Lines Matching full:radv

42 - Add support for VK_VALVE_mutable_descriptor_type on RADV
46 - VK_KHR_fragment_shading_rate on RADV (RDNA2 only)
56 - Sparse memory support on RADV
58 - Rapid packed math (16bit-vectorization) on RADV
71 - \[RADV\] Nioh 2 - The Complete Edition: "Bloom" on lights
72 - \[RADV][BISECTED\] The Surge 2 (644830) - In-game assets do not render correctly since 20.3.4.
75 - RADV: robustBufferAccessUpdateAfterBind is not exposed
76 - \[RADV/DXVK\] Shadow artifacts with different games
82 - RADV: Extreme overhead in vkQueueSubmit
85 - occasional corruption issue with RADV in multiple games, disappears after using amdvlk
92 - \[RADV][REGRESSION][BISECTED\] radv_GetMemoryFdPropertiesKHR returns no valid memory types for va…
95 - \[RADV][ACO\] Overwatch game crash: amd/compiler/aco_insert_exec_mask.cpp: Failed Assertion
97 - \[RADV\] Severe performance drop when exceeding VRAM compared to AMDVLK
99 - RADV: Occlusion query hangs Big Navi GPU
106 - \[RADV/ACO\] Death Stranding cause a GPU hung (\*ERROR\* Waiting for fences timed out!)
118 - radv: dEQP-VK.sparse_resources.\* failures on GFX9
119 - radv: dEQP-VK.sparse_resources.\* failures on GFX9
126 - BUG: After issues playing World of Warcraft with RADV
129 - radv: Some MSAA tests fail when DCC is forced.
130 - \[RADV/ACO/SIENNA_CICHLID\] Into the game Shadow of the Tomb Raider the flickering artifacts are …
132 - radv/aco: "Failed to allocate registers" in AC:Valhalla
145 - radv: blit/copy tests with A2B10G10R10 SNORM fail when DCC is forced on GFX9
169 - RADV: Strange clear behavior with multisample arrays
177 - radv,aco: CTS image robustness tests fail to compile
200 - zink+radv: corruption on pre-game menu in quake3
459 - ci: update some radv trace checksums
460 - ci: update some radv trace checksums
499 - radv: Do the sample check for tiling earlier.
511 - radv: Use internal drm_fourcc.h
513 - radv: Dump BO VA ranges on hang.
514 - radv: Fix RB+ blending for VK_FORMAT_E5B9G9R9_UFLOAT_PACK32.
515 - radv: Fix a hang on CB change by adding flushes.
516 - radv: Deal with unused attachments in mip flush
517 - radv: Don't invalidate the SCACHE for image barriers.
518 - radv: Don't skip layout transitions that only differ in render loop.
519 - radv: Never allow fast clears on DCC images that are not compressed.
520 - radv: Add option to disable DCC in renderpasses without layout.
521 - radv: Disable DCC explicitly for incompatible copies.
522 - radv: Enable DCC in the GENERAL layout on GFX10+.
523 - radv: Use VRAM for upload buffers if entire VRAM is CPU-visible.
524 - radv: Put commandbuffers in VRAM if all VRAM is CPU visible.
525 - radv: Use VRAM for the initial gfx cmdbuffer.
530 - radv/winsys: Fix inequality for sparse buffer remapping.
531 - radv/winsys: Fix offset in range merging.
532 - radv: Create sparse images.
533 - radv: Add image sparse memory update implementation.
534 - radv: Add sparse image queries.
535 - radv: Enable sparse buffer and image support.
536 - radv: Add Android module info to linker script.
538 - radv: Remove redundant WB_L2 flush.
539 - radv: Invalidate CB on SHADER_WRITE for meta operations.
540 - radv: Do dst invalidations for write accesses.
541 - radv: Use access helpers for flushing with meta operations.
542 - radv: Use L2 for CP DMA on GFX9+.
543 - radv: Use L2 coherency on GFX9+.
545 - radv: Do not use a pipe offset for aliased sparse images.
546 - radv: Use stricter HW resolve swizzle compat check.
547 - radv: Do not hash vk_object_base in descriptor set layout.
548 - radv: Improve spilling on discrete GPUs.
549 - radv: Fix vram override with fully visible VRAM.
550 - radv: Ignore WC flags for VRAM.
