Lines Matching full:radv
37 - VK_EXT_image_2d_view_of_3d on RADV
45 - VK_EXT_primitives_generated_query on RADV
47 - VK_EXT_non_seamless_cube_map on RADV, ANV, lavapipe
49 - VK_EXT_border_color_swizzle on lavapipe, ANV, turnip, RADV
51 - VK_EXT_shader_module_identifier on RADV
63 - VK_EXT_attachment_feedback_loop_layout on RADV
98 - [Bisected][RADV] Rendering issues and performance drop with GFX10 on RPCS3
118 - radv: line rasterization mode affects multisampling
121 - radv: dynamic vertex input failure
122 - navi21\\radv\\amdgpu Halo Infinite screen corruption and then hang
134 - radv: some zink cts fails
136 - [RADV, bisected] addrlib changes cause Navi CTS regressions
141 - [RADV][bisected] after commit 124848bf9efa1cc5d8a3cdfc654e0c03d57540e3 the games which use vulkan…
148 - radv: failure in image.sample_texture.*_compressed_format*
151 - radv: Enabling support for instance extensions inside vkCreateInstance does not work
166 - [RADV/ACO] Star Citizen Flickeriing Lights
183 - radv/NIR: Suboptimal codegen for AccessChain on buffer device addresses
187 - Use marketing_name for the Vulkan device string in RADV
189 - radv: Khronos ray tracing sample issues
204 - radv: Conditional rendering on compute queue
215 - radv: no nir validation in radv_link_shaders
224 - radv: Indirect unaligned dispatch
225 - [radv] DCC causes artifacts in Senran Kagura Shinovi Versus
230 - Radv leaks a descriptor set layout
251 - radv/aco: xfb bug
259 - radv,aco: Regression with global atomics with negative offsets
806 - radv/ci: update vkd3d-proton results for AMD's Kabini
847 - radv: always enable NGG on GFX11
848 - radv: Mark all formats as DCC compatible with each other on gfx11.
849 - radv: gfx11 register changes.
850 - radv: Adjust for new SWIZZLE_ENABLE.
851 - radv: Add gfx11 DCC fast clear support.
852 - radv: Disable transform feedback for gfx11.
853 - radv: update the initialization of SGPR0/1 registers for HS and GS on GFX11
854 - radv: Deal with derefs from opaque types in function parameters.
855 - radv: Add performance counter reg write.
856 - radv: move ac_perfcounters to physical_device.
857 - radv: Add pstate locking for perfcounters.
858 - radv: Rename SPM functions.
859 - radv: Make inhibiting clockgating not static.
860 - radv: Add helper BO for perf counters.
861 - radv: Add perf counter lock/unlock commandbuffers.
862 - radv: Add a list of performance counters.
863 - radv: Implement support for querying performance counters.
864 - radv: Implement listing performance counters.
865 - radv: Expose VK_KHR_performance_query.
866 - radv: Add ability to override the build id for the cache.
867 - radv: Add a 32bit memory type.
868 - radv: Skip setting empty index buffers to avoid hang
869 - radv: Expose function to write vertex descriptors for dgc.
870 - radv: Always store stride in the vbo descriptor.
871 - radv: Require 32bit memory for indirect buffers.
872 - radv: Expose helper for base pa_su_sc_mode_cntl.
873 - radv: Make radv_get_vgt_index_size non-static.
874 - radv: Add helper to write scissors.
875 - radv: Add flushing for DGC.
876 - radv: Add DGC meta shader.
877 - radv: Implement DGC generated command layout structure.
878 - radv: Implement DGC cmdbuffer generation.
879 - radv: Implement CmdExecuteGeneratedCommandsNV.
880 - radv: Add stub for vkCmdBindPipelineShaderGroupNV.
881 - radv: Expose VK_NV_device_generated_commands.
