On 6/21/22 22:00, Robert Beckett wrote:
By default i915_ttm_cache_level() decides I915_CACHE_LLC if HAS_SNOOP.
This is divergent from existing backends code which only considers
HAS_LLC.
Testing shows that trusting snooping on gen5- is unreliable and bsw via
ggtt mappings, so limit DGFX for now and maintain previous behaviour.
Yeah, IIRC Matthew mentioned that HAS_SNOOP() can be overridden in
various ways, but not on DGFX, (at least not for DG1). So this looks
correct to me.
Signed-off-by: Robert Beckett <bob.beckett@xxxxxxxxxxxxx>
Reviewed-by: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx>
---
drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
index 4c1de0b4a10f..40249fa28a7a 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
@@ -46,7 +46,9 @@ static enum i915_cache_level
i915_ttm_cache_level(struct drm_i915_private *i915, struct ttm_resource *res,
struct ttm_tt *ttm)
{
- return ((HAS_LLC(i915) || HAS_SNOOP(i915)) &&
+ bool can_snoop = HAS_SNOOP(i915) && IS_DGFX(i915);
+
+ return ((HAS_LLC(i915) || can_snoop) &&
!i915_ttm_gtt_binds_lmem(res) &&
ttm->caching == ttm_cached) ? I915_CACHE_LLC :
I915_CACHE_NONE;