551 - radv: Do pipe misalignment check per plane.
553 - radv: Don't use dedicated memory info to indicate sharing.
554 - radv: Expose robustBufferAccessUpdateAfterBind correctly.
829 - radv: optimize idiv_const for small bitsizes
830 - radv: call nir_opt_algebraic_late() after lowering idiv for small bitsizes
831 - radv: don't lower_pack() after load-store-vectorization
832 - radv: enable .lower_ineg
842 - radv: vectorize 16bit instructions
868 - CI: Collapse radv & radeonsi stages into one
912 - u_blitter: port radv 3D blit coords logic.
966 - radv: move queue object to a common base object
974 - radv: fix divide by zero with no tesselation params
1436 - radv: Implement VK_VALVE_mutable_descriptor_type.
1616 - radv: Fix radv_queue_init failure handling
1621 - util,radv: Cross-platform monotonic condition variable
1622 - radv: Const aco_compiler_statistic_info usage
1649 - radv: Exclude amdgpu driver files for Windows
1650 - radv: Update build defines for Windows
1651 - radv: Replace VLAs with alloca
1652 - radv: Wrap pragmas with \__GNUC_\_ to fix MSVC
1653 - radv: Use os_localtime instead of localtime_r
1654 - radv: Don't return value in void function
1655 - radv: Ignore radv_printflike on Windows
1656 - radv: Update radv_assert for MSVC
1657 - radv: Fix callback signatures
1658 - radv: Fix leak in radv_amdgpu_winsys_destroy()
1659 - radv: Fix function parameter types
1660 - radv: Use standard \__VA_ARGS_\_ macro
1661 - radv: Create shader cache if ENABLE_SHADER_CACHE
1662 - radv: Use unsigned with u_bit_scan for MSVC
1663 - radv: Replace pthread mutex with mtx_t
1664 - radv: Replace pthread thread with thrd_t
1665 - radv: Use portable ffs and util_bitcount macros
2332 - android: radv: add libcutils shared dependency
2343 - android: ac/radv: fix typo in ac_rgp.h listed in Makefile.sources
2493 - radv: avoid oob read during clear
2518 - radv: null bo list pointer for null descriptors on update
2519 - radv: zero the bo descriptor array when allocating a new set
2656 - ac/radv: move sqtt structs and helpers to amd/common
2657 - ac/radv: move radv_rgp.c to ac
2676 - radv/winsys: set has_dedicated_vram in the null winsys
2708 - radv/llvm,aco/ngg: fix large shift exponent in ngg_gs_vertex_lds_addr
2722 - radv: use intrinsic builders
2731 - radv: use FALLTHROUGH macro
2780 - radv: implement CREATE_REQUIRE_FULL_SUBGROUPS_BIT with cswave32
2782 - radv: workaround games which assume full subgroups if cswave32 is enabled
2786 - radv: vectorize shader I/O
2787 - nir,radv: add and use nir_vectorize_tess_levels()
2797 - radv: implement is_sparse_texels_resident and sparse_residency_code_and
2798 - radv: support SpvCapabilitySparseResidency
2799 - radv/winsys: set has_packed_math_16bit in null winsys
2804 - radv: add RADV_DEBUG=invariantgeom
2805 - radv: set invariantgeom for Shadow of the Tomb Raider
2810 - radv,aco: don't use MUBUF for multi-channel loads on GFX8 with robustness2
2812 - radv: round-up num_records division in radv_flush_vertex_descriptors
2813 - radv: correctly enable WGP_MODE for NGG and GS
2814 - radv: correctly enable WGP_MODE for tessellation control
2821 - radv: don't shrink image stores for The Surge 2
2822 - radv: don't set sx_blend_opt_epsilon for V_028C70_COLOR_10_11_11
2962 - ci: update the list of expected failures for RADV
2964 - radv: do VGT_FLUSH when switching NGG -\> legacy on Sienna Cichlid
2965 - radv: fix applying the NGG minimum vertex count requirement
2966 - radv: don't count unusable vertices to the NGG LDS size
2967 - radv: don't subtract max_verts_per_prim from hw_max_esverts on gfx10.3
2969 - radv/winsys: fill real PCIID for Sienna Cichlid and Navy Flounder
2970 - radv/winsys: add missing Van Gogh and Dimgrey Cavefish in the null winsys
2972 - radv: ignore other blend targets if dual-source blending is enabled
2973 - radv: print more debug messages when generating a hang report