894 - radv: fix memory leak of descriptor set layout
895 - radv: refactor image binding into struct
896 - radv: consolidate descriptor set buffer counts
897 - radv: implement disjoint memory for multiplane images
898 - radv: expose disjoint image support
1017 - radv: Kill unused fs_m local var in radv_create_shaders()
1018 - radv: Use vk_pipeline_hash_shader_stage()
1318 - tu: Rewrite cache UUID based on radv
1406 - radv: move two NIR passes out of tight optimization loop
1407 - radv: call nir_lower_vars_to_ssa later
1413 - radv: don't lower vectorized instructions to 32bit
1414 - radv: use callback for nir_lower_to_scalar
1415 - radv: vectorize nir_op_fdiv & nir_op_flrp & nir_op_ineg
1416 - radv: vectorize nir_op_fabs
1417 - radv: fix swizzles after nir_opt_algebraic_late
1418 - radv: reorder some NIR passes
1420 - radv/shader_info: fix load_frag_coord and load_sample_pos read masks
1426 - radv: Only create noop-fs for graphics pipelines
1427 - radv: create RT traversal as separate shader
1428 - radv/rt: use derefs for the traversal stack
1429 - radv/rt: use stage ID as handle for general and closestHit shaders
1430 - radv/rt: Don't load ClosestHit SBT on every hit, but only once after traversal
1431 - radv/rt: fix nir_builder cursor in lower_rt_instructions()
1432 - radv/rt: simplify lower_rt_instructions()
1516 - radv: precalculate hs offchip parameters.
1517 - radv: precalculate tess ring sizes/offsets.
1518 - radv/ac: introduce a new common function for hs calcs.
1523 - radv: cache the slab bo pointer in the pipeline.
1524 - radv: only dirty line stipple when necessary.
1525 - radv: optimised command buffer reset of vertex bindings.
1529 - aco: remove radv specific streamout info
1532 - aco/radv: convert to aco shader info at the radv level.
1533 - aco: remove radv vs prolog key from aco internals.
1534 - aco: drop unused radv include
1535 - ac/radv: drop info pointer from the ac and radv shader structs
1536 - aco/radv: drop radv_nir_compiler_options from aco.
1554 - aco: refactor the radv binary builder out of the core aco fn.
1555 - aco/radv: provide a callback from aco shader building to build binary
1556 - aco/radv: provide a vs prolog callback from aco to radv.
1558 - radv: add a dynamic vertex format cache.
2009 - radv: use updated tokens from vk.xml
2246 - radv: Use nir_fold_16bit_image_load_store_conversions.
2247 - radv: Run copy_prop and dce after folding 16bit sampling/load/store.
2254 - radv: Don't retile read-only images.
2262 - radv: Lower mul_32x16.
2263 - radv: Lower hadd/hradd.
2264 - radv: Lower 8bit isub_sat/usub_sat.
2265 - radv: Enable VK_INTEL_shader_integer_functions2.
2266 - radv, aco: Packed usub_sat/isub_sat.
2268 - radv, aco: Lower txf offset in NIR.
2270 - radv, aco: Round texture array layer in NIR.
2272 - radv: Implement VK_EXT_non_seamless_cube_map.
2276 - radv: Remove redundant memset in radv_descriptor_set_create.
2277 - radv: Check descriptor pool entry count before allocating a new set.
2278 - radv: Don't check if we need to copy immutable samplers for non push templates.
2294 - radv,aco: Don't lower and vectorize 16bit iabs.
2295 - radv: Remove ineg from lower_bit_size_callback callback.
2296 - radv, aco: Don't lower 16bit isign.
2304 - radv: Fold 16bit tex sources.
2469 - radv: Fix RTPSO hashing of pGroups.
2472 - radv: Implement VK_EXT_shader_module_identifier.
2483 - radv: fix assertion for the count of xfb output
2815 - radv: Only use PLANE_0 in meta when actually needed
2816 - radv: Use vk_image as the base for radv_image
2817 - radv: Use vk_image_view as the base for radv_image_view
2824 - radv: Add a sqtt entrypoint for CmdBindVertexBuffers2
2825 - radv: Use the common CmdBindVertexBuffers wrapper
2848 - radv: Use nir_gather_xfb_info
2858 - radv: Use the common image offset/extent sanitize helpers
2859 - vulkan,radv: Move vk_format_get_plane_format to common code
2863 - vulkan,radv: Steal some image offset/extent helpers from radv
2864 - radv: Use vk_image_buffer_copy_layout
2881 - radv: Drop create_sync_for_memory
2966 - radv: Set uses_sample_shading for copy shaders
3022 - radv: Use both aspects for depth/stencil blit destinations
3438 - radv: Fix ray queries with !15854
3439 - radv: Do not discard hits with t=tmax
3440 - radv: Enable rt primitive culling for spirv2nir
3442 - radv: Use VAs in radv_meta_buffer.c
3443 - radv: Make fill_buffer_shader non-static
3444 - radv: Format radv_acceleration_structure.c
3445 - radv: Add the fuchsia radix sort
3446 - radv: Build accaleration structures using LBVH
3447 - radv: Clean up the accel-struct build shaders
3448 - radv: Fix lowering ignore_ray_intersection
3450 - radv: Copy shader modules to avoid use after free
3451 - radv/radix_sort: Add missing entry points
3452 - radv/radix_sort: Make variable names consistent
3453 - radv: Use the entire morton code as sort key
3455 - radv: Move RADV_RT_STAGE_BITS to radv_private.h
3456 - radv: Treat rt stages like compute stages
3457 - radv: Use RADV_RT_STAGE_BITS more often
3459 - radv: Allow indirect dispatches without bo
3460 - radv: Indirect unaligned dispatches
3461 - radv,aco: Use ray_launch_size_addr
3462 - radv: Implement vkCmdTraceRaysIndirectKHR
3463 - radv: Advertise indirect ray tracing
3464 - radv: Allow radv_fill_buffer to work with VAs only
3465 - radv: Use radv_fill_buffer for accel struct builds
3466 - Revert "radv: Make fill_buffer_shader non-static"
3467 - radv: Add and use radv_bvh_node_type
3468 - radv: Replace nir_vector_extract with nir_channel
3469 - radv/ci: Skip slow/crashing ray tracing tests
3470 - radv: Enable ray queries by default
3471 - radv: Add fail/flake lists for navi22
3472 - radv: Fix handling of primitiveOffset
3473 - radv: Ignore transformOffset if transformData is 0
3475 - radv: Use vk_buffer
3476 - radv: Add and use radv_cp_dma_wait_for_stages
3477 - radv: Some acceleration structure cleanups
3478 - radv: Require an alignment of 64 for accel structs
3479 - radv: Fix the accel struct size calculation
3482 - radv: ray cull mask
3483 - radv: Add accel struct queries for maintenance1
3484 - radv: ACCESS_2_SHADER_BINDING_TABLE_READ_BIT_KHR
3485 - radv: PIPELINE_STAGE_2_ACCELERATION_STRUCTURE_COPY
3487 - radv: vkCmdTraceRaysIndirect2KHR
3488 - radv: Advertise VK_KHR_ray_tracing_maintenance1
3491 - radv: Move some rt intrinsics to the top
3492 - radv: Introduce an enum for the any hit status
3493 - radv: Fix the t comparison for triangles
3494 - radv: Refactor lower_rt_derefs
3496 - radv: Use nir_test_mask instead of i2b(iand)
3502 - radv: Use two bools for ahit_status
3503 - radv: Skip pipeline_no_null_shaders_flag tests
3504 - radv: Fix acceleration structure size queries
3505 - radv: Use a global address for ray_launch_size
3506 - radv: Use a global address for sbt_base
3507 - radv: Only set rt stack size for dynamic stacks
3510 - radv: Use nir_gen_rect_vertices
3517 - radv: Remove format desc null checks
3518 - radv: Use desc->format
3536 - radv: Add meta saving/restoring for predicating
3537 - radv: Use RADV_META_SUSPEND_PREDICATING
3538 - radv: Simplify the meta init fail path
3539 - radv: Fix out of bounds access in RTPSO creation
3542 - radv: Use vk_descriptor_set_layout
3543 - radv: Fix stack size calculation with stage ids
4145 - radv/ci: add a Van Gogh dEQP runner
4147 - ci/radv: remove amdgpu.gpu_recovery=0 from the kernel cmdline
4148 - radv/ci: reboot the DUTs when hitting a ring timeout
4149 - radv/ci: reduce the execution timeout
4150 - radv/ci: re-enable fossils testing
4151 - radv/ci: enable the experimental support for mesh shaders
4152 - radv/ci: update the list of flakes
4153 - radv/ci: allow polaris10 to retry more times
4154 - radv/ci: bump the timeout of vangogh/renoir
4155 - radv/ci: update the vkcts expectations for the CI machines
4156 - radv/ci: remove an expected fail from all families
4157 - radv/ci: try to guess failure expectations for Fiji/Raven/Navi14/Navi22
4161 - radv/ci: test vkd3d on kabini
4162 - ci/radv: move a test from the fail list to the flake list
4163 - ci/radv: add a missing flake
4164 - ci/radv: enable vkcts testing on kabini
4166 - radv/ci: test the llvm backend on navi21
4167 - radv/ci: allow specifying a warning pattern
4168 - radv/ci: do not reboot on soft-recovered hangs, just warn
4269 - zink: radv baseline fix
4351 - zink: radv ci updates
4382 - radv: emit fewer framebuffer registers
4469 - zink: update radv baseline
4499 - zink: radv ci updates
4502 - zink: update radv piglit fails
4528 - zink: fix radv piglit fails
4645 - zink: radv baseline updates
4881 - radv: fix return type for meta resolve shaders
5086 - radv: no need to do gs_alloc_req for newer chips in ngg vs/tes
5169 - radv: remove unnecessary align_mul/offset initialization
5170 - radv: initialize meta shader options earlier
5171 - radv: use radv_meta_build_nir_vs_generate_vertices() more
5172 - radv: use nir_op_imm helpers
5173 - radv: use inot(a) instead of ine(a, true)/ieq(a, false)
5175 - radv: always use nir_lower_io_lower_64bit_to_32
5177 - radv: fix clearing of TRUNC_COORD with tg4 and immutable samplers
5178 - radv: consider radix sort shaders to be internal/meta
5181 - radv: add radv_has_shader_buffer_float_minmax
5182 - radv: allow LBVH on GFX6+
5183 - radv: remove subtractions in address calculations
5184 - radv: use SMEM for a few load_global
5194 - radv: call nir_metadata_preserve in various lowering passes
5195 - radv: add missing NIR_PASS() and switch from NIR_PASS_V()
5196 - radv: validate shaders after linking passes
5208 - radv,aco: disable shader cache when ACO debug options are used
5220 - radv/ci: re-enable vega10 fossils testing
5235 - radv,aco,ac/llvm: use nir_op_f{sin,cos}_amd
5249 - radv: fix vbo_bound_mask indexing
5250 - radv: don't update misaligned_mask in CmdBindVertexBuffers2
5254 - radv: remove claimed support for sRGB vertex buffer formats
5255 - radv: fix 16-bit support in radv_lower_vs_input
5369 - radv: exclude PRIMITIVE_{COUNT,INDICES} from the per-vertex output mask
5370 - radv: use shader_info::writes_memory
5371 - radv: use shader_info::gs::active_stream_mask
5372 - radv: use shader_info::inputs_read/outputs_written for FS IO
5373 - radv: use shader_info::outputs_written/per_primitive_outputs for VS outputs
5374 - radv: use shader_info::system_values_read
5375 - radv: fix enabling adjust_frag_coord_z and apply per-pipeline
5376 - radv: add support for independent descriptor set layouts
5377 - radv: add support for VkShaderModuleCreateInfo as pNext
5378 - radv: fix initializing pipeline_key::topology for GFX9 and older
5379 - radv: mark all active stages earlier in the pipeline creation path
5380 - radv: use the hardware primitive topology everywhere
5381 - radv: initialize the vertex input interface state in only one place
5382 - radv: do not support UNIFORM_TEXEL_BUFFER with SRGB
5383 - radv: only apply enable_mrt_output_nan_fixup for 32-bit float MRTs
5384 - radv/ci: update the flakes list for GFX9 chips
5388 - radv: clarify why STAGE_2_CLEAR_BIT needs to wait for CP DMA to be idle
5389 - radv: remove redundant VK_PIPELINE_STAGE_2_TRANSFER_BIT for CP DMA idle
5390 - radv: add an SQTT workaround for chips with disabled RBs
5391 - radv: fix handling divisor == 0 with dynamic vertex input state
5392 - radv: allow to disable sinking of load inputs for FS via drirc
5393 - radv: enable radv_disable_sinking_load_input_fs for Grid Autosport
5394 - radv: use correct push constants range for internal operations
5395 - radv/ci: update list of expected failures for Bonaire (GFX7)
5396 - radv/ci: remove empty flakes lists for GFX9
5397 - radv: fix the number of generated primitive queries with NGG GS vs legacy
5398 - radv/ci: stop skipping dEQP-VK.synchronization.* on Bonaire
5399 - radv: use 3D views for 3D internal operations on GFX6-8
5400 - radv: re-emit dynamic line stipple state if the primitive topology changed
5401 - radv/radix: handle intentional allocation failures properly
5403 - radv/sqtt: fix configuring AUTO_FLUSH_MODE on GFX10.3
5404 - radv: only init acceleration structure if RT is enabled
5405 - radv: disable instance packing to fix pipeline query stats
5406 - radv: remove useless check against nir_texop_samples_identical
5407 - radv/winsys: allow to use RADV_FORCE_FAMILY=gfx1100
5408 - radv: disable DCC for Senra Kagura Shinovi Versus
5411 - radv,aco: add support for packed threadID VGPRs on GFX11
5412 - radv,aco: use the new TCS WaveID SGPR to compute vs_rel_patch_id on GFX11
5420 - radv,aco: export alpha-to-coverage via MRTZ on GFX11
5422 - radv,aco: do not implicitly export the primitive ID for mesh shaders
5424 - radv,aco: add a workaround for binding 2D views of a 3D image on GFX9
5425 - radv: implement VK_EXT_image_2d_view_of_3d
5426 - radv: advertise VK_EXT_image_2d_view_of_3d
5427 - radv: do not declare the scratch offset on GFX11
5429 - radv: do not align VGPRS to 8 or 16 on GFX11
5430 - radv: use the new format table on GFX11
5431 - radv: update scratch buffer registers on GFX11
5432 - radv: update sampler registers on GFX11
5433 - radv: update buffer descriptor registers on GFX11
5434 - radv: update image descriptor registers on GFX11
5435 - radv: use PIXEL_PIPE_STATE_DUMP event instead of ZPASS_DONE on GFX11
5436 - radv: configure DB_RENDER_CONTROL on GFX11
5437 - radv: do not emit non-existent CP_COHER_START_DELAY on GFX11
5438 - radv: do not emit FLUSH_AND_INV_DB_META on GFX11
5439 - radv: update TF_RING_size to a per-SE size on GFX11
5440 - radv: update VRS registers on GFX11
5441 - radv: apply a workaround for CB perf counters on GFX11
5442 - radv: limit CP DMA to max 32KB sizes on GFX11
5443 - radv: more register changes on GFX11
5444 - radv: do not lower loading TESS/ESGS rings using the ABI for LLVM
5446 - radv: rename shader compile functions to spirv_to_nir/nir_to_asm
5447 - radv: do not try to dump the NIR of the trap handler shader
5448 - radv: add new pipeline helpers for NIR->ASM compilation
5449 - radv: init states from pTessellationState at only one place
5450 - radv: init states from pViewportState at only one place
5451 - radv: init states from pRasterizationState at only one place
5453 - radv: determine if shaders use NGG before linking
5454 - radv: export implicit primitive ID in NIR for legacy VS or TES
5455 - radv: constify radv_pipeline in more radv_pipeline_generate_XXX() helpers
5456 - radv: rename radv_pipeline_generate_XXX() to radv_pipeline_emit_XXX()
5457 - radv: remove couple of useless pCreateInfo pointers in the emit path
5458 - radv: remove unused radv_get_conservative_raster_mode() helper
5459 - radv: move gs_table_depth to the physical_device
5460 - radv: cleanup using device/physical_device during pipeline creation
5461 - radv: split radv_pipeline into radv_{graphics,compute,library}pipeline
5462 - radv: introduce radv_graphics_pipeline_info and split existing info structs
5463 - radv/ci: fix fails list for NAVI21
5464 - radv: move HS info and task_num_entries to the physical device
5465 - radv: rename radv_cmd_state::pipeline to graphics_pipeline
5466 - radv: unify radv_pipeline_has_XXX() helpers
