summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorSamuel Pitoiset <[email protected]>2018-04-11 21:34:43 +0200
committerSamuel Pitoiset <[email protected]>2018-04-12 09:58:46 +0200
commit9eac49246cdc501530418e8bd2a3e6d47173332b (patch)
tree7b25007a8dd14de384d02f8895eba4dee8d6cdce
parentab0e625a671d19bb6a736d663bbc6fa5fd9ecd06 (diff)
radv: fix radv_layout_dcc_compressed() when image doesn't have DCC
num_dcc_levels means that DCC is supported, but this doesn't mean that it's enabled by the driver. Instead, we should rely on radv_image_has_dcc(). This fixes some multisample regressions since 0babc8e5d66 ("radv: fix picking the method for resolve subpass") on Vega. This is because the resolve method changed from HW to FS, but those fails are totally unexpected, so there might some differences between Polaris and Vega here. Fixes: 44fcf587445 ("radv: Disable DCC for GENERAL layout and compute transfer dest.") Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
-rw-r--r--src/amd/vulkan/radv_image.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/src/amd/vulkan/radv_image.c b/src/amd/vulkan/radv_image.c
index acb569203d4..a14e7c18b29 100644
--- a/src/amd/vulkan/radv_image.c
+++ b/src/amd/vulkan/radv_image.c
@@ -1241,7 +1241,7 @@ bool radv_layout_dcc_compressed(const struct radv_image *image,
(queue_mask & (1u << RADV_QUEUE_COMPUTE)))
return false;
- return image->surface.num_dcc_levels > 0 && layout != VK_IMAGE_LAYOUT_GENERAL;
+ return radv_image_has_dcc(image) && layout != VK_IMAGE_LAYOUT_GENERAL;
}