[PATCH 5.4 035/118] drm/ttm: flush the fence on the bo after we individualize the reservation object

From: Greg Kroah-Hartman
Date: Wed Apr 22 2020 - 06:43:44 EST


From: xinhui pan <xinhui.pan@xxxxxxx>

[ Upstream commit 1bbcf69e42fe7fd49b6f4339c970729d0e343753 ]

As we move the ttm_bo_individualize_resv() upwards, we need flush the
copied fence too. Otherwise the driver keeps waiting for fence.

run&Kill kfdtest, then perf top.

25.53% [ttm] [k] ttm_bo_delayed_delete
24.29% [kernel] [k] dma_resv_test_signaled_rcu
19.72% [kernel] [k] ww_mutex_lock

Fix: 378e2d5b("drm/ttm: fix ttm_bo_cleanup_refs_or_queue once more")
Signed-off-by: xinhui pan <xinhui.pan@xxxxxxx>
Reviewed-by: Christian KÃnig <christian.koenig@xxxxxxx>
Link: https://patchwork.freedesktop.org/series/72339/
Signed-off-by: Christian KÃnig <christian.koenig@xxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/gpu/drm/ttm/ttm_bo.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index f078036998092..abf165b2f64fc 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -517,8 +517,10 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)

dma_resv_unlock(bo->base.resv);
}
- if (bo->base.resv != &bo->base._resv)
+ if (bo->base.resv != &bo->base._resv) {
+ ttm_bo_flush_all_fences(bo);
dma_resv_unlock(&bo->base._resv);
+ }

error:
kref_get(&bo->list_kref);
--
2.20.1