5467 - radv: move streamout_shader to radv_graphics_pipeline
5468 - radv: move active_stages to radv_graphics_pipeline
5469 - radv: rename VERT_GRP_SIZE to VERTS_PER_SUBGRP on GFX11
5470 - radv: do not emit more non-existent registers on GFX11
5471 - radv: update framebuffer registers on GFX11
5472 - radv: update VRS rates on GFX11
5473 - radv: use the fragment resolve path by default on GFX11
5474 - radv: report adjusted LDS size for fragment shaders on GFX11
5475 - radv: fix configuring COLOR_INVALID on GFX11
5476 - radv: fix VK_BLEND_FACTOR_CONSTANT_COLOR translation on GFX11
5477 - radv: configure DB_Z_INFO.NUM_SAMPLES correctly on GFX11
5478 - radv: disable VK_AMD_shader_fragment_mask on GFX11
5479 - radv: disable attachementFragmentShadingRate on GFX11
5480 - radv: use pipeline->slab_bo in more places
5481 - radv: fix writing buffer markers with non-zero memory offset
5482 - radv: copy viewport/scissor when initializing radv_viewport_info
5483 - radv: init states from pMultisampleState at only one place
5484 - radv: init states from pDepthStencilState at only one place
5485 - radv: init states from VkPipelineRenderingCreateInfo at only one place
5486 - radv: use AMD values for computing blend related state
5487 - radv: init states from pColorBlendState at only one place
5488 - radv: init states from VkAttachmentSampleCountInfo at only one place
5489 - radv: init states from VkPipelineDiscardRectangleState at only one place
5490 - radv: init states from VkPipelineFragmentShadingRateState at only one place
5491 - radv: ignore DYNAMIC_STENCIL_OP if stencil test isn't enabled
5492 - radv: reduce radv_tessellation_info::patch_control_points to 8-bit
5493 - radv: reduce radv_input_assembly_info::primitive_topology to 8-bit
5494 - radv: reduce radv_rasterization_info::polygon_mode to 8-bit
5495 - radv: reduce radv_discard_rectangle_info::count to 8-bit
5496 - radv: fix lowering GS intrinsics if NGG is disabled per pipeline
5497 - radv/ci: update list of failures for Pitcairn
5498 - radv: enable radv_zero_vram for Hammerting
5499 - radv/winsys: add get_chip_name() to the null winsys
5500 - radv: report the marketing name as part of the device name
5501 - radv: rename ngg_gs_state to ngg_query_state
5502 - radv: track if primitives generated query features are enabled
5503 - radv: add primitives_generated_query to the graphics pipeline key
5504 - radv: declare the NGG query argument for primitives generated query
5506 - radv: flush the NGG query state when the argument is declared
5507 - radv: add few helpers related to streamout
5508 - radv: implement VK_EXT_primitives_generated_query
5509 - radv: advertise VK_EXT_primitives_generated_query
5510 - radv/ci: rename deqp-XXX jobs to vkcts-XXX
5511 - zink/ci: skip two KHR-GL46 tests that timeout most of the time with RADV
5512 - zink/ci: update list of expected failures with RADV
5513 - radv: remove redundant check when importing vertex input info
5514 - radv: update radv_is_vrs_enabled() to use radv_graphics_pipeline_info
5515 - radv: update the check to determine if dynamic discard rectangle is needed
5516 - radv: update the check to determine if dynamic sample location is needed
5517 - radv: update the check to determine if rasterization is enabled
5518 - radv: remove remaining unused pCreateInfo pointers
5523 - radv/ci: add GLES/GLCTS testing with Zink on NAVI10
5525 - radv/ci: refactor test machines and dEQP rules to reduce copy&paste config
5529 - radv/ci: add vkd3d-proton testing
5530 - radv: advertise VK_EXT_border_color_swizzle on GFX10+
5532 - radv/amdgpu: make sure to reset the number of BO when there is no ranges
5533 - radv/amdgpu: do not add "global" BO to the virtual BO list
5535 - radv,aco: rename radv_prolog_binary/radv_shader_prolog to shader_part
5536 - radv/llvm: always emit a null export even if the FS doesn't discard
5537 - radv: disable small primitive culling for user sample locations
