a full reset in-between.In QEMU, when
PCI_COMMAND_MASTERis disabled we disable the corresponding IOMMU memory region, so DMA accesses (including to vring fields like idx/flags) will no longer undergo the necessary translation. Normally we wouldn't expect this to happen since it would be misbehavior on the driver side to continue driving DMA requests.However, in the case of pseries, with iommu_platform=on, we trigger the following sequence when tearing down the virtio-blk dataplane ioeventfd in response to the guest unsetting
PCI_COMMAND_MASTER:#2
0x0000555555922651invirtqueue_map_desc(vdev=vdev@entry=0x555556dbcfb0, p_num_sg=p_num_sg@entry=0x7fffe657e1a8, addr=addr@entry=0x7fffe657e240, iov=iov@entry=0x7fffe6580240, max_num_sg=max_num_sg@entry=1024, is_write=is_write@entry=false, pa=0, sz=0)at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:757 #3 0x0000555555922a89 in virtqueue_pop (vq=vq@entry=0x555556dc8660, sz=sz@entry=184) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:950 #4 0x00005555558d3eca in virtio_blk_get_request (vq=0x555556dc8660, s=0x555556dbcfb0) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:255 #5 0x00005555558d3eca in virtio_blk_handle_vq (s=0x555556dbcfb0, vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:776 #6 0x000055555591dd66 in virtio_queue_notify_aio_vq (vq=vq@entry=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1550 #7 0x000055555591ecef in virtio_queue_notify_aio_vq (vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1546 #8 0x000055555591ecef in virtio_queue_host_notifier_aio_poll (opaque=0x555556dc86c8) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:2527 #9 0x0000555555d02164 in run_poll_handlers_once (ctx=ctx@entry=0x55555688bfc0, timeout=timeout@entry=0x7fffe65844a8) at /home/mdroth/w/qemu.git/util/aio-posix.c:520 #10 0x0000555555d02d1b in try_poll_mode (timeout=0x7fffe65844a8, ctx=0x55555688bfc0) at /home/mdroth/w/qemu.git/util/aio-posix.c:607 #11 0x0000555555d02d1b in aio_poll (ctx=ctx@entry=0x55555688bfc0, blocking=blocking@entry=true) at /home/mdroth/w/qemu.git/util/aio-posix.c:639 #12 0x0000555555d0004d in aio_wait_bh_oneshot (ctx=0x55555688bfc0, cb=cb@entry=0x5555558d5130 <virtio_blk_data_plane_stop_bh>, opaque=opaque@entry=0x555556de86f0) at /home/mdroth/w/qemu.git/util/aio-wait.c:71 #13 0x00005555558d59bf in virtio_blk_data_plane_stop (vdev=<optimized out>) at /home/mdroth/w/qemu.git/hw/block/dataplane/virtio-blk.c:288 #14 0x0000555555b906a1 in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:245 #15 0x0000555555b90dbb in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:237 #16 0x0000555555b92a8e in virtio_pci_stop_ioeventfd (proxy=0x555556db4e40) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:292 #17 0x0000555555b92a8e in virtio_write_config (pci_dev=0x555556db4e40, address=<optimized out>, val=1048832, len=<optimized out>) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:613I.e. the calling code is only scheduling a one-shot BH for
virtio_blk_data_plane_stop_bh, but somehow we end up trying to process an additional virtqueue entry before we get there. This is likely due to the following check invirtio_queue_host_notifier_aio_poll:static bool virtio_queue_host_notifier_aio_poll(void *opaque) { EventNotifier *n = opaque; VirtQueue *vq = container_of(n, VirtQueue, host_notifier); bool progress;if (!vq->vring.desc || virtio_queue_empty(vq)) { return false; } progress = virtio_queue_notify_aio_vq(vq);namely the call to
virtio_queue_empty(). In this case, since no new requests have actually been issued,shadow_avail_idx==last_avail_idx, so we actually try to access the vring viavring_avail_idx()to get the latest non-shadowed idx:int virtio_queue_empty(VirtQueue *vq) { bool empty; ...if (vq->shadow_avail_idx != vq->last_avail_idx) { return 0; } rcu_read_lock(); empty = vring_avail_idx(vq) == vq->last_avail_idx; rcu_read_unlock(); return empty;but since the IOMMU region has been disabled we get a bogus value (0 usually), which causes
virtio_queue_empty()to falsely report that there are entries to be processed, which causes errors such as:"virtio: zero sized buffers are not allowed"
or
"virtio-blk missing headers"
and puts the device in an error state.
