[Git][xorg-team/lib/mesa][upstream-experimental] 170 commits: .pick_status.json: Update to d49a00bdf15fd48b31af93aaf5feed3eebcbda12
"Timo Aaltonen (@tjaalton)" <[email protected]> Mon, 03 Aug 2026 15:18:18 +0000
| Newsgroups | gmane.linux.debian.devel.x |
|---|---|
| Message-ID | <[email protected]> |
----==_mimepart_6a70b13ab73b_5515b98ebc885935 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Timo Aaltonen pushed to branch upstream-experimental at X Strike Force / = lib / mesa Commits: f505ac9e by Eric Engestrom at 2026-07-22T10:52:04+02:00 .pick_status.json: Update to d49a00bdf15fd48b31af93aaf5feed3eebcbda12 Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 56e7e13b by Alyssa Rosenzweig at 2026-07-22T10:52:04+02:00 intel: fuse off Jay in Mesa 26.2 Jay is under heavy development and is not considered released. It is available in upstream Mesa for developers to hack on but is not part of the 26.1 release. Add a comment acting like a chicken bit to fuse off the= compiler while minimizing conflicts with backports (which is why we don't= remove Jay wholesale from the release). Signed-off-by: Alyssa Rosenzweig <[email protected]> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 060ff1ef by JaeHoon Lee at 2026-07-22T10:52:04+02:00 v3dv: make room in the descriptor map for the no-sampler entries pipeline_lower_nir() reserves two "no sampler" entries before applying th= e pipeline layout, so real samplers start at index 2. The map is only V3D_MAX_TEXTURE_SAMPLERS long and we advertise exactly that many, so the last two samplers we allow write past the end of it, silently in release builds. Fixes: fc17231b2b4e ("v3dv/pipeline: reduce descriptor_map size") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 9d534e38a2e20d1076e2dbd4b82d8c9e6b65e932) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - cb7ea5bf by JaeHoon Lee at 2026-07-22T10:52:04+02:00 v3dv: use the binning VS variant for the binning VPM config compute_vpm_config() reads the render VS into vs_bin_variant, so the binning VPM config comes from the wrong program data. The binning VS writes less, so this only ever under-sizes the binning VPM allocation. Fixes: 9e7d9a6efb67 ("v3dv: add support for geometry shaders to pipelines= ") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 771dfe55fcb62b62212b418d7f702ab8d42ea07e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - ffb925ef by JaeHoon Lee at 2026-07-22T10:52:05+02:00 v3dv: record the multiview geometry shader with its Vulkan stage bit active_stages is a VkShaderStageFlags, but MESA_SHADER_GEOMETRY is an ordinal, so this sets VERTEX|TESSELLATION_CONTROL instead of GEOMETRY, an= d vkGetPipelineExecutable*KHR skips the shader. Use mesa_to_vk_shader_stage= (), as the rest of the file does. Fixes: 5872c7ca7ba7 ("v3dv: inject a custom passthrough geometry shader f= or multiview pipelines") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 1d9201e377eb6f595c83d68757b054818bd4176a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 0b5641de by JaeHoon Lee at 2026-07-22T10:52:05+02:00 v3dv: record the no-op fragment shader with its Vulkan stage bit Same as the previous commit: MESA_SHADER_FRAGMENT sets TESSELLATION_EVALUATION instead of FRAGMENT. Fixes: 02de1d9ab16f ("v3dv: add a no-op fragment shader if we don't have = one") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit fb7300c6da2b740b0605b1562fe569391f33be94) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - b75b5645 by Pavel Ondra=C4=8Dka at 2026-07-22T10:52:05+02:00 nv50_ir_ra: align B96 spill slots to vec4 This fixes nouveau NVE4 shader-db godot crash exposed by vectorizing commutative ALU sources. Fixes: d12e16fc3b5 ("nouveau: Handle unaligned tlsBase during spills") Suggested-by: Karol Herbst <[email protected]> Reviewed-by: Karol Herbst <[email protected]> Reviewed-by: Mel Henning <[email protected]> Signed-off-by: Pavel Ondra=C4=8Dka <[email protected]> (cherry picked from commit 05f9f41fc294f6adb6d9b8ffed9a801bb232f40e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - ed764e64 by Caleb Callaway at 2026-07-22T10:52:05+02:00 anv: hide Intel vendor ID for Cyberpunk Works around XeSS misrendering Fixes #15739 Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit d021138b4100bf1dc876e930645f3a382d1ed78d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - f1ee264a by Michael Cheng at 2026-07-22T10:52:05+02:00 intel: Fix DEBUG_FS_SIMD mask Spell out the FS modes like the sibling DEBUG_{CS,TS,MS,RT}_SIMD masks. Fixes: 1728c055d4b ("intel: Fix multipolygon flags in INTEL_SIMD_DEBUG de= fault handling") Signed-off-by: Michael Cheng <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> (cherry picked from commit 7635d56fba9bbcafd0b0b448951947917211bb34) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 9f9dd9e7 by Michael Cheng at 2026-07-22T10:52:05+02:00 intel: Fix operator precedence in intel_simd_overridden | binds more tightly than ?:, so this chain of ternaries wasn't OR'd together but parsed as a nested ternary that returns only the first matching stage's bit. With more than one stage requested, the later stages were silently dropped (e.g. INTEL_SIMD_DEBUG=3Dcs32,ts16 gave 0x20, COMPUTE only, instead of 0x60). Wrap each ternary in parentheses. Fixes: baf2f0ea4a9 ("intel: Replace INTEL_DEBUG=3Ddo32 with INTEL_SIMD_DE= BUG") Signed-off-by: Michael Cheng <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> (cherry picked from commit 48297c8344669c4a867fae28e2cafb646218016e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 2ef226bd by Timothy Arceri at 2026-07-22T10:52:05+02:00 zink: fix swap interval changes being dropped db9dbcbec00 unconditionally restores the previous present mode before attempting to update the swapchain, which turns runtime swap interval changes into a no-op: * the immediate path (no currently acquired image) recreates the swapchain with the old present mode still set, so nothing changes * the deferred path added to zink_kopper_present_queue() can never trigger since cdt->present_mode always matches the current swapchain's mode The only remaining effect is pointless swapchain recreation whenever an app changes the swap interval while no image is acquired. Keep the new present mode instead, and only restore the old one if the immediate swapchain update fails. Fixes: db9dbcbec00 ("zink: defer swapchain updates for interval changes i= f acquired image is active") Generated-by: Claude Code (Fable 5) Reviewed-by: Mike Blumenkrantz <[email protected]> (cherry picked from commit 09180f0eed603872b75f44b398048db06f236f54) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 3f9bbe20 by Benjamin Cheng at 2026-07-22T10:52:05+02:00 ac/vcn_enc: Disable var slice for preencode with VCN4 An internal FW change was never upstreamed, so the version numbers that were previously put here became inaccurate. Disable it until the FW fix is available. Signed-off-by: Benjamin Cheng <[email protected]> Cc: mesa-stable Reviewed-by: David Rosca <[email protected]> (cherry picked from commit c2ecb433fda48ee5d0ef295b3aa79f23143ceeca) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - d0e687b6 by Karol Herbst at 2026-07-22T10:52:05+02:00 mesa_clc: support 32 bit targets Backport-to: 26.2 Reviewed-by: Alyssa Rosenzweig <[email protected]> (cherry picked from commit f264cb38182a2c2d75ec84d150fc28f4fe00598b) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - b120fc9a by Karol Herbst at 2026-07-22T10:52:05+02:00 meson/rusticl: fix typo in depfile for builtin shaders Backport-to: 26.2 (cherry picked from commit b44f878eb7b90de9a5ed45d4844313dc85c8e50c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - b27cf38f by Karol Herbst at 2026-07-22T10:52:05+02:00 rusticl/meta: mark builtin kernels SPIR-V as 64 bit Backport-to: 26.2 (cherry picked from commit 4130bbe1502ab27e3eec4bc4580dbc89ab579ab8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - f8a2af79 by Karol Herbst at 2026-07-22T10:52:05+02:00 rusticl/mesa: compile 32 bit version Backport-to: 26.2 (cherry picked from commit be269027d76853d1a75cae09e99c0c2b719fbe2b) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 4d1809f6 by Karol Herbst at 2026-07-22T10:52:05+02:00 rusticl/program: add Program::from_spirv_with_devs Backport-to: 26.2 (cherry picked from commit 5ab128c585e988db20ffa948b8c285e53320c6cd) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 41dbdc7f by Karol Herbst at 2026-07-22T10:52:05+02:00 rusticl/meta: support 32 bit devices Backport-to: 26.2 Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15869 (cherry picked from commit a5ea6c5a02bb1e19b8c47d78e3339ba920a9e4e0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 1321fd2e by Karol Herbst at 2026-07-22T10:52:05+02:00 rusticl/meta: split out ulong kernels Backport-to: 26.2 (cherry picked from commit b2c832063be1b2f56019a6b9739bae42649541e6) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 421dcdf8 by Christoph Pillmayer at 2026-07-22T10:52:05+02:00 pan: Fix BIFROST_MESA_DUMP_DIR Don't try to fclose when we didn't fopen successfully. Fixes: 712ebaa4 ("pan: Add BIFROST_MESA_DUMP_DIR option to dump shader bi= naries") Reviewed-by: Lars-Ivar Hesselberg Simonsen <[email protected]> (cherry picked from commit 306762b57411eb09d0eb09376a0b5a876497d14f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - cff1fce8 by Lionel Landwerlin at 2026-07-22T10:52:06+02:00 anv: flush accumulated barriers for top of TOP_OF_PIPE If there is a barrier preceding the write timestamp, we need to flush it to make sure the timestamp is collected after the barrier is executed. Signed-off-by: Lionel Landwerlin <[email protected]> Cc: mesa-stable Reviewed-by: Tapani P=C3=A4lli <[email protected]> (cherry picked from commit c6f6c9b02da991c9c878d8992b4074bbba70aa73) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 7fae50bb by Georg Lehmann at 2026-07-22T10:52:06+02:00 nir/opt_dead_write_vars: handle atomics as reads As long as there is no cross invocation access, write, followed by atomic= , followed by write is valid. We can not remove the first write in this cas= e: shared_array[invocation_index] =3D 0; res =3D atomicAdd(shared_array[invocation_index], 2); shared_array[invocation_index] =3D 0; Cc: mesa-stable Reviewed-by: Rhys Perry <[email protected]> (cherry picked from commit 1bc7b09d583b4f623fe18983f048014c508a2364) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 5d59ca7e by Lo=C3=AFc Molinari at 2026-07-22T10:52:06+02:00 panfrost/ci: Mark "spec@!opengl 1.4@copy-pixels" as flake This is a random failure on g57/g610/g925. See Mesa bug #15880. Signed-off-by: Lo=C3=AFc Molinari <[email protected]> Reviewed-by: Erik Faye-Lund <[email protected]> (cherry picked from commit b71fab7fbe1ab859f38032eafd9b75341373413f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 90907cfe by Yiwei Zhang at 2026-07-22T11:25:40+02:00 venus: recycle idle sfb cmds only after async wait This fixes a regression introduced by a code refactor. The correct order should be: 1. see sync feedback signaled to a new point 2. emit renderer side fence/semaphore wait for the new point 3. recycle sync feedback idle cmds In (3), those exceeding the cache capacity will be purged, ending up with renderer side cmd free, which must be executed after (2) for validity of the renderer side Vulkan usage. Fixes: f4a3d6941fd ("venus: extract sync feedback host write and query") (cherry picked from commit 97c9a4eaa40b41739c20948117a1b977704f5c50) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 185650d6 by Ian Romanick at 2026-07-22T11:25:40+02:00 brw: Handle empty top block in brw_nir_move_interpolation_to_top If the block is empty, nir_block_first_instr returns NULL. nir_before_instr(NULL) creates an invalid cursor that eventually results in a segfault. This was discovered while testing !42933 on Ratchet and Clank - Rift Apart. I also considered using nir_before_block(top). That would require moving the instructions in reverse order because nir_before_block always added at the top. That really demands a comment explaining the reasoning. Instead, I decided to do the obvious thing that is slightly more code. Fixes: 3037603b703 ("intel/fs: Linked list micro optimizations in brw_nir= _move_interpolation_to_top") Reviewed-by: Michael Cheng <[email protected]> (cherry picked from commit a7cdad725932c60e8b760b61115574d8fb935a3c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 8e480fdd by Lionel Landwerlin at 2026-07-22T11:25:40+02:00 anv: fix Wa_18040903259 The description above says we need CS_STALL on both PIPE_CONTROL, it was missing for the first one. Signed-off-by: Lionel Landwerlin <[email protected]> Fixes: 243c01c703 ("anv/iris: implement Wa_18040903259") Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15844 Reviewed-by: Michael Cheng <[email protected]> Reviewed-by: Tapani P=C3=A4lli <[email protected]> (cherry picked from commit be696377a17b63228fc4ff78284ac1379506d0ae) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 8cbece81 by Marek Ol=C5=A1=C3=A1k at 2026-07-22T11:25:40+02:00 radv: don't expose memory types from AMD_device_coherent_memory without t= he ext Fixes: 7fa1b487556 - radv: disable AMD_device_coherent_memory on gfx12 du= e to out of order behavior Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit 5f4595a5c07a2e428a9c4416ed4c94659598bf69) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 052c4212 by Iv=C3=A1n Briano at 2026-07-22T11:25:40+02:00 anv: fix 2d-array to 3d blits It was accounting for the depth center offset if src_image is 3D, assuming dst_image would be too, but with maintenance8 that may not be the case, so we need to check both. Fixes new tests: dEQP-VK.api.copy_and_blit.core.blit_image.simple_tests.2d_array_to_3d.rev= erse_blit_* Cc: mesa-stable Signed-off-by: Iv=C3=A1n Briano <[email protected]> Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit 1c6e7e82912fa43acd538cb2e30a4d132c296dac) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - b5ee82b3 by Adam Stylinski at 2026-07-22T11:25:40+02:00 nv30: fix an issue when this push buffer is NULL Fixing an issue when an empty context is destroyed. When a context is destroyed prior to this buffer being initialized, a NULL pointer dereference occurs. The nv50 & nvc0 implementations do this same thing. Reviewed-by: Karol Herbst <[email protected]> Tested-by: Filip Gawin <[email protected]> Backport-to: * (cherry picked from commit 4ce4feff89bad1ffd2ac7933ab25b44b6cca38d9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 949268fe by Filip Gawin at 2026-07-22T11:25:40+02:00 nv30: fix 1 << 31 issues Reviewed-by: Karol Herbst <[email protected]> Backport-to: * (cherry picked from commit 51a7e9723538833e74ea9aebc251db22743eaf7f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 0889c4db by Filip Gawin at 2026-07-22T11:25:40+02:00 nv30: fix another left shift cannot be represented in type 'int' Reviewed-by: Karol Herbst <[email protected]> Backport-to: * (cherry picked from commit 81dca3bf64c086bbae6c542824f5a82f34a0ed3f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 5ae3c235 by Lionel Landwerlin at 2026-07-22T11:25:40+02:00 vulkan/runtime: fixup vk_shader leak on RT group recompile If you compile the same RT groups in multiple processes, there is a chance one process writes the disk cache with one of the shader in the group. Another process will pickup on the disk cache entry but not have all the entries for the group and therefore trigger a recompile. In that unlikely cases that is probably only reachable while running the CTS in multiple processes, there will be a shader leak that will materialize later at cache destroy : deqp-vk: ../src/vulkan/runtime/vk_pipeline_cache.c:687: vk_pipeline_cac= he_destroy: Assertion `cache->object_cache->entries =3D=3D 0' failed. Signed-off-by: Lionel Landwerlin <[email protected]> Cc: mesa-stable Reviewed-by: Ivan Briano <[email protected]> (cherry picked from commit 879365ec9f06adef47f116d11b88632264c1f271) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 09b725f7 by Timothy Arceri at 2026-07-22T11:25:40+02:00 nir/opt_dead_write_vars: handle memcpy_deref as reads memcpy_deref reads its source and writes its destination, but the pass did not account for it. A store feeding an OpCopyMemorySized copy could be wrongly eliminated by a later store to the same deref, since nothing marked the intervening access as a use. memcpy is a sized copy that walks the derefs in a loop, so it can touch memory beyond the src/dst deref itself. A deref comparison is therefore not sufficient - conservatively clear all unused writes in the accessed modes. Reviewed-by: Georg Lehmann <[email protected]> Assisted-by: Claude Code (Claude Opus 4.8) Cc: mesa-stable (cherry picked from commit 0a12d316e7ecaad7474ee782636675b72fcf9ffb) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 3995f83d by Timothy Arceri at 2026-07-22T11:25:40+02:00 util: add Blockland workaround for crash The game only initialises the glFramebufferTextureLayerEXT entry point as part of the EXT_geometry_shader4 / NV_geometry_program4 function groups, so without one of those extensions advertised the game calls a NULL pointer when "Shader Quality" is set to high and crashes. The function is provided by EXT_texture_array which Mesa does support, so advertise one of the extensions to avoid the crash. Cc: mesa-stable Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/13311 (cherry picked from commit df73bd930ae57bc1928beb1fdfb30bc3ab41b684) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 71001521 by Valentine Burley at 2026-07-22T11:25:40+02:00 ir3: Fix ballot_components for subgroups smaller than 32 ballot_components becomes 0 when max_subgroup_size is smaller than 32. Clamp it to a minimum of 1. Fixes: 639d7946c71 ("freedreno: Add initial A702 support") Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit ff5352b4900737d62d8e3fa5878ebc6c381aace0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 168a591a by Valentine Burley at 2026-07-22T11:25:41+02:00 ir3: Derive max_variable_workgroup_size from device limits max_variable_workgroup_size was hardcoded to 1024, which is incorrect for= devices with a smaller threadsize_base such as A702. Use nir_options.max_workgroup_invocations to override the hardcoded value. Fixes: 639d7946c71 ("freedreno: Add initial A702 support") Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit be71f85a78c5d106b7adae912a7f22811ad8954e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - aae5e129 by Valentine Burley at 2026-07-22T11:25:41+02:00 ir3: Compute subgroup_size from threadsize_base ir3_info::subgroup_size was hardcoded to 64 (or 128 with double threadsize), which is incorrect for devices with a smaller threadsize_base such as A702 (16/32). Fixes: 639d7946c71 ("freedreno: Add initial A702 support") Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit 5e66bfd939a34093cc9db22c64a7698955f98df9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 5c6b5619 by Valentine Burley at 2026-07-22T11:25:41+02:00 tu: Use computed subgroup size Some turnip paths still hardcoded a subgroup size of 64 (or 128 with