Lines Matching full:radv

30 - VK_KHR_workgroup_memory_explicit_layout on Intel, RADV
34 - VK_KHR_zero_initialize_workgroup_memory on Intel, RADV
50 - VRS attachment on RADV
59 - Graphics corruption and GPU hang with RADV/LLVM
60 - old kernels (4.19) support in radv
69 - RADV: TRUNC_COORD breaks gather operations
70 - [RADV] corruption in avatar after dying in Heroes of the Storm
79 - RADV - Vertex explosion in DIRT 5 on RDNA2
101 - radv: dEQP regressions after addrlib update
113 - [RADV] Halo: The Master Chief Collection: Crash in Halo Reach Firefight
117 - subgroupBallotFindMSB() broken in RADV/ACO 20.3.4
122 - [RADV][RDNA2] Red Dead Redemption 2 image glitches during menu/overlay menu transitions
125 - radv: VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT exposed for VK_FORMAT_R64_SFLOAT
134 - [RADV] Nioh 2 - The Complete Edition: "Bloom" on lights
135 - [RADV] Oblivion: Poor Performance while MSAA Is Enabled
137 - [RADV][BISECTED] The Surge 2 (644830) - In-game assets do not render correctly since 20.3.4.
146 - RADV: robustBufferAccessUpdateAfterBind is not exposed
148 - [RADV/DXVK] Shadow artifacts with different games
154 - RADV/ACO - DCC causing garbled output on RX570
157 - RADV - Path of Exile: Shimmering outlines where water and other objects meet
162 - RADV: Extreme overhead in vkQueueSubmit
165 - occasional corruption issue with RADV in multiple games, disappears after using amdvlk
180 - [RADV][REGRESSION][BISECTED] radv_GetMemoryFdPropertiesKHR returns no valid memory types for vaap…
186 - [RADV][ACO] Overwatch game crash: amd/compiler/aco_insert_exec_mask.cpp: Failed Assertion
189 - [RADV] Severe performance drop when exceeding VRAM compared to AMDVLK
192 - RADV: Occlusion query hangs Big Navi GPU
201 - RADV: NonUniform OpArrayLength on SSBO ignores NonUniform.
203 - [RADV/ACO] Death Stranding cause a GPU hung (\*ERROR* Waiting for fences timed out!)
222 - radv: dEQP-VK.sparse_resources.* failures on GFX9
224 - radv: dEQP-VK.sparse_resources.* failures on GFX9
916 - radv: Do not use a pipe offset for aliased sparse images.
917 - radv: Add a trivial implementation of VK_KHR_deferred_host_operation
918 - radv: Use stricter HW resolve swizzle compat check.
919 - radv: Expose VK_KHR_workgroup_memory_explicit_layout.
920 - radv: Do not hash vk_object_base in descriptor set layout.
922 - radv: Extract DCC format support handling.
923 - radv: Use the surface offset from ac_surface instead of a plane offset.
924 - radv: Don't relayout images with modifiers.
925 - radv: Add format modifier format queries.
926 - radv: Add drm format modifier queries.
927 - radv: Add image layout with drm format modifiers.
928 - radv: Enable DRM format modifiers on GFX9+.
929 - radv: Enable modifiers with the WSI.
930 - radv: Add modifier fails for CTS bug.
931 - radv: Fix assert.
932 - radv: Implement VK_KHR_zero_initialize_workgroup_memory.
933 - radv: Improve spilling on discrete GPUs.
934 - radv: Fix vram override with fully visible VRAM.
935 - radv: Remove custom icd json generation.
936 - radv: Define supported extensions in C.
937 - radv: Ignore WC flags for VRAM.
938 - radv: Determine swizzles correctly.
939 - radv: Add plane width/height helpers.
940 - radv: Use u_format helpers when possible.
941 - radv: Remove VK_SWIZZLE_*.
942 - radv: Do no use vk_format for getting divisors.
943 - radv: Do not use generated table for plane formats.
944 - radv: Stop checking for MULTIPLANE layout.
945 - radv: Stop using plane_count.
946 - radv: Only support format with a PIPE_FORMAT.
947 - radv: Start using util_format_description for everything.
948 - radv: Remove the format table.
949 - radv: Remove vk_format_has_stencil/depth helpers.