2974 - radv: append a time string to the hang report dump directory
2975 - radv: dump application info in the GPU hang report
2976 - radv: add RADV_DEBUG=noumr to disable UMR logs during GPU hang detection
2977 - radv: dump BO ranges into bo_ranges.log instead of stderr
2980 - radv: add new vk_format_is_*() helpers
2981 - ac,radv: use better export formats for 8-bit when RB+ isn't allowed
2988 - radv: use 32-bit predication for conditional rendering on GFX10.3+
2989 - radv: always use 32-bit predication on compute queues
2990 - radv: fix missing initialization of the predication value
2991 - radv/winsys: fix the sysmem submission path for GFX6
2992 - radv: disable SQTT support for unsupported GPUs
2993 - radv: fix using bitfields for debug/perftest options
2994 - radv: save and dump vertex descriptors during GPU hang detection
2995 - radv: enable NGG on GFX10.3 APUs by default
2996 - radv: only disable CU2 & CU3 when NGG is enabled
2997 - radv: only mask 1 CU for GS/VS waves on GFX10.3
2998 - radv: disable WGP_MODE for NGG on GFX10.3
2999 - radv/llvm,aco: always split typed vertex buffer loads on GFX6 and GFX10+
3001 - Revert "radv/llvm,aco: always split typed vertex buffer loads on GFX6 and GFX10+"
3005 - radv: reduce maxTransformFeedbackBufferDataSize to 512
3006 - radv: mark GFX10.3 as a non-conformant Vulkan implementation
3007 - radv: fix exporting multiviews with NGG
3008 - radv: set the predication boolean as 32-bit if necessary
3009 - radv: use 32-bit predication for skipping FCE on GFX10.3+
3010 - radv: fix using FS sample shading if the linker optimized inputs away
3011 - ci: update the list of expected failures for RADV/FIJI
3012 - radv: enable using MSAA2x and MSAA4x sample locations on GFX10+
3013 - radv: advertise VK_EXT_sample_locations on GFX10+
3015 - radv: fix clearing FMASK for layered MSAA images on GFX9+
3016 - radv: disable alphaToOne feature
3018 - radv: add VK_KHR_fragment_shading_rate but leave it disabled
3019 - radv: implement VK_KHR_fragment_shading_rate
3020 - radv/llvm: implement fragment shading rate
3022 - radv: track if VRS is enabled to apply a workaround on GFX10.3
3023 - radv/llvm: implement a workaround for gl_FragCoord.z with VRS on GFX10.3
3025 - radv: advertise VK_KHR_fragment_shading_rate on GFX10.3+
3026 - radv: add support for resolving layered depth/stencil images
3027 - radv: add missing DB flush after depth/stencil resolve operations
3028 - radv: enable TC-compat HTILE for D32_SFLOAT+MSAA on GFX10+
3029 - radv: adjust the maximum number of coverage samples for VRS
3030 - radv: fix maxFragmentShadingRateRasterizationSamples
3031 - radv: remove useless push constants data when resolving ds attachments
3032 - radv: ignore the mutable bit for TC-compatible HTILE
3033 - radv: enable VK_EXT_line_rasterization on GFX9
3034 - radv: sort the extension table like Khronos
3035 - radv: add code that checks if the extension table is sorted correctly
3036 - radv: make sure FMASK compression is enabled for MSAA copies
3037 - Revert "radv: use 32-bit predication for skipping FCE on GFX10.3+"
3038 - radv: dump VA ranges history when a GPU hang is detected
3039 - radv: add a Python script to check if a VA was ever valid
3040 - radv: disable stippledBresenhamLines on GFX9
3042 - radv: disable A2 SNORM/SSCALED/SINT for texel buffers & images on all gens
3043 - radv: fix clearing images with vkCmdClear{Color,DepthStencil}Image()
3044 - radv: remove unused radv_image::aspects
3045 - radv: always clear the SR0/SR1 bits of the HTILE buffer
3046 - radv: fix potential HTILE issues for TC-compat images on GFX8
3047 - radv: add radv_htile_get_initial_value() and document the HTILE dword
3048 - radv: fix TC-compat HTILE images with DST_OPTIMAL on the compute queue
3049 - radv: clean up radv_layout_is_htile_compressed()