double_threadsize), which is incorrect for devices with a smaller threadsize_base such as A702. Fixes: 639d7946c71 ("freedreno: Add initial A702 support") Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit 69416494db271747d015b6ab010f3e348d7f76da) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - e672979e by Jan Meisel at 2026-07-22T11:25:41+02:00 nir/range_analysis: handle msad_4x8 in unsigned upper bound get_alu_uub() already had logic for nir_op_msad_4x8, but the opcode was missing from the early ALU opcode allowlist. This made the existing case unreachable and forced the analysis to return the default UINT32_MAX boun= d. Add nir_op_msad_4x8 to that supported-opcode switch. Fixes: 0477421f7db ("nir: add msad_4x8") Reviewed-by: Georg Lehmann <[email protected]> Assisted-by: Codex Signed-off-by: Jan Meisel <[email protected]> (cherry picked from commit dec6f98c7bea75d9fade316e57ea272d2dd218a8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 91545029 by Mary Guillemard at 2026-07-22T11:25:41+02:00 nvk: Do not take base into account when lowering emulated attributes We were previously adding the base indice to the shared memory offset. That was fine before because the base was always 0, but this was just plain wrong. After calling nir_recompute_io_bases, this would cause misaligned address faults as the base would represent the current slot number used for that specific varying. This fixes "dEQP-VK.mesh_shader.ext.builtin.viewport_index" with nir_opt_varyings_bulk. Signed-off-by: Mary Guillemard <[email protected]> Fixes: cdb0dea462e ("nvk: Lower mesh and task shaders") Reviewed-by: Mel Henning <[email protected]> (cherry picked from commit 794cd9d334584cf034084d611a1d8ede10b6b6bc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 644eac3a by Connor Abbott at 2026-07-22T11:25:41+02:00 tu: Fix resetting command streams with writeable BOs We were forgetting to reset the start pointer that is used when switching back to writeable, which meant that when using the CS again and again making the CS writeable, we would keep going where we left off in the BO instead of actually reusing it. We still threw away the other BOs, but this meant that we would eventually run out of space and grow the BO unnecessarily, and if every BO hit the maximum size of 4MB then we could easily run out of memory eventually in games with hundreds of command buffers. Fixes: ebb8e104a56 ("tu/cs: Add support for CS patching") (cherry picked from commit c140fdd2d163bc3cefc29dd3f457c3a131fd1bc0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 0922498e by Lorenzo Rossi at 2026-07-22T11:25:41+02:00 panfrost/midgard: Fix fau max not initialized In the refactor we forgot to initialize the FAU size for Midgard. Signed-off-by: Lorenzo Rossi <[email protected]> Fixes: d1247f55790 ("panfrost: Unify FAU constants and relocation handlin= g") Acked-by: Erik Faye-Lund <[email protected]> (cherry picked from commit 07e937fdb73195e1962c5cbf306f6e427178f9ab) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 6ffaa0ca by Jan Meisel at 2026-07-22T11:25:41+02:00 radeonsi/vcn: fail feedback for truncated encodes VCN stops writing encoded output at allowed_max_bitstream_size, but can still report the full coded frame size in feedback. If that size is large= r than the bitstream buffer VA clients can read past the mapped buffer when= copying VACodedBufferSegment::buf. Record the bitstream size left after packed headers, clamp feedback to that limit, and flag the encode as failed if the reported size exceeds it= . The VA frontend then reports the frame as bad instead of exposing a truncated bitstream as valid output. Analysis and patch by Maksim Nikolaev. Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15774 Cc: mesa-stable Reviewed-by: David Rosca <[email protected]> Assisted-by: Codex (cherry picked from commit 15e06548daba70c7adb222fdffbdf2f0d7f1e0d1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 072dd59d by Yiwei Zhang at 2026-07-22T11:25:41+02:00 venus: properly check sync2 enablement We might be able to upgrade instance api version or simply enable sync2 behind the app, but let's still forward for now. Fixes: 07cee75c39c ("venus: layer vkQueueSubmit2 over vkQueueSubmit w/o s= ync2") (cherry picked from commit 64cbdd73a878f44952ed640e4d3f7d6aff442b36) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 5a3ec210 by Yiwei Zhang at 2026-07-22T11:25:41+02:00 venus: host image copy to scrub present_src layout if needed Fixes: c13da1c744a ("venus: implement host image copy commands") (cherry picked from commit f3ea8314c1b2f5a1edc2a323dcf8735221238e88) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 13e953f8 by Jeremy Gebben at 2026-07-22T11:25:41+02:00 kk: Fix Vulkan to Metal stage translation for timestamps A typo in the stage lookup table was causing vertex stages to be mishandled in vkCmdWriteTimestamp2(). Fixes: ed807097ebd ("kk: Implement VK_QUERY_TYPE_TIMESTAMP") Reviewed-by: Aitor Camacho <[email protected]> (cherry picked from commit 7f540b051c40cfc0c5ed0c48fd70c0309fb82b37) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 35811c56 by David Rosca at 2026-07-22T11:25:41+02:00 vulkan/video: Fix coding AV1 decoder/encoder_buffer_delay Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> (cherry picked from commit 71f8d76c7d6eb355edb77344ceed23889fe673a1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 2e2b3ed3 by David Rosca at 2026-07-22T11:25:41+02:00 vulkan/video: Don't code AV1 decoder model info when not present We always coded decoder model info, with all values being zero if we got NULL pStdDecoderModelInfo from app. Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> (cherry picked from commit ff12737333ae32b11e784739cdcb569fe61c6db2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - a3000525 by David Rosca at 2026-07-22T11:25:41+02:00 vulkan/video: Fix coding AV1 operating points We always coded the default operating point, because the app input wasn't copied. Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> (cherry picked from commit 5e1cf2cb5da1f51782ea69b2083c5e61deb635d6) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 28201f10 by David Rosca at 2026-07-22T11:25:41+02:00 d3d12/video: Don't reset batches in fence_wait va frontend unlocks driver mutex while waiting on fences, so there is a race accessing codec when trying to reset batches. Cc: mesa-stable Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15886 (cherry picked from commit 42f636bee95409f8ac89a453ab0edcbe58530180) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - b273d82d by David Rosca at 2026-07-22T11:25:41+02:00 vulkan/video: Fix coding H265 ref pic list modification lists This should code num_ref_idx_l*_active entries, not minus 1. Change it to use the minus1 value to match spec. Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> Reviewed-by: Hyunjun Ko <[email protected]> (cherry picked from commit 90e7282527a5f3907264ed4b146c6004092fa183) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - bec00022 by David Rosca at 2026-07-22T11:25:41+02:00 vulkan/video: Fix coding H265 SPS pcm block sizes, inter ref pic set and = lt refs Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> (cherry picked from commit 35a89677c45d79a7d08682d8afc4454009ca27d4) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - ff8f9eee by Valentine Burley at 2026-07-22T11:25:41+02:00 panvk/ci: Update expectations for g610-vk-asan following VK CTS uprev Fixes: 4c8de8c202d ("ci/deqp: Update VK CTS to 1.4.6.1") Signed-off-by: Valentine Burley <[email protected]> Acked-by: Erik Faye-Lund <[email protected]> (cherry picked from commit c85e3522b592344ecb6de4bb00362108c0a80cd2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - e842b025 by Timothy Arceri at 2026-07-22T11:25:41+02:00 util/mesa: add workaround to zero invalidated buffers Little Inferno streams its vertex data through glMapBufferRange(GL_MAP_WRITE_BIT | GL_MAP_INVALIDATE_RANGE_BIT), but only partially writes each 32-byte vertex, relying on the bytes it doesn't write reading back as zero. llvmpipe happens to satisfy this: it maps buffer storage directly and its allocations start out as kernel-zeroed pages, so the never-written bytes remain zero. radeonsi and iris recycle buffer memory (BO cache and staging upload buffers) without clearing it, so those bytes contain stale data and the game renders corrupt geometry. The game is relying on undefined behaviour; per the spec: "MAP_INVALIDATE_RANGE_BIT indicates that the previous contents of the specified range may be discarded. Data within this range are undefined with the exception of subsequently written data." Here we work around this with a new driconf option (zero_invalidated_buff= ers) that zeroes the memory returned by write-only invalidating maps, and enable it for Little Inferno. Cc: mesa-stable Generated-by: Claude (Fable 5) Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/7903 Reviewed-by: Pierre-Eric Pelloux-Prayer <[email protected]= om> (cherry picked from commit 8a6f0fca20cd2e9ce47f3f562a44fa6a0c8f25e1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 09423397 by Juan A. Suarez Romero at 2026-07-22T11:25:41+02:00 v3d: save fragment constants on sand8/sand30 blit These blit functions were setting a fragment constant buffer for their own operation, but relied on current constant buffer to be saved before, which actually didn't happen as the BLIT operation didn't include it. So they when restoring the constant buffer, nothing was saved to restore, which leads to crashes. Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15857 Fixes: a69aaadec1b ("v3d: add more blitter ops for saving resources") Signed-off-by: Juan A. Suarez Romero <[email protected]> Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 8b79064907bb8b0677fdae942483e20b54cf4a3d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 2c986ebd by Tacodiva at 2026-07-22T11:25:41+02:00 vulkan/runtime: Fix bad assumption in GetPipelineBinaryDataKHR pPipelineBinaryDataSize is not guaranteed to be 0-initialized when callin= g `vkGetPipelineBinaryDataKHR` without an output buffer. This commit change= s the check to see if the buffer is null instead of the possibly uninitialized length being 0. Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15901 Fixes: 5c47ac640b ("vulkan/runtime: implement VK_KHR_pipeline_binary") Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit 53a453b3ef7f3eb6a0e7ae37bd93985916497885) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - ab26aea4 by Sushma Venkatesh Reddy at 2026-07-22T11:25:41+02:00 intel/dev: Clamp PTL+ CS workgroup threads to 32 PTL introduces compute workgroup limits that depend on SIMD width and GRF usage. As a simple workaround suitable for stable releases, clamp max_cs_workgroup_threads to 32 on PTL+. This causes brw_simd_should_compile() to reject SIMD16 for large compute workgroups requiring more than 32 hardware threads and select SIMD32 instead, avoiding unsupported configurations while preserving the advertised 1024-invocation workgroup limit. This is a conservative workaround until PTL workgroup limits are modeled more accurately. Cc: mesa-stable Reviewed-by: Francisco Jerez <[email protected]> Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit ea77d4e261806c642f168beb5aac671b626bb2c5) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 9c94df5e by Danylo Piliaiev at 2026-07-22T11:25:41+02:00 tu: Custom resolve should always use AVOID_CCU layout E.g. dynamic subpass with custom resolve always has resolve_count =3D 0, but still needs TU_GMEM_LAYOUT_AVOID_CCU. Fixes: ad84ae27190 ("tu: Implement VK_QCOM_subpass_shader_resolve") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit 02035be13a3492c0a5a639a21a956693848c5c1f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - 54d67238 by Timothy Arceri at 2026-07-22T11:25:41+02:00 util: add workaround for Riddick using round() in glsl 1.20 Round is only advertised in 1.30 but the game uses 1.20 so force 1.30 Also fix up stray tab on the disable_uniform_array_resize line while we are updating this games workarounds. Cc: mesa-stable Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/11083 Reviewed-by: Pierre-Eric Pelloux-Prayer <[email protected]= om> (cherry picked from commit efd508cf90d597bb11c874674366c8ee8641745d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - f863cd46 by Timothy Arceri at 2026-07-22T11:25:41+02:00 llvmpipe: emit FS input vertex attributes in driver location order compute_vertex_info() emitted the vertex attributes by walking the FS input variables in declaration order, while the FS reads the input with driver location i from attrib i + 1. Since dbbb46aa38c the two orders can disagree: the FS inputs PRIMITIVE_ID, VIEWPORT and LAYER are assigned driver locations after all other inputs while the variable list is sorted by location, so the FS read its inputs from the wrong attribs when gl_PrimitiveID was used together with user varyings without a geometry shader. Emit the attributes indexed by driver location instead. This also removes the slot_emitted dedup, which could skip an attrib and shift all following attribs away from their FS input mapping; variables that share a slot (packed components) collapse onto the same driver location naturally. Fixes: dbbb46aa38c ("nir: compute io base for fragment shader inputs whic= h maybe per primitive") Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/15660 Reviewed-by: Dave Airlie <[email protected]> Generated-by: Claude Code (Claude Fable 5) (cherry picked from commit 90c66c69767add9a374445ebf1052fa683edab39) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43133= > - - - - - a5ab5f0c by Eric Engestrom at 2026-07-22T15:37:39+02:00 VERSION: bump for 26.2.0-rc2 - - - - - 8b44a451 by Eric Engestrom at 2026-07-29T13:35:26+02:00 .pick_status.json: Update to 8b00adbe72f2705985146b057f6fde9256d0dcb0 Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 2ea5bad2 by Eric Engestrom at 2026-07-29T13:35:26+02:00 .pick_status.json: Mark 47efd739121d51e2f9049cec715a70c13767a67c as denom= inated Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 732feb86 by Eric Engestrom at 2026-07-29T13:35:26+02:00 .pick_status.json: Mark 7999060992e9cee91d1962faf65dc4e5c6fe4f69 as denom= inated Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 84108887 by Eric Engestrom at 2026-07-29T13:35:26+02:00 .pick_status.json: Mark 2515024a5919ed14fe05471e3f1f89c54a454610 as denom= inated Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - edf586c5 by Eric Engestrom at 2026-07-29T13:35:26+02:00 .pick_status.json: Mark 4fd93a0039a07ec2027f2a6d1d252c73ed033e0a as denom= inated Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 17cded78 by Konstantin Seurer at 2026-07-29T13:35:26+02:00 radv/bvh: Fix updating acceleration structures containing AABBs If the node type is not set, child_index will not be calculated correctly. Fixes: 405c93c ("radv: Optimize BVH4 acceleration structure updates") Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15848 Reviewed-by: Natalie Vock <[email protected]> (cherry picked from commit 67eae5e5fd8c918e16d671a2255426a4af4e66c0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - ae48afbe by David Rosca at 2026-07-29T13:35:26+02:00 va: Fix leak when vlVaUploadImage fails Fixes: 153b2f21082 ("va: Add vlVaPostProc and use it instead of composito= r and vid engine blit") Reviewed-by: Pierre-Eric Pelloux-Prayer <[email protected]= om> (cherry picked from commit 0e20116a2fa88a19630dd7e0f1092d1ce7bf9cb5) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - a8da95a8 by David Rosca at 2026-07-29T13:35:26+02:00 va: Ensure templat is valid for temporary surfaces This is later used by vlVaVideoSurfaceSize and vlVaRegionDefault to get the unaligned surface size. Fixes: 153b2f21082 ("va: Add vlVaPostProc and use it instead of composito= r and vid engine blit") Reviewed-by: Pierre-Eric Pelloux-Prayer <[email protected]= om> (cherry picked from commit a187e191c0c0947d171f6c0f646e70a85c8600f0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 22400dee by Sonny Jiang at 2026-07-29T13:35:26+02:00 radeonsi: always set is_format_supported in screen create On GFX12 ASICs where neither ACO nor LLVM reports shader compiler support, si_init_gfx_screen() returns early before reaching si_init_screen_state_functions(), leaving screen->is_format_supported as NULL. When a VA-API client calls vaQueryImageFormats, vl_video_buffer_is_format_supported() dereferences the NULL pointer, causing a segfault. Fix by assigning is_format_supported directly in radeonsi_screen_create_impl(), ensuring it is always set regardless of shader compiler availability. We intentionally avoid calling the full si_init_screen_state_functions() here, as its vertex_state_cache init would leak =E2=80=94 the corresponding deinit only runs from si_fini_gfx_= screen which is skipped on the no-compute path. Signed-off-by: Sonny Jiang <[email protected]> Reviewed-by: [email protected] (cherry picked from commit 34190c25c149d97fdd0bded229f21261c8232831) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 690b449c by Sonny Jiang at 2026-07-29T13:35:26+02:00 radeonsi/vcn: Add vcn_5_0_2 support Signed-off-by: Sonny Jiang <[email protected]> Reviewed-by: Boyuan Zhang <[email protected]> (cherry picked from commit ddbdc9826344d121f608fc77ede21a0d9cc2eb2f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - e8589ba2 by Connor Abbott at 2026-07-29T13:35:26+02:00 tu: Fix condition for skipping emitting aprons We were trying to detect the case where both tiles are not subsampled and they touch each other, i.e. for tiles in the high-density region in the center, and we therefore must not setup an apron. However we only checked if their edges have the same x or y coordinate, and that is not enough to say their edges touch, because we have to check the other coordinate as well. For example, when processing horizontally-adjecent tiles, we may have a situation like this in subsampled space: +--------------+--------------+ | | | | | <- other_tile in | |------+ subsampled space | | | | other_tile | +--------------+ size=3D(2, 2) | | | pos=3D(0, 1) | | tile | area=3D(4, 4) | | size=3D(1, 1) | | | pos=3D(1, 0) | | | area=3D(1, 1) | | | | | +--------------+--------------+--------------+ Because their vertical edges are at the same y coordinate, we'd assume they're touching in subsampled space when actually they're not. Fixes: 4b87df29b3a ("tu: Implement subsampled images") (cherry picked from commit 9e40eb6c80cdd88c12326b3bd6ffda8bd6305d3d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 73884644 by JaeHoon Lee at 2026-07-29T13:35:26+02:00 nvk: free copy_memory_indirect_temps on command buffer destroy nvk_create_cmd_buffer() initializes two util_dynarrays, cmd->pushes and cmd->copy_memory_indirect_temps. nvk_destroy_cmd_buffer() finalizes the former but not the latter. Both are zero-initialized (mem_ctx =3D=3D NULL= ), so their backing storage comes from realloc() and is released by util_dynarray_fini(); nvk_reset_cmd_buffer() clears them without