This patch works around the issue by introducing
virtio_set_disabled(), which piggy-backs off thevdev->brokenflag we already use to bypass checks likevirtio_queue_empty(), and sets/unsets it in response to enabling/disabling bus-mastering.NOTES:
- It's possible we could also work around this in SLOF by doing a full reset instead of relying on
PCI_COMMANDto enable/disable, but in any case the QEMU behavior seems wrong.
- This leaves some other oddities in play, like the fact thatDRIVER_OKalso gets unset in response to bus-mastering being disabled, but not restored (however the device seems to continue working)
- Similarly, we disable the host notifier viavirtio_bus_stop_ioeventfd(), which seems to move the handling out of virtio-blk dataplane and back into the main IO thread, and it ends up staying there till a reset (but otherwise continues working normally)Cc David Gibson <[email protected]>,Cc Alexey Kardashevskiy <[email protected]>Cc "Michael S. Tsirkin" <[email protected]>Signed-off-by Michael Roth <[email protected]>--- hw/virtio/virtio-pci.c | 12 ++++++++---- hw/virtio/virtio.c | 7 ++++++- include/hw/virtio/virtio.h | 1 + 3 files changed, 15 insertions(+), 5 deletions(-)hw/virtio/virtio-pci.c +8
-4@@ -608,10 +608,14 @@ static void virtio_write_config(PCIDevice *pci_dev, uint32_t address,pcie_cap_flr_write_config (pci_dev, address, val, len); } -if (range_covers_byte (address, len,PCI_COMMAND ) && - !(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { -virtio_pci_stop_ioeventfd (proxy); -virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); +if (range_covers_byte (address, len,PCI_COMMAND )) { +if (!(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { +virtio_set_disabled (vdev, true); +virtio_pci_stop_ioeventfd (proxy); +virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); + }else { +virtio_set_disabled (vdev, false); + } } if (proxy->config_cap &&hw/virtio/virtio.c +0
-0@@ -2575,6 +2575,11 @@ void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) vdev->bus_name =g_strdup (bus_name); }Quoted 15 lines
hw/virtio/virtio.c void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) · line 2585
Quoted 8 lines
hw/virtio/virtio.c +5
-0+void virtio_set_disabled (VirtIODevice *vdev,bool disable) +{ + vdev->broken = disable; +} + void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...) {va_list ap;Hmm. I think just clear and immediate set of bus master while device was not doing any DMA actually should be fine and should not require a reset. Now it's true that right now guests reset first thing which will clear the broken flag, but I'd say it's wrong to require this specific order.
I think the easiest thing is to add a "disabled" flag.Agreed. Some comments on that below.
Quoted 22 lines
Quoted 22 lines
hw/virtio/virtio.c +1
-1@@ -2588,7 +2593,7 @@ void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...)virtio_notify_config (vdev); } - vdev->broken = true; +virtio_set_disabled (vdev, true); }static void virtio_memory_listener_commit (MemoryListener *listener)include/hw/virtio/virtio.h +1
-0@@ -168,6 +168,7 @@ void virtio_init(VirtIODevice *vdev, const char *name, uint16_t device_id, size_t config_size);void virtio_cleanup (VirtIODevice *vdev); +void virtio_set_disabled (VirtIODevice *vdev,bool disable); void virtio_error(VirtIODevice *vdev, const char *fmt, ...) GCC_FMT_ATTR(2, 3);Quoted 6 lines
include/hw/virtio/virtio.h void virtio_init(VirtIODevice *vdev, const char *name, · line 174
Quoted 1 line
include/hw/virtio/virtio.h +0
-0/* Set the child bus name. */ One open question here is cross version migration. What exactly happens if we migrate to an old qemu? from an old qemu?
Currently
vdev->brokenis already migrated via a subsection if it is set.So, with the current proposed patch I would expect the following:
old -> new:
dev->brokenset due to proposedvirtio_set_disabled():
- N/A
dev->brokenset due to other/existing reasons:
- value is migrated, migration succeeds
- device continues not working until a reset, or bus-mastering enabled/re-enabled. (which is good, but also makes me realize that a device that was put into broken state for reasons other thanvirtio_set_disabled()should not get 'unbroken' simply because bus-master bit was cycled. a separate flag is probably needed)
- PASS
dev->brokennot set:
- not migrated
- PASS
new -> old:
dev->brokenset due to proposedvirtio_set_disabled():
- migration succeeds to any QEMU that already has handling fordev->broken.
