On Thu, Aug 24, 2017, at 11:10, Xavier Bouchoux wrote: > Otherwise radv_cmd_state_setup_attachments() will complain it has no > clearvalues, > when called via radv_process_depth_image_inplace(). > > Signed-off-by: Xavier Bouchoux <[email protected]> > --- > src/amd/vulkan/radv_meta_decompress.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/src/amd/vulkan/radv_meta_decompress.c > b/src/amd/vulkan/radv_meta_decompress.c > index f68ce8d2b0..f932b4c492 100644 > --- a/src/amd/vulkan/radv_meta_decompress.c > +++ b/src/amd/vulkan/radv_meta_decompress.c > @@ -38,10 +38,13 @@ create_pass(struct radv_device *device, > const VkAllocationCallbacks *alloc = &device->meta_state.alloc; > VkAttachmentDescription attachment; > > + attachment.flags = 0; > attachment.format = VK_FORMAT_D32_SFLOAT_S8_UINT; > attachment.samples = samples; > attachment.loadOp = VK_ATTACHMENT_LOAD_OP_LOAD; > attachment.storeOp = VK_ATTACHMENT_STORE_OP_STORE; > + attachment.stencilLoadOp = VK_ATTACHMENT_LOAD_OP_DONT_CARE; > + attachment.stencilStoreOp = VK_ATTACHMENT_STORE_OP_DONT_CARE;
I think we should make these LOAD/STORE instead of DONT_CARE, since HTILE decompression needs to preserve stencil. > attachment.initialLayout = > VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL; > attachment.finalLayout = > VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL; > > -- > 2.14.1 > > _______________________________________________ > mesa-dev mailing list > [email protected] > https://lists.freedesktop.org/mailman/listinfo/mesa-dev _______________________________________________ mesa-dev mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/mesa-dev