950 - radv: Properly handle modifier import failure.
951 - radv: Do pipe misalignment check per plane.
952 - radv: Don't use dedicated memory info to indicate sharing.
955 - radv: Handle UMD metadata on import.
956 - radv: Use shared code for setting opaque metadata.
958 - radv: Implement initialization of displayable DCC.
959 - radv: Implement displayable DCC retiling.
960 - radv: Add DCC info to the metadata.
961 - radv: Use ac_surface DCC settings for shareable images.
962 - radv: Enable displayable DCC.
963 - radv: Disable displayable DCC for GFX8 properly.
965 - radv: Expose robustBufferAccessUpdateAfterBind correctly.
967 - radv: Enable linear sampling for depth textures.
968 - radv: Add sam option.
969 - radv: Add nodisplaydcc option.
970 - radv: Use correct DCC compressed block size for sampling.
971 - radv: Dedupe winsyses per device.
972 - radv: Allow extra planes for DCC.
973 - radv: Enable sharing with DCC with modifiers.
974 - radv: Ensure we never decompress or FCE read-only textures.
975 - radv: Allow DCC for images with modifiers that are read-only.
976 - radv: Use 8x8 meta compute workgroups.
977 - radv: Enable DCC for image stores on GFX10.
978 - radv: Only set WRITE_COMPRESS_ENABLE on supported HW.
983 - radv: Flush caches for shader read operations.
988 - radv: Support DCC without a fast clear value.
989 - radv: Support DCC without DCC/FCE predicates.
990 - radv: Add retiling for foreign queues.
991 - radv: Support DCC modifiers fully.
992 - radv: Add clang-format for AMD code.
993 - radv: Format.
994 - radv: Update editorconfig.
995 - radv: Re-enable retiling.
996 - radv: Refactor cs_domain to be a winsys function.
997 - radv: Use VRAM cmdbuffers in more situations.
998 - radv/winsys: Remove use_local_bos
999 - radv: Fix memory leak on descriptor pool reset with  layout_size=0.
1001 - radv: Only require DRM 3.23.
1280 - radv/meta: Use sized types for nir_tex_instr::dest_type
1350 - radv: vectorize 16bit instructions
1367 - radv: don't vectorize shift operations
1489 - radv: move queue object to a common base object
1490 - radv: remove all entrypoint enabled debug option
1491 - radv: move to subclassed instance/physical_device structs
1492 - radv: port to using common dispatch code.
1514 - radv: reset object base on recycled command buffers
1638 - radv: fix divide by zero with no tesselation params
2241 - radv: Take image alignment into account when allocating MUTABLE pool.
2242 - radv: Allocate buffer list for MUTABLE descriptor types as well.
2561 - radv: Use linker pragma to simulate weak functions
2562 - radv: Remove unnecessary headers
2563 - radv: Stub sections that don't have _WIN32 support
2564 - radv: Modify radv_descriptor_set for MSVC legality
2565 - radv: Pointer arithmetic on char/uint8_t, not void
2566 - radv: Disable calibrated timestamps on Windows
2567 - radv: Use typed outarray API
2568 - radv: Fix struct initialization for MSVC
2572 - radv: Use typed outarray API
2574 - radv: Pointer arithmetic without void*
2575 - radv: Update JSON generator if Windows
2631 - radv: Properly clean up vk_device
2666 - radv: Use common entrypoints for VK_EXT_private_data
2667 - radv: Switch to the common VK_EXT_debug_report
2672 - anv,radv: Use common entrypoints for VK_KHR_deferred_operation
2730 - radv/meta: Use CreateRenderPass2
2731 - radv: Drop CreateRenderPass
3701 - android: radv: port to using common dispatch code.
