linux.git/drivers/virtio/virtio_ring.c, branch v5.16

virtio_ring: Fix querying of maximum DMA mapping size for virtio device

2021-12-08T20:04:06+00:00

virtio_max_dma_size() returns the maximum DMA mapping size of the virtio
device by querying dma_max_mapping_size() for the device when the DMA
API is in use for the vring. Unfortunately, the device passed is
initialised by register_virtio_device() and does not inherit the DMA
configuration from its parent, resulting in SWIOTLB errors when bouncing
is enabled and the default 256K mapping limit (IO_TLB_SEGSIZE) is not
respected:

  | virtio-pci 0000:00:01.0: swiotlb buffer is full (sz: 294912 bytes), total 1024 (slots), used 725 (slots)

Follow the pattern used elsewhere in the virtio_ring code when calling
into the DMA layer and pass the parent device to dma_max_mapping_size()
instead.

Cc: Marc Zyngier 
Cc: Quentin Perret 
Cc: "Michael S. Tsirkin" 
Cc: Jason Wang 
Signed-off-by: Will Deacon 
Link: https://lore.kernel.org/r/20211201112018.25276-1-will@kernel.org
Acked-by: Jason Wang 
Tested-by: Suzuki K Poulose 
Fixes: e6d6dd6c875e ("virtio: Introduce virtio_max_dma_size()")
Cc: Joerg Roedel 
Cc: Konrad Rzeszutek Wilk 
Cc: Christoph Hellwig 
Cc: Robin Murphy 
Signed-off-by: Steven Price 
Signed-off-by: Suzuki K Poulose 
Cc: stable@vger.kernel.org
Signed-off-by: Michael S. Tsirkin

Revert "virtio_ring: validate used buffer length"

2021-11-24T23:47:27+00:00

This reverts commit 939779f5152d161b34f612af29e7dc1ac4472fcf.

Attempts to validate length in the core did not work out: there turn out
to exist multiple broken devices, and in particular legacy devices are
known to be broken in this respect.

We have ideas for handling this better in the next version but for now
let's revert to a known good state to make sure drivers work for people.

Signed-off-by: Michael S. Tsirkin

virtio_ring: validate used buffer length

2021-11-01T09:26:48+00:00

This patch validate the used buffer length provided by the device
before trying to use it. This is done by record the in buffer length
in a new field in desc_state structure during virtqueue_add(), then we
can fail the virtqueue_get_buf() when we find the device is trying to
give us a used buffer length which is greater than the in buffer
length.

Since some drivers have already done the validation by themselves,
this patch tries to makes the core validation optional. For the driver
that doesn't want the validation, it can set the
suppress_used_validation to be true (which could be overridden by
force_used_validation module parameter). To be more efficient, a
dedicate array is used for storing the validate used length, this
helps to eliminate the cache stress if validation is done by the
driver.

Signed-off-by: Jason Wang 
Link: https://lore.kernel.org/r/20211027022107.14357-2-jasowang@redhat.com
Signed-off-by: Michael S. Tsirkin

virtio_ring: fix typos in vring_desc_extra

2021-11-01T09:26:48+00:00

We're actually tracking descriptor address and length instead of the
buffer.

Signed-off-by: Jason Wang 
Link: https://lore.kernel.org/r/20211019070152.8236-7-jasowang@redhat.com
Signed-off-by: Michael S. Tsirkin

virtio_ring: check desc == NULL when using indirect with packed

2021-11-01T09:26:47+00:00

When using indirect with packed, we don't check for allocation failures.
This patch checks that and fall back on direct.

Fixes: 1ce9e6055fa0 ("virtio_ring: introduce packed ring support")
Signed-off-by: Xuan Zhuo 
Link: https://lore.kernel.org/r/20211020112323.67466-3-xuanzhuo@linux.alibaba.com
Signed-off-by: Michael S. Tsirkin

virtio_ring: make virtqueue_add_indirect_packed prettier

2021-11-01T09:26:47+00:00

Align the arguments of virtqueue_add_indirect_packed() to the open ( to
make it look prettier.

Signed-off-by: Xuan Zhuo 
Acked-by: Jason Wang 
Link: https://lore.kernel.org/r/20211020112323.67466-2-xuanzhuo@linux.alibaba.com
Signed-off-by: Michael S. Tsirkin

virtio-ring: fix DMA metadata flags

2021-10-27T19:54:34+00:00

The flags are currently overwritten, leading to the wrong direction
being passed to the DMA unmap functions.

Fixes: 72b5e8958738aaa4 ("virtio-ring: store DMA metadata in desc_extra for split virtqueue")
Signed-off-by: Vincent Whitchurch 
Link: https://lore.kernel.org/r/20211026133100.17541-1-vincent.whitchurch@axis.com
Signed-off-by: Michael S. Tsirkin 
Acked-by: Jason Wang

virtio_ring: pull in spinlock header

2021-08-11T10:44:24+00:00

we use a spinlock now pull in the correct header to
make virtio_ring.c self sufficient.

Signed-off-by: Michael S. Tsirkin

virtio: Protect vqs list access

2021-08-10T15:50:55+00:00

VQs may be accessed to mark the device broken while they are
created/destroyed. Hence protect the access to the vqs list.

Fixes: e2dcdfe95c0b ("virtio: virtio_break_device() to mark all virtqueues broken.")
Signed-off-by: Parav Pandit 
Link: https://lore.kernel.org/r/20210721142648.1525924-4-parav@nvidia.com
Signed-off-by: Michael S. Tsirkin

virtio: Keep vring_del_virtqueue() mirror of VQ create

2021-08-10T15:50:55+00:00

Keep the vring_del_virtqueue() mirror of the create routines.
i.e. to delete list entry first as it is added last during the create
routine.

Signed-off-by: Parav Pandit 
Link: https://lore.kernel.org/r/20210721142648.1525924-3-parav@nvidia.com
Signed-off-by: Michael S. Tsirkin