diff options
| author | Chris Wilson <chris@chris-wilson.co.uk> | 2019-05-28 10:29:51 +0100 |
|---|---|---|
| committer | Chris Wilson <chris@chris-wilson.co.uk> | 2019-05-28 12:45:29 +0100 |
| commit | 6951e5893b4821f68a48022842f67c3033ca7b30 (patch) | |
| tree | bc7d39fb7ae0b8ba3967646059ac40b931ae3ed8 /drivers/gpu/drm/i915/i915_gem_render_state.c | |
| parent | 37d63f8fdb4941b6a82ef9e59bee62a494225c7e (diff) | |
| download | linux-6951e5893b4821f68a48022842f67c3033ca7b30.tar.gz linux-6951e5893b4821f68a48022842f67c3033ca7b30.tar.bz2 linux-6951e5893b4821f68a48022842f67c3033ca7b30.zip | |
drm/i915: Move GEM object domain management from struct_mutex to local
Use the per-object local lock to control the cache domain of the
individual GEM objects, not struct_mutex. This is a huge leap forward
for us in terms of object-level synchronisation; execbuffers are
coordinated using the ww_mutex and pread/pwrite is finally fully
serialised again.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190528092956.14910-10-chris@chris-wilson.co.uk
Diffstat (limited to 'drivers/gpu/drm/i915/i915_gem_render_state.c')
| -rw-r--r-- | drivers/gpu/drm/i915/i915_gem_render_state.c | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/drivers/gpu/drm/i915/i915_gem_render_state.c b/drivers/gpu/drm/i915/i915_gem_render_state.c index f3b42b026fff..706ed71468e8 100644 --- a/drivers/gpu/drm/i915/i915_gem_render_state.c +++ b/drivers/gpu/drm/i915/i915_gem_render_state.c @@ -222,7 +222,9 @@ int i915_gem_render_state_emit(struct i915_request *rq) goto err_unpin; } + i915_vma_lock(so.vma); err = i915_vma_move_to_active(so.vma, rq, 0); + i915_vma_unlock(so.vma); err_unpin: i915_vma_unpin(so.vma); err_vma: |