3702 - android: radv: fix building error in radv_android.c
3856 - zink: enable WSI-faking for RADV too
3927 - radv: print image array size in debug mode
3931 - radv: null bo list pointer for null descriptors on update
3932 - radv: zero the bo descriptor array when allocating a new set
4119 - radv: for_each_bit -> foreach_bit
4270 - radv: use common interfaces for shader modules
4479 - radv: stop zeroing radv_draw_info during draw
4480 - radv: refactor draw dispatch
4481 - radv: track whether gl_BaseInstance is used
4482 - radv: simplify vs draw param counting during setup
4483 - radv: set gfx pipeline vtx_emit_num to the number of sgprs
4484 - radv: track whether drawid is used on the pipeline struct
4485 - radv: track whether baseinstance is used on the pipeline struct
4486 - radv: break out vertex shader param emission into separate function
4487 - radv: make vertex param sgpr count more explicit
4488 - radv: reorder vertex shader params
4489 - radv: don't emit baseinstance and drawid if neither is used
4490 - radv: don't reset vertex state params on pipeline bind if reg layout matches
4659 - ac/rgp: move radv/sqtt functions to ac
4703 - radv: fix max_waves estimation on GFX10.3
4711 - radv,aco: don't use MUBUF for multi-channel loads on GFX8 with robustness2
4713 - radv,aco: use deref_buffer_array_length
4714 - radv: use nir_opt_access
4716 - radv: sink load_ssbo
4734 - radv,ac/nir: implement non-uniform get_ssbo_size
4736 - radv: round-up num_records division in radv_flush_vertex_descriptors
4737 - radv: correctly enable WGP_MODE for NGG and GS
4738 - radv: correctly enable WGP_MODE for tessellation control
4746 - radv: use a more relaxed alignment for upload buffer allocations
4747 - radv: fix max_lds_per_simd on GFX10
4748 - radv: switch MaxWaves statistic to wave32 waves
4750 - radv: use lds_{encode,alloc}_granularity
4751 - radv: round up max_lds_per_simd / lds_per_wave
4754 - radv,aco: add radv_nir_compiler_options::wgp_mode
4758 - radv: relax shared alignment requirements in mem_vectorize_callback
4759 - radv,aco: allow unaligned LDS access on GFX9+
4771 - Revert "radv,aco: allow unaligned LDS access on GFX9+"
4774 - radv: don't shrink image stores for The Surge 2
4775 - radv: don't set sx_blend_opt_epsilon for V_028C70_COLOR_10_11_11
4776 - radv,aco: remove aco_compiler_statistics
4777 - radv: cache pipeline statistics
4779 - radv/llvm: fix enabled_channels for compressed exports
4802 - radv: lower variables to ssa before nir_propagate_invariant
4803 - radv: lower view_index to zero if multiview is disabled
4804 - ci: add expected fail for RADV
4818 - radv: remove second nir_lower_idiv
4823 - radv: don't use fp16 for 8-bit division lowering before GFX9
4827 - radv: fix conditions for running nir_opt_vectorize
4830 - radv: fix barrier in radv_decompress_dcc_compute shader
4831 - radv: fix clearing DCC-compressed e5b9g9r9 images
4837 - radv: disable VK_FORMAT_R64_SFLOAT
4839 - radv: fix use-after-free upon GS copy shader cache hits
4840 - radv,ac/llvm: use a dword alignment for descriptor loads
5035 - radv: do not invalidate the L2 metadata cache on compute queues
5037 - radv: flush L2 metadata as part of CB/DB flush instead of CS_DONE on GFX9
5038 - radv: add a comment explaining the micro tile mode resolve
5039 - radv: enable TC-compat HTILE with D32S8 and MSAA on GFX9+
5040 - radv: enable TC-compat HTILE for D16S8 on GFX9+
5041 - radv: restore invalidating the vector cache for internal meta operations
5042 - radv: flush L2 for images affected by the pipe misaligned issue on GFX10+
5043 - ci: exclude one CTS test that timeout most of the time for RADV CI
5044 - radv: remove redundant check in radv_process_depth_stencil()
5045 - radv: remove unnecessary radv_image::tc_compatible_htile
5046 - radv: remove redundant check in depth_view_can_fast_clear()
5047 - radv: fix a sync issue with geometry shader primitives query on GFX10+
5048 - radv: fix overflow when computing the SQTT buffer size
5049 - radv: inhibit clock gating when tracing with SQTT