freeing,= keeping the allocation for reuse. copy_memory_indirect_temps grows a heap buffer the first time vkCmdCopyMemoryIndirectKHR is recorded (util_dynarray_ensure_cap() in nvk_cmd_indirect.c), and that buffer is not freed: destroying the command= buffer frees the struct but not the storage the dynarray points at. The nvk_cmd_mem entries it held are tracked separately in cmd->owned_mem and are freed there, so the leak is the pointer array itself. Finalize it alongside cmd->pushes. Fixes: cbec12627bb ("nvk: VK_KHR_copy_memory_indirect") Reviewed-by: Mel Henning <[email protected]> (cherry picked from commit 84c7fef4581a4bd6673923240dab1a0fb6795e7a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 7fbd3d93 by Valentine Burley at 2026-07-29T13:35:26+02:00 panvk: Disable VK_KHR_internally_synchronized_queues on Vulkan 1.0 The extension depends on Vulkan 1.1. Fixes: 3b1f86d437f ("panvk: Advertise VK_KHR_internally_synchronized_queu= es") Signed-off-by: Valentine Burley <[email protected]> Reviewed-by: Christian Gmeiner <[email protected]> Reviewed-by: Aksel Hjerpbakk <[email protected]> (cherry picked from commit 79293b23366199314c33172e3ded126913e799f8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 22912cba by Lorenzo Rossi at 2026-07-29T13:35:26+02:00 panfrost: Fix wrong layout reuse in user clip planes OpenGL user clip planes reuse the same fragment shader with a different vertex shader even when it was linked. Since the FS uses the previous memory layout, but the vertex shader emitting it is different, this causes all sort of problems. This commit fixes the bug by disabling memory layout optimizations when UCP is enabled. Fixes: 32ca5f7515d ("panfrost: Plumb VS varying_layout in FS") Signed-off-by: Lorenzo Rossi <[email protected]> Reviewed-by: Faith Ekstrand <[email protected]> Reviewed-by: Jakob Sinclair <[email protected]> (cherry picked from commit 73e7b3847f3231ac40da194ed384ac781d8cc6c2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 12ce20ba by JaeHoon Lee at 2026-07-29T13:35:26+02:00 v3dv: avoid restoring stale descriptor state after a meta op v3dv_cmd_buffer_meta_state_push() skips the descriptor_state memcpy when the current gfx descriptor_state is not valid. The skip path doesn't propagate that invalid state into state->meta, so state->meta's copy keeps whatever a previous meta op's memcpy left there. has_descriptor_state is still unconditionally set to true, so pop cannot tell push saw an invalid state this time: it reads the stale valid flag as current and, if nonzero, restores a descriptor set pointer left over from an earlier operation instead of the app's actual state. Clear meta's descriptor_state.valid in the skip branch so pop can observe the invalid state instead of reading what was left behind. Also fix a "subpass" typo in the neighboring attachment-state comment in the same function, found while reading through this code. Fixes: e59e7069283 ("v3dv: don't reset descriptor state after a meta oper= ation") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 0ff8b37a44ed8d8e76beb7c5df48e5a0867c7053) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - d67e05f8 by Benjamin Cheng at 2026-07-29T13:35:27+02:00 vl/proc: Guard compositor creation with HAVE_GFX_COMPUTE Decode-only builds were broken because the vl_compositor was being created outside of stubbed entrypoints. Fixes: 1c8cec690cc ("va: Use vl_proc for processing context") Signed-off-by: Benjamin Cheng <[email protected]> Reviewed-by: David Rosca <[email protected]> Reviewed-by: Boyuan Zhang <[email protected]> (cherry picked from commit f083b55523ce72ed3654c35ac415f7045ed55f69) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - c5411590 by JaeHoon Lee at 2026-07-29T13:35:27+02:00 nvk: report fills from memory correctly The pipeline executable statistic for fills from memory reports the spill= count. Change it to report fills from memory instead. Fixes: 2a0302967feb ("nak: Add spill/fill statistics") Reviewed-by: Mel Henning <[email protected]> (cherry picked from commit eadcadf067ede525b5f511d22215272bb31eaa41) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 4432d3d3 by Christian Gmeiner at 2026-07-29T13:35:27+02:00 mesa/st: Invalidate FS sampler views after PBO texture transfers Both PBO paths unbind all FS sampler views when they finish but the invalidate mask lost ST_NEW_FS_SAMPLER_VIEWS in the conversion from explicit dirtying. A following draw that does not rebind its textures then samples from no bound view and renders black. Seen as a black screen in dosbox, which uploads via glTexSubImage2D(..) from a PBO and then draws a textured quad from a display list without touching any texture state in between. Fixes: 62efee18607 ("mesa/st: Convert st_cb_texture to use st_context_inv= alidate_state") Signed-off-by: Christian Gmeiner <[email protected]> Reported-by: David Witbrodt <[email protected]> Tested-by: David Witbrodt <[email protected]> Reviewed-by: Marek Ol=C5=A1=C3=A1k <[email protected]> (cherry picked from commit 479773c7e4264506f2d9ec4bf15c6bf677f0d67a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 380dd8b9 by Dmitry Baryshkov at 2026-07-29T13:35:27+02:00 tu: limit KHR_internally_synchronized_queues to Vulkan 1.1+ The extension depends on VK 1.1. Fixes: 879dd9ca8c3 ("tu: Advertise VK_KHR_internally_synchronized_queues"= ) Signed-off-by: Dmitry Baryshkov <[email protected]> (cherry picked from commit 67a59892af13e821ab90a99e9c19132f9d5869f2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 6027574a by Dmitry Baryshkov at 2026-07-29T13:35:27+02:00 tu: limit VALVE_fragment_density_map_layered to Vulkan 1.1 devices VK_VALVE_fragment_density_map_layered requires Vulkan 1.4, VK_KHR_extended_flags or VK_KHR_maintenance5 (which in turn requires at least Vulkan 1.1). Limit the VALVE_fragment_density_map_layered extension to Vulkan 1.1 until the driver implements KHR_extended_flags extension. Fixes: a3a53b7cee4a ("tu: Implement VK_VALVE_fragment_density_map_layered= ") Signed-off-by: Dmitry Baryshkov <[email protected]> (cherry picked from commit 8116745e085bcf2d66575f9e6d25857352abee89) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 652b088a by Frank Bouwer at 2026-07-29T13:35:27+02:00 pvr: Fix for depth stencil 2d array writes. The base array layer was not taken into account. This fixes a reverted co= mmit. original commit: d058d2adb315e67bb62f4ff2b292bfd751f8be7f revert commit: 7f01ea0094dda7614287912c9922ad7d3e761507 Fix dEQP-GLES: dEQP-GLES31.functional.texture.multisample.samples_*.use_texture_dept= h_2d_array Fix regressions introduced by original commit DEQP-VK: dEQP-VK.api.image_clearing.core.clear_depth_stencil_attachment.multiple_l= ayers.* dEQP-VK.api.image_clearing.core.partial_clear_depth_stencil_attachment.mu= ltiple_layers.* dEQP-VK.api.image_clearing.dedicated_allocation.clear_depth_stencil_attac= hment.multiple_layers.* dEQP-VK.api.image_clearing.dedicated_allocation.partial_clear_depth_stenc= il_attachment.multiple_layers.* Backport-to: 26.1 Signed-off-by: Frank Bouwer <[email protected]> Reviewed-by: Frank Binns <[email protected]> (cherry picked from commit 1d899a489de21796a28075b8a4aed8a5954b0057) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - ca6ef3b7 by Danylo Piliaiev at 2026-07-29T13:35:27+02:00 tu: Fix subsampled metadata and blit emission for separate stencil Fixes: 4b87df29b3a ("tu: Implement subsampled images") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit bf70ab68a2105b24e211e3862337e95c7c2c97fd) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - ebf01f0b by Toshinari Morikawa at 2026-07-29T13:35:27+02:00 egl: avoid calling loader_get_driver_for_fd with fd =3D -1 platform_android.c, platform_surfaceless.c does not initialize fd_display_gpu before calling dri2_create_screen so that "libEGL warning: failed to get driver name for fd -1" was always printed on log. Cc: mesa-stable Tested-by: Dieter N=C3=BCtzel [email protected] Reviewed-by: Pierre-Eric Pelloux-Prayer <[email protected]= om> Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/13850 Signed-off-by: Toshinari Morikawa <[email protected]> (cherry picked from commit 0bc2efab05ce6cf1dc90aeeff7e064482c127b2d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 4fe4782d by Georg Lehmann at 2026-07-29T13:35:27+02:00 nir: fix divergence for deref_cast The old check was broken because deref->var is a union member, and not valid for deref_cast. Either it's a trivial bit size changing cast, or it's a pointer to deref cast. In the first case we can rely on the parent divergence, in the second we have to check if the deref could be of a mode where even a uniform pointer would mean a different address per invocation. Fixes: df86c5ffb35 ("nir: add divergence analysis pass.") Reviewed-by: Marek Ol=C5=A1=C3=A1k <[email protected]> (cherry picked from commit 2c3c8b7fc267a226ad5d38035b9e86526d6063d2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 0ffb53c2 by Peng_Lx at 2026-07-29T13:35:27+02:00 turnip/kgsl: close the dma-buf fd of our own allocations The three internal allocation paths -- dma-heap, ION and legacy ION -- obtain a dma-buf fd from the allocator and hand it straight to tu_bo_init_dmabuf() as the function's return value. tu_bo_init_dmabuf() only borrows that fd: kgsl_bo_init_dmabuf() keeps a dup of its own in bo->shared_fd, and that dup is what kgsl_bo_finish() closes. Borrowing is the right contract for an imported dma-buf, where the fd belongs to the caller, but these three allocators created the fd themselves, so the reference they hold is never dropped. The result is that every internally allocated shareable BO leaks one dma-buf fd. kgsl_bo_finish() frees the KGSL handle and closes its dup, but the dma-buf refcount never reaches zero, so the memory stays allocated for the lifetime of the process. It is only reachable for exportable BOs, which makes it easy to miss: ordinary offscreen textures are unaffected. It shows up on a Wayland compositor, whose EGL window back buffers are exportable and are reallocated whenever Mesa trims the ones that went idle. Toggling the niri overview on an Adreno 830 (KGSL, 3840x2560 panel) leaked one 37.5 MB buffer per toggle, growing without bound -- 80 toggles took dma-buf usage from 202 MB to 3956 MB. With this patch the same run stays flat at 202-239 MB, and the allocate/close ledger balances exactly. Fixes: 17c12a9924a0 ("turnip/kgsl: Support external memory via ION/DMABUF= buffers") Cc: mesa-stable Signed-off-by: Peng_Lx <[email protected]> (cherry picked from commit 9f955af118d0cdabae776d08e36f99f7aa92317b) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - b4e4d4ac by Caio Oliveira at 2026-07-29T13:35:27+02:00 nir: Account for cmat memory accesses in copy_prop_vars Cc: mesa-stable Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15932 Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit b277a4925a161592801d711a8258433e2e2cda42) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 71378f32 by Marek Ol=C5=A1=C3=A1k at 2026-07-29T13:35:27+02:00 radv: fix determining the raster prim for guardband No change in correctness, but this is more accurate and faster because we= now have a tighter guardband around the viewport more often. The following commits that are fixes depend on this. Cc: mesa-stable Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit c61fdeb64c4765927708a1426426055266ba8776) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - c9ffe755 by Marek Ol=C5=A1=C3=A1k at 2026-07-29T13:35:27+02:00 radv: fix determining the raster prim for line mode Cc: mesa-stable Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit 6b534309ead43c157b11f467bbdc059dd1816dcc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 6ac27d7f by Marek Ol=C5=A1=C3=A1k at 2026-07-29T13:35:27+02:00 radv: fix determining the dynamic raster prim for FS barycentrics Cc: mesa-stable Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit 9a77ee629d75f35a048649462c44fdc3f1e1d676) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - f7b88a24 by Marek Ol=C5=A1=C3=A1k at 2026-07-29T13:35:27+02:00 radv: fix determining the static raster prim for FS barycentrics and fron= t_face This also happens to fix the replacement of front_face with a constant wh= en face culling is enabled, improving performance. No front_face replacement= was done before this commit according to gpu-ratemeter. That may have bee= n because the unknown_rast_prim condition was incorrect and its result was unnecessarily true most of the time. Cc: mesa-stable Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit edeb6dd258b216bc45c9b2e7d260b01cce81e3af) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - d0ccc7b8 by Karol Herbst at 2026-07-29T13:35:27+02:00 rusticl/memory: return 0 for CL_IMAGE_SLICE_PITCH also for 2d images Backport-to: * Tested-by: Rob Clark <[email protected]> Reviewed-by: Rob Clark <[email protected]> (cherry picked from commit 1b0f0e0a83c4b20a0a24df41bafc6d0ca88d5fcb) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - bea70de7 by Ryan Mckeever at 2026-07-29T13:35:27+02:00 docs: advertise VK_KHR_multiview support for Bifrost Fixes: a1376449c80 ("panvk: Expose support for multiview on v7") Reviewed-by: Lorenzo Rossi <[email protected]> Reviewed-by: Erik Faye-Lund <[email protected]> (cherry picked from commit f11071ce865144ca3ff759c9d1a6c46ede06a16e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 8bb6d464 by Caio Oliveira at 2026-07-29T13:35:27+02:00 anv: Include build and device identity in shader binary UUID In anv_physical_device.c we document that shaderBinaryUUID includes the driver build identity, but anv_shader_init_uuid() only hashed compile= r knobs and workaround settings. Add the build id and device info to the U= UID so driver or device-specific shader binary ABI changes invalidate VK_EXT_shader_object binaries. Fixes: 4314c891f4c ("anv: expose VK_EXT_shader_object") Assisted-by: Pi coding agent (GPT-5.5) Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit c4039495967db0bff1ffaf2c955bc8586ce9c269) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 06845a1b by Yiwei Zhang at 2026-07-29T13:35:27+02:00 venus: clean up sync2 treatment leftovers (cherry picked from commit ebf0bafa7dae458208c4fa2c75f074dd22fe27b5) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - d083a51d by Yiwei Zhang at 2026-07-29T13:35:27+02:00 venus: simplify sync feedback tracking (cherry picked from commit 5c46ee7379e5f3f95d48402b1f7b9560ac3276aa) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - e9464e48 by Yiwei Zhang at 2026-07-29T13:35:27+02:00 venus: update pnext fix tracking Also move pnext alloc and assign right after main batch. (cherry picked from commit 9dd3d12f2d1ddabdd90992f3a27398d8c5d0d63e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 82c84250 by Yiwei Zhang at 2026-07-29T13:35:27+02:00 venus: explicitly track if need to fix batch This is to prepare for later fixes that don't need a batch: e.g. renderer sync preparations (cherry picked from commit 0406c3b0ae07b98a5842d0f6164ce71f752bdcce) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 60b5537d by Yiwei Zhang at 2026-07-29T13:35:27+02:00 venus: flatten sync feedback cmd counting This change: 1. move ffb cmd counting to ffb detection time 2. early return after suspending ffb and sfb if unable to do feedback 3. flatten qfb and later device group submit detection (cherry picked from commit b1f44187bc8ed1b1ab6e91553ac4cf5cd0326c66) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - ff18906a by Yiwei Zhang at 2026-07-29T13:35:27+02:00 venus: deprecate fence feedback This puts fence back to synchronous driver<->renderer polling temporarily. (cherry picked from commit 4c1938c8adbe0e328f9db69362e79105c1e52d56) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 12632852 by Yiwei Zhang at 2026-07-29T13:35:27+02:00 venus: split ring submission to vn_queue_submission_do_submit (cherry picked from commit 2841b9b173c78cbe76d586070315fdfe05d64d2a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - e18e363c by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: skip empty batch submission With do_submission focusing on ring submission, now we can skip the empty batch after fully resolving the batch. This is to prepare for fence and sempahore being dropped from submission. (cherry picked from commit ca69ef0660963b9023dbe0265d1af9d9544de764) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 82d4f280 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: drop vn_sync_payload_external from submission tracking struct vn_sync_payload_external will be deprecated later after migrating external fence and semaphore over to syncobj. (cherry picked from commit 044c099959c22a949fa9866de51060c30c35c5e7) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 36205ba4 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: extract vn_timeout_to_poll_timeout (cherry picked from commit c90f8dfd01d45e6efbc3878492bba08d5ec1ebcc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 5ab4de36 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus/virtgpu: only signal non-zero initial value (cherry picked from commit 29abb028d912130bc4f57fef19d84a1f1cb4cc18) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - a34ca686 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus/virtgpu/vtest: drop initial value from sync reset No use cases to reset to non-zero. (cherry picked from commit cfe2a5987f4e9009a2ea8dcf45c5df3febff58e6) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 00b415ff by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: prepare for VN_SYNC_TYPE_SYNC This submits the syncs to renderer with a dependency on the last queue submit ring seqno. (cherry picked from commit d921c271c8b6ef53f9b0a52e9021403739ad3c0f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 275c5916 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: migrate fence over to VN_SYNC_TYPE_SYNC This change: - stops creating renderer fence object - drops fence->external_payload trick for SYNC_FD fence export - adopts vn_renderer_sync for fence support in: - vn_CreateFence/vn_DestroyFence/vn_ResetFences - vn_GetFenceStatus/vn_WaitForFences - vn_ImportFenceFdKHR/vn_GetFenceFdKHR - skips ring submission if submitting just a fence with empty batch To be noted, VN_SYNC_TYPE_IMPORTED_SYNC_FD is preserved in favor of: - handling of -1 across drm_syncobj and sim_syncobj - vtest backend compatibility - avoiding unnecessary virtgpu ioctls (cherry picked from commit 6aded3b9d65b8af85f1684d042249f1a0b32a9c9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - c073f335 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: relax SYNC_FD fence export requirement No longer need such requirement after fence being migrated to driver side syncobj. (cherry picked from commit 772cb76cc80ed5bde9c8a3972f0852d68fa54f9d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - d32fa110 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: deprecate queue idle wait workaround Now that fence payload is submitted via the renderer, it has fenced prior submissions on the same