5538 - radv: use RADEON_FLAG_VA_UNCACHED for the trace BO
5539 - radv: fix command line for dumping waves with UMR
5540 - radv: dump UMR waves before UMR rings
5541 - radv: do not emit SQTT user data packets on the SDMA queue
5542 - radv/ci: add CI lists for LLVM on NAVI21
5543 - radv: use LOAD_CONTEXT_REG to load the opaque buffer size on GFX10+
5544 - radv: do not abort if SPM isn't supported for the current GPU
5545 - radv: remove old workaround for HTILE layers with F1 2021
5546 - zink/ci: update list of failures for piglit with RADV
5547 - radv: disable DCC for Melty Blood Actress Again Current Code
5548 - radv: always enable VK_EXT_debug_utils
5550 - radv/ci: update list of failures against CTS 1.3.3.0
5552 - radv: fix dumping VS prologs assembly
5553 - radv/ci: enable fossils testing for GFX1100
5554 - radv: remove the radv_report_apu_as_dgpu workaround for Red Dead Redemption 2
5555 - radv: fix wide points/lines by configuring the guardband correctly
5556 - zink/ci: update list of CTS flakes for RADV
5558 - radv,aco: track if a fragment shader needs an epilog
5559 - radv,aco: introduce {radv,aco}_ps_epilog_key
5560 - radv: declare a new user SGPR arg in FS for the epilog PC
5561 - radv: add a function that declares PS epilog shader arguments
5567 - radv,aco: rename radv_aco_build_prolog to radv_aco_build_shader_part
5570 - radv: implement PS epilogs
5571 - radv: disable viewport depth clamping only when necessary
5573 - radv: only force 1x sample for Bresenham lines when pipeline draws lines
5575 - radv: implement VK_EXT_attachment_feedback_loop_layout
5576 - radv: ignore out-of-order rasterization if stencil write mask is dynamic
5577 - radv: fix gathering XFB info if there is dead outputs
5578 - radv: fix cleaning the meta query state if an error occured
5579 - radv: re-emit viewports if negative one to one or depth clamp mode changed
5580 - radv: fix bogus assertion with RADV_FORCE_VRS
5581 - radv: fix pipelineStageCreationFeedbackCount when it's 0
5703 - radv: Fix redundant subpass barriers due to erroneous comparison
5704 - radv/ci: skip dEQP-VK.fragment_operations.transient_attachment_bit
5705 - radv/ci: Move transient_attachment_bit from fail to skip list
5706 - radv/ci: skip image.sample_texture.*_compressed_format*
5707 - radv/ci: Move sample_texture.*_compressed_format_* to faillist for gfx<=9
5708 - radv: Fix vkCmdCopyQueryResults -> vkCmdResetPool hazard.
5710 - radv: Only set pstate for the first hw_ctx.
5711 - radv: Implement radv_flush_before_query_copy to workaround UE Vulkan bugs.
5799 - radv: Minor formatting fix in radv_device.c file.
5800 - radv: Move queue submit sparse bindings to a separate function.
5801 - radv: Move empty queue submit code path to a separate function.
5802 - radv: Move normal (non-empty) queue submit to a separate function.
5803 - radv: Move up early exit for transfer queues in radv_get_preamble_cs.
5804 - radv: Simplify some boolean code in radv_get_preamble_cs.
5805 - radv: Initialize BO pointers when creating preambles.
5806 - radv: Refactor cache flush code for the initial preambles.
5807 - radv: Don't create continue preamble when it's not needed.
5808 - radv: Rename fill_geom_tess_rings to radv_fill_shader_rings.
5809 - radv: Don't use pointers to pointers when updating the preambles.
5810 - radv: Only emit what is appropriate to the queue family for preambles.
5812 - radv: Move radv_nggc_settings enum out of radv_cmd_buffer.
5814 - radv: Lower ABI in NIR for tess/ESGS/NGG shader arguments.
5818 - radv/amdgpu: Group queue submit info into a structure.
5819 - radv/amdgpu: Initialize semaphore info with a designated initializer.
5820 - radv/amdgpu: Pass new queue submit info structure to internal function.
5821 - radv/amdgpu: Pass correct struct type instead of repeating the cast.
5822 - radv/amdgpu: Use scheduled dependency for submitting to multiple queues.
