Re: [Intel-gfx] [PATCH 7/8] drm/i915/fbc: Implement Wa_16011863758 for icl+
Maybe that TODO comment could be moved into the code instead of leaving it just into commit message? Either way, patch look ok to me. Reviewed-by: Juha-Pekka Heikkila On 2.7.2021 23.46, Ville Syrjala wrote: From: Ville Syrjälä There's some kind of weird corner cases in FBC which requires FBC segments to be separated by at least one extra cacheline. Make sure that is present. TODO: the formula laid out in the spec seem to be semi-nonsense so this is mostly my interpretation on what it is actually trying to say. Need to wait for clarification from the hw folks to know for sure. Signed-off-by: Ville Syrjälä --- drivers/gpu/drm/i915/display/intel_fbc.c | 10 ++ 1 file changed, 10 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index 2da5295092e7..daf2191dd3f6 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -88,6 +88,16 @@ static unsigned int intel_fbc_cfb_stride(struct drm_i915_private *i915, { unsigned int stride = _intel_fbc_cfb_stride(cache); + /* +* Wa_16011863758: icl+ +* CFB segment stride needs at least one extra cacheline. +* We make sure each line has an extra cacheline so that +* the 4 line segment will have one regarless of the +* compression limit we choose later. +*/ + if (DISPLAY_VER(i915) >= 11) + stride = max(stride, cache->plane.src_w * 4 + 64u); + /* * At least some of the platforms require each 4 line segment to * be 512 byte aligned. Aligning each line to 512 bytes guarantees
[Intel-gfx] [PATCH 7/8] drm/i915/fbc: Implement Wa_16011863758 for icl+
From: Ville Syrjälä There's some kind of weird corner cases in FBC which requires FBC segments to be separated by at least one extra cacheline. Make sure that is present. TODO: the formula laid out in the spec seem to be semi-nonsense so this is mostly my interpretation on what it is actually trying to say. Need to wait for clarification from the hw folks to know for sure. Signed-off-by: Ville Syrjälä --- drivers/gpu/drm/i915/display/intel_fbc.c | 10 ++ 1 file changed, 10 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index 2da5295092e7..daf2191dd3f6 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -88,6 +88,16 @@ static unsigned int intel_fbc_cfb_stride(struct drm_i915_private *i915, { unsigned int stride = _intel_fbc_cfb_stride(cache); + /* +* Wa_16011863758: icl+ +* CFB segment stride needs at least one extra cacheline. +* We make sure each line has an extra cacheline so that +* the 4 line segment will have one regarless of the +* compression limit we choose later. +*/ + if (DISPLAY_VER(i915) >= 11) + stride = max(stride, cache->plane.src_w * 4 + 64u); + /* * At least some of the platforms require each 4 line segment to * be 512 byte aligned. Aligning each line to 512 bytes guarantees -- 2.31.1 ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx