[Intel-gfx] [PATCH 8/9] drm/i915: mark up internal objects with start_cpu_write

Matthew Auld matthew.auld at intel.com
Mon Oct 18 17:45:07 UTC 2021


While the pages can't be swapped out, they can be discarded by the shrinker.
Normally such objects are marked with __I915_MADV_PURGED, which can't be
unset, and therefore requires a new object. For kernel internal objects
this is not true, since the madv hint is reset for our special volatile
objects, such that we can re-acquire new pages, if so desired, without
needing a new object. As a result we should probably be paranoid here
and put the object back into the CPU domain when discarding the pages,
and also correctly set cache_dirty, if required.

Signed-off-by: Matthew Auld <matthew.auld at intel.com>
Cc: Thomas Hellström <thomas.hellstrom at linux.intel.com>
---
 drivers/gpu/drm/i915/gem/i915_gem_internal.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_internal.c b/drivers/gpu/drm/i915/gem/i915_gem_internal.c
index e5ae9c06510c..a57a6b7013c2 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_internal.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_internal.c
@@ -134,6 +134,8 @@ static void i915_gem_object_put_pages_internal(struct drm_i915_gem_object *obj,
 	internal_free_pages(pages);
 
 	obj->mm.dirty = false;
+
+	__start_cpu_write(obj);
 }
 
 static const struct drm_i915_gem_object_ops i915_gem_object_internal_ops = {
-- 
2.26.3



More information about the Intel-gfx mailing list