5051 - ac,radv: add SQTT support on GFX10.3
5052 - radv: enable SQTT support on GFX10.3
5053 - radv: fix separate depth/stencil layout in render pass
5054 - radv: add multi-layer support to FMASK color expand
5055 - radv: use the range aspect mask in FMASK color expand
5056 - radv: use a workgroup size of 8x8 for FMASK color expand
5057 - radv: only decompress the depth/stencil aspect that needs to be resolved
5058 - radv: enable sparseImageInt64Atomics/sparseImageFloat32Atomics
5059 - radv,aco: fix shifting input VGPRs for the LS VGPR init bug on GFX9
5060 - radv: synchronize Cmd{Set,Write}Event() using PS_DONE/CS_DONE events
5061 - radv: add support for emitting PS_DONE/CS_DONE on GFX6-8
5062 - radv: remove radv_util.h
5063 - radv: remove stub() macros
5064 - radv: remove unused EMPTY constant in radv_descript_set.c
5067 - radv: use the image view range when fast clearing depth
5068 - radv: check if HTILE is enabled per-level instead of the entire image
5069 - radv: do not decompress/resummarize levels without HTILE
5070 - radv: remove mipmaps related assertions when initializing HTILE
5071 - radv: add support for fast clearing levels of the HTILE buffer
5072 - radv: teach radv_htile_enabled() about the number of HTILE levels
5073 - radv: enable TC-compat HTILE for mipmaps on GFX10+
5074 - radv: re-disable TC-compat HTILE for D32S8 on all generations
5075 - radv: fix centroid with VRS coarse shading
5076 - radv/winsys: move the initial BO domain to radeon_winsys_bo
5077 - radv: prefer CP DMA for GTT buffer copies/clears on dGPUs due to slow PCIe
5078 - radv: fix waiting on the last enabled RB for occlusion queries
5079 - radv/winsys: use an array for the global BO list instead of a list
5080 - radv/winsys: remove the radv_amdgpu_winsys_bo::ws indirection
5081 - radv/winsys: remove useless continue preamble CS for IBs path
5082 - radv/winsys: remove useless is_local check in radv_amdgpu_cs_add_buffer()
5083 - radv/winsys: remove unused radeon_bo_usage enum
5084 - radv/winsys: simplify the user fence logic for submission
5085 - radv/winsys: remove unused fields in radv_amdgpu_cs_request
5086 - radv/winsys: stop zeroing radv_amdgpu_cs_request
5087 - radv: use less AMDGPU contexts by creating only one per queue priority
5088 - radv: add radeon_winsys_bo::use_global_list
5089 - radv: stop using VM_ALWAYS_VALID on APUs
5090 - radv/winsys: move the debug_all_bos check outside of the add/del helpers
5091 - radv/winsys: set use_global_list to avoid adding a BO twice
5092 - radv/winsys: add buffer_make_resident() to the API
5093 - radv/winsys: add the resident BOs to the list of BOs at submit time
5094 - radv/winsys: enable the global BO list unconditionally
5095 - radv: use the global BO list from the winsys
5096 - radv: fix printing the debug option names
5097 - radv: fix double free when creating a fence failed
5098 - radv: stop allocating useless ESGS scratch BO on GFX10+
5099 - radv: fix memory leaks if a submission fails
5100 - radv: do not overallocate the SQTT buffer
5101 - radv: adjust an error message related to the SQTT buffer size
5102 - radv: add support for resizing the SQTT buffer automatically
5104 - radv: emit pipeline bind markers for SQTT
5105 - radv: only make the WSI images resident if the global BO list is used
5106 - radv/winsys: set use_global_list inside the critical section
5107 - radv: only apply the MRT output NaN fixup to non-meta shaders
5108 - radv: create the start/stop CS for SQTT dynamically
5109 - radv: move SQTT parameters initialization to radv_thread_trace_init()
5110 - radv: remove an outdated TODO about SQTT cache flushes
5111 - radv: make sure to allocate enough space when emitting SQTT userdata
5112 - radv: stop emitting pipeline bind markers
5113 - radv: do not allow to capture SQTT on the compute queue
5114 - radv: add support for user event markers with SQTT
5115 - radv: only emit pipeline bind markers for application pipelines
5116 - radv: use the pipeline key as hash for pipeline bind markers
5117 - radv: set correct value for OFFCHIP_BUFFERING on GFX10+
