Matt Roper <matthew.d.roper@xxxxxxxxx> writes: > On gen11 the XY_FAST_COPY_BLT command has some size restrictions on its > usage. Although this instruction is mainly used by userspace, i915 also > uses it to copy object contents during some selftests, so let's ensure > the restrictions are followed. > > Bspec: 6544 > Signed-off-by: Matt Roper <matthew.d.roper@xxxxxxxxx> > --- > drivers/gpu/drm/i915/gem/i915_gem_object_blt.c | 14 +++++++++++++- > 1 file changed, 13 insertions(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c > index 39b8a055d80a..e00792158f13 100644 > --- a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c > +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c > @@ -196,6 +196,17 @@ int i915_gem_object_fill_blt(struct drm_i915_gem_object *obj, > return err; > } > > +/* Wa_1209644611:icl,ehl */ > +static bool wa_1209644611_applies(struct drm_i915_private *i915, u32 size) > +{ > + u32 height = size >> PAGE_SHIFT; > + > + if (!IS_GEN(i915, 11)) > + return false; > + > + return height % 4 == 3 && height <= 8; The workaround description matches with this. However the original sighting was only for Y-Tiled surfaces. I asked for clarification. If we won't get one, we can play it safe and use this as it is. -Mika > +} > + > struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce, > struct i915_vma *src, > struct i915_vma *dst) > @@ -237,7 +248,8 @@ struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce, > size = min_t(u64, rem, block_size); > GEM_BUG_ON(size >> PAGE_SHIFT > S16_MAX); > > - if (INTEL_GEN(i915) >= 9) { > + if (INTEL_GEN(i915) >= 9 && > + !wa_1209644611_applies(i915, size)) { > *cmd++ = GEN9_XY_FAST_COPY_BLT_CMD | (10 - 2); > *cmd++ = BLT_DEPTH_32 | PAGE_SIZE; > *cmd++ = 0; > -- > 2.24.1 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@xxxxxxxxxxxxxxxxxxxxx > https://lists.freedesktop.org/mailman/listinfo/intel-gfx _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx