Commit Graph

Select branches

Hide Pull Requests

main

Mono Color

• 02a1ec1021 kk: Hash vertex input state Aitor Camacho 2025-10-21 16:32:23 -06:00
• dc0dcc993b nvk: implement VK_EXT_discard_rectangles Lorenzo Rossi 2025-10-22 00:19:29 +02:00
• 4c62e09505 vulkan: increase MESA_VK_MAX_DISCARD_RECTANGLES Lorenzo Rossi 2025-10-22 00:18:57 +02:00
• b18421ae3d amd/lower_mem_access_bit_sizes: fix shared access when bytes<bit_size/8 Rhys Perry 2025-10-15 15:34:06 +01:00
• 64ec757688 nir/lower_mem_access_bit_sizes: increase chunk limit Rhys Perry 2025-10-15 15:31:34 +01:00
• e89b22280f amd/lower_mem_access_bit_sizes: be more careful with 8/16-bit scratch load Rhys Perry 2025-10-17 15:39:20 +01:00
• 8829fc3bd6 amd/lower_mem_access_bit_sizes: improve subdword/unaligned SMEM lowering Rhys Perry 2025-10-20 10:45:49 +01:00
• 79b2fa785d amd/lower_mem_access_bit_sizes: don't create subdword UBO loads with LLVM Rhys Perry 2025-10-20 10:44:07 +01:00
• 38e1a43f53 intel/mda: Fix potential underflow in printing code Dylan Baker 2025-10-07 18:36:22 +00:00
• f25e59b951 intel/mda/tests: use an ASSERT on fread() Dylan Baker 2025-10-07 20:10:30 +00:00
• 28fbc6addb nvk: VK_DEPENDENCY_ASYMMETRIC_EVENT_BIT_KHR Mel Henning 2025-10-20 16:07:56 -04:00
• e7dca5a6ca nak: fix MMA latencies on Ampere Karol Herbst 2025-10-21 13:28:29 +02:00
• cf4df97093 nak: improve fp16 latencies on Ampere Karol Herbst 2025-10-21 13:27:49 +02:00
• 85480200f8 nak: simplify SM80 HMMA latency categorization Karol Herbst 2025-10-21 10:59:01 +02:00
• 3bbf3f7826 nak: ensure deref has a ptr_stride in cmat load/store lowering Karol Herbst 2025-10-20 15:49:42 +02:00
• f632bfc715 nak: extract cmat load/store element offset calculation Karol Herbst 2025-10-20 14:58:35 +02:00
• d423554e9e radv/bvh: Pair compress triangles in more cases Konstantin Seurer 2025-08-23 20:55:37 +02:00
• c0f332f1cb vulkan/bvh: Add leaf.h to vk_bvh_includes Konstantin Seurer 2025-05-06 22:12:58 +02:00
• 020bd86d30 vulkan: Remove the vk_ir_triangle_node::id field Konstantin Seurer 2025-05-04 12:51:49 +02:00
• c18a7d0e2b radv: Emit compressed primitive nodes on GFX12 Konstantin Seurer 2025-04-07 22:19:42 +02:00
• c5f9fe5e3b radv/rra/gfx12: Properly validate geometry indices Konstantin Seurer 2025-08-23 20:56:13 +02:00
• 82728380a2 vulkan/bvh: Add some debug helpers Konstantin Seurer 2025-09-12 17:35:12 +02:00
• 639cc4d937 vulkan: Bump MAX_ENCODE_PASSES to 4 Konstantin Seurer 2025-05-02 14:45:24 +02:00
• 6a53aae6b2 vulkan: Add vk_ir_header::driver_internal Konstantin Seurer 2025-05-02 14:44:48 +02:00
• 2ee8bfefe6 radv/bvh: Add radv_first_active_invocation Konstantin Seurer 2025-05-02 14:49:32 +02:00
• bd53bbbc57 panvk: support VK_EXT_external_memory_acquire_unmodified Yiwei Zhang 2025-10-20 18:19:25 -07:00
• 990f1868ec vulkan/cmd_queue: Free all elements of struct arrays Konstantin Seurer 2025-10-05 09:50:41 +02:00
• a3e77fe5d2 vulkan/cmd_queue: Fix indentation for struct array copies Konstantin Seurer 2025-10-05 09:50:12 +02:00
