Lines Matching full:radv
32 - VK_EXT_provoking_vertex on RADV.
34 - VK_EXT_extended_dynamic_state2 on RADV.
36 - VK_EXT_global_priority_query on RADV.
38 - VK_EXT_physical_device_drm on RADV.
40 - VK_KHR_shader_subgroup_uniform_control_flow on Intel and RADV.
42 - VK_EXT_color_write_enable on RADV.
48 - VK_EXT_acquire_drm_display on RADV and ANV.
56 - VK_EXT_multi_draw on ANV, lavapipe, and RADV
64 - NGG shader based primitive culling is now supported by RADV.
91 - [RADV] FSR in Resident Evil: Village looks very pixelated on Polaris
99 - radv: FP16 mode in FidelityFX FSR doesn't look right
102 - [RADV] Blocky corruption in Scarlet Nexus and vkd3d-proton 2.4
109 - [radv] Textures missing in Doom (2016) w/ any temporal AA setting enabled
115 - radv: VBO range check issues with odd strides and sizes
119 - [radv] [regression] Textures missing in Doom (2016)
121 - radv: fd leak in Android WSI radv_AcquireImageANDROID
141 - RADV: Resident Evil Village Freezes during a specific cutscene
144 - radv: front face and cull mode dynamic state can desync
145 - radv: GPU hang in Cyberpunk 2077 on Big Navi
155 - radv: ACO miscompiles a specific DIRT 5 shader
168 - [RADV] - Path of Exile (238960) - Ground decals are missing or corrupted using the Vulkan rendere…
173 - radv: significant overhead from radv_pipeline_has_ngg()
178 - [radv] Revert !7207 once BG3 is fixed.
184 - RADV - Regression - CPU side 'hang' in metro exodus
195 - radv: RoTR fails on Raven APU
198 - Graphics corruption and GPU hang with RADV/LLVM
199 - old kernels (4.19) support in radv
215 - RADV: TRUNC_COORD breaks gather operations
216 - [RADV] corruption in avatar after dying in Heroes of the Storm
1024 - ci: update some radv trace checksums
1025 - ci: update some radv trace checksums
1026 - ci: update radv's trace job tag for Raven
1027 - ci: remove radv's trace job for Polaris10
1030 - ci: add radv's trace job for Navy Flounder
1033 - ci: add VKD3D-Proton testsuite job for radv's Navy Flounder
1039 - ci: update some radv trace checksums
1114 - radv: Fix memory leak on descriptor pool reset with layout_size=0.
1116 - radv: Only require DRM 3.23.
1119 - radv: Add sbt descriptors user SGPR input.
1121 - radv: Use global BO list with raytracing.
1122 - radv: Add support for RT bind point.
1123 - radv: Add RT pipeline bind.
1124 - radv: Implement vkCmdTraceRays.
1125 - radv: Use correct border swizzle on GFX9+.
1129 - radv: Don't skip barriers that only change queues.
1130 - radv: Actually return correct value for read-only DCC compressedness.
1131 - radv: Allow DCC images to be compressed with foreign queues.
1134 - radv: Use the global BO list for acceleration structures.
1135 - radv: Add initial CPU BVH building.
1136 - radv: Implement device-side BVH building.
1137 - radv: Add acceleration structure descriptor set support.
1138 - radv: Convert lower_intrinsics to a switch statement
1139 - radv: Implement load_vulkan_descriptor for acceleration structures.
1140 - radv: Expose formats for acceleration structure.
1141 - radv: Add rt perftest flag.
1142 - radv: Enable VK_KHR_acceleration_structure with RADV_PERFTEST=rt.
1143 - radv: Add -Wpointer-arith.
1151 - radv/winsys: Return vulkan errors for buffer creation.
1152 - radv/winsys: Add support for a fixed VA address for replay.
1153 - radv: Support address capture and replay.
1295 - radv: Merge dEQP default skips into all generation-specific skip lists
1296 - radv: Add a STONEY baseline for dEQP.
1297 - radv: Provide a toggle to avoid warnings about unsupported devices.
1376 - radv: use vk_default_allocator
1555 - radv: call nir_copy_prop() after load-store vectorization
1558 - radv,aco: scalarize all phis via nir_lower_phis_to_scalar()
2370 - ci: fix source-deps for radv on windows
2454 - radv: Fix compatible image handle type for dmabufs.