- migration fails for any QEMU that doesn't have handling fordev->brokenas it does now, but will a higher chance of triggering
- device won't work until it is reset. for most guests we will probably get a reset before the device is used again anyway. for the SLOF case the device will stay broken after bus-mastering re-enabled, but that's the case for existing QEMU now anyway
- PASS, but with increased chance of migration breakage for QEMU's that don't have handling fordev->broken.
dev->brokenset due to other/existing reasons:
- migration succeeds to any QEMU that already has handling fordev->broken. device will require a reset as it does now.
- migration fails for any QEMU that doesn't have handling fordev->brokenas it does now
- PASS
dev->brokennot set:
- not migrated
- PASSWith a new
dev->disabledflag, which we'd likely send using a subsection like withdev->broken, I would expect the following:old -> new:
dev->disabledset:
- N/A
dev->disablednot set:
- not migrated
- if source disables BM just before migration we will likely go into an error state that either setsdev->brokenor puts the device in some other possibly bad state. either case would require subsequent reset, just as they would without migration
- PASS
new -> old:
dev->disabledset:
- migration stream fails
- FAIL, but as expected, and doesn't seem common currently outside of SLOF early-boot else we'd probably have more reports of breakage from vring access while BM isn't set.
dev->brokennot set:
- not migrated
- PASSSo re-using 'broken' is slightly better from a migration standpoint, but as noted above it is probably wrong to unset 'broken' just because BM bit gets cycled, so a new 'disabled' flag is probably needed.
We could probably get by with just adding a check for
dev->disabledinvirtio_queue_empty(), or even earlier invirtio_queue_host_notifier_aio_poll(), but it seems more proper to also add it in most of the same places we currently check fordev->broken.
Quoted 4 lines
That seems somewhat redundant though, so I think maybe the best approach is to:
- replace most
dev->brokenchecks with checks fordev->disabled
I propose calling an inline functions so we are not tied to a specific field.
Quoted 13 lines
- set
dev->disabledwheneverdev->brokengets set
- add a check invirtio_set_disabled()that only allows us to re-enable ifdev->brokenhasn't also been set.I'll work on a follow-up using that approach if it seems reasonable to you.
Thanks!
Quoted 4 lines
Quoted 3 lines
Signature · 2 lines
-- 2.17.1
Quoted 283 lines · Michael S. Tsirkin
Quoted 282 lines · Michael Roth
Quoting Michael S. Tsirkin (2019-11-13 04:09:02)
Quoted 160 lines · Michael Roth
Quoted 159 lines
Currently the SLOF firmware for pseries guests will disable/re-enable a PCI device multiple times via IO/MEM/MASTER bits of
PCI_COMMANDregister after the initial probe/feature negotiation, as it tends to work with a single device at a time at various stages like probing and running block/network bootloaders without doing a full reset in-between.In QEMU, when
PCI_COMMAND_MASTERis disabled we disable the corresponding IOMMU memory region, so DMA accesses (including to vring fields like idx/flags) will no longer undergo the necessary translation. Normally we wouldn't expect this to happen since it would be misbehavior on the driver side to continue driving DMA requests.However, in the case of pseries, with iommu_platform=on, we trigger the following sequence when tearing down the virtio-blk dataplane ioeventfd in response to the guest unsetting
PCI_COMMAND_MASTER:#2
0x0000555555922651invirtqueue_map_desc(vdev=vdev@entry=0x555556dbcfb0, p_num_sg=p_num_sg@entry=0x7fffe657e1a8, addr=addr@entry=0x7fffe657e240, iov=iov@entry=0x7fffe6580240, max_num_sg=max_num_sg@entry=1024, is_write=is_write@entry=false, pa=0, sz=0)at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:757 #3 0x0000555555922a89 in virtqueue_pop (vq=vq@entry=0x555556dc8660, sz=sz@entry=184) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:950 #4 0x00005555558d3eca in virtio_blk_get_request (vq=0x555556dc8660, s=0x555556dbcfb0) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:255 #5 0x00005555558d3eca in virtio_blk_handle_vq (s=0x555556dbcfb0, vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:776 #6 0x000055555591dd66 in virtio_queue_notify_aio_vq (vq=vq@entry=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1550 #7 0x000055555591ecef in virtio_queue_notify_aio_vq (vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1546 #8 0x000055555591ecef in virtio_queue_host_notifier_aio_poll (opaque=0x555556dc86c8) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:2527 #9 0x0000555555d02164 in run_poll_handlers_once (ctx=ctx@entry=0x55555688bfc0, timeout=timeout@entry=0x7fffe65844a8) at /home/mdroth/w/qemu.git/util/aio-posix.c:520 #10 0x0000555555d02d1b in try_poll_mode (timeout=0x7fffe65844a8, ctx=0x55555688bfc0) at /home/mdroth/w/qemu.git/util/aio-posix.c:607 #11 0x0000555555d02d1b in aio_poll (ctx=ctx@entry=0x55555688bfc0, blocking=blocking@entry=true) at /home/mdroth/w/qemu.git/util/aio-posix.c:639 #12 0x0000555555d0004d in aio_wait_bh_oneshot (ctx=0x55555688bfc0, cb=cb@entry=0x5555558d5130 <virtio_blk_data_plane_stop_bh>, opaque=opaque@entry=0x555556de86f0) at /home/mdroth/w/qemu.git/util/aio-wait.c:71 #13 0x00005555558d59bf in virtio_blk_data_plane_stop (vdev=<optimized out>) at /home/mdroth/w/qemu.git/hw/block/dataplane/virtio-blk.c:288 #14 0x0000555555b906a1 in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:245 #15 0x0000555555b90dbb in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:237 #16 0x0000555555b92a8e in virtio_pci_stop_ioeventfd (proxy=0x555556db4e40) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:292 #17 0x0000555555b92a8e in virtio_write_config (pci_dev=0x555556db4e40, address=<optimized out>, val=1048832, len=<optimized out>) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:613I.e. the calling code is only scheduling a one-shot BH for
virtio_blk_data_plane_stop_bh, but somehow we end up trying to process an additional virtqueue entry before we get there. This is likely due to the following check invirtio_queue_host_notifier_aio_poll:static bool virtio_queue_host_notifier_aio_poll(void *opaque) { EventNotifier *n = opaque; VirtQueue *vq = container_of(n, VirtQueue, host_notifier); bool progress;if (!vq->vring.desc || virtio_queue_empty(vq)) { return false; } progress = virtio_queue_notify_aio_vq(vq);namely the call to
virtio_queue_empty(). In this case, since no new requests have actually been issued,shadow_avail_idx==last_avail_idx, so we actually try to access the vring viavring_avail_idx()to get the latest non-shadowed idx:int virtio_queue_empty(VirtQueue *vq) { bool empty; ...if (vq->shadow_avail_idx != vq->last_avail_idx) { return 0; } rcu_read_lock(); empty = vring_avail_idx(vq) == vq->last_avail_idx; rcu_read_unlock(); return empty;but since the IOMMU region has been disabled we get a bogus value (0 usually), which causes
virtio_queue_empty()to falsely report that there are entries to be processed, which causes errors such as:"virtio: zero sized buffers are not allowed"
or
"virtio-blk missing headers"
and puts the device in an error state.
This patch works around the issue by introducing
virtio_set_disabled(), which piggy-backs off thevdev->brokenflag we already use to bypass checks likevirtio_queue_empty(), and sets/unsets it in response to enabling/disabling bus-mastering.NOTES:
- It's possible we could also work around this in SLOF by doing a full reset instead of relying on
PCI_COMMANDto enable/disable, but in any case the QEMU behavior seems wrong.
- This leaves some other oddities in play, like the fact thatDRIVER_OKalso gets unset in response to bus-mastering being disabled, but not restored (however the device seems to continue working)
- Similarly, we disable the host notifier viavirtio_bus_stop_ioeventfd(), which seems to move the handling out of virtio-blk dataplane and back into the main IO thread, and it ends up staying there till a reset (but otherwise continues working normally)Cc David Gibson <[email protected]>,Cc Alexey Kardashevskiy <[email protected]>Cc "Michael S. Tsirkin" <[email protected]>Signed-off-by Michael Roth <[email protected]>--- hw/virtio/virtio-pci.c | 12 ++++++++---- hw/virtio/virtio.c | 7 ++++++- include/hw/virtio/virtio.h | 1 + 3 files changed, 15 insertions(+), 5 deletions(-)hw/virtio/virtio-pci.c +8
-4@@ -608,10 +608,14 @@ static void virtio_write_config(PCIDevice *pci_dev, uint32_t address,pcie_cap_flr_write_config (pci_dev, address, val, len); } -if (range_covers_byte (address, len,PCI_COMMAND ) && - !(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { -virtio_pci_stop_ioeventfd (proxy); -virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); +if (range_covers_byte (address, len,PCI_COMMAND )) { +if (!(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { +virtio_set_disabled (vdev, true); +virtio_pci_stop_ioeventfd (proxy); +virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); + }else { +virtio_set_disabled (vdev, false); + } } if (proxy->config_cap &&hw/virtio/virtio.c +0
-0@@ -2575,6 +2575,11 @@ void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) vdev->bus_name =g_strdup (bus_name); }Quoted 15 lines
hw/virtio/virtio.c void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) · line 2585
Quoted 8 lines
hw/virtio/virtio.c +5
-0+void virtio_set_disabled (VirtIODevice *vdev,bool disable) +{ + vdev->broken = disable; +} + void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...) {va_list ap;Hmm. I think just clear and immediate set of bus master while device was not doing any DMA actually should be fine and should not require a reset. Now it's true that right now guests reset first thing which will clear the broken flag, but I'd say it's wrong to require this specific order.