• 38950083ae panvk: Fix integer dot product properties Faith Ekstrand 2025-10-21 10:11:03 -04:00
• d380e54422 d3d12: Fix d3d12_video_enc.cpp(4794,33): Error C4244: initializing: conversion from uint64_t to SIZE_T, possible loss of data Silvio Vilerino 2025-10-21 09:09:33 -04:00
• 44d8e999e2 mediafoundation: Also set pSyncObjectQueue = m_spStagingQueue when DX11 input sample Silvio Vilerino 2025-10-20 19:46:08 -04:00
• d2cbbccaaa mediafoundation: Only wait on pSyncObjectQueue for stats completion if any stat was enabled Silvio Vilerino 2025-10-20 19:33:50 -04:00
• 4f7aa40222 mediafoundation: Allocate pro-rated buffer sizes for multi-slice encoding Silvio Vilerino 2025-10-20 17:38:38 -04:00
• b454c35318 mediafoundation: Only use sliced mode when CODECAPI_AVEncSliceGenerationMode is set, disregarding num slices configured Silvio Vilerino 2025-10-20 09:32:54 -04:00
• 71aecf4a93 mediafoundation: SliceGeneration=1: Zero copy IMFSample output with wrapped ID3D12Resource frame/slice buffers Silvio Vilerino 2025-10-18 17:12:13 -04:00
• 45e56e4c96 d3d12: Only check for GetDeviceRemovedReason in debug builds Silvio Vilerino 2025-10-19 10:46:21 -04:00
• 4e1bb2111f d3d12: d3d12_promote_to_permanent_residency to accept res array batch Silvio Vilerino 2025-10-19 10:41:11 -04:00
• 07224f6d15 d3d12: Make output metadata frame buffer READBACK and use direct Map() in get_feedback Silvio Vilerino 2025-10-19 10:13:17 -04:00
• e4d8a49fcd d3d12: Only check H264 video caps if configuration changed between frames Silvio Vilerino 2025-10-19 00:39:24 -04:00
• 8fd82cb339 d3d12: d3d12_video_encoder_get_slice_bitstream_data use regular Map/Unmap Silvio Vilerino 2025-10-18 18:12:59 -04:00
• 1dc76fcaa8 d3d12: Use readback heaps for staging bitstream allocations Silvio Vilerino 2025-10-18 17:36:21 -04:00
• 9b131f1407 d3d12: Video Encode - Make some parameters const & instead of by value Silvio Vilerino 2025-10-18 13:02:32 -04:00
• 1ffefc3e32 d3d12: Use cached heap allocations for output bitstreams instead of allocating per frame Silvio Vilerino 2025-10-18 12:39:48 -04:00
• adbb07e927 d3d12: Use cached heap allocations for barriers instead of allocating per frame Silvio Vilerino 2025-10-18 12:35:12 -04:00
• b076cfdf22 d3d12: Remove unused d3d12_video_encoder::m_transitionsBeforeCloseCmdList Silvio Vilerino 2025-10-18 11:20:08 -04:00
• 6f9c49f6f5 d3d12: Only check HEVC video caps if configuration changed between frames Silvio Vilerino 2025-10-18 10:51:01 -04:00
• e3ab866fea d3d12: Only call CheckFeatureSupport(D3D12_FEATURE_FORMAT_INFO when video format changes Silvio Vilerino 2025-10-18 10:27:31 -04:00
• ca2a1e470a d3d12: Remove per frame allocation slice_sizes(picture->num_slice_descriptors) Silvio Vilerino 2025-10-18 10:20:16 -04:00
• 53e07e78c7 d3d12: Cache ID3D12VideoEncodeCommandList4 instance if supported Silvio Vilerino 2025-10-18 10:01:31 -04:00
• b1ea2b06eb d3d12: Cache ID3D12VideoEncoderHeap1 instance if supported Silvio Vilerino 2025-10-18 09:53:23 -04:00
