linux.git/drivers/dma/idxd, branch v5.15

dmaengine: idxd: remove interrupt disable for dev_lock

2021-08-29T13:44:20+00:00

The spinlock is not being used in hard interrupt context. There is no need
to disable irq when acquiring the lock. The interrupt thread handler also
is not in bottom half context, therefore we can also remove disabling of
the bh. Convert all dev_lock acquisition to plain spin_lock() calls.

Reviewed-by: Dan Williams 
Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162984026772.1939166.11504067782824765879.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: remove interrupt disable for cmd_lock

2021-08-29T13:44:20+00:00

The cmd_lock spinlock is not being used in hard interrupt context. There is
no need to disable irq when acquiring the lock. Convert all cmd_lock
acquisition to plain spin_lock() calls.

Reviewed-by: Dan Williams 
Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162984027930.1939209.15758413737332339204.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: fix setting up priv mode for dwq

2021-08-29T13:44:20+00:00

DSA spec says WQ priv bit is 0 if the Privileged Mode Enable field of the
PCI Express PASID capability is 0 and pasid is enabled. Make sure that the
WQCFG priv field is set correctly according to usage type. Reject config if
setting up kernel WQ type and no support. Also add the correct priv setup
for a descriptor.

Fixes: 484f910e93b4 ("dmaengine: idxd: fix wq config registers offset programming")
Cc: Ramesh Thomas 
Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162939084657.903168.14160019185148244596.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: set descriptor allocation size to threshold for swq

2021-08-25T11:30:24+00:00

Since submission is sent to limited portal, the actual wq size for shared
wq is set by the threshold rather than the wq size. When the wq type is
shared, set the allocated descriptors to the threshold.

Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162827151733.3459223.3829837172226042408.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: make submit failure path consistent on desc freeing

2021-08-25T11:30:24+00:00

The submission path for dmaengine API does not do descriptor freeing on
failure. Also, with the abort mechanism, the freeing of descriptor happens
when the abort callback is completed. Therefore free descriptor on all
error paths for submission call to make things consistent. Also remove the
double free that would happen on abort in idxd_dma_tx_submit() call.

Fixes: 6b4b87f2c31a ("dmaengine: idxd: fix submission race window")
Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162827146072.3459011.10255348500504659810.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: remove interrupt flag for completion list spinlock

2021-08-06T16:20:00+00:00

The list lock is never acquired in interrupt context. Therefore there is no
need to disable interrupts. Remove interrupt flags for lock operations.

Reviewed-by: Dan Williams 
Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162826417450.3454650.3733188117742416238.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: make I/O interrupt handler one shot

2021-08-06T13:53:06+00:00

The interrupt thread handler currently loops forever to process outstanding
completions. This causes either an "irq X: nobody cared" kernel splat or
the NMI watchdog kicks in due to running too long in the function. The irq
thread handler is expected to run again after exiting if there are
interrupts fired while the thread handler is running. So the handler code
can process all the completed I/O in a single pass and exit without losing
the follow on completed I/O.

Reviewed-by: Dan Williams 
Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162802977005.3084234.11836261157026497585.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: clear block on fault flag when clear wq

2021-08-06T13:53:06+00:00

The block on fault flag is not cleared when we disable or reset wq. This
causes it to remain set if the user does not clear it on the next
configuration load. Add clear of flag in dxd_wq_disable_cleanup()
routine.

Fixes: da32b28c95a7 ("dmaengine: idxd: cleanup workqueue config after disabling")
Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162803023553.3086015.8158952172068868803.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: add capability check for 'block on fault' attribute

2021-08-06T13:53:05+00:00

The device general capability has a bit that indicate whether 'block on
fault' is supported. Add check to wq sysfs knob to check if cap exists
before allowing user to toggle.

Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/162802992615.3084999.12539468940404102898.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: idxd: Remove unused status variable in irq_process_work_list()

2021-08-06T13:45:58+00:00

status is no longer used within this block:

drivers/dma/idxd/irq.c:255:6: warning: unused variable 'status'
[-Wunused-variable]
                u8 status = desc->completion->status & DSA_COMP_STATUS_MASK;
                   ^
1 warning generated.

Fixes: b60bb6e2bfc1 ("dmaengine: idxd: fix abort status check")
Signed-off-by: Nathan Chancellor 
Acked-by: Dave Jiang 
Link: https://lore.kernel.org/r/20210802175820.3153920-1-nathan@kernel.org
Signed-off-by: Vinod Koul