I think the easiest thing is to add a "disabled" flag.Agreed. Some comments on that below.
Quoted 22 lines
Quoted 22 lines
hw/virtio/virtio.c +1
-1@@ -2588,7 +2593,7 @@ void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...)virtio_notify_config (vdev); } - vdev->broken = true; +virtio_set_disabled (vdev, true); }static void virtio_memory_listener_commit (MemoryListener *listener)include/hw/virtio/virtio.h +1
-0@@ -168,6 +168,7 @@ void virtio_init(VirtIODevice *vdev, const char *name, uint16_t device_id, size_t config_size);void virtio_cleanup (VirtIODevice *vdev); +void virtio_set_disabled (VirtIODevice *vdev,bool disable); void virtio_error(VirtIODevice *vdev, const char *fmt, ...) GCC_FMT_ATTR(2, 3);Quoted 6 lines
include/hw/virtio/virtio.h void virtio_init(VirtIODevice *vdev, const char *name, · line 174
Quoted 1 line
include/hw/virtio/virtio.h +0
-0/* Set the child bus name. */ One open question here is cross version migration. What exactly happens if we migrate to an old qemu? from an old qemu?
Currently
vdev->brokenis already migrated via a subsection if it is set.So, with the current proposed patch I would expect the following:
old -> new:
dev->brokenset due to proposedvirtio_set_disabled():
- N/A
dev->brokenset due to other/existing reasons:
- value is migrated, migration succeeds
- device continues not working until a reset, or bus-mastering enabled/re-enabled. (which is good, but also makes me realize that a device that was put into broken state for reasons other thanvirtio_set_disabled()should not get 'unbroken' simply because bus-master bit was cycled. a separate flag is probably needed)
- PASS
dev->brokennot set:
- not migrated
- PASS
new -> old:
dev->brokenset due to proposedvirtio_set_disabled():
- migration succeeds to any QEMU that already has handling fordev->broken.
- migration fails for any QEMU that doesn't have handling fordev->brokenas it does now, but will a higher chance of triggering
- device won't work until it is reset. for most guests we will probably get a reset before the device is used again anyway. for the SLOF case the device will stay broken after bus-mastering re-enabled, but that's the case for existing QEMU now anyway
- PASS, but with increased chance of migration breakage for QEMU's that don't have handling fordev->broken.
dev->brokenset due to other/existing reasons:
- migration succeeds to any QEMU that already has handling fordev->broken. device will require a reset as it does now.
- migration fails for any QEMU that doesn't have handling fordev->brokenas it does now
- PASS
dev->brokennot set:
- not migrated
- PASSWith a new
dev->disabledflag, which we'd likely send using a subsection like withdev->broken, I would expect the following:old -> new:
dev->disabledset:
- N/A
dev->disablednot set:
- not migrated
- if source disables BM just before migration we will likely go into an error state that either setsdev->brokenor puts the device in some other possibly bad state. either case would require subsequent reset, just as they would without migration
- PASS
new -> old:
dev->disabledset:
- migration stream fails
- FAIL, but as expected, and doesn't seem common currently outside of SLOF early-boot else we'd probably have more reports of breakage from vring access while BM isn't set.
dev->brokennot set:
- not migrated
- PASSSo re-using 'broken' is slightly better from a migration standpoint, but as noted above it is probably wrong to unset 'broken' just because BM bit gets cycled, so a new 'disabled' flag is probably needed.
We could probably get by with just adding a check for
dev->disabledinvirtio_queue_empty(), or even earlier invirtio_queue_host_notifier_aio_poll(), but it seems more proper to also add it in most of the same places we currently check fordev->broken.
Quoted 14 lines
Quoted 4 lines
That seems somewhat redundant though, so I think maybe the best approach is to:
- replace most
dev->brokenchecks with checks fordev->disabledI propose calling an inline functions so we are not tied to a specific field.
Quoted 6 lines
- set
dev->disabledwheneverdev->brokengets set
- add a check invirtio_set_disabled()that only allows us to re-enable ifdev->brokenhasn't also been set.I'll work on a follow-up using that approach if it seems reasonable to you.
Also, since the new field is not migrated anyway, how about a flag that old machine types set, and behave the way we do now?
Quoted 7 lines
Quoted 7 lines
Thanks!
Quoted 4 lines
Quoted 3 lines
Signature · 2 lines
-- 2.17.1
Quoting Michael S. Tsirkin (2019-11-14 03:12:00)
Quoted 284 lines · Michael S. Tsirkin
Quoted 283 lines · Michael Roth
Quoted 282 lines
Quoting Michael S. Tsirkin (2019-11-13 04:09:02)
Quoted 160 lines · Michael Roth
Quoted 159 lines
Currently the SLOF firmware for pseries guests will disable/re-enable a PCI device multiple times via IO/MEM/MASTER bits of
PCI_COMMANDregister after the initial probe/feature negotiation, as it tends to work with a single device at a time at various stages like probing and running block/network bootloaders without doing a full reset in-between.In QEMU, when
PCI_COMMAND_MASTERis disabled we disable the corresponding IOMMU memory region, so DMA accesses (including to vring fields like idx/flags) will no longer undergo the necessary translation. Normally we wouldn't expect this to happen since it would be misbehavior on the driver side to continue driving DMA requests.However, in the case of pseries, with iommu_platform=on, we trigger the following sequence when tearing down the virtio-blk dataplane ioeventfd in response to the guest unsetting
PCI_COMMAND_MASTER:#2
0x0000555555922651invirtqueue_map_desc(vdev=vdev@entry=0x555556dbcfb0, p_num_sg=p_num_sg@entry=0x7fffe657e1a8, addr=addr@entry=0x7fffe657e240, iov=iov@entry=0x7fffe6580240, max_num_sg=max_num_sg@entry=1024, is_write=is_write@entry=false, pa=0, sz=0)at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:757 #3 0x0000555555922a89 in virtqueue_pop (vq=vq@entry=0x555556dc8660, sz=sz@entry=184) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:950 #4 0x00005555558d3eca in virtio_blk_get_request (vq=0x555556dc8660, s=0x555556dbcfb0) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:255 #5 0x00005555558d3eca in virtio_blk_handle_vq (s=0x555556dbcfb0, vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/block/virtio-blk.c:776 #6 0x000055555591dd66 in virtio_queue_notify_aio_vq (vq=vq@entry=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1550 #7 0x000055555591ecef in virtio_queue_notify_aio_vq (vq=0x555556dc8660) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:1546 #8 0x000055555591ecef in virtio_queue_host_notifier_aio_poll (opaque=0x555556dc86c8) at /home/mdroth/w/qemu.git/hw/virtio/virtio.c:2527 #9 0x0000555555d02164 in run_poll_handlers_once (ctx=ctx@entry=0x55555688bfc0, timeout=timeout@entry=0x7fffe65844a8) at /home/mdroth/w/qemu.git/util/aio-posix.c:520 #10 0x0000555555d02d1b in try_poll_mode (timeout=0x7fffe65844a8, ctx=0x55555688bfc0) at /home/mdroth/w/qemu.git/util/aio-posix.c:607 #11 0x0000555555d02d1b in aio_poll (ctx=ctx@entry=0x55555688bfc0, blocking=blocking@entry=true) at /home/mdroth/w/qemu.git/util/aio-posix.c:639 #12 0x0000555555d0004d in aio_wait_bh_oneshot (ctx=0x55555688bfc0, cb=cb@entry=0x5555558d5130 <virtio_blk_data_plane_stop_bh>, opaque=opaque@entry=0x555556de86f0) at /home/mdroth/w/qemu.git/util/aio-wait.c:71 #13 0x00005555558d59bf in virtio_blk_data_plane_stop (vdev=<optimized out>) at /home/mdroth/w/qemu.git/hw/block/dataplane/virtio-blk.c:288 #14 0x0000555555b906a1 in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:245 #15 0x0000555555b90dbb in virtio_bus_stop_ioeventfd (bus=bus@entry=0x555556dbcf38) at /home/mdroth/w/qemu.git/hw/virtio/virtio-bus.c:237 #16 0x0000555555b92a8e in virtio_pci_stop_ioeventfd (proxy=0x555556db4e40) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:292 #17 0x0000555555b92a8e in virtio_write_config (pci_dev=0x555556db4e40, address=<optimized out>, val=1048832, len=<optimized out>) at /home/mdroth/w/qemu.git/hw/virtio/virtio-pci.c:613I.e. the calling code is only scheduling a one-shot BH for
