linux_dsm_epyc7002/drivers/dma-buf
Dominik Behr ea4d5a270b dma-buf/sw_sync: force signal all unsignaled fences on dying timeline
To avoid hanging userspace components that might have been waiting on the
active fences of the destroyed timeline we need to signal with error all
remaining fences on such timeline.

This restore the default behaviour of the Android sw_sync framework, which
Android still relies on. It was broken on the dma fence conversion a few
years ago and never fixed.

v2: Do not bother with cleanup do the list (Chris Wilson)

Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Dominik Behr <dbehr@chromium.org>
Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170907190246.16425-2-gustavo@padovan.org
2017-09-08 15:30:02 -03:00
..
dma-buf.c dma-buf: Adjust a null pointer check in dma_buf_attach() 2017-05-08 20:50:21 +05:30
dma-fence-array.c dma-fence: add dma_fence_match_context helper 2017-03-17 23:51:36 +05:30
dma-fence.c Linux 4.13-rc2 2017-07-27 08:15:43 +10:00
Kconfig dma-buf: Rename struct fence to dma_fence 2016-10-25 14:40:39 +02:00
Makefile dma-buf: Rename struct fence to dma_fence 2016-10-25 14:40:39 +02:00
reservation.c dma-buf: fix reservation_object_wait_timeout_rcu to wait correctly v2 2017-08-14 13:01:25 -04:00
seqno-fence.c dma-buf: Rename struct fence to dma_fence 2016-10-25 14:40:39 +02:00
sw_sync.c dma-buf/sw_sync: force signal all unsignaled fences on dying timeline 2017-09-08 15:30:02 -03:00
sync_debug.c Linux 4.13-rc2 2017-07-27 08:15:43 +10:00
sync_debug.h dma-buf/sw-sync: Use an rbtree to sort fences in the timeline 2017-06-29 18:54:07 -03:00
sync_file.c dma-buf/sync_file: Allow multiple sync_files to wrap a single dma-fence 2017-07-29 11:10:52 -03:00
sync_trace.h dma-buf/sw_sync: de-stage SW_SYNC 2016-08-15 16:41:46 +02:00