summaryrefslogtreecommitdiff
path: root/src/amd/vulkan
AgeCommit message (Expand)AuthorFilesLines
18 hoursradv: only decompress the depth/stencil aspect that needs to be resolvedSamuel Pitoiset2-13/+26
20 hoursradv: use a workgroup size of 8x8 for FMASK color expandSamuel Pitoiset1-2/+2
20 hoursradv: use the range aspect mask in FMASK color expandSamuel Pitoiset1-1/+1
20 hoursradv: add multi-layer support to FMASK color expandSamuel Pitoiset1-46/+54
21 hoursradv: fix separate depth/stencil layout in render passSamuel Pitoiset1-30/+68
22 hoursradv: enable SQTT support on GFX10.3Samuel Pitoiset1-7/+2
22 hoursac,radv: add SQTT support on GFX10.3Samuel Pitoiset1-29/+34
22 hoursradv: inhibit clock gating when tracing with SQTTSamuel Pitoiset1-0/+23
22 hoursradv: fix overflow when computing the SQTT buffer sizeSamuel Pitoiset1-1/+1
43 hoursradv: sink load_ssboRhys Perry1-3/+3
43 hoursradv: use nir_opt_accessRhys Perry1-0/+6
2 daysradv,aco: use deref_buffer_array_lengthRhys Perry1-0/+1
2 daysradv: don't vectorize shift operationsDaniel Schürmann1-3/+3
2 daysradv: fix a sync issue with geometry shader primitives query on GFX10+Samuel Pitoiset1-14/+16
3 daysradv,aco: don't use MUBUF for multi-channel loads on GFX8 with robustness2Rhys Perry4-2/+13
3 daysradv: remove redundant check in depth_view_can_fast_clear()Samuel Pitoiset1-2/+1
3 daysradv: remove unnecessary radv_image::tc_compatible_htileSamuel Pitoiset2-6/+3
3 daysradv: remove redundant check in radv_process_depth_stencil()Samuel Pitoiset1-3/+0
3 daysradv: fix max_waves estimation on GFX10.3Rhys Perry1-0/+2
4 daysradv: flush L2 for images affected by the pipe misaligned issue on GFX10+Samuel Pitoiset3-5/+77
4 daysradv: restore invalidating the vector cache for internal meta operationsSamuel Pitoiset4-0/+5
4 daysradv: enable TC-compat HTILE for D16S8 on GFX9+Samuel Pitoiset1-14/+16
4 daysradv: enable TC-compat HTILE with D32S8 and MSAA on GFX9+Samuel Pitoiset1-1/+2
4 daysradv: add a comment explaining the micro tile mode resolveSamuel Pitoiset1-0/+4
4 daysradv: flush L2 metadata as part of CB/DB flush instead of CS_DONE on GFX9Samuel Pitoiset1-17/+6
5 daysradv: Add a trivial implementation of VK_KHR_deferred_host_operationBas Nieuwenhuizen2-0/+41
5 daysradv: Do not use a pipe offset for aliased sparse images.Bas Nieuwenhuizen1-1/+2
8 daysradv: do not invalidate the L2 metadata cache on compute queuesSamuel Pitoiset1-0/+1
10 daysnir/loop_unroll: unroll more aggressively if it can improve load schedulingRhys Perry1-0/+1
10 daysradv: vectorize 16bit instructionsDaniel Schürmann2-0/+36
10 daysradv: enable DCC for MSAA on GFX10+21.0-branchpointSamuel Pitoiset1-6/+4
10 daysradv: Use L2 coherency on GFX9+.Bas Nieuwenhuizen3-64/+94
10 daysradv: Use L2 for CP DMA on GFX9+.Bas Nieuwenhuizen1-4/+13
10 daysradv: Use access helpers for flushing with meta operations.Bas Nieuwenhuizen10-72/+96
10 daysradv: Do dst invalidations for write accesses.Bas Nieuwenhuizen1-1/+7
10 daysradv: Invalidate CB on SHADER_WRITE for meta operations.Bas Nieuwenhuizen1-0/+12
10 daysradv: Remove redundant WB_L2 flush.Bas Nieuwenhuizen1-1/+0
10 daysradv: disable VK_EXT_sample_locations again on GFX10+Samuel Pitoiset1-1/+1
10 daysradv: enable DCC for mipmaps on GFX10+Samuel Pitoiset1-7/+2
10 daysradv: do not enable DCC for 3D images with mipmaps on GFX10+Samuel Pitoiset1-0/+6
10 daysradv: add support for fast-clearing DCC levels on GFX10+Samuel Pitoiset1-5/+4
10 daysradv: prevent fast-clearing uncompressed DCC levelsSamuel Pitoiset1-0/+4
10 daysradv: skip fast-clear eliminate for CMASK based on a predicateSamuel Pitoiset1-1/+3
10 daysradv: update the FCE predicate for fast clears using CMASKSamuel Pitoiset1-4/+8
10 daysradv: allocate and initialize the FCE predicate value for CMASK tooSamuel Pitoiset2-4/+6
10 daysradv: only use predication if the FCE value is allocatedSamuel Pitoiset1-1/+1
10 daysradv: fix clearing DCC on GFX9Samuel Pitoiset1-1/+5
11 daysradv: Add Android module info to linker script.Bas Nieuwenhuizen1-0/+5
11 daysradv: enable .lower_inegDaniel Schürmann1-0/+1
11 daysradv: don't lower_pack() after load-store-vectorizationDaniel Schürmann1-4/+1