ring_idx. We can safely drop this semi-working roundtrip. (cherry picked from commit e966e8293b6d5ae0519c6cae00a606b48de9a2ea) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 20a0c785 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: rename to be explicit about sync fd semaphore Let's call out sync fd explicitly. The special treatment isn't specific to imported sync fd but will also apply to exportable sync fd semaphore. We can't preserve the renderer object without requirement more on the renderer side due to the copy transference nature of sync fd. (cherry picked from commit f69094c19828feba0e65ba5c2530c00e5a2c8f2d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 34af8e27 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: vn_semaphore_(is|wait)_sync_fd to support VN_SYNC_TYPE_SYNC (cherry picked from commit 919b44d331305461cde32a6646df479e9fdf615c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 91fa1d93 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: rename existing queue submission wait semaphore tracking Make it specific to wait semaphores. (cherry picked from commit 25fdd7a27ef446ee052b9861a8b7f0d460f27323) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 840221a3 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: count and prepare storage to scrub SYNC_FD signal semaphore (cherry picked from commit 90f58caf8cde8853499eaa1163f96321696be2be) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - c0ceec78 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: extract syncs and scrub SYNC_FD signal semaphores 1. properly fix the pNext chain for those to be dropped 2. extract the syncs from the SYNC_FD semaphores 3. drop those semaphores from the ring submission batch (cherry picked from commit 6f3a570d4181e4c762388afed8a12a2f6f92c5de) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - d5c11ff3 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: ensure renderer sync fence is submitted between queue batches Previously, the later queue batches might end up submitted before the sync fence submission on the renderer side. Now we submit roundtrip and have the followed queue batch submission via the ring to properly wait for the submitted roundtrip, which ensures the completion of sync fence submission. (cherry picked from commit 36f79c3c6459bd3442519b223be071e15582b15c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - c23bf808 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: migrate SYNC_FD semaphore over to VN_SYNC_TYPE_SYNC This change: - stops creating renderer semaphore object for SYNC_FD semaphore - drops sem->external_payload trick for SYNC_FD semaphore export To be noted, VN_SYNC_TYPE_IMPORTED_SYNC_FD is preserved in favor of: - handling of -1 across drm_syncobj and sim_syncobj - vtest backend compatibility - avoiding unnecessary virtgpu ioctls (cherry picked from commit c5eda762851d1ec8226e10ecd20795fe2511ea6d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 88114a51 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: relax SYNC_FD semaphore export requirement (cherry picked from commit abbaca283bbde0d208cc1b5ff5c39ea609799a34) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 885c401d by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus/virtgpu: hide has_timeline_sync behind a new perf option This change: 1. adds a new VN_PERF=3Dno_timeline_sync perf option 2. refactors to use renderer instead of gpu->base indirection (cherry picked from commit b17c963ec861abc74900b245915d680d8fe69408) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - dbdd5b7c by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus/vtest: advertise timeline syncobj support The vtest interface does support timeline sync. Similarly, hide behind the new perf option. (cherry picked from commit b64c5743837d0564131ef901e5b073e3dcfdbad8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - c9fd788b by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: drop vn_renderer_sync_flags At this point, I'm pretty sure we don't need this flag at all. Meanwhile, drop some obsolete TODOs. (cherry picked from commit 23c8ebd60c79c43599675042dc8d613095df34eb) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 6b1a82e8 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: add VN_SYNC_TYPE_TIMELINE_SYNC (cherry picked from commit 0ed627fed42316bc3e4ff0c962507b632c8a9d2c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 975a86f6 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: track queue internal array index within vn_device::queues This will be used to map queue timeline to timeline semaphore syncs. (cherry picked from commit 71732947f30be3d9def4a43884394b38ca86e580) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 9dccd7a6 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: count and init syncs from timeline semaphores For timeline semaphore, the wait and signal semaphores are both forwarded to the renderer side for device operation. So we only count the signal timeline semaphores and submit those extracted syncs. (cherry picked from commit fe293af9f18a5bb87b15bb85cf96bfb5c80e0961) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - f707650b by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: implement host signal for TIMELINE_SYNC Only need to signal the cpu timeline sync (back of the payload syncs) to the requested counter value. (cherry picked from commit 4219bb97c5ae5fbcdfe84d122017bfec74b99e92) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 8c2b91be by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: implement counter query for TIMELINE_SYNC One of the TIMELINE_SYNC payload syncs has the current point of the timeline semaphore, so we query each of them to return the greatest. (cherry picked from commit cf06a29d206ae140a5dd15c8a2bfea7f51767726) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - b34e5b2d by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: add vn_wait_semaphores_legacy for legacy wait Add a simple sem payload type test to prepare for timeline sync. (cherry picked from commit 85543981c2f002555f1ae42cf4d77b99dbfbcc87) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 5fce4ae9 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: implement semaphore wait for TIMELINE_SYNC For WAIT_ANY, collect all syncs from all semaphores and wait any one of them to signal with the passed timeout in one pass. Otherwise, wait for individual semaphores one by one. For each of the semaphores, wait for any one of the semaphore timeline syncs to signal. (cherry picked from commit f72c0018463eaed412eee984537d2f94d3511aac) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 284dca2d by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: migrate timeline semaphore to VN_SYNC_TYPE_TIMELINE_SYNC (cherry picked from commit 0959c8ec29776c054bb07d48609d71093e37f2b1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - be915e54 by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: document timeline semaphore implementation (cherry picked from commit 8b794a5167a729758cac9bde45fbe1886eb89c91) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - d1a61399 by Eric Engestrom at 2026-07-29T13:35:28+02:00 pick-ui: turn commit.date into a (cached) property This is a property of the commit, not an action performed on the commit, but it's expensive to compute so let's cache it. (cherry picked from commit b6386e96e5a81e074a90dff724ef3c35b1a9d1e7) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - dc260c14 by Eric Engestrom at 2026-07-29T13:35:28+02:00 pick-ui: show MR number for additional context Shows which commits belong together, and shows context for problematic commits (doesn't backport, doesn't build, fails the unit tests, breaks a CI test). (cherry picked from commit 2086df71501a2e2934d36a33d5ea9087057423e9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - e7b165db by Karol Herbst at 2026-07-29T13:35:28+02:00 rusticl/kernel: add libclc source hash to kernel shader keys Backport-to: * (cherry picked from commit 780727e68adc62f446dc1df7bcaf86f20b6c03d1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 882dcedd by Christoph Pillmayer at 2026-07-29T13:35:28+02:00 pan/kraid: Fix 32bit hw_runner builds On armhf, unsigned long is uint32_t, so passing a u64 into drmIoctl doesnt work. Also don't pull in everything into bindgen. Fixes: 1d171a41 ("kraid: Add hw_runner") Reviewed-by: Lars-Ivar Hesselberg Simonsen <[email protected]> (cherry picked from commit 6cc038ecd1fdf6b654d7b976752a68bc60ba7b50) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 80747bae by Christoph Pillmayer at 2026-07-29T13:35:28+02:00 pan/perf: Fix 32bit build of panquick Fixes: 5c15ef50 ("pan/perf: Transition to auto-generated derived counters= ") Reviewed-by: Erik Faye-Lund <[email protected]> (cherry picked from commit 6015d5e538d531f230bb102e0af10eed566e6900) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - b7dc7ca1 by Rhys Perry at 2026-07-29T13:35:28+02:00 ac/nir/ngg: don't shrink device-scope memory barriers Signed-off-by: Rhys Perry <[email protected]> Backport-to: * Reviewed-by: Marek Ol=C5=A1=C3=A1k <[email protected]> (cherry picked from commit 027a2b45fcb69ed5e421e6a3f829ea1fd7c65787) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 26eb9ed2 by Rhys Perry at 2026-07-29T13:35:28+02:00 ac/nir/ngg: track uniformity with multiple set_vertex_and_primitive_count= This can be in uniform control flow. Signed-off-by: Rhys Perry <[email protected]> Fixes: 863fdcf5196 ("ac/nir/ngg: Use workgroup divergence analysis for me= sh output counts.") Reviewed-by: Marek Ol=C5=A1=C3=A1k <[email protected]> (cherry picked from commit ba8dfc5aa8ac3af757d7330f69e1cd8573f5772e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - bd94a96c by Yiwei Zhang at 2026-07-29T13:35:28+02:00 venus: ensure cached vn_ring_submit batches are bounded The prior fix can still end up with unbounded leak since the actual shmem count in the storage isn't tracked. This change reverts the prior fix while avoiding caching over-sized batch. Practically, there's no need to cache more since the malloc overhead is trivial as compared to indirect uploads. Fixes: 2cf1f6cb508 ("venus: fix unbound malloc leak in vn_ring_get_submit= s") (cherry picked from commit 09fb7ca8d824616712e8d3b2639240087800e3d5) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - b6130bce by Danylo Piliaiev at 2026-07-29T13:35:28+02:00 tu: Fix gfx_write_access checking for TRANSFORM_FEEDBACK_COUNTER_READ_BIT= It should check for WRITE_BIT. Fixes: 59259a01671 ("tu: Convert to sync2 entrypoints") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit 221d58dc47746e344d10b3149322131b4e0f2b27) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 3a789ca9 by Lionel Landwerlin at 2026-07-29T13:35:28+02:00 anv: fix push buffer descriptor address relocation Fixes a few tests on < Gfx12.5 like : ANV_DEBUG=3Dexperimental ./deqp-vk -n dEQP-VK.dgc.ext.compute.misc.desc= riptor_buffer_push_descriptor_with_ies Signed-off-by: Lionel Landwerlin <[email protected]> Fixes: 1ffdbe276d ("anv: Replace va.indirect_descriptor_pool access with = a function") Reviewed-by: Tapani P=C3=A4lli <[email protected]> (cherry picked from commit 849eb7067fbaddf07c350a5d00760d700e27de86) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - fbed83c8 by Janne Grunau at 2026-07-29T13:35:28+02:00 asahi: nir: lower flrp64 Fixes unhandled ALU instruction asserts | Unhandled ALU op flrp | src/asahi/compiler/agx_compile.c:2160: agx_emit_alu: Assertion `!"Unhan= dled ALU instruction"' failed. observed since commit ac679a85049 ("glsl: translate fma as fma_weak") in following OpenGL CTS tests: - KHR-GL46.gpu_shader_fp64.builtin.mix_double - KHR-GL46.gpu_shader_fp64.builtin.mix_dvec2 - KHR-GL46.gpu_shader_fp64.builtin.mix_dvec3 - KHR-GL46.gpu_shader_fp64.builtin.mix_dvec4 Backport-to: 26.2 Signed-off-by: Janne Grunau <[email protected]> (cherry picked from commit ed30ed7c39d2fae442c4a353144ba1112e9f0d3d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 6ebbe5d7 by Danylo Piliaiev at 2026-07-29T13:35:28+02:00 tu: Fix blit_cache_cleaned never being set to true tu6_emit_flushes clears cache's flush_bits, resulting in the follow up `flush_bits & TU_CMD_FLAG_BLIT_CACHE_CLEAN` never being true, leading to excessive flushing when dynamic input attachments were used. Fixes: a99600322c1 ("tu: Track possible feedback loops for dynamic render= passes") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit ba8de990e36474c9b46a103ca274c52fc15c87b8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 409d6f1e by David Rosca at 2026-07-29T13:35:28+02:00 radeonsi: Stop forcing GTT with no gfx/compute This is no longer needed as staging texture is now never used without gfx/compute. Backport-to: 26.2 Reviewed-by: Pierre-Eric Pelloux-Prayer <[email protected]= om> (cherry picked from commit 1e70c9859d16a9fcca3d9d70c545c8e10c81c8e2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 119a930a by David Rosca at 2026-07-29T13:35:28+02:00 ac/video: Fix number of AV1 single refs 2 refs are only supported with compound. Fixes: 914b4a700cc ("ac/info: Add video codec caps") Reviewed-by: Ruijing Dong <[email protected]> (cherry picked from commit 6f0f36e3b1f9462335d717a9363d0880ae5726be) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 0b940b05 by Jaakko Jokinen at 2026-07-29T13:35:28+02:00 nir: Add cases to nir_get_io_offset_src_number() Added cases nir_intrinsic_store_shared_unlock_nv and nir_intrinsic_load_shared_lock_nv. This fixes segmentation faults caused by the unexpected default return value -1. Fixes: 47f6c74b71c ("nir,nak: Add KeplerB shared atomics intrinsics and l= owering") Reviewed-by: Mary Guillemard <[email protected]> (cherry picked from commit 61f6c264bcf917468854a82b2a0c9a0912bc6c1a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - f56d024f by Danylo Piliaiev at 2026-07-29T13:35:28+02:00 tu: Fix LRZ handling for VK_EXT_custom_resolve Custom resolve can resolve depth attachments, this case is a bit narrower that general case of RP using several depth attachment. Here we should have LRZ enabled for the main part of RP and disabling it only for custom resolve. Note, if custom resolve resolves depth we still disable LRZ for the main depth attachment _after_ the RP, which is not strictly necessary but I don't see a point in handling this case. Fixes: 520e3f3a473 ("tu: Implement VK_EXT_custom_resolve") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit adca1f255fd5cb9702db1e0a2bc25d13a521ae5f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - a6058026 by Danylo Piliaiev at 2026-07-29T13:35:28+02:00 tu: Dirty LRZ after changing attachment locations disable LRZ writes We called tu_lrz_disable_write_for_rp, but didn't trigger LRZ state reemission. Fixes: 7ee9835475c ("tu/lrz: Disable LRZ when CmdSetRenderingAttachmentLo= cations is used") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit 3c0803f5a17f13bad5c2f2da7cdeaee4f792eeb2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - ac675d7e by Lionel Landwerlin at 2026-07-29T13:35:28+02:00 brw: add missing INTEL_FS_CONFIG_PER_PRIMITIVE_REMAPPING handling Fixes tests in dEQP-VK.mesh_shader.ext.pipeline.builtin.fast_lib.* in DG2= . Signed-off-by: Lionel Landwerlin <[email protected]> Fixes: df5a6d7b87 ("brw/jay: move some coarse lowering to NIR") Acked-by: Alyssa Rosenzweig <[email protected]> (cherry picked from commit 3cd07aacb9ebe839bd73c9c5e07afff4db0e0d00) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 1c7425fa by Jan Meisel at 2026-07-29T13:35:28+02:00 radv: fix RADV_PERFTEST=3Dnircache enablement RADV_PERFTEST_NIR_CACHE is parsed from RADV_PERFTEST, but enable_nir_cache reads debug_flags after radv_compiler_info was introduced. This makes RADV_PERFTEST=3Dnircache a no-op. It also accidentally ties the feature to RADV_DEBUG=3Dinfo because both enums use bit 13. Use perftest_flags so the documented perftest flag enables per-stage graphics NIR caching again. On an RX 9070 XT (GFX1201), a local graphics pipeline benchmark with 1000 vertex-input permutations improved from 368.345 ms to 255.500 ms with RADV_PERFTEST=3Dnircache (-30.636%, n=3D5, interleaved cold-cache runs). Fixes: 4a91fd8babce ("radv: add a radv_compiler_info object") Reviewed-by: Konstantin Seurer <[email protected]> Assisted-by: Codex Signed-off-by: Jan Meisel <[email protected]> (cherry picked from commit ad0cdded0e439515155677941cada1cb4bb4d388) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 261a270c by utzcoz at 2026-07-29T13:35:28+02:00 virtio: magma-gpu-rs: accept a null device in virtgpu_kumquat_finish virtgpu_kumquat_init leaves the caller's pointer untouched when it fails,= and VirtGpuKumquatDevice releases the device unconditionally from its destructor, so virtgpu_kumquat_finish is reached with a null pointer on every failed initialization. Box::from_raw on a null pointer raises a non-unwinding panic, which aborts the process and takes with it the error= that explained why initialization failed. The surrounding catch_unwind cannot intercept that. Make the release side tolerate null instead of fixing the one caller. This is a C init/finish pair, and C's canonical release function treats null as a no-op, so C callers reasonably expect that; it also covers callers we do not control. Fixes: f2b07903abaf ("mesa: import virtgpu_kumquat_ffi") (cherry picked from commit 0989349030b36e6ce587cf70f2a5c7615465f04f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - dcfeffe0 by Caio Oliveira at 2026-07-29T13:35:28+02:00 brw: Match fill/spill optimization scratch accesses by logical offset brw_opt_fill_and_spill matches and invalidates scratch accesses by physical offset but removes dead spills by logical offset. With scratch slot reuse these two views disagree. For example, values A, B and C take turns using the same physical slot, the first operations on B and C are partial writes (which trigger read-modify-write sequences), and C is defined by the same `mad` that last reads B. ``` fill A ; A's last use mad X, A, ... fill B ; B's RMW fill mad B:hf, ... spill B ... fill C ; C's RMW fill (C was spilled before B) fill B ; B's last use mad C:hf, B, ... spill C ``` The duplicate fill elimination matches the first `fill B` with `fill A` and the second one with `fill C`. Same physical offsets, even though all their logical offsets differ: ``` fill A mad X, A, ... mov B, A ; was `fill B`, matched by physical offset mad B:hf, ... spill B ... fill C mov B, C ; was `fill B`, matched by physical offset mad C:hf, B, ... spill C ``` Then, with no fills of B left, `spill B` looks like a dead write based on logical offsets and is removed: ``` fill A mad X, A, ... mov B, A mad B:hf, ... ; `spill B` was removed ... fill C mov B, C mad C:hf, B, ... spill C ``` But `spill B` was not dead: its value is still used by the last `mad`. The `mov`s hid that use. B's value now flows through `fill C` and `mov B, C`, so without the spill, `fill C` reads A's old bytes and the last `mad` gets a stale B. The right thing here would be to not change those `fill B`s; they should