5825 - radv: Add radv_pipeline_has_task helper.
5826 - radv: Set user data register for task shaders.
5827 - radv: Set wave size for task shaders.
5828 - radv: Fill task shader info.
5829 - radv: Add task shader arguments.
5830 - radv: Add task ring entry argument for mesh shaders.
5831 - radv: Implement task shader intrinsics in the ABI.
5832 - radv: Enable nir_opt_offsets for task shaders.
5833 - radv: Use I/O lowering for task and mesh shaders.
5834 - radv: Postprocess task shader configuration.
5835 - radv: Allow linking task shaders.
5836 - radv: Lower shared and task_payload variables in task/mesh shaders.
5837 - radv: Print task shader stage name before disasm.
5838 - radv: Fix task shader info.
5839 - radv: Fix loading task shader ring buffer addresses.
5841 - radv: Introduce ring info structure for queues, refactor preamble generation.
5842 - radv: Use a ws variable during preamble creation.
5843 - radv: Pass radv_device pointer to some functions.
5844 - radv: Refactor queue state to separate structure.
5845 - radv: Clarify emitting graphics shader pointers.
5847 - radv: Implement conditional rendering for async compute queue.
5850 - radv: Remove trailing whitespace.
5851 - radv: Disable predication for supass clear and image clears.
5854 - radv: Add mesh and task stage names to pipeline executable properties.
5856 - radv, ac/nir: Fix multiview layer export for mesh shaders.
5862 - radv: Implement task shader draw and payload rings.
5867 - radv: Implement mesh shader scratch ring.
5868 - radv: Enable NGG wave ID for mesh shader scratch ring.
5870 - radv: Lower mesh shader 3D workgroup ID to 1D index.
5872 - radv: Always use 3D block ID and grid size in task shaders.
5879 - radv: Don't assign driver locations to mesh shader outputs.
5880 - radv: Add CULL_PRIMITIVE to special output mask.
5881 - radv: Use NIR optimization to move discards to the top.
5882 - radv: Only initialize DGC state when DGC is enabled.
5883 - radv: Remove trailing whitespace introduced by DGC commits.
5884 - radv: Refactor predication for compute queues.
5885 - radv: Refactor some CP DMA functions to work with radeon_cmdbuf.
5886 - radv: Refactor radv_emit_userdata_address to work with radeon_cmdbuf.
5887 - radv: Refactor radv_emit_descriptor_pointers to work with radeon_cmdbuf.
5888 - radv: Refactor radv_emit_inline_push_consts to work with radeon_cmdbuf.
5889 - radv: Refactor view index emit to use a per-stage function.
5890 - radv: Move inline push constants to a new function.
5891 - radv: Minor cleanup of radv_queue_submit_normal.
5892 - radv: Allow reusing pipeline compute state emit functions.
5893 - radv: Add dispatch_initiator_task field to radv_device.
5894 - radv: Create internal cmdbuf when a graphics pipeline needs compute.
5895 - radv: Flush descriptors and push constants for task shaders.
5896 - radv: Introduce radv_before_taskmesh_draw.
5897 - radv: Implement mesh shading draw calls with task shaders.
5898 - radv: Synchronization for task shaders.
5899 - radv: Support task shaders in secondary cmd buffers.
5900 - radv: Workaround MEC taskmesh dispatch hang when count buffer has zero.
5901 - radv: Submit internal compute cmdbuf.
5902 - radv: Copy BO list to ACE internal CS.
5903 - radv: Enable task shader feature for NV_mesh_shader.
5913 - radv/llvm: Remove incorrect hardcoded workgroup size from NGG GS.
5914 - radv: Properly set LDS size for LLVM NGG shaders.
5915 - radv: Use NIR-based NGG lowering with LLVM.
5945 - radv: always check entry count in descriptor pool when allocating
5962 - radv: Remove unnecessary null check.
5997 - radv: remove VkPhysicalDeviceMemoryProperties2 workaround
6061 - radv: fixes msvc compiling error
6062 - radv: Fixes compiling error with msvc
6063 - radv: Add losing member hw_ip_version_major and hw_ip_version_minor for win32
6064 - radv: Add macros for win32 that accessed
6070 - radv: Remove usage of \`cnd_monotonic.h`