5118 - radv: make the border color BO a resident buffer
5119 - radv: make the trace BO a resident buffer
5120 - radv: make the TMA/TBA BOs resident buffers
5121 - radv: emit the trap handler registers earlier
5122 - radv: rework radv_cmd_buffer_resolve_subpass() a bit
5123 - radv: emit missing subpass resolve marker for SQTT
5125 - radv: store a pointer to the code in radv_shader_variant
5126 - radv: add support for exporting pipelines with RGP
5127 - radv: add support for instruction timing with RGP
5128 - radv: do not scale the depth bias for D16_UNORM depth surfaces
5134 - radv: remove duplicate REG_INCLUDE_CONTEXT setting for SQTT
5135 - radv: always select the first active CU when profiling with SQTT
5136 - radv: fix exporting SQTT pipelines with LLVM
5137 - radv: exclude perf counters for SQTT also on GFX10.3
5138 - Revert "radv: do not overallocate the SQTT buffer"
5139 - radeonsi,radv: do not overallocate the SQTT buffer size
5140 - radv: remove useless decompression of the DS resolve attachment
5141 - radv: do not trace inactive shader engines with SQTT
5143 - radv: double the SQTT buffer size when it is resized
5144 - radv: trigger a new SQTT capture automatically after resizing the buffer
5145 - radv: bump the initial SQTT buffer size to 32MB per SE
5146 - radv: fix RGP barrier layout transition for TC-compatible CMASK images
5147 - Revert "radv: stop using VM_ALWAYS_VALID on APUs"
5148 - radv: cleanup enabling TC-compat HTILE for depth surfaces
5149 - radv: remove useless check about mips+layers for TC-compat HTILE images
5150 - radv: skip useless FCE when fast-clearing MSAA images with DCC enabled
5151 - radv: re-enable TC-compat HTILE for MSAA D32S8 images on GFX9+
5152 - radv: do not declare push constants for DCC decompress on compute
5153 - radv: check if dynamic VRS state changed
5154 - radv: check if dynamic line stipple state changed
5155 - radv: disable sampling with VK_FORMAT_R64_SFLOAT
5156 - radv: fix meta save/restore state with non renderable images
5157 - radv: fix potential clears with non renderable images on GFX9+
5158 - radv: fix initialization of disable_compression when clearing color image
5159 - radv: add missing SQTT events for copy_commands2/create_renderpass2
5160 - radv: remove useless DCC disable check for 3D images on GFX10+
5161 - radv: rework radv_use_dcc_for_image() a bit
5163 - radv: use common entrypoints for VK_KHR_copy_commands2
5164 - radv: do not enable TC-compat CMASK if the image isn't readable by a shader
5165 - radv: remove redundant check when enabling TC-compat CMASK
5166 - radv: make sure FMASK is enabled for TC-compat CMASK
5167 - radv: only configure the CMASK tiling for TC-compat on GFX8
5168 - radv: initialize TC-compat CMASK images with the DCC clear code
5169 - radv: enable TC-compat CMASK on GFX10+
5170 - radv: add notccompatcmask debug option
5171 - radv: extend the dirty bits to 64-bit
5173 - radv: fix clearing CMASK layers on GFX9+
5174 - radv: initialize CMASK with correct clear codes
5175 - radv: restore previous MRT CB_SHADER_MASK logic
5176 - radv: gather if the FS uses perspective or linear interpolations
5177 - radv: determine if a pipeline is candidate for flat shading
5178 - radv: enable VRS 2x2 coarse shading for flat shading on GFX10.3+
5179 - radv: add RADV_DEBUG=novrsflatshading option
5180 - ci: update list of expected CTS failures for RADV
5182 - radv: use common entrypoints for VK_KHR_create_renderpass2
5186 - radv: report that degenerated triangles are not culled
5187 - radv: require DRM 3.35+
5189 - radv: do not fixup DCC after compute color resolves if DCC stores enabled
5190 - radv: only set WRITE_COMPRESS_ENABLE for storage image descriptors
5191 - radv: use a sampled image descriptor for reads for the MSAA color decompress
5192 - radv: compress FMASK for all layouts except GENERAL
5193 - radv: cleanup FMASK expand transitions
5194 - radv: do not force enable FMASK during MSAA blits
5195 - radv: use COLOR_ATTACHMENT_OPTIMAL for fast clear/hw resolve operations