2458 - radv: Implement VK_EXT_global_priority_query.
2524 - radv: Take image alignment into account when allocating MUTABLE pool.
2525 - radv: Allocate buffer list for MUTABLE descriptor types as well.
2815 - radv: ignore redundant variable descriptor counts (v2)
2878 - radv: Use ac_drm_fourcc.h
2880 - meson: Disable libdrm for RADV on Windows
2883 - radv: Add _WIN32 guard in radv_check_gpu_hangs
2884 - radv: Fix unused label warning on Windows
2885 - radv: Add <io.h> on WIndows for missing close()
2888 - ci: Update Windows image to build RADV
3137 - radv: Handle unnormalized samplers in YCbCr lowering
3576 - radeonsi,radv: fix a late alloc deadlock with <= 6 CUs per SA
3610 - ci: add the dEQP expectations for radv on Renoir
3897 - radv: make radv_pipeline::attrib_ends 32bit
3898 - radv: set maxVertexInputAttributeOffset to UINT32_MAX
3944 - radv: declare index_va in a single call for indexed draw packet emit
3945 - radv: explicitly load a desc set layout struct member during set allocate
4062 - radv: move pipe_misaligned and l2_coherent image checks to flags set on init
4128 - radv: pre-calc vertex buffer descriptor size on pipeline object
4169 - radv: use multidraw iteration for direct draws
4170 - radv: emit NOT_EOP for multi indexed draws
4171 - radv: emit drawid for multidraws
4172 - radv: determine if hardware can emit NOT_EOP before emitting
4173 - radv: split indexed draw cases based on whether drawid is used
4174 - radv: add a gfx10 bug workaround for NOT_EOP
4175 - radv: implement VK_EXT_multi_draw
4222 - features: mark off VK_EXT_multi_draw for radv
4392 - radv: fix barrier in radv_decompress_dcc_compute shader
4393 - radv: fix clearing DCC-compressed e5b9g9r9 images
4405 - radv: improve vectorization callback for small bit sizes
4406 - radv: only set robust_modes if robustBufferAccess2 is enabled
4407 - radv: disable VK_FORMAT_R64_SFLOAT
4408 - radv: cleanup LLVM implementation of vulkan_descriptor_index
4409 - radv: implement vulkan_resource_reindex
4411 - radv,aco: use nir_address_format_vec2_index_32bit_offset
4413 - radv: fix use-after-free upon GS copy shader cache hits
4414 - radv: fix possible use-after-free when inserting GS copy shader from cache
4415 - radv,ac/llvm: use a dword alignment for descriptor loads
4417 - radv,aco: use per-attribute vertex descriptors for robustness
4418 - Revert "radv,aco: don't use MUBUF for multi-channel loads on GFX8 with robustness2"
4419 - radv,aco: compact vertex buffer descriptors
4427 - radv: don't allocate DCC predicate if the image doesn't use DCC
4428 - radv: add radv_absolute_depth_bias
4429 - radv: workaround incorrect depthBiasConstantFactor by Path of Exile
4430 - radv: fix formatting of radv_dri_options
4431 - radv: make attrib_end variable in radv_flush_vertex_descriptors 32-bit
4445 - radv: use byte/word extract/insert instructions
4452 - radv: improve LDS alignment check for load/store vectorization
4456 - radv: increase maxComputeSharedMemorySize
4461 - radv,aco: use all attributes in a binding to obtain an alignment for fetch
4471 - radv: allow VK_FORMAT_R8G8_SRGB sampling
4477 - radv: enable VK_KHR_shader_subgroup_uniform_control_flow
4478 - radv: don't ever convert num_records to bytes if it's zero
4479 - radv: adjust num_records when offset>stride
4480 - radv: use null vertex descriptor if num_records=0
4720 - radv: keep DCC compressed for clears on compute with image stores
4722 - radv: fix fast clearing depth-only or stencil-only aspects with HTILE
4723 - radv: fix emitting depth bias when beginning a command buffer
4724 - radv: remove radv_image_iview::bo
4725 - radv: remove radv_image_iview::multiplane_planes
4726 - radv: allow concurrent MSAA images to be FMASK compressed
4727 - radv: fix emitting default depth bounds state on GFX6
4728 - radv/winsys: remove set but never used use_llvm
4729 - radv: remove old comment about LLVM <= 8