virtio_blk_data_plane_stop_bh, but somehow we end up trying to process an additional virtqueue entry before we get there. This is likely due to the following check invirtio_queue_host_notifier_aio_poll:static bool virtio_queue_host_notifier_aio_poll(void *opaque) { EventNotifier *n = opaque; VirtQueue *vq = container_of(n, VirtQueue, host_notifier); bool progress;if (!vq->vring.desc || virtio_queue_empty(vq)) { return false; } progress = virtio_queue_notify_aio_vq(vq);namely the call to
virtio_queue_empty(). In this case, since no new requests have actually been issued,shadow_avail_idx==last_avail_idx, so we actually try to access the vring viavring_avail_idx()to get the latest non-shadowed idx:int virtio_queue_empty(VirtQueue *vq) { bool empty; ...if (vq->shadow_avail_idx != vq->last_avail_idx) { return 0; } rcu_read_lock(); empty = vring_avail_idx(vq) == vq->last_avail_idx; rcu_read_unlock(); return empty;but since the IOMMU region has been disabled we get a bogus value (0 usually), which causes
virtio_queue_empty()to falsely report that there are entries to be processed, which causes errors such as:"virtio: zero sized buffers are not allowed"
or
"virtio-blk missing headers"
and puts the device in an error state.
This patch works around the issue by introducing
virtio_set_disabled(), which piggy-backs off thevdev->brokenflag we already use to bypass checks likevirtio_queue_empty(), and sets/unsets it in response to enabling/disabling bus-mastering.NOTES:
- It's possible we could also work around this in SLOF by doing a full reset instead of relying on
PCI_COMMANDto enable/disable, but in any case the QEMU behavior seems wrong.
- This leaves some other oddities in play, like the fact thatDRIVER_OKalso gets unset in response to bus-mastering being disabled, but not restored (however the device seems to continue working)
- Similarly, we disable the host notifier viavirtio_bus_stop_ioeventfd(), which seems to move the handling out of virtio-blk dataplane and back into the main IO thread, and it ends up staying there till a reset (but otherwise continues working normally)Cc David Gibson <[email protected]>,Cc Alexey Kardashevskiy <[email protected]>Cc "Michael S. Tsirkin" <[email protected]>Signed-off-by Michael Roth <[email protected]>--- hw/virtio/virtio-pci.c | 12 ++++++++---- hw/virtio/virtio.c | 7 ++++++- include/hw/virtio/virtio.h | 1 + 3 files changed, 15 insertions(+), 5 deletions(-)hw/virtio/virtio-pci.c +8
-4@@ -608,10 +608,14 @@ static void virtio_write_config(PCIDevice *pci_dev, uint32_t address,pcie_cap_flr_write_config (pci_dev, address, val, len); } -if (range_covers_byte (address, len,PCI_COMMAND ) && - !(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { -virtio_pci_stop_ioeventfd (proxy); -virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); +if (range_covers_byte (address, len,PCI_COMMAND )) { +if (!(pci_dev->config[PCI_COMMAND ] &PCI_COMMAND_MASTER )) { +virtio_set_disabled (vdev, true); +virtio_pci_stop_ioeventfd (proxy); +virtio_set_status (vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK ); + }else { +virtio_set_disabled (vdev, false); + } } if (proxy->config_cap &&hw/virtio/virtio.c +0
-0@@ -2575,6 +2575,11 @@ void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) vdev->bus_name =g_strdup (bus_name); }Quoted 15 lines
hw/virtio/virtio.c void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name) · line 2585
Quoted 8 lines
hw/virtio/virtio.c +5
-0+void virtio_set_disabled (VirtIODevice *vdev,bool disable) +{ + vdev->broken = disable; +} + void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...) {va_list ap;Hmm. I think just clear and immediate set of bus master while device was not doing any DMA actually should be fine and should not require a reset. Now it's true that right now guests reset first thing which will clear the broken flag, but I'd say it's wrong to require this specific order.
I think the easiest thing is to add a "disabled" flag.Agreed. Some comments on that below.