only match other accesses of B. So change the rest of brw_opt_fill_and_spill subpasses to also use logical offsets. This only narrows the matches (same logical offset implies same physical offset), and the invalidation checks remain correct since only one value at a time can access its slot. Found by the llama.cpp unit test on BMG: ``` test-backend-ops test -o FLASH_ATTN_EXT -p hsk=3D512 ``` No changes to fossil-db on BMG. Fixes: a9b67ad0954b ("brw: Reuse scratch slots between non-interfering sp= illed VGRFs") Assisted-by: Claude Code (Fable 5) Reviewed-by: Ian Romanick <[email protected]> (cherry picked from commit 28f0804ea039c333d2294bdd6565332dbdfe7979) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - ad94120b by Janne Grunau at 2026-07-29T13:35:28+02:00 asahi: ci: Drop no longer failing VK.wsi.xcb.present_timing test "dEQP-VK.wsi.xcb.present_timing.basic.timing_queue" no longer fails after implementing present timing for X11 WSI in https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39551 Fixes: 1546dabd4a9 ("wsi/x11: Implement main body of present timing.") Reviewed-by: Eric Engestrom <[email protected]> Signed-off-by: Janne Grunau <[email protected]> (cherry picked from commit c523afe2b0b258a701ad0e0cf2851524be93c7c1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - e151dd11 by Janne Grunau at 2026-07-29T13:35:28+02:00 panfrost: ci: Drop no longer failing VK.wsi.xcb.present_timing test "dEQP-VK.wsi.xcb.present_timing.basic.timing_queue" no longer fails after implementing present timing for X11 WSI in https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39551 Fixes: 1546dabd4a9 ("wsi/x11: Implement main body of present timing.") Reviewed-by: Eric Engestrom <[email protected]> Signed-off-by: Janne Grunau <[email protected]> (cherry picked from commit aa4a378f301079faec3a042081ba19ba6588c7fc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - cfdb81f4 by Janne Grunau at 2026-07-29T13:35:28+02:00 asahi: ci: Add failing b10g11r11 and e5b9g9r9 copy tests These tests are new in VKCTS 1.4.5.3 and the tested functionality was likely always broken. It is not immediately clear why the tests fail so mark them as fails for now until they can be fixed in one way or another. This simplifies chasing other VK CTS regressions for honeycomb. Backport-to: 26.2 Acked-by: Eric Engestrom <[email protected]> Signed-off-by: Janne Grunau <[email protected]> (cherry picked from commit c276444b7190c2feaa722168253baaf9f3c737e9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - ae843758 by Karol Herbst at 2026-07-29T13:35:29+02:00 vtn/opencl: fix libclc needing fp16 lowering to fp32 fp16 builtins inside libclc might be implemented on top of other fp16 builtins. However some of those might not have a fp16 version in which case we have to fall back to the fp32 one. This might be a libclc build system configuration issue, but we can easily workaround this by looking up in the current shaders whether we find a fp32 version of a builtin. Backport-to: * Reviewed-by: Zolt=C3=A1n B=C3=B6sz=C3=B6rm=C3=A9nyi <[email protected]> (cherry picked from commit 60654ceb9139ab8f289c6252bc12659fcf655e1c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 2bce69ca by Christian Gmeiner at 2026-07-29T13:35:29+02:00 nir/opt_vectorize: Remove the phis that were combined Combining two phis rewrote their uses but left the phis themselves in the block. They still hash equal and still dominate, so the next run of the pass combined them again, reported progress again and added another dead phi. The ALU path already drops its inputs, in rewrite_uses(). Fixes: 850768bacbf ("nir/opt_vectorize: prefer to swizzle vector phis at = the destination, not the source") Signed-off-by: Christian Gmeiner <[email protected]> Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit 829e6d16301d3daf83743f4edc777567cf76d8f4) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - d5c0266b by Jeremy Huddleston at 2026-07-29T13:35:29+02:00 glx/apple: only skip AppleGL election on macOS 26.0 through 26.5 The OpenGL.framework breakage that made AppleGL unusable on macOS 26 was = addressed in macOS 26.6, so limit the fallback to Gallium to the affected releases and elect AppleGL as usu= al on newer systems. Fixes: fa0974ee1cbf ("glx/apple: skip AppleGL election on macOS 26 and ne= wer") Signed-off-by: Jeremy Huddleston Sequoia <[email protected]> (cherry picked from commit c810f97ad70db582277b74326d610ec62528eeb9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - a2642c78 by Frank Binns at 2026-07-29T13:35:29+02:00 pvr: rearrange some functions in pvr_arch_border.c This is in preparation for a follow on commit. Backport-to: 26.1 Signed-off-by: Frank Binns <[email protected]> Reviewed-by: Luigi Santivetti <[email protected]> (cherry picked from commit 311207091a7d9eecc5a0be6f9828e942709fa351) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 1beb8749 by Frank Binns at 2026-07-29T13:35:29+02:00 pvr: setup all format fields for custom border color entries This fixes an issue seen with depth/stencil formats where Zink creates a = sampler with either a depth only or stencil only custom border color, but then us= es this with an image with a combined depth/stencil format, taking either th= e depth or stencil aspect. This results in a mismatch between the format field th= at gets written in the border color entry and the one that gets read from the bor= der color entry. For example, if a custom border color is setup for VK_FORMAT_S8_UINT and = used in combination with a VK_FORMAT_D32_SFLOAT_S8_UINT image, taking the stencil= aspect, then the format in the image state words will be X24G8X32. This r= esults in the value in the X24G8X32 format field being looked up in the border c= olor entry rather than the S8 field. One way to resolve this would be to add additional handling for depth/ste= ncil formats to write values to all relevant format fields in a border color entry. However, a simpler approach is to setup all format fields for cust= om border color entries, as is already done for built-in border colors. This= allows the code to be simplified, since the same function can be used for both c= ases. Fixes tests in dEQP-GLES31.functional.texture.border_clamp.*. Fixes: bf9c84991fc ("pvr: Add support for custom border colors") Signed-off-by: Frank Binns <[email protected]> Reviewed-by: Luigi Santivetti <[email protected]> (cherry picked from commit ea6e5b6b38322599bd36890ee1a4dec08771faf1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 8d5f3952 by Lionel Landwerlin at 2026-07-29T13:35:29+02:00 brw: fix wa_18019110168 lowering If the linked mesh shader to the fragment shader doesn't have the workaround active, there is nothing to do. Signed-off-by: Lionel Landwerlin <[email protected]> Fixes: fcf4401824 ("brw: handle wa_18019110168 with independent shader co= mpilation") Reviewed-by: Ivan Briano <[email protected]> (cherry picked from commit f8801d555b58d1965edab147018250830a3483a3) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 637a4773 by Natalie Vock at 2026-07-29T13:35:29+02:00 radv/nir: Clean up descriptor index lowering This switches to a forward iteration over descriptor-related intrinsics. The biggest advantage of this is that we can now change the result types of vulkan_resource_(re)index, which also allows us to nuke all the special-casing for raytracing descriptors. Aside from just looking better, this will also allow us to implement mutable acceleration structure descriptors. Since this is needed to fix Unreal Engine 5.8, it should be backported even though it's a cleanup. Cc: mesa-stable (cherry picked from commit 0be03076dcdc2131f8084a67f5db4c07579a76fb) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 504d077c by Natalie Vock at 2026-07-29T13:35:29+02:00 radv: Expose mutable acceleration structure descriptors These Just Work(tm) now. Unreal Engine 5.8 needs it. Cc: mesa-stable (cherry picked from commit dcb6045ac5b45a0d086d37904588a4fc05262233) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 70b85dc2 by Danylo Piliaiev at 2026-07-29T13:35:29+02:00 nir: Include scalarized component offsets in UBO ranges When scalarizing UBO loads, we should take into account per-component offset when calculating base and range. This is important for dynamically index row major matrices, where without this fix each row's base/range is calculated as if it starts from offset 0. ``` layout(std140, row_major, offset =3D 512) mat4 mat_arr[2]; ... =3D mat_arr[gl_ViewIndex] ``` Without the fix it was scalarized, then vectorized back into: ``` 32x4 %1266 =3D @load_ubo (%2638, %1297) (access=3Dnone, align_mul=3D= 64, align_offset=3D0, range_base=3D512, range=3D92) 32 %2725 =3D load_const (0x00000210 =3D 528) 32 %2726 =3D iadd %2725 (0x210), %1295 32x4 %1273 =3D @load_ubo (%2638, %2726) (access=3Dnone, align_mul=3D= 64, align_offset=3D16, range_base=3D512, range=3D92) 32 %2727 =3D load_const (0x00000220 =3D 544) 32 %2728 =3D iadd %2727 (0x220), %1295 32x4 %1280 =3D @load_ubo (%2638, %2728) (access=3Dnone, align_mul=3D= 64, align_offset=3D32, range_base=3D512, range=3D92) 32 %2729 =3D load_const (0x00000230 =3D 560) 32 %2730 =3D iadd %2729 (0x230), %1295 32x4 %1287 =3D @load_ubo (%2638, %2730) (access=3Dnone, align_mul=3D= 64, align_offset=3D48, range_base=3D512, range=3D92) ``` Notice the same base, which caused ir3 to not preload the whole 128b. After the fix we get: ``` 32x4 %1266 =3D @load_ubo (%2638, %1297) (access=3Dnone, align_mul=3D= 64, align_offset=3D0, range_base=3D512, range=3D80) 32 %2725 =3D load_const (0x00000210 =3D 528) 32 %2726 =3D iadd %2725 (0x210), %1295 32x4 %1273 =3D @load_ubo (%2638, %2726) (access=3Dnone, align_mul=3D= 64, align_offset=3D16, range_base=3D528, range=3D80) 32 %2727 =3D load_const (0x00000220 =3D 544) 32 %2728 =3D iadd %2727 (0x220), %1295 32x4 %1280 =3D @load_ubo (%2638, %2728) (access=3Dnone, align_mul=3D= 64, align_offset=3D32, range_base=3D544, range=3D80) 32 %2729 =3D load_const (0x00000230 =3D 560) 32 %2730 =3D iadd %2729 (0x230), %1295 32x4 %1287 =3D @load_ubo (%2638, %2730) (access=3Dnone, align_mul=3D= 64, align_offset=3D48, range_base=3D560, range=3D80) ``` Which results in load of the correct UBO range. Cc: mesa-stable Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit d9d106a5295dd4a220eb7918e1d0f16524ebb3d4) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - cc44702a by Emma Anholt at 2026-07-29T13:35:29+02:00 freedreno: Don't force image component A=3D1 substitution on R/RG texture= s. This is a no-op for normal texture sampling, but with QCOM_image_processing the image component substitution happens before filtering (successfully, due to the underlying format), so re-substitutio= n during the texture component swizzle stage causes incorrect A=3D1 when yo= u should be getting the filtered value. This also means we need to refuse weight image sampling for BC1 RGB, where we have no choice but to do the A=3D1 substitution in the swizzle. Cc: mesa-stable (cherry picked from commit 3bdfd556b1ca602bef7936dc548f624708d1a5d7) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 3c47d782 by Pohsiang (John) Hsu at 2026-07-29T13:35:29+02:00 d3d12: fix msvc build warning C4819 Fixes: 42f636bee95 ("d3d12/video: Don't reset batches in fence_wait") Reviewed-by: Sil Vilerino <[email protected]> (cherry picked from commit a25e41df17602b3c3962805be6fe748ad226abbc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 76589dcd by Caio Oliveira at 2026-07-29T13:35:29+02:00 intel/gen: Fix Gfx11 3-src accumulator file encoding The decoder was correct, but encoder was wrongly using IMM instead of ARF in two fields. Fixes: 03c32e90a98f ("intel/gen: Add gen encoding module") Reviewed-by: Ian Romanick <[email protected]> Reviewed-by: Jordan Justen <[email protected]> (cherry picked from commit 1c0e6ad8cbf1a36b5eb22bfbd0389264955b8b60) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 5d440d08 by Lionel Landwerlin at 2026-07-29T13:35:29+02:00 anv: fix leak in RT binding point Signed-off-by: Lionel Landwerlin <[email protected]> Cc: mesa-stable Reviewed-by: Michael Cheng <[email protected]> (cherry picked from commit b287c8fd44bfae03b9e7ef2585bf28c30d5bc67f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - d6d8b826 by Georg Lehmann at 2026-07-29T13:35:29+02:00 aco/live_var_analysis: make sure shared vgprs are within the encodable vg= prs If we are fine with a low number of waves, vgprs will be greater than 256= before we clamp to the vgpr limit. Cc: mesa-stable Reviewed-by: Daniel Sch=C3=BCrmann <[email protected]> (cherry picked from commit da3e65d22d7a678160bfa68c4eee106282e1eea4) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - a293c85a by Valentine Burley at 2026-07-29T13:35:29+02:00 tu: Report correct maxFragmentInputComponents limits maxFragmentInputComponents doesn't need to reserve space for builtin variables. Remove the conservative adjustment and report the actual hardware limits. Cc: mesa-stable Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit 42694cb11015d3efea30abb3358b8a0b3875f71a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43295= > - - - - - 5db7111f by Eric Engestrom at 2026-07-29T18:32:23+02:00 VERSION: bump for 26.2.0-rc3 - - - - - 158 changed files: - + .pick_status.json - VERSION - bin/pick/core.py - bin/pick/ui.py - docs/features.txt - src/amd/common/ac_gpu_info.c - src/amd/common/ac_surface.c - src/amd/common/ac_vcn_dec.c - src/amd/common/ac_vcn_enc.c - src/amd/common/ac_video.c - src/amd/common/amd_family.h - src/amd/common/nir/ac_nir_lower_ngg_mesh.c - src/amd/compiler/aco_live_var_analysis.cpp - src/amd/vulkan/bvh/update.comp - src/amd/vulkan/nir/radv_nir.h - src/amd/vulkan/nir/radv_nir_lower_abi.c - src/amd/vulkan/nir/radv_nir_lower_descriptors.c - src/amd/vulkan/nir/radv_nir_lower_fs_barycentric.c - src/amd/vulkan/nir/radv_nir_opt_fs_builtins.c - src/amd/vulkan/radv_cmd_buffer.c - src/amd/vulkan/radv_cmd_buffer.h - src/amd/vulkan/radv_descriptors.c - src/amd/vulkan/radv_device.c - src/amd/vulkan/radv_physical_device.c - src/amd/vulkan/radv_pipeline_graphics.c - src/amd/vulkan/radv_pipeline_graphics.h - src/amd/vulkan/radv_shader.h - src/amd/vulkan/radv_shader_args.c - src/amd/vulkan/radv_shader_object.c - src/asahi/ci/asahi-agx2-fails.txt - src/asahi/compiler/agx_compile.h - src/broadcom/vulkan/v3dv_cmd_buffer.c - src/broadcom/vulkan/v3dv_descriptor_set.h - src/broadcom/vulkan/v3dv_pipeline.c - src/compiler/clc/mesa_clc.c - src/compiler/nir/nir_divergence_analysis.c - src/compiler/nir/nir_lower_explicit_io.c - src/compiler/nir/nir_lower_io.c - src/compiler/nir/nir_opt_copy_prop_vars.c - src/compiler/nir/nir_opt_dead_write_vars.c - src/compiler/nir/nir_opt_vectorize.c - src/compiler/nir/nir_range_analysis.c - src/compiler/spirv/vtn_opencl.c - src/egl/drivers/dri2/egl_dri2.c - src/freedreno/fdl/fd6_view.cc - src/freedreno/ir3/ir3.c - src/freedreno/ir3/ir3_compiler.c - src/freedreno/ir3/ir3_nir.c - src/freedreno/vulkan/tu_acceleration_structure.cc - src/freedreno/vulkan/tu_clear_blit.cc - src/freedreno/vulkan/tu_clear_blit.h - src/freedreno/vulkan/tu_cmd_buffer.cc - src/freedreno/vulkan/tu_cs.cc - src/freedreno/vulkan/tu_device.cc - src/freedreno/vulkan/tu_formats.cc - src/freedreno/vulkan/tu_knl_kgsl.cc - src/freedreno/vulkan/tu_lrz.cc - src/freedreno/vulkan/tu_shader.cc - src/freedreno/vulkan/tu_subsampled_image.cc - src/gallium/auxiliary/pipe-loader/driinfo_gallium.h - src/gallium/auxiliary/util/u_driconf.c - src/gallium/auxiliary/vl/vl_proc.c - src/gallium/drivers/d3d12/d3d12_video_dec.cpp - src/gallium/drivers/d3d12/d3d12_video_enc.cpp - src/gallium/drivers/llvmpipe/ci/llvmpipe-fails.txt - src/gallium/drivers/llvmpipe/lp_state_derived.c - src/gallium/drivers/nouveau/codegen/nv50_ir_ra.cpp - src/gallium/drivers/nouveau/nouveau_mm.c - src/gallium/drivers/nouveau/nv30/nv30_context.c - src/gallium/drivers/nouveau/nv30/nv30_state_validate.c - src/gallium/drivers/nouveau/nv30/nvfx_fragprog.c - src/gallium/drivers/nouveau/nv30/nvfx_shader.h - src/gallium/drivers/panfrost/pan_shader.c - src/gallium/drivers/radeonsi/mm/radeon_vcn_enc.c - src/gallium/drivers/radeonsi/mm/radeon_vcn_enc.h - src/gallium/drivers/radeonsi/si_buffer.c - src/gallium/drivers/radeonsi/si_pipe.c - src/gallium/drivers/radeonsi/si_state.c - src/gallium/drivers/radeonsi/si_state.h - src/gallium/drivers/v3d/v3d_blit.c - src/gallium/drivers/zink/ci/zink-tu-a750-fails.txt - src/gallium/drivers/zink/zink_kopper.c - src/gallium/frontends/rusticl/api/memory.rs - src/gallium/frontends/rusticl/builtin_kernels/copy.cl - + src/gallium/frontends/rusticl/builtin_kernels/copy_ulong.cl - src/gallium/frontends/rusticl/builtin_kernels/lib.rs - src/gallium/frontends/rusticl/builtin_kernels/meson.build - src/gallium/frontends/rusticl/core/device.rs - src/gallium/frontends/rusticl/core/kernel.rs - src/gallium/frontends/rusticl/core/meta.rs - src/gallium/frontends/rusticl/core/program.rs - src/gallium/frontends/rusticl/mesa/compiler/nir.rs - src/gallium/frontends/va/image.c - src/gallium/include/frontend/api.h - src/glx/glxext.c - src/imagination/vulkan/pvr_arch_border.c - src/imagination/vulkan/pvr_arch_cmd_buffer.c - src/imagination/vulkan/pvr_arch_queue.c - src/imagination/vulkan/pvr_job_render.h - src/intel/compiler/brw/brw_compile_fs.cpp - src/intel/compiler/brw/brw_nir.c - src/intel/compiler/brw/brw_opt_fill_spill.cpp - src/intel/compiler/gen/gen_encoding_pre_xe.cpp - src/intel/dev/intel_debug.c - src/intel/dev/intel_device_info.c - src/intel/vulkan/00-anv-defaults.conf - src/intel/vulkan/anv_blorp.c - src/intel/vulkan/anv_cmd_buffer.c - src/intel/vulkan/anv_shader.c - src/intel/vulkan/anv_shader_compile.c - src/intel/vulkan/genX_cmd_buffer.c - src/intel/vulkan/genX_query.c - src/kosmickrisp/vulkan/kk_query_pool.c - src/mesa/main/bufferobj.c - src/mesa/state_tracker/st_cb_texture.c - src/nouveau/compiler/nak_nir_lower_mesh_intrinsics.c - src/nouveau/vulkan/nvk_cmd_buffer.c - src/nouveau/vulkan/nvk_shader.c - src/panfrost/ci/panfrost-g52-fails.txt - src/panfrost/ci/panfrost-g57-fails.txt - src/panfrost/ci/panfrost-g57-flakes.txt - src/panfrost/ci/panfrost-g610-fails.txt - src/panfrost/ci/panfrost-g610-flakes.txt - src/panfrost/ci/panfrost-g610-skips.txt - src/panfrost/ci/panfrost-g925-fails.txt - src/panfrost/ci/panfrost-g925-flakes.txt - src/panfrost/compiler/bifrost/bifrost_compile.c - src/panfrost/compiler/kraid/hw_runner/device.rs - src/panfrost/compiler/kraid/hw_runner/meson.build - src/panfrost/compiler/midgard/midgard_compile.c - src/panfrost/perf/quick.c - src/panfrost/vulkan/panvk_vX_physical_device.c - src/util/00-mesa-defaults.conf - src/util/driconf.h - src/virtio/magma-gpu-rs/ffi/virtgpu_kumquat_ffi/lib.rs - src/virtio/vulkan/vn_common.c - src/virtio/vulkan/vn_common.h - src/virtio/vulkan/vn_device.c - src/virtio/vulkan/vn_device.h - src/virtio/vulkan/vn_feedback.c - src/virtio/vulkan/vn_feedback.h - src/virtio/vulkan/vn_host_copy.c - src/virtio/vulkan/vn_physical_device.c - src/virtio/vulkan/vn_physical_device.h - src/virtio/vulkan/vn_queue.c - src/virtio/vulkan/vn_queue.h - src/virtio/vulkan/vn_renderer.h - src/virtio/vulkan/vn_renderer_sim_syncobj.c - src/virtio/vulkan/vn_renderer_sim_syncobj.h - src/virtio/vulkan/vn_renderer_util.c - src/virtio/vulkan/vn_renderer_virtgpu.c - src/virtio/vulkan/vn_renderer_vtest.c - src/virtio/vulkan/vn_ring.c - src/virtio/vulkan/vn_sync.c - src/virtio/vulkan/vn_sync.h - src/vulkan/runtime/vk_pipeline.c - src/vulkan/runtime/vk_video.c - src/vulkan/runtime/vk_video.h The diff was not included because it is too large. View it on GitLab: https://salsa.debian.org/xorg-team/lib/mesa/-/compare/= 57017725151dc7e96ba82eeedc35baf1fc13911f...5db7111ffb9f932815baf7b299cd4d= b772e6d301 -- = View it on GitLab: https://salsa.debian.org/xorg-team/lib/mesa/-/compare/= 57017725151dc7e96ba82eeedc35baf1fc13911f...5db7111ffb9f932815baf7b299cd4d= b772e6d301 You're receiving this email because of your account on salsa.debian.org. = Manage all notifications: https://salsa.debian.org/-/profile/notification= s | Help: https://salsa.debian.org/help ----==_mimepart_6a70b13ab73b_5515b98ebc885935 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www= .w3.org/TR/REC-html40/loose.dtd"> <html lang=3D"en" style=3D'--code-editor-font: var(--default-mono-font, "= GitLab Mono"), JetBrains Mono, Menlo, DejaVu Sans Mono, Liberation Mono, = Consolas, Ubuntu Mono, Courier New, andale mono, lucida console, monospac= e;'> <head> <meta content=3D"text/html; charset=3Dutf-8" http-equiv=3D"Content-Type">= <title> GitLab </title> <style data-premailer=3D"ignore" type=3D"text/css"> a { color: #1068bf; } </style> <style>img { max-width: 100%; height: auto; } body { font-size: .875rem; } body { -webkit-text-shadow: hsla(0,0%,100%,.01) 0 0 1px; } body { font-family: "GitLab Sans",-apple-system,BlinkMacSystemFont,"Segoe UI",Ro= boto,"Noto Sans",Ubuntu,Cantarell,"Helvetica Neue",sans-serif,"Apple Colo= r Emoji","Segoe UI Emoji","Segoe UI Symbol","Noto Color Emoji"; font-size= : inherit; } </style> </head> <body style=3D'font-size: inherit; -webkit-text-shadow: hsla(0,0%,100%,.0= 1) 0 0 1px; font-family: "GitLab Sans",-apple-system,BlinkMacSystemFont,"= Segoe UI",Roboto,"Noto Sans",Ubuntu,Cantarell,"Helvetica Neue",sans-serif= ,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol","Noto Color Emoji= ";'> <div class=3D"content"> <h3 style=3D"margin-top: 20px; margin-bottom: 10px;"> Timo Aaltonen pushed to branch upstream-experimental at <a href=3D"https:= //salsa.debian.org/xorg-team/lib/mesa">X Strike Force / lib / mesa</a> </h3> <h4 style=3D"margin-top: 10px; margin-bottom: 10px;"> Commits: </h4> <ul> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/f505ac9eb998ffd2fc55991f5d6a63276c8f8197">f505= ac9e</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-22T10:52:04+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>.pick_status.json: = Update to d49a00bdf15fd48b31af93aaf5feed3eebcbda12 Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/56e7e13b7ebbc0ede342e70302c58ca8ee67b805">56e7= e13b</a></strong> <div> <span> by Alyssa Rosenzweig </span> <i> at 2026-07-22T10:52:04+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>intel: fuse off Jay= in Mesa 26.2 Jay is under heavy development and is not considered released. It is available in upstream Mesa for developers to hack on but is not part of the 26.1 release. Add a comment acting like a chicken bit to fuse off the= compiler while minimizing conflicts with backports (which is why we don't= remove Jay wholesale from the release). Signed-off-by: Alyssa Rosenzweig <[email protected]> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/060ff1ef657de5af25facad6cd53ab955feac7ef">060f= f1ef</a></strong> <div> <span> by JaeHoon Lee </span> <i> at 2026-07-22T10:52:04+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>v3dv: make room in = the descriptor map for the no-sampler entries pipeline_lower_nir() reserves two "no sampler" entries before applying th= e pipeline layout, so real samplers start at index 2. The map is only V3D_MAX_TEXTURE_SAMPLERS long and we advertise exactly that many, so the last two samplers we allow write past the end of it, silently in release builds. Fixes: fc17231b2b4e ("v3dv/pipeline: reduce descriptor_map size") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 9d534e38a2e20d1076e2dbd4b82d8c9e6b65e932) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/cb7ea5bfd4f2562f3441294a4baf60dbdbbd46ad">cb7e= a5bf</a></strong> <div> <span> by JaeHoon Lee </span> <i> at 2026-07-22T10:52:04+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>v3dv: use the binni= ng VS variant for the binning VPM config compute_vpm_config() reads the render VS into vs_bin_variant, so the binning VPM config comes from the wrong program data. The binning VS writes less, so this only ever under-sizes the binning VPM allocation. Fixes: 9e7d9a6efb67 ("v3dv: add support for geometry shaders to pipelines= ") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 771dfe55fcb62b62212b418d7f702ab8d42ea07e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ffb925ef5251d753ab3b64852b6c2958847bc9d8">ffb9= 25ef</a></strong> <div> <span> by JaeHoon Lee </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>v3dv: record the mu= ltiview geometry shader with its Vulkan stage bit active_stages is a VkShaderStageFlags, but MESA_SHADER_GEOMETRY is an ordinal, so this sets VERTEX|TESSELLATION_CONTROL instead of GEOMETRY, an= d vkGetPipelineExecutable*KHR skips the shader. Use mesa_to_vk_shader_stage= (), as the rest of the file does. Fixes: 5872c7ca7ba7 ("v3dv: inject a custom passthrough geometry shader f= or multiview pipelines") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 1d9201e377eb6f595c83d68757b054818bd4176a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/0b5641decc09cb67a3b7334f0d6841f078dd3d78">0b56= 41de</a></strong> <div> <span> by JaeHoon Lee </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>v3dv: record the no= -op fragment shader with its Vulkan stage bit Same as the previous commit: MESA_SHADER_FRAGMENT sets TESSELLATION_EVALUATION instead of FRAGMENT. Fixes: 02de1d9ab16f ("v3dv: add a no-op fragment shader if we don't have = one") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit fb7300c6da2b740b0605b1562fe569391f33be94) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b75b5645462e09af42bee195ba6b09f6e35c277d">b75b= 5645</a></strong> <div> <span> by Pavel Ondra=C4=8Dka </span> <i> at 2026-07-22T10:52:05+02:00 </= i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nv50_ir_ra: align B= 96 spill slots to vec4 This fixes nouveau NVE4 shader-db godot crash exposed by vectorizing commutative ALU sources. Fixes: d12e16fc3b5 ("nouveau: Handle unaligned tlsBase during spills") Suggested-by: Karol Herbst <[email protected]> Reviewed-by: Karol Herbst <[email protected]> Reviewed-by: Mel Henning <[email protected]> Signed-off-by: Pavel Ondra=C4=8Dka <[email protected]> (cherry picked from commit 05f9f41fc294f6adb6d9b8ffed9a801bb232f40e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ed764e642332992c502781e5caeaf3d00af904c7">ed76= 4e64</a></strong> <div> <span> by Caleb Callaway </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>anv: hide Intel ven= dor ID for Cyberpunk Works around XeSS misrendering Fixes #15739 Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit d021138b4100bf1dc876e930645f3a382d1ed78d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/f1ee264a2354930fd1c6c6c64e14451b6c646ad6">f1ee= 264a</a></strong> <div> <span> by Michael Cheng </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>intel: Fix DEBUG_FS= _SIMD mask Spell out the FS modes like the sibling DEBUG_{CS,TS,MS,RT}_SIMD masks. Fixes: 1728c055d4b ("intel: Fix multipolygon flags in INTEL_SIMD_DEBUG de= fault handling") Signed-off-by: Michael Cheng <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> (cherry picked from commit 7635d56fba9bbcafd0b0b448951947917211bb34) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/9f9dd9e7a9a87d780f69e847e037cc087c5cae7c">9f9d= d9e7</a></strong> <div> <span> by Michael Cheng </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>intel: Fix operator= precedence in intel_simd_overridden | binds more tightly than ?:, so this chain of ternaries wasn't OR'd together but parsed as a nested ternary that returns only the first matching stage's bit. With more than one stage requested, the later stages were silently dropped (e.g. INTEL_SIMD_DEBUG=3Dcs32,ts16 gave 0x20, COMPUTE only, instead of 0x60). Wrap each ternary in parentheses. Fixes: baf2f0ea4a9 ("intel: Replace INTEL_DEBUG=3Ddo32 with INTEL_SIMD_DE= BUG") Signed-off-by: Michael Cheng <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]> (cherry picked from commit 48297c8344669c4a867fae28e2cafb646218016e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/2ef226bd50fa1799548a08c1ee8887945e0e1476">2ef2= 26bd</a></strong> <div> <span> by Timothy Arceri </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>zink: fix swap inte= rval changes being dropped db9dbcbec00 unconditionally restores the previous present mode before attempting to update the swapchain, which turns runtime swap interval changes into a no-op: * the immediate path (no currently acquired image) recreates the swapchain with the old present mode still set, so nothing changes * the deferred path added to zink_kopper_present_queue() can never trigger since cdt->present_mode always matches the current swapchain's mode The only remaining effect is pointless swapchain recreation whenever an app changes the swap interval while no image is acquired. Keep the new present mode instead, and only restore the old one if the immediate swapchain update fails. Fixes: db9dbcbec00 ("zink: defer swapchain updates for interval changes i= f acquired image is active") Generated-by: Claude Code (Fable 5) Reviewed-by: Mike Blumenkrantz <[email protected]> (cherry picked from commit 09180f0eed603872b75f44b398048db06f236f54) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/3f9bbe2033a9587f6536dcd7dfd8eb0feece203d">3f9b= be20</a></strong> <div> <span> by Benjamin Cheng </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>ac/vcn_enc: Disable= var slice for preencode with VCN4 An internal FW change was never upstreamed, so the version numbers that were previously put here became inaccurate. Disable it until the FW fix is available. Signed-off-by: Benjamin Cheng <[email protected]> Cc: mesa-stable Reviewed-by: David Rosca <[email protected]> (cherry picked from commit c2ecb433fda48ee5d0ef295b3aa79f23143ceeca) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d0e687b69063443e3edd50d2fb39e01f3cdc00a9">d0e6= 87b6</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>mesa_clc: support 3= 2 bit targets Backport-to: 26.2 Reviewed-by: Alyssa Rosenzweig <[email protected]> (cherry picked from commit f264cb38182a2c2d75ec84d150fc28f4fe00598b) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b120fc9a77c6d2bb897cdebf95d50bb0a675fdec">b120= fc9a</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>meson/rusticl: fix = typo in depfile for builtin shaders Backport-to: 26.2 (cherry picked from commit b44f878eb7b90de9a5ed45d4844313dc85c8e50c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b27cf38f852d53a5cc5da02dcf27eb1101fa6d08">b27c= f38f</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>rusticl/meta: mark = builtin kernels SPIR-V as 64 bit Backport-to: 26.2 (cherry picked from commit 4130bbe1502ab27e3eec4bc4580dbc89ab579ab8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/f8a2af7914d5ad2c40c09bdad9af5ef5f7c95603">f8a2= af79</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>rusticl/mesa: compi= le 32 bit version Backport-to: 26.2 (cherry picked from commit be269027d76853d1a75cae09e99c0c2b719fbe2b) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/4d1809f676366ba63a9a795a4fbaa61dfcdb25df">4d18= 09f6</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>rusticl/program: ad= d Program::from_spirv_with_devs Backport-to: 26.2 (cherry picked from commit 5ab128c585e988db20ffa948b8c285e53320c6cd) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/41dbdc7f872eab6f98280559f1ac7103c29e2beb">41db= dc7f</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>rusticl/meta: suppo= rt 32 bit devices Backport-to: 26.2 Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15869 (cherry picked from commit a5ea6c5a02bb1e19b8c47d78e3339ba920a9e4e0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/1321fd2ec6c66afc172096ed3f8f785e213b1be9">1321= fd2e</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-22T10:52:05+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>rusticl/meta: split= out ulong kernels Backport-to: 26.2 (cherry picked from commit b2c832063be1b2f56019a6b9739bae42649541e6) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/421dcdf8da27a809516f08359f918c6da463bd6c">421d= cdf8</a></strong> <div> <span> by Christoph Pillmayer </span> <i> at 2026-07-22T10:52:05+02:00 </= i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>pan: Fix BIFROST_ME= SA_DUMP_DIR Don't try to fclose when we didn't fopen successfully. Fixes: 712ebaa4 ("pan: Add BIFROST_MESA_DUMP_DIR option to dump shader bi= naries") Reviewed-by: Lars-Ivar Hesselberg Simonsen <[email protected]= > (cherry picked from commit 306762b57411eb09d0eb09376a0b5a876497d14f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/cff1fce835d8c578b8f7b23a54aca7eb9f4ab583">cff1= fce8</a></strong> <div> <span> by Lionel Landwerlin </span> <i> at 2026-07-22T10:52:06+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>anv: flush accumula= ted barriers for top of TOP_OF_PIPE If there is a barrier preceding the write timestamp, we need to flush it to make sure the timestamp is collected after the barrier is executed. Signed-off-by: Lionel Landwerlin <[email protected]> Cc: mesa-stable Reviewed-by: Tapani P=C3=A4lli <[email protected]> (cherry picked from commit c6f6c9b02da991c9c878d8992b4074bbba70aa73) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/7fae50bbd9847a37c8f7725369671b9dd752448a">7fae= 50bb</a></strong> <div> <span> by Georg Lehmann </span> <i> at 2026-07-22T10:52:06+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nir/opt_dead_write_= vars: handle atomics as reads As long as there is no cross invocation access, write, followed by atomic= , followed by write is valid. We can not remove the first write in this cas= e: shared_array[invocation_index] =3D 0; res =3D atomicAdd(shared_array[invocation_index], 2); shared_array[invocation_index] =3D 0; Cc: mesa-stable Reviewed-by: Rhys Perry <[email protected]> (cherry picked from commit 1bc7b09d583b4f623fe18983f048014c508a2364) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/5d59ca7ee9e1fbc8793fd2b57c7ec8ebedcb6711">5d59= ca7e</a></strong> <div> <span> by Lo=C3=AFc Molinari </span> <i> at 2026-07-22T10:52:06+02:00 </i= > </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>panfrost/ci: Mark "= spec@!opengl 1.4@copy-pixels" as flake This is a random failure on g57/g610/g925. See Mesa bug #15880. Signed-off-by: Lo=C3=AFc Molinari <[email protected]> Reviewed-by: Erik Faye-Lund <[email protected]> (cherry picked from commit b71fab7fbe1ab859f38032eafd9b75341373413f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/90907cfee87b68f9443d838e212ce73343fcac42">9090= 7cfe</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: recycle idle= sfb cmds only after async wait This fixes a regression introduced by a code refactor. The correct order should be: 1. see sync feedback signaled to a new point 2. emit renderer side fence/semaphore wait for the new point 3. recycle sync feedback idle cmds In (3), those exceeding the cache capacity will be purged, ending up with renderer side cmd free, which must be executed after (2) for validity of the renderer side Vulkan usage. Fixes: f4a3d6941fd ("venus: extract sync feedback host write and query") (cherry picked from commit 97c9a4eaa40b41739c20948117a1b977704f5c50) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/185650d64ceb92c11ed2cba36443276c62dd9acc">1856= 50d6</a></strong> <div> <span> by Ian Romanick </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>brw: Handle empty t= op block in brw_nir_move_interpolation_to_top If the block is empty, nir_block_first_instr returns NULL. nir_before_instr(NULL) creates an invalid cursor that eventually results in a segfault. This was discovered while testing !42933 on Ratchet and Clank - Rift Apart. I also considered using nir_before_block(top). That would require moving the instructions in reverse order because nir_before_block always added at the top. That really demands a comment explaining the reasoning. Instead, I decided to do the obvious thing that is slightly more code. Fixes: 3037603b703 ("intel/fs: Linked list micro optimizations in brw_nir= _move_interpolation_to_top") Reviewed-by: Michael Cheng <[email protected]> (cherry picked from commit a7cdad725932c60e8b760b61115574d8fb935a3c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/8e480fdd373713aebfe6184a9982a55c8dbeb971">8e48= 0fdd</a></strong> <div> <span> by Lionel Landwerlin </span> <i> at 2026-07-22T11:25:40+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>anv: fix Wa_1804090= 3259 The description above says we need CS_STALL on both PIPE_CONTROL, it was missing for the first one. Signed-off-by: Lionel Landwerlin <[email protected]> Fixes: 243c01c703 ("anv/iris: implement Wa_18040903259") Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15844 Reviewed-by: Michael Cheng <[email protected]> Reviewed-by: Tapani P=C3=A4lli <[email protected]> (cherry picked from commit be696377a17b63228fc4ff78284ac1379506d0ae) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/8cbece813a060c12e87fe630ead0eb51101ac59e">8cbe= ce81</a></strong> <div> <span> by Marek Ol=C5=A1=C3=A1k </span> <i> at 2026-07-22T11:25:40+02:00 = </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv: don't expose = memory types from AMD_device_coherent_memory without the ext Fixes: 7fa1b487556 - radv: disable AMD_device_coherent_memory on gfx12 du= e to out of order behavior Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit 5f4595a5c07a2e428a9c4416ed4c94659598bf69) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/052c4212072cc65b2c6d3d4c483bec2f5ee61d44">052c= 4212</a></strong> <div> <span> by Iv=C3=A1n Briano </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>anv: fix 2d-array t= o 3d blits It was accounting for the depth center offset if src_image is 3D, assuming dst_image would be too, but with maintenance8 that may not be the case, so we need to check both. Fixes new tests: dEQP-VK.api.copy_and_blit.core.blit_image.simple_tests.2d_array_to_3d.rev= erse_blit_* Cc: mesa-stable Signed-off-by: Iv=C3=A1n Briano <[email protected]> Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit 1c6e7e82912fa43acd538cb2e30a4d132c296dac) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b5ee82b3fec12569387499bbe541b613c5c80df2">b5ee= 82b3</a></strong> <div> <span> by Adam Stylinski </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nv30: fix an issue = when this push buffer is