4731 - radv: cleanup LLVM related includes
4732 - radv: remove RADV_DEBUG=nothreadllvm
4733 - radv/winsys: fix allocating the number of CS in the sysmem path
4734 - radv/winsys: fix resetting the number of padded IB words
4735 - radv: make sure CP DMA is idle before executing secondary command buffers
4736 - radv: remove warnings about RADV_PERFTEST=aco,llvm
4737 - radv/llvm: implement the image load DCC bug
4738 - radv: enable DCC stores with the LLVM backend
4739 - radv: re-introduce missing skip list for Polaris10
4740 - radv: fix various CMASK regressions on GFX9
4741 - radv: add the provoking vertex mode to the pipeline/shader keys
4742 - radv/llvm: adjust NGG if provoking vertex mode is last
4744 - radv: implement VK_EXT_provoking_vertex
4745 - radv: enable TC-compat CMASK on GFX8-9
4746 - radv: fix computation of the number of user SGPRS for NGG GS state
4747 - radv: check if DCC is enabled when resolving different levels
4748 - radv: only keep concurrent MSAA images compressed if TC-compat CMASK
4749 - radv/winsys: add GFX6_MAX_CS_SIZE instead of using a magic value
4750 - radv/winsys: fix executing huge secondary command buffers on GFX6
4751 - radv: implement RADV_FORCE_VRS for the LLVM backend
4753 - radv: adjust the computation of the total usage of memory used
4754 - radv: expose 2/3rd of total memory as VRAM and 1/3rd as GTT on APUs
4755 - radv: fix missing ITERATE_256 for D/S MSAA images that are TC-compat HTILE
4756 - radv: declare VK_EXT_extended_dynamic_state2 but leave it disabled
4757 - radv: declare new dynamic states for VK_EXT_extended_dynamic_state2
4758 - radv: implement dynamic depth bias enable
4759 - radv: implement dynamic primitive restart enable
4760 - radv: implement dynamic rasterizer discard enable
4761 - radv: advertise VK_EXT_extended_dynamic_state2
4762 - radv: fix extending the dirty bits to 64-bit
4763 - radv: dump the trap handler shader with RADV_DEBUG=metashaders
4765 - radv: add missing entrypoints for VK_EXT_extended_dynamic_state2
4766 - radv: enable DCC stores on RDNA2
4768 - Revert "radv: Do not access set layout during vkCmdBindDescriptorSets."
4769 - radv: fix heap indices when computing the budget
4771 - radv: fix fast clearing DCC if one level can't be compressed on GFX10+
4772 - radv: simplify radv_pipeline_has_gs_copy_shader()
4773 - radv: remove small overhead of radv_pipeline_has_ngg()
4774 - radv: ignore dynamic blend constants if blend isn't enabled
4775 - radv: remove an useless TODO for dynamic line width
4776 - radv: pass an image range to radv_layout_dcc_compressed()
4777 - radv: remove redundant call to radv_dcc_enabled()
4778 - radv: only mark DCC as compressed when drawing if layout allows it
4779 - radv: only init DCC if compressed in the HW resolve path
4780 - radv: do not decompress DCC for partial resolves if stores are supported
4781 - radv: use radv_dcc_enabled() for the FB mip flush workaround
4783 - radv: stop reporting ACO from the device name
4784 - radv: remove DFSM
4786 - radv: add few new drirc options
4787 - util/drirc: use application_name_match for the SotTR RADV workaround
4788 - radv: move all game workarounds to drirc
4789 - radv: fix missing default state for DB_DFSM_CONTROL
4790 - radv: fix generating hang reports if mutable descriptors are used
4791 - radv: enable RADV_DEBUG=invariantgeom for Monster Hunter World
4794 - radv: enable RADV_DEBUG=invariantgeom for SotTR DX11/DX12 versions
4799 - radv/winsys: allow to reserve a VMID
4800 - radv: emit PA_SC_CONSERVATIVE_RASTERIZATION_CNTL only on GFX9+
4803 - radv: dump SPIR-V instead of using spirv-dis when generating a hang report
4805 - ci: update list of expected failures for Pitcairn/Oland (RADV)
4806 - radv: do not launch an IB2 for secondary cmdbuf with INDIRECT_MULTI on GFX7
4807 - radv/winsys: add a small comment explaining the CHAIN bit
4808 - ci: add expected list of failures for Bonaire (RADV)
4809 - radv: fix aligning the image offset by using align64()
4810 - radv/winsys: adjust some error messages
4811 - radv/winsys: remove useless errno.h includes
4812 - radv: fix dynamic rasterizer discard enable state
4813 - radv: reject binding buffer/image when the device memory is too small
4814 - radv: always decompress both aspects of a depth/stencil image
4815 - radv: create only one pipeline for decompressing depth/stencil images
4816 - radv: fix dynamic culling and depth/stencil related dynamic states
4822 - radv: implement dynamic logic op
4823 - radv: advertise extendedDynamicState2LogicOp
4824 - radv: fix RADV_FORCE_VRS for 2x1 and 1x2
4825 - radv: fix fd leak in vkAcquireImageANDROID()
4826 - radv: disable DCC for DOOM 2016 and Wolfenstein II
4827 - radv: implement VK_EXT_color_write_enable
4828 - radv: advertise VK_EXT_color_write_enable
4829 - radv: add support for more HTILE clear codes
4830 - radv: prevent fast clearing HTILE depth for unrestricted ranges
4831 - radv: allow more fast clears for depth surfaces without TC-compat HTILE
4832 - ci: update list of expected failures against CTS 1.2.6.2 for RADV
4833 - ci: remove few CTS that are now skipped with RADV
4837 - radv: fix applying radv_disable_dcc for DOOM and Wolfenstein II
4839 - ac,radv: implement the cs_regalloc_hang HW bug workaround
4840 - radv: fix applying radv_disable_dcc for DOOM 2016 again
4841 - radv: lower primitive shading rate in NIR
4842 - radv: only init the TC-compat ZRANGE metadata for the depth aspect
4843 - radv: fix bounds checking for zero vertex stride on GFX6-7
4844 - radv: report APUs as discrete GPUs for Red Dead Redemption 2
4845 - radv: fix specifying the stencil layout for separate depth/stencil layouts
4846 - radv: allow unused VkSpecializationMapEntries
4847 - radv: fix selecting the first active CU when profiling with SQTT
4848 - radv: fix missing cache flushes when clearing HTILE levels on GFX10+
4864 - radv: stop special-casing multi-planar formats in radv_get_modifier_flags
4870 - radv: implement VK_EXT_physical_device_drm
4888 - radv: Implement VK_EXT_acquire_drm_display
4974 - radv: Ignore GS copy shader when determining NGG GS wave size.
4975 - radv: Properly enable Wave32 mode for NGG GS.
4983 - radv: Fill shader info earlier.
4984 - radv: Gather NGG info sooner.
4987 - radv: Use new NGG NIR lowering for VS/TES when ACO is used.
4989 - radv: Use new NIR lowering of NGG GS when ACO is used.
5002 - radv/cmd_buffer: Fix warning by initializing instance count.
5010 - radv: Remove duplicate code for getting GS info.
5011 - radv: Don't generate GS copy shader when the pipeline has NGG.
5012 - radv: Assert that there is no GS copy shader when the pipeline has NGG.
5025 - radv/llvm: Emit s_barrier at the beginning of NGG non-GS shaders.
5028 - radv: Add last_vgt_api_stage and use it to simplify some code.
5029 - radv: Move radv_optimize_nir_algebraic to a separate function.
5030 - radv: Allow enabling vertex grouping, fix NGG info with it disabled.
5031 - radv: Set parameter cache oversubscription according to the PC lines.
5036 - radv: Expose radv_get_viewport_xform in radv_private.h
5037 - radv: New shader args for NGG culling settings and viewport.
5039 - radv: Support NGG culling with new perftest environment variable.
5040 - radv: Run algebraic optimizations before NGG lowering.
5044 - radv, aco, ac/nir: Tweak position export scheduling for NGG culling.
5045 - radv: Don't compile NGG culling into shaders that write viewport index.
5046 - radv: Remove num_viewports from radv_skip_ngg_culling.
5067 - radv/ci: Test on Stoney on CI
5104 - radv: Remove assert about pDepthStencilState
5188 - radv: fix AHB leak upon exportable allocation
5224 - radv: fix build errors after commit 8b7ff784
5244 - radv: set RADEON_FLAG_GTT_WC flag for prime memory