• a51c3b5bd0 d3d12: Cache ID3D12VideoDevice4 instance if supported Silvio Vilerino 2025-10-18 09:45:36 -04:00
• 702e299f33 d3d12: Remove multiple index calc in d3d12_video_encoder_prepare_input_buffers Silvio Vilerino 2025-10-18 09:33:32 -04:00
• 2ffaa9a9f4 d3d12: Remove multiple index calc in d3d12_video_encoder_begin_frame Silvio Vilerino 2025-10-18 09:28:39 -04:00
• 712607bbec d3d12: Optimize d3d12_video_encoder_flush Silvio Vilerino 2025-10-18 09:20:26 -04:00
• 63db1254ff d3d12: Use a separate queue for encode resolve operations Silvio Vilerino 2025-10-16 23:56:57 -04:00
• 416398aecb mediafoundation: Only attach stats to last slice mfsample Silvio Vilerino 2025-10-16 20:21:16 -04:00
• 80c8994455 mediafoundation: Add pLastSliceFence shortcircuit wait for auto slice mode async slices mode Silvio Vilerino 2025-10-16 17:33:33 -04:00
• fd546c1cde mediafoundation: Refactor frame, multi slice and combine slice IMFSample emission to make it simpler Silvio Vilerino 2025-10-16 16:47:50 -04:00
• f4f619e26e d3d12: Implement last slice signal by splitting Encode/Resolve in two ECL Silvio Vilerino 2025-10-16 15:52:31 -04:00
• 134274488d mediafoundation: Attach stats deferred buffers to all samples for simplicity Silvio Vilerino 2025-10-16 13:32:34 -04:00
• b4592bbb77 mediafoundation: Add some more trace logging Silvio Vilerino 2025-10-16 08:36:03 -04:00
• 43112ce1a4 mediafoundation: Emit multiple MFSamples per slice when CODECAPI_AVEncSliceGenerationMode = 1i Silvio Vilerino 2025-10-15 18:22:47 -04:00
• 3de675eef5 mediafoundation: Prepare for multi sample multi slice Silvio Vilerino 2025-10-15 17:38:36 -04:00
• e612709141 mediafoundation: Add support for setting CODECAPI_AVEncSliceGenerationMode Silvio Vilerino 2025-10-15 17:27:39 -04:00
• a6f84de5b0 mediafoundation: emit subregions samples before pAsyncFence wait to reduce latency Silvio Vilerino 2025-10-15 16:19:58 -04:00
• 4bc6ace7a1 mediafoundation: Attach async stats DXGI buffers without CPU fence wait Silvio Vilerino 2025-10-15 16:00:03 -04:00
• d7384e9f65 mediafoundation: Set ConfigureBitstreamOutputSampleAttributes earlier for async subregion notifications do not need resolved metadata for it Silvio Vilerino 2025-10-15 15:30:45 -04:00
• 791a966510 mediafoundation: Add Resolve completion fence to stats IDXGIBuffers Silvio Vilerino 2025-10-15 14:50:36 -04:00
• 88492ee80b mediafoundation: Helpers ConfigureBitstreamOutputSampleAttributes/ConfigureStatsMetadataOutputSampleAttributes Silvio Vilerino 2025-10-15 12:23:10 -04:00
• 932401ddf0 mediafoundation: add stats resource pool so we can use pool for QP map as well Pohsiang (John) Hsu 2025-10-14 21:32:18 -07:00
• daa3ed7fea ci/deqp: Also print logs to logcat on Android Valentine Burley 2025-10-17 08:16:00 +02:00
• 7cd12e5c6a amd: move CP emit helpers to ac_cmdbuf_cp.c/h Samuel Pitoiset 2025-10-15 11:42:01 +02:00
• e0ffc41d9a amd,radv: move SDMA utility helpers to common code Samuel Pitoiset 2025-10-10 09:06:53 +02:00