NULL Fixing an issue when an empty context is destroyed. When a context is destroyed prior to this buffer being initialized, a NULL pointer dereference occurs. The nv50 & nvc0 implementations do this same thin= g. Reviewed-by: Karol Herbst <[email protected]> Tested-by: Filip Gawin <[email protected]> Backport-to: * (cherry picked from commit 4ce4feff89bad1ffd2ac7933ab25b44b6cca38d9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/949268fef4b999bd571cd5fcfd59a2ad09a5b5ec">9492= 68fe</a></strong> <div> <span> by Filip Gawin </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nv30: fix 1 <<= ; 31 issues Reviewed-by: Karol Herbst <[email protected]> Backport-to: * (cherry picked from commit 51a7e9723538833e74ea9aebc251db22743eaf7f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/0889c4db1f1e05bbd1e2ae90eb649a9be84fa4da">0889= c4db</a></strong> <div> <span> by Filip Gawin </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nv30: fix another l= eft shift cannot be represented in type 'int' Reviewed-by: Karol Herbst <[email protected]> Backport-to: * (cherry picked from commit 81dca3bf64c086bbae6c542824f5a82f34a0ed3f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/5ae3c235e93dbd32b4ca0682a6b094d21a0e5cee">5ae3= c235</a></strong> <div> <span> by Lionel Landwerlin </span> <i> at 2026-07-22T11:25:40+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vulkan/runtime: fix= up vk_shader leak on RT group recompile If you compile the same RT groups in multiple processes, there is a chance one process writes the disk cache with one of the shader in the group. Another process will pickup on the disk cache entry but not have all the entries for the group and therefore trigger a recompile. In that unlikely cases that is probably only reachable while running the CTS in multiple processes, there will be a shader leak that will materialize later at cache destroy : deqp-vk: ../src/vulkan/runtime/vk_pipeline_cache.c:687: vk_pipeline_cac= he_destroy: Assertion `cache->object_cache->entries =3D=3D 0' faile= d. Signed-off-by: Lionel Landwerlin <[email protected]> Cc: mesa-stable Reviewed-by: Ivan Briano <[email protected]> (cherry picked from commit 879365ec9f06adef47f116d11b88632264c1f271) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/09b725f7565e8dd22385479576160a4c362cf719">09b7= 25f7</a></strong> <div> <span> by Timothy Arceri </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nir/opt_dead_write_= vars: handle memcpy_deref as reads memcpy_deref reads its source and writes its destination, but the pass did not account for it. A store feeding an OpCopyMemorySized copy could be wrongly eliminated by a later store to the same deref, since nothing marked the intervening access as a use. memcpy is a sized copy that walks the derefs in a loop, so it can touch memory beyond the src/dst deref itself. A deref comparison is therefore not sufficient - conservatively clear all unused writes in the accessed modes. Reviewed-by: Georg Lehmann <[email protected]> Assisted-by: Claude Code (Claude Opus 4.8) Cc: mesa-stable (cherry picked from commit 0a12d316e7ecaad7474ee782636675b72fcf9ffb) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/3995f83da39b16639ed3f445ca8ecef34f07322c">3995= f83d</a></strong> <div> <span> by Timothy Arceri </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>util: add Blockland= workaround for crash The game only initialises the glFramebufferTextureLayerEXT entry point as part of the EXT_geometry_shader4 / NV_geometry_program4 function groups, so without one of those extensions advertised the game calls a NULL pointer when "Shader Quality" is set to high and crashes. The function is provided by EXT_texture_array which Mesa does support, so advertise one of the extensions to avoid the crash. Cc: mesa-stable Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/13311 (cherry picked from commit df73bd930ae57bc1928beb1fdfb30bc3ab41b684) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/71001521c6b32d27c1a450189d238afb0dcdd537">7100= 1521</a></strong> <div> <span> by Valentine Burley </span> <i> at 2026-07-22T11:25:40+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>ir3: Fix ballot_com= ponents for subgroups smaller than 32 ballot_components becomes 0 when max_subgroup_size is smaller than 32. Clamp it to a minimum of 1. Fixes: 639d7946c71 ("freedreno: Add initial A702 support") Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit ff5352b4900737d62d8e3fa5878ebc6c381aace0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/168a591a24f80fc7ffbfdab5f16c26176f12528f">168a= 591a</a></strong> <div> <span> by Valentine Burley </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>ir3: Derive max_var= iable_workgroup_size from device limits max_variable_workgroup_size was hardcoded to 1024, which is incorrect for= devices with a smaller threadsize_base such as A702. Use nir_options.max_workgroup_invocations to override the hardcoded value. Fixes: 639d7946c71 ("freedreno: Add initial A702 support") Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit be71f85a78c5d106b7adae912a7f22811ad8954e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/aae5e1293c15cd71952851d51fbb45489db245fb">aae5= e129</a></strong> <div> <span> by Valentine Burley </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>ir3: Compute subgro= up_size from threadsize_base ir3_info::subgroup_size was hardcoded to 64 (or 128 with double threadsize), which is incorrect for devices with a smaller threadsize_base such as A702 (16/32). Fixes: 639d7946c71 ("freedreno: Add initial A702 support") Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit 5e66bfd939a34093cc9db22c64a7698955f98df9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/5c6b56195a7c40b1573fe99cf01efb09e849a0ac">5c6b= 5619</a></strong> <div> <span> by Valentine Burley </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Use computed su= bgroup size Some turnip paths still hardcoded a subgroup size of 64 (or 128 with double_threadsize), which is incorrect for devices with a smaller threadsize_base such as A702. Fixes: 639d7946c71 ("freedreno: Add initial A702 support") Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit 69416494db271747d015b6ab010f3e348d7f76da) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/e672979ea69f3fb7980a1127f72a11ab070635fd">e672= 979e</a></strong> <div> <span> by Jan Meisel </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nir/range_analysis:= handle msad_4x8 in unsigned upper bound get_alu_uub() already had logic for nir_op_msad_4x8, but the opcode was missing from the early ALU opcode allowlist. This made the existing case unreachable and forced the analysis to return the default UINT32_MAX boun= d. Add nir_op_msad_4x8 to that supported-opcode switch. Fixes: 0477421f7db ("nir: add msad_4x8") Reviewed-by: Georg Lehmann <[email protected]> Assisted-by: Codex Signed-off-by: Jan Meisel <[email protected]> (cherry picked from commit dec6f98c7bea75d9fade316e57ea272d2dd218a8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/915450298f4feb7d14d1ecc9ce9e13cf6e2d3fe9">9154= 5029</a></strong> <div> <span> by Mary Guillemard </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nvk: Do not take ba= se into account when lowering emulated attributes We were previously adding the base indice to the shared memory offset. That was fine before because the base was always 0, but this was just plain wrong. After calling nir_recompute_io_bases, this would cause misaligned address faults as the base would represent the current slot number used for that specific varying. This fixes "dEQP-VK.mesh_shader.ext.builtin.viewport_index" with nir_opt_varyings_bulk. Signed-off-by: Mary Guillemard <[email protected]> Fixes: cdb0dea462e ("nvk: Lower mesh and task shaders") Reviewed-by: Mel Henning <[email protected]> (cherry picked from commit 794cd9d334584cf034084d611a1d8ede10b6b6bc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/644eac3a69440d005e87a3c485130f570c149884">644e= ac3a</a></strong> <div> <span> by Connor Abbott </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Fix resetting c= ommand streams with writeable BOs We were forgetting to reset the start pointer that is used when switching back to writeable, which meant that when using the CS again and again making the CS writeable, we would keep going where we left off in the BO instead of actually reusing it. We still threw away the other BOs, but this meant that we would eventually run out of space and grow the BO unnecessarily, and if every BO hit the maximum size of 4MB then we could easily run out of memory eventually in games with hundreds of command buffers. Fixes: ebb8e104a56 ("tu/cs: Add support for CS patching") (cherry picked from commit c140fdd2d163bc3cefc29dd3f457c3a131fd1bc0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/0922498e61726ccbf4870e8121369105f8008c98">0922= 498e</a></strong> <div> <span> by Lorenzo Rossi </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>panfrost/midgard: F= ix fau max not initialized In the refactor we forgot to initialize the FAU size for Midgard. Signed-off-by: Lorenzo Rossi <[email protected]> Fixes: d1247f55790 ("panfrost: Unify FAU constants and relocation handlin= g") Acked-by: Erik Faye-Lund <[email protected]> (cherry picked from commit 07e937fdb73195e1962c5cbf306f6e427178f9ab) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/6ffaa0ca648295ac7917eae9a4042c789e035bba">6ffa= a0ca</a></strong> <div> <span> by Jan Meisel </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radeonsi/vcn: fail = feedback for truncated encodes VCN stops writing encoded output at allowed_max_bitstream_size, but can still report the full coded frame size in feedback. If that size is large= r than the bitstream buffer VA clients can read past the mapped buffer when= copying VACodedBufferSegment::buf. Record the bitstream size left after packed headers, clamp feedback to that limit, and flag the encode as failed if the reported size exceeds it= . The VA frontend then reports the frame as bad instead of exposing a truncated bitstream as valid output. Analysis and patch by Maksim Nikolaev. Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15774 Cc: mesa-stable Reviewed-by: David Rosca <[email protected]> Assisted-by: Codex (cherry picked from commit 15e06548daba70c7adb222fdffbdf2f0d7f1e0d1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/072dd59dd86cae7c8963acec4b1da2b4cbd8351f">072d= d59d</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: properly che= ck sync2 enablement We might be able to upgrade instance api version or simply enable sync2 behind the app, but let's still forward for now. Fixes: 07cee75c39c ("venus: layer vkQueueSubmit2 over vkQueueSubmit w/o s= ync2") (cherry picked from commit 64cbdd73a878f44952ed640e4d3f7d6aff442b36) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/5a3ec210082e7c87b63adb73b8617afcbaef17da">5a3e= c210</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: host image c= opy to scrub present_src layout if needed Fixes: c13da1c744a ("venus: implement host image copy commands") (cherry picked from commit f3ea8314c1b2f5a1edc2a323dcf8735221238e88) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/13e953f848f5a49758506d12c3c3be4d134d31d1">13e9= 53f8</a></strong> <div> <span> by Jeremy Gebben </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>kk: Fix Vulkan to M= etal stage translation for timestamps A typo in the stage lookup table was causing vertex stages to be mishandled in vkCmdWriteTimestamp2(). Fixes: ed807097ebd ("kk: Implement VK_QUERY_TYPE_TIMESTAMP") Reviewed-by: Aitor Camacho <[email protected]> (cherry picked from commit 7f540b051c40cfc0c5ed0c48fd70c0309fb82b37) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/35811c56383e6dc6e51beea0b8d5ddf0b515bea8">3581= 1c56</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vulkan/video: Fix c= oding AV1 decoder/encoder_buffer_delay Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> (cherry picked from commit 71f8d76c7d6eb355edb77344ceed23889fe673a1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/2e2b3ed323b25df331c387cf3863b8673a9fe466">2e2b= 3ed3</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vulkan/video: Don't= code AV1 decoder model info when not present We always coded decoder model info, with all values being zero if we got NULL pStdDecoderModelInfo from app. Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> (cherry picked from commit ff12737333ae32b11e784739cdcb569fe61c6db2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/a300052534472b0f532cc4cba72ac16a5f370269">a300= 0525</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vulkan/video: Fix c= oding AV1 operating points We always coded the default operating point, because the app input wasn't copied. Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> (cherry picked from commit 5e1cf2cb5da1f51782ea69b2083c5e61deb635d6) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/28201f10e4b07b0bfc395865a1ea4d890fbbf26c">2820= 1f10</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>d3d12/video: Don't = reset batches in fence_wait va frontend unlocks driver mutex while waiting on fences, so there is a race accessing codec when trying to reset batches. Cc: mesa-stable Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15886 (cherry picked from commit 42f636bee95409f8ac89a453ab0edcbe58530180) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b273d82dc634dbbce65859db7534a73153ca7543">b273= d82d</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vulkan/video: Fix c= oding H265 ref pic list modification lists This should code num_ref_idx_l*_active entries, not minus 1. Change it to use the minus1 value to match spec. Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> Reviewed-by: Hyunjun Ko <[email protected]> (cherry picked from commit 90e7282527a5f3907264ed4b146c6004092fa183) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/bec000221f6089ffd51a8b6ed794ad188d5dccb6">bec0= 0022</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vulkan/video: Fix c= oding H265 SPS pcm block sizes, inter ref pic set and lt refs Cc: mesa-stable Reviewed-by: Benjamin Cheng <[email protected]> (cherry picked from commit 35a89677c45d79a7d08682d8afc4454009ca27d4) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ff8f9eeef7c2c921583bf543cb847900283872bd">ff8f= 9eee</a></strong> <div> <span> by Valentine Burley </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>panvk/ci: Update ex= pectations for g610-vk-asan following VK CTS uprev Fixes: 4c8de8c202d ("ci/deqp: Update VK CTS to 1.4.6.1") Signed-off-by: Valentine Burley <[email protected]> Acked-by: Erik Faye-Lund <[email protected]> (cherry picked from commit c85e3522b592344ecb6de4bb00362108c0a80cd2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/e842b0250b4ecd619bb7aadc36a167d9de3bb2b6">e842= b025</a></strong> <div> <span> by Timothy Arceri </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>util/mesa: add work= around to zero invalidated buffers Little Inferno streams its vertex data through glMapBufferRange(GL_MAP_WRITE_BIT | GL_MAP_INVALIDATE_RANGE_BIT), but only partially writes each 32-byte vertex, relying on the bytes it doesn't write reading back as zero. llvmpipe happens to satisfy this: it maps buffer storage directly and its allocations start out as kernel-zeroed pages, so the never-written bytes remain zero. radeonsi and iris recycle buffer memory (BO cache and staging upload buffers) without clearing it, so those bytes contain stale data and the game renders corrupt geometry. The game is relying on undefined behaviour; per the spec: "MAP_INVALIDATE_RANGE_BIT indicates that the previous contents of the specified range may be discarded. Data within this range are undefined with the exception of subsequently written data." Here we work around this with a new driconf option (zero_invalidated_buff= ers) that zeroes the memory returned by write-only invalidating maps, and enable it for Little Inferno. Cc: mesa-stable Generated-by: Claude (Fable 5) Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/7903 Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@am= d.com> (cherry picked from commit 8a6f0fca20cd2e9ce47f3f562a44fa6a0c8f25e1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/09423397ae6eca6291ea5a2646899d59bac86488">0942= 3397</a></strong> <div> <span> by Juan A. Suarez Romero </span> <i> at 2026-07-22T11:25:41+02:00 = </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>v3d: save fragment = constants on sand8/sand30 blit These blit functions were setting a fragment constant buffer for their own operation, but relied on current constant buffer to be saved before, which actually didn't happen as the BLIT operation didn't include it. So they when restoring the constant buffer, nothing was saved to restore, which leads to crashes. Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15857 Fixes: a69aaadec1b ("v3d: add more blitter ops for saving resources") Signed-off-by: Juan A. Suarez Romero <[email protected]> Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 8b79064907bb8b0677fdae942483e20b54cf4a3d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/2c986ebde6f68456f921a8edeeeba6a00fcdbaa1">2c98= 6ebd</a></strong> <div> <span> by Tacodiva </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vulkan/runtime: Fix= bad assumption in GetPipelineBinaryDataKHR pPipelineBinaryDataSize is not guaranteed to be 0-initialized when callin= g `vkGetPipelineBinaryDataKHR` without an output buffer. This commit change= s the check to see if the buffer is null instead of the possibly uninitialized length being 0. Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15901 Fixes: 5c47ac640b ("vulkan/runtime: implement VK_KHR_pipeline_binary") Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit 53a453b3ef7f3eb6a0e7ae37bd93985916497885) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ab26aea4cf6015d4951e08abe08c1e2ebf36b4c3">ab26= aea4</a></strong> <div> <span> by Sushma Venkatesh Reddy </span> <i> at 2026-07-22T11:25:41+02:00= </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>intel/dev: Clamp PT= L+ CS workgroup threads to 32 PTL introduces compute workgroup limits that depend on SIMD width and GRF usage. As a simple workaround suitable for stable releases, clamp max_cs_workgroup_threads to 32 on PTL+. This causes brw_simd_should_compile() to reject SIMD16 for large compute workgroups requiring more than 32 hardware threads and select SIMD32 instead, avoiding unsupported configurations while preserving the advertised 1024-invocation workgroup limit. This is a conservative workaround until PTL workgroup limits are modeled more accurately. Cc: mesa-stable Reviewed-by: Francisco Jerez <[email protected]> Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit ea77d4e261806c642f168beb5aac671b626bb2c5) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/9c94df5e6540f04756b046355a299e0f9f067b35">9c94= df5e</a></strong> <div> <span> by Danylo Piliaiev </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Custom resolve = should always use AVOID_CCU layout E.g. dynamic subpass with custom resolve always has resolve_count =3D 0, but still needs TU_GMEM_LAYOUT_AVOID_CCU. Fixes: ad84ae27190 ("tu: Implement VK_QCOM_subpass_shader_resolve") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit 02035be13a3492c0a5a639a21a956693848c5c1f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/54d67238c910c2167d9df97cfc3102879105b901">54d6= 7238</a></strong> <div> <span> by Timothy Arceri </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>util: add workaroun= d for Riddick using round() in glsl 1.20 Round is only advertised in 1.30 but the game uses 1.20 so force 1.30 Also fix up stray tab on the disable_uniform_array_resize line while we are updating this games workarounds. Cc: mesa-stable Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/11083 Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@am= d.com> (cherry picked from commit efd508cf90d597bb11c874674366c8ee8641745d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/f863cd467ebcc32bf87e7f07e38a236420bf2ff9">f863= cd46</a></strong> <div> <span> by Timothy Arceri </span> <i> at 2026-07-22T11:25:41+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>llvmpipe: emit FS i= nput vertex attributes in driver location order compute_vertex_info() emitted the vertex attributes by walking the FS input variables in declaration order, while the FS reads the input with driver location i from attrib i + 1. Since dbbb46aa38c the two orders can disagree: the FS inputs PRIMITIVE_ID, VIEWPORT and LAYER are assigned driver locations after all other inputs while the variable list is sorted by location, so the FS read its inputs from the wrong attribs when gl_PrimitiveID was used together with user varyings without a geometry shader. Emit the attributes indexed by driver location instead. This also removes the slot_emitted dedup, which could skip an attrib and shift all following attribs away from their FS input mapping; variables that share a slot (packed components) collapse onto the same driver location naturally. Fixes: dbbb46aa38c ("nir: compute io base for fragment shader inputs whic= h maybe per primitive") Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/15660 Reviewed-by: Dave Airlie <[email protected]> Generated-by: Claude Code (Claude Fable 5) (cherry picked from commit 90c66c69767add9a374445ebf1052fa683edab39) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 133> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/a5ab5f0c57a696552b5d7a2c6a239b391d977d3e">a5ab= 5f0c</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-22T15:37:39+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>VERSION: bump for 2= 6.2.0-rc2 </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/8b44a451a8220c09f8191c1dd88377deee7dd58c">8b44= a451</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>.pick_status.json: = Update to 8b00adbe72f2705985146b057f6fde9256d0dcb0 Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/2ea5bad2751ade9b27f808cd51a121d75f57cf0c">2ea5= bad2</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>.pick_status.json: = Mark 47efd739121d51e2f9049cec715a70c13767a67c as denominated Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/732feb864dd588c76f28b0c8089caf7654d5d3ca">732f= eb86</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>.pick_status.json: = Mark 7999060992e9cee91d1962faf65dc4e5c6fe4f69 as denominated Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/8410888707c6c1f78614c177c392fe7994ac3c2c">8410= 8887</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>.pick_status.json: = Mark 2515024a5919ed14fe05471e3f1f89c54a454610 as denominated Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/edf586c5f06ce58233e823d730abf32707f517e3">edf5= 86c5</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>.pick_status.json: = Mark 4fd93a0039a07ec2027f2a6d1d252c73ed033e0a as denominated Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/17cded78e0328e43b44b29cedae183e2531183d4">17cd= ed78</a></strong> <div> <span> by Konstantin Seurer </span> <i> at 2026-07-29T13:35:26+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv/bvh: Fix updat= ing acceleration structures containing AABBs If the node type is not set, child_index will not be calculated correctly. Fixes: 405c93c ("radv: Optimize BVH4 acceleration structure updates") Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15848 Reviewed-by: Natalie Vock <[email protected]> (cherry picked from commit 67eae5e5fd8c918e16d671a2255426a4af4e66c0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ae48afbe2dc5cf2e0037ce492c232a2d96efa6ea">ae48= afbe</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>va: Fix leak when v= lVaUploadImage fails Fixes: 153b2f21082 ("va: Add vlVaPostProc and use it instead of composito= r and vid engine blit") Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@am= d.com> (cherry picked from commit 0e20116a2fa88a19630dd7e0f1092d1ce7bf9cb5) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/a8da95a8f261a86418efc6f3561e1529ac6ad969">a8da= 95a8</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>va: Ensure templat = is valid for temporary surfaces This is later used by vlVaVideoSurfaceSize and vlVaRegionDefault to get the unaligned surface size. Fixes: 153b2f21082 ("va: Add vlVaPostProc and use it instead of composito= r and vid engine blit") Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@am= d.com> (cherry picked from commit a187e191c0c0947d171f6c0f646e70a85c8600f0) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/22400dee8459e7730bba55ff7f785bc20fe34113">2240= 0dee</a></strong> <div> <span> by Sonny Jiang </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radeonsi: always se= t is_format_supported in screen create On GFX12 ASICs where neither ACO nor LLVM reports shader compiler support, si_init_gfx_screen() returns early before reaching si_init_screen_state_functions(), leaving screen->is_format_supported as NULL. When a VA-API client calls vaQueryImageFormats, vl_video_buffer_is_format_supported() dereferences the NULL pointer, causing a segfault. Fix by assigning is_format_supported directly in radeonsi_screen_create_impl(), ensuring it is always set regardless of shader compiler availability. We intentionally avoid calling the full si_init_screen_state_functions() here, as its vertex_state_cache init would leak =E2=80=94 the corresponding deinit only runs from si_fini_gfx_= screen which is skipped on the no-compute path. Signed-off-by: Sonny Jiang <[email protected]> Reviewed-by: [email protected] (cherry picked from commit 34190c25c149d97fdd0bded229f21261c8232831) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/690b449c3229bda3699398abc53e577a2e701f9c">690b= 449c</a></strong> <div> <span> by Sonny Jiang </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radeonsi/vcn: Add v= cn_5_0_2 support Signed-off-by: Sonny Jiang <[email protected]> Reviewed-by: Boyuan Zhang <[email protected]> (cherry picked from commit ddbdc9826344d121f608fc77ede21a0d9cc2eb2f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/e8589ba2e6533176e913279f84e3aa953200bdf9">e858= 9ba2</a></strong> <div> <span> by Connor Abbott </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Fix condition f= or skipping emitting aprons We were trying to detect the case where both tiles are not subsampled and they touch each other, i.e. for tiles in the high-density region in the center, and we therefore must not setup an apron. However we only checked if their edges have the same x or y coordinate, and that is not enough to say their edges touch, because we have to check the other coordinate as well. For example, when processing horizontally-adjecent tiles, we may have a situation like this in subsampled space: +--------------+--------------+ | | | | | <- other_tile in | |------+ subsampled space | | | | other_tile | +--------------+ size=3D(2, 2) | | | pos=3D(0, 1) | | tile | area=3D(4, 4) | | size=3D(1, 1) | | | pos=3D(1, 0) | | | area=3D(1, 1) | | | | | +--------------+--------------+--------------+ Because their vertical edges are at the same y coordinate, we'd assume they're touching in subsampled space when actually they're not. Fixes: 4b87df29b3a ("tu: Implement subsampled images") (cherry picked from commit 9e40eb6c80cdd88c12326b3bd6ffda8bd6305d3d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/7388464496a2fa0249b7314f2168e2bcad959230">7388= 4644</a></strong> <div> <span> by JaeHoon Lee </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nvk: free copy_memo= ry_indirect_temps on command buffer destroy nvk_create_cmd_buffer() initializes two util_dynarrays, cmd->pushes an= d cmd->copy_memory_indirect_temps. nvk_destroy_cmd_buffer() finalizes th= e former but not the latter. Both are zero-initialized (mem_ctx =3D=3D NULL= ), so their backing storage comes from realloc() and is released by util_dynarray_fini(); nvk_reset_cmd_buffer() clears them without freeing,= keeping the allocation for reuse. copy_memory_indirect_temps grows a heap buffer the first time vkCmdCopyMemoryIndirectKHR is recorded (util_dynarray_ensure_cap() in nvk_cmd_indirect.c), and that buffer is not freed: destroying the command= buffer frees the struct but not the storage the dynarray points at. The nvk_cmd_mem entries it held are tracked separately in cmd->owned_mem a= nd are freed there, so the leak is the pointer array itself. Finalize it alongside cmd->pushes. Fixes: cbec12627bb ("nvk: VK_KHR_copy_memory_indirect") Reviewed-by: Mel Henning <[email protected]> (cherry picked from commit 84c7fef4581a4bd6673923240dab1a0fb6795e7a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/7fbd3d9359a5225d20e3e26005dbb97ea232229b">7fbd= 3d93</a></strong> <div> <span> by Valentine Burley </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>panvk: Disable VK_K= HR_internally_synchronized_queues on Vulkan 1.0 The extension depends on Vulkan 1.1. Fixes: 3b1f86d437f ("panvk: Advertise VK_KHR_internally_synchronized_queu= es") Signed-off-by: Valentine Burley <[email protected]> Reviewed-by: Christian Gmeiner <[email protected]> Reviewed-by: Aksel Hjerpbakk <[email protected]> (cherry picked from commit 79293b23366199314c33172e3ded126913e799f8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/22912cba2fd9033cd9e01aed16dcaca97832ce58">2291= 2cba</a></strong> <div> <span> by Lorenzo Rossi </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>panfrost: Fix wrong= layout reuse in user clip planes OpenGL user clip planes reuse the same fragment shader with a different vertex shader even when it was linked. Since the FS uses the previous memory layout, but the vertex shader emitting it is different, this causes all sort of problems. This commit fixes the bug by disabling memory layout optimizations when UCP is enabled. Fixes: 32ca5f7515d ("panfrost: Plumb VS varying_layout in FS") Signed-off-by: Lorenzo Rossi <[email protected]> Reviewed-by: Faith Ekstrand <[email protected]> Reviewed-by: Jakob Sinclair <[email protected]> (cherry picked from commit 73e7b3847f3231ac40da194ed384ac781d8cc6c2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/12ce20bac0bc7fee70660cad4ac9bbea0d4a5490">12ce= 20ba</a></strong> <div> <span> by JaeHoon Lee </span> <i> at 2026-07-29T13:35:26+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>v3dv: avoid restori= ng stale descriptor state after a meta op v3dv_cmd_buffer_meta_state_push() skips the descriptor_state memcpy when the current gfx descriptor_state is not valid. The skip path doesn't propagate that invalid state into state->meta, so state->meta's copy keeps whatever a previous meta op's memcpy left there. has_descriptor_state is still unconditionally set to true, so pop cannot tell push saw an invalid state this time: it reads the stale valid flag as current and, if nonzero, restores a descriptor set pointer left over from an earlier operation instead of the app's actual state. Clear meta's descriptor_state.valid in the skip branch so pop can observe the invalid state instead of reading what was left behind. Also fix a "subpass" typo in the neighboring attachment-state comment in the same function, found while reading through this code. Fixes: e59e7069283 ("v3dv: don't reset descriptor state after a meta oper= ation") Reviewed-by: Iago Toral Quiroga <[email protected]> (cherry picked from commit 0ff8b37a44ed8d8e76beb7c5df48e5a0867c7053) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d67e05f82fc8f615907fca8d153cc14913245c16">d67e= 05f8</a></strong> <div> <span> by Benjamin Cheng </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vl/proc: Guard comp= ositor creation with HAVE_GFX_COMPUTE Decode-only builds were broken because the vl_compositor was being created outside of stubbed entrypoints. Fixes: 1c8cec690cc ("va: Use vl_proc for processing context") Signed-off-by: Benjamin Cheng <[email protected]> Reviewed-by: David Rosca <[email protected]> Reviewed-by: Boyuan Zhang <[email protected]> (cherry picked from commit f083b55523ce72ed3654c35ac415f7045ed55f69) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/c54115900a4f30a5cb027bd00f8cf268df0a1ae0">c541= 1590</a></strong> <div> <span> by JaeHoon Lee </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nvk: report fills f= rom memory correctly The pipeline executable statistic for fills from memory reports the spill= count. Change it to report fills from memory instead. Fixes: 2a0302967feb ("nak: Add spill/fill statistics") Reviewed-by: Mel Henning <[email protected]> (cherry picked from commit eadcadf067ede525b5f511d22215272bb31eaa41) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/4432d3d3363a34c27d97a4c734ee38c0a7ff17aa">4432= d3d3</a></strong> <div> <span> by Christian Gmeiner </span> <i> at 2026-07-29T13:35:27+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>mesa/st: Invalidate= FS sampler views after PBO texture transfers Both PBO paths unbind all FS sampler views when they finish but the invalidate mask lost ST_NEW_FS_SAMPLER_VIEWS in the conversion from explicit dirtying. A following draw that does not rebind its textures then samples from no bound view and renders black. Seen as a black screen in dosbox, which uploads via glTexSubImage2D(..) from a PBO and then draws a textured quad from a display list without touching any texture state in between. Fixes: 62efee18607 ("mesa/st: Convert st_cb_texture to use st_context_inv= alidate_state") Signed-off-by: Christian Gmeiner <[email protected]> Reported-by: David Witbrodt <[email protected]> Tested-by: David Witbrodt <[email protected]> Reviewed-by: Marek Ol=C5=A1=C3=A1k <[email protected]> (cherry picked from commit 479773c7e4264506f2d9ec4bf15c6bf677f0d67a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/380dd8b9aff7e8525f9cd34d256abc178a869809">380d= d8b9</a></strong> <div> <span> by Dmitry Baryshkov </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: limit KHR_inter= nally_synchronized_queues to Vulkan 1.1+ The extension depends on VK 1.1. Fixes: 879dd9ca8c3 ("tu: Advertise VK_KHR_internally_synchronized_queues"= ) Signed-off-by: Dmitry Baryshkov <[email protected]>= (cherry picked from commit 67a59892af13e821ab90a99e9c19132f9d5869f2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/6027574a39c2ae2db06eb6cb1827f882dc102bdb">6027= 574a</a></strong> <div> <span> by Dmitry Baryshkov </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: limit VALVE_fra= gment_density_map_layered to Vulkan 1.1 devices VK_VALVE_fragment_density_map_layered requires Vulkan 1.4, VK_KHR_extended_flags or VK_KHR_maintenance5 (which in turn requires at least Vulkan 1.1). Limit the VALVE_fragment_density_map_layered extension to Vulkan 1.1 until the driver implements KHR_extended_flags extension. Fixes: a3a53b7cee4a ("tu: Implement VK_VALVE_fragment_density_map_layered= ") Signed-off-by: Dmitry Baryshkov <[email protected]>= (cherry picked from commit 8116745e085bcf2d66575f9e6d25857352abee89) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/652b088aae6bce49207aab23e7f9ce35b1c78283">652b= 088a</a></strong> <div> <span> by Frank Bouwer </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>pvr: Fix for depth = stencil 2d array writes. The base array layer was not taken into account. This fixes a reverted co= mmit. original commit: d058d2adb315e67bb62f4ff2b292bfd751f8be7f revert commit: 7f01ea0094dda7614287912c9922ad7d3e761507 Fix dEQP-GLES: dEQP-GLES31.functional.texture.multisample.samples_*.use_texture_dept= h_2d_array Fix regressions introduced by original commit DEQP-VK: dEQP-VK.api.image_clearing.core.clear_depth_stencil_attachment.multiple_l= ayers.* dEQP-VK.api.image_clearing.core.partial_clear_depth_stencil_attachment.mu= ltiple_layers.* dEQP-VK.api.image_clearing.dedicated_allocation.clear_depth_stencil_attac= hment.multiple_layers.* dEQP-VK.api.image_clearing.dedicated_allocation.partial_clear_depth_stenc= il_attachment.multiple_layers.* Backport-to: 26.1 Signed-off-by: Frank Bouwer <[email protected]> Reviewed-by: Frank Binns <[email protected]> (cherry picked from commit 1d899a489de21796a28075b8a4aed8a5954b0057) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ca6ef3b7f0c414e68aaafbf03b83d6488b6f92d2">ca6e= f3b7</a></strong> <div> <span> by Danylo Piliaiev </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Fix subsampled = metadata and blit emission for separate stencil Fixes: 4b87df29b3a ("tu: Implement subsampled images") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit bf70ab68a2105b24e211e3862337e95c7c2c97fd) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ebf01f0b0ce9693eb82ccea182ec78f855c20232">ebf0= 1f0b</a></strong> <div> <span> by Toshinari Morikawa </span> <i> at 2026-07-29T13:35:27+02:00 </i= > </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>egl: avoid calling = loader_get_driver_for_fd with fd =3D -1 platform_android.c, platform_surfaceless.c does not initialize fd_display_gpu before calling dri2_create_screen so that "libEGL warning: failed to get driver name for fd -1" was always printed on log. Cc: mesa-stable Tested-by: Dieter N=C3=BCtzel [email protected] Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@am= d.com> Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/13850 Signed-off-by: Toshinari Morikawa <[email protected]= > (cherry picked from commit 0bc2efab05ce6cf1dc90aeeff7e064482c127b2d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/4fe4782d5b32d0c792a4e0948de2f8c00510a184">4fe4= 782d</a></strong> <div> <span> by Georg Lehmann </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nir: fix divergence= for deref_cast The old check was broken because deref->var is a union member, and not valid for deref_cast. Either it's a trivial bit size changing cast, or it's a pointer to deref cast. In the first case we can rely on the parent divergence, in the second we have to check if the deref could be of a mode where even a uniform pointer would mean a different address per invocation. Fixes: df86c5ffb35 ("nir: add divergence analysis pass.") Reviewed-by