Quoted 22 lines
Quoted 22 lines
hw/virtio/virtio.c +1
-1@@ -2588,7 +2593,7 @@ void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...)virtio_notify_config (vdev); } - vdev->broken = true; +virtio_set_disabled (vdev, true); }static void virtio_memory_listener_commit (MemoryListener *listener)include/hw/virtio/virtio.h +1
-0@@ -168,6 +168,7 @@ void virtio_init(VirtIODevice *vdev, const char *name, uint16_t device_id, size_t config_size);void virtio_cleanup (VirtIODevice *vdev); +void virtio_set_disabled (VirtIODevice *vdev,bool disable); void virtio_error(VirtIODevice *vdev, const char *fmt, ...) GCC_FMT_ATTR(2, 3);Quoted 6 lines
include/hw/virtio/virtio.h void virtio_init(VirtIODevice *vdev, const char *name, · line 174
Quoted 1 line
include/hw/virtio/virtio.h +0
-0/* Set the child bus name. */ One open question here is cross version migration. What exactly happens if we migrate to an old qemu? from an old qemu?
Currently
vdev->brokenis already migrated via a subsection if it is set.So, with the current proposed patch I would expect the following:
old -> new:
dev->brokenset due to proposedvirtio_set_disabled():
- N/A
dev->brokenset due to other/existing reasons:
- value is migrated, migration succeeds
- device continues not working until a reset, or bus-mastering enabled/re-enabled. (which is good, but also makes me realize that a device that was put into broken state for reasons other thanvirtio_set_disabled()should not get 'unbroken' simply because bus-master bit was cycled. a separate flag is probably needed)
- PASS
dev->brokennot set:
- not migrated
- PASS
new -> old:
dev->brokenset due to proposedvirtio_set_disabled():
- migration succeeds to any QEMU that already has handling fordev->broken.
- migration fails for any QEMU that doesn't have handling fordev->brokenas it does now, but will a higher chance of triggering
- device won't work until it is reset. for most guests we will probably get a reset before the device is used again anyway. for the SLOF case the device will stay broken after bus-mastering re-enabled, but that's the case for existing QEMU now anyway
- PASS, but with increased chance of migration breakage for QEMU's that don't have handling fordev->broken.
dev->brokenset due to other/existing reasons:
- migration succeeds to any QEMU that already has handling fordev->broken. device will require a reset as it does now.
- migration fails for any QEMU that doesn't have handling fordev->brokenas it does now
- PASS
dev->brokennot set:
- not migrated
- PASSWith a new
dev->disabledflag, which we'd likely send using a subsection like withdev->broken, I would expect the following:old -> new:
dev->disabledset:
- N/A
dev->disablednot set:
- not migrated
- if source disables BM just before migration we will likely go into an error state that either setsdev->brokenor puts the device in some other possibly bad state. either case would require subsequent reset, just as they would without migration
- PASS
new -> old:
dev->disabledset:
- migration stream fails
- FAIL, but as expected, and doesn't seem common currently outside of SLOF early-boot else we'd probably have more reports of breakage from vring access while BM isn't set.
dev->brokennot set:
- not migrated
- PASSSo re-using 'broken' is slightly better from a migration standpoint, but as noted above it is probably wrong to unset 'broken' just because BM bit gets cycled, so a new 'disabled' flag is probably needed.
We could probably get by with just adding a check for
dev->disabledinvirtio_queue_empty(), or even earlier invirtio_queue_host_notifier_aio_poll(), but it seems more proper to also add it in most of the same places we currently check fordev->broken.
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
That seems somewhat redundant though, so I think maybe the best approach is to:
- replace most
dev->brokenchecks with checks fordev->disabledI propose calling an inline functions so we are not tied to a specific field.
Makes sense.
Quoted 10 lines
Quoted 6 lines
Quoted 6 lines
- set
dev->disabledwheneverdev->brokengets set
- add a check invirtio_set_disabled()that only allows us to re-enable ifdev->brokenhasn't also been set.I'll work on a follow-up using that approach if it seems reasonable to you.
Also, since the new field is not migrated anyway, how about a flag that old machine types set, and behave the way we do now?
Hmm, you're right that in the above cases there's only one case where we would migrate (new->old, when dev->disabled is set), and this would break the migration stream anyway, so there's no reason to migrate. But I forgot to include the more common case of new->new:
new -> new:
dev->disabled set:
- migrated
- PASS
dev->disabled not set:
- not migrated
- PASS
So I was planning on migrating it via subsection. But it does make sense to only use it for newer machine types so we don't trigger the failure for (new->old && dev->disabled == true). Maybe that's what you were suggesting in the first place?
I'll give that a shot.
Quoted 8 lines
Quoted 7 lines
Quoted 7 lines
Thanks!
Quoted 4 lines
Quoted 3 lines
Signature · 2 lines
-- 2.17.1