• 4989b6e6b9 amd,radv,radeonsi: add ac_emit_cp_write_data_{head}() Samuel Pitoiset 2025-10-15 10:59:45 +02:00
• 1c3754a271 radv: use ac_emit_cp_copy_data() more Samuel Pitoiset 2025-10-15 10:26:01 +02:00
• ed7f9df864 amd: add a predicate parameter to ac_emit_cp_copy_data() Samuel Pitoiset 2025-10-15 10:24:07 +02:00
• 29c2d02d64 amd,radv,radeonsi: add ac_emit_cp_load_context_reg_index() Samuel Pitoiset 2025-10-15 10:20:07 +02:00
• c7c237dd27 amd,radv,radeonsi: add ac_emit_cp_nop() Samuel Pitoiset 2025-10-21 13:31:13 +02:00
• 5801986f53 amd: add missing _cp_ to some emit helpers Samuel Pitoiset 2025-10-15 09:48:37 +02:00
• a0117b5e74 amd,radv: add ac_emit_cp_atomic_mem() Samuel Pitoiset 2025-10-15 09:41:45 +02:00
• 93f3b36b55 radv: use ac_emit_cp_copy_data() more for perfcounters Samuel Pitoiset 2025-10-15 09:35:44 +02:00
• dae502a780 treewide: use nir_load_global_constant alias of nir_build_load_global_constant Georg Lehmann 2025-10-20 17:48:47 +02:00
• cf4ab485ea nir: remove manual nir_load_global_constant Georg Lehmann 2025-10-20 17:47:15 +02:00
• 654bd74c60 treewide: use nir_store_global alias of nir_build_store_global Georg Lehmann 2025-10-20 17:27:55 +02:00
• 2306cba65b nir: remove manual nir_store_global Georg Lehmann 2025-10-20 17:26:20 +02:00
• 9e41a7c139 treewide: use nir_load_global alias of nir_build_load_global Georg Lehmann 2025-10-20 16:51:09 +02:00
• 77540cac8c nir: remove manual nir_load_global Georg Lehmann 2025-10-20 16:46:58 +02:00
• 9ebda88e34 radv: remove an obsolete comment about SMEM stores Samuel Pitoiset 2025-10-14 14:23:19 +02:00
• 63099554ac radv: simplify L2 cache flushes on < GFX12 Samuel Pitoiset 2025-10-14 14:21:52 +02:00
• 52e036fa9f radv: remove useless parameter to gfx10_cs_emit_cache_flush() Samuel Pitoiset 2025-10-14 16:24:14 +02:00
• c8de5a7479 radv: pass int_sel to radv_cs_emit_write_event_eop() Samuel Pitoiset 2025-10-14 15:56:55 +02:00
• eb65d17e00 radv: simplify error handling when creating descriptor pools Samuel Pitoiset 2025-10-16 18:07:51 +02:00
• 11daa11c4e radv: use vk_zalloc2() for allocating the descriptor pool Samuel Pitoiset 2025-10-16 18:04:20 +02:00
• 6106d9e818 radv: simplify allocating pool entries for descriptor sets Samuel Pitoiset 2025-10-16 11:50:20 +02:00
• d4cb7d160d radv: add a small helper to destroy descriptor pool entries Samuel Pitoiset 2025-10-16 11:24:16 +02:00
• d29959f50d radv: remove an useless check when destroying descriptor sets Samuel Pitoiset 2025-10-15 16:49:24 +02:00
• 2b6c75423e anv: reenable VK_KHR_maintenance8 on pre-Xe2 platforms Lionel Landwerlin 2025-10-16 17:16:27 +03:00
• c5d313a2a8 brw: handling dynamic programmable offsets pre-Xe2 Lionel Landwerlin 2025-10-16 17:14:34 +03:00
• d37c6ff4ed brw: mark divergence data as valid for debug purposes Lionel Landwerlin 2025-10-17 10:51:03 +03:00
• 255d1e883d nir/divergence: fix handling of intel uniform block load Lionel Landwerlin 2025-10-17 10:49:21 +03:00