5197 - radv: change RADV_FORCE_FAMILY to use family name instead of LLVM processor name
5198 - radv: try to keep HTILE compressed with DEPTH_STENCIL_READ_ONLY_OPTIMAL
5199 - radv: clean up fence syncobj code
5203 - radv: allow DCC for storage images on GFX10.3 with RADV_PERFTEST=dccstores
5204 - radv: handle implicit subpass dependencies per attachment
5205 - radv: init CMASK/FMASK/DCC in parallel
5206 - radv: perform MSAA color decompression for storage images with DCC
5207 - radv: enable DCC stores with MSAA 4x/8x on GFX10+
5208 - radv: simplify a check when enabling DCC for concurrent images
5209 - radv: enable DCC for concurrent images on GFX10
5210 - radv: make sure FMASK decompress and FCE are performed on gfx queue
5211 - radv: add MSAA support to ClearColorImage() on compute queue
5212 - radv: do not clamp framebuffer dimensions to the minimum dimension
5213 - radv: add MSAA support to CopyImage() on compute queue
5214 - radv: use explicit VRS mode when configuring PA_CL_VRS_CNTL
5215 - radv: allow to force VRS rates on GFX10.3 with RADV_FORCE_VRS
5216 - radv: fix needed dynamic state for VRS
5225 - radv: expose R8_UINT as the only supported format for VRS attachments
5226 - radv: do not allow MSAA with fragment shading rate attachments
5227 - radv: do not enable DCC for fragment shading rate attachments
5228 - radv: determine if attachment VRS is enabled
5229 - radv: configure the VRS HTILE encoding size
5230 - radv: do not use the whole HTILE buffer for depth when VRS is used
5231 - radv: update the HTILE clear word when VRS is used
5232 - radv: allow HTILE for very small images if VRS attachment is used
5233 - radv: create an image for VRS if no depth/stencil attachment is bound
5234 - radv: handle the VRS attachment subpass
5235 - radv: bind our internal depth buffer when not provided by the app
5236 - radv: add support for copying VRS rates into HTILE
5237 - radv: copy VRS rates to HTILE when beginning a subpass
5238 - radv: configure the VRS combiners when an attachment is used
5239 - radv: advertise attachmentFragmentShadingRate on GFX10.3
5241 - radv: keep DCC compressed for clears on compute with image stores
5243 - radv: fix fast clearing depth-only or stencil-only aspects with HTILE
5244 - radv: fix emitting depth bias when beginning a command buffer
5245 - radv: fix emitting default depth bounds state on GFX6
5246 - radv/winsys: fix allocating the number of CS in the sysmem path
5247 - radv/winsys: fix resetting the number of padded IB words
5248 - radv: make sure CP DMA is idle before executing secondary command buffers
5249 - radv: fix various CMASK regressions on GFX9
5250 - radv: fix computation of the number of user SGPRS for NGG GS state
5251 - radv: check if DCC is enabled when resolving different levels
5252 - radv/winsys: fix executing huge secondary command buffers on GFX6
5274 - radv: fix format feature reporting for modifiers
5351 - radv: Only enable sparse features on Polaris and newer.
5354 - radv/llvm: Fix reporting LDS stats of tess control shaders.
5385 - radv: Lower IO and set driver locations earlier.
5386 - radv: Save I/O usage data to both shader infos for merged stages.
5387 - radv: Calculate tess patches and LDS use outside the backend compilers.
5388 - radv: Determine tcs_in_out_eq in radv_pipeline instead of the compiler.
5389 - radv: Fill some tess shader info earlier.
5390 - radv: Reorder some NIR optimizations in preparation for the I/O changes.
5391 - radv: Use new, NIR-based I/O lowering.
5392 - radv/llvm: Only store TCS outputs where they are really needed.
5393 - radv/llvm: Delete superfluous tess and ESGS I/O code.
5450 - radv: Fix improper max_index_count argument for indexed draws
5452 - radv: Skip 0-sized index buffers only when necessary
5460 - ci: skip pipeline_barrier tests that currently crash on RADV
5461 - gitlab: rename RADV bug report template
5519 - radv: memset the alignment hole in cache_entry to 0
5552 - ac/rgp,radeonsi,radv: pass struct thread_trace_data to ac_sqtt_dump_data()