3050 - radv: only load the DS fast clear values for compressed rendering
3051 - radv: enable TC-compat HTILE in GENERAL on GFX10+
3053 - radv/llvm,aco: always split typed vertex buffer loads on GFX6 and GFX10+
3054 - radv: configure the texture descriptor for TC-compat CMASK on GFX10+
3055 - radv: fix enabling TC-compat HTILE in GENERAL for writes on GFX10+
3056 - radv: fix performance regression by restoring TC-compat HTILE in GENERAL
3057 - radv: determine at creation if an image view can be fast cleared
3058 - radv: do not predicate FMASK decompression when DCC+MSAA is used
3060 - radv: fix crashes when fast-clearing in a secondary command buffer
3061 - radv: disable TC-compat HTILE in GENERAL for Detroit: Become Human
3062 - radv: re-initialize HTILE properly after depth/stencil compute resolves
3063 - radv: only re-initialize HTILE after ds compute resolves if compressed
3065 - radv: add support for fast-clearing DCC layers on GFX9+
3066 - radv: clean up radv_decompress_dcc_compute()
3067 - radv: do not use predication when the range doesn't cover the whole image
3068 - radv: enable DCC for layered color images on GFX10+
3069 - radv: mark VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT as unsupported on GFX6-7
3072 - radv: mark some sparse texture CTS as expected failures on GFX9
3073 - radv: set depth to 1 for subpass resolves using the compute path
3074 - radv: decompress DCC for partial resolves using the compute path
3075 - radv: fixup DCC after color resolves using the compute path
3076 - radv: fix color resolves if the dest image has DCC
3077 - radv: fix clearing DCC on GFX9
3078 - radv: only use predication if the FCE value is allocated
3079 - radv: allocate and initialize the FCE predicate value for CMASK too
3080 - radv: update the FCE predicate for fast clears using CMASK
3081 - radv: skip fast-clear eliminate for CMASK based on a predicate
3083 - radv: prevent fast-clearing uncompressed DCC levels
3084 - radv: add support for fast-clearing DCC levels on GFX10+
3085 - radv: do not enable DCC for 3D images with mipmaps on GFX10+
3086 - radv: enable DCC for mipmaps on GFX10+
3087 - radv: disable VK_EXT_sample_locations again on GFX10+
3088 - radv: enable DCC for MSAA on GFX10+
3089 - radv: do not invalidate the L2 metadata cache on compute queues
3090 - radv: flush L2 metadata as part of CB/DB flush instead of CS_DONE on GFX9
3091 - radv: restore invalidating the vector cache for internal meta operations
3092 - radv: flush L2 for images affected by the pipe misaligned issue on GFX10+
3093 - ci: exclude one CTS test that timeout most of the time for RADV CI
3094 - radv: fix a sync issue with geometry shader primitives query on GFX10+
3095 - radv: fix overflow when computing the SQTT buffer size
3096 - radv: inhibit clock gating when tracing with SQTT
3097 - radv: fix separate depth/stencil layout in render pass
3098 - radv,aco: fix shifting input VGPRs for the LS VGPR init bug on GFX9
3100 - radv: fix centroid with VRS coarse shading
3101 - radv: fix waiting on the last enabled RB for occlusion queries
3102 - radv: only apply the MRT output NaN fixup to non-meta shaders
3103 - radv: set correct value for OFFCHIP_BUFFERING on GFX10+
3104 - radv: do not scale the depth bias for D16_UNORM depth surfaces
3114 - radv: add img debug flag
3118 - radv: fix access to uninitialized radeon_bo_metadata
3120 - radv: only set BO metadata for the first plane
3166 - radv: Only enable sparse features on Polaris and newer.
3168 - radv/llvm: Fix reporting LDS stats of tess control shaders.
3186 - radv: Fix -Wshadow warnings
3187 - radv,aco: Compile with -Wshadow when available
3188 - radv/query: Avoid hardcoding array size constants
3189 - radv/winsys: Fix use of nonexisting struct type in sizeof
3191 - radv,aco: Compile with -Wimplicit-fallthrough when available
3192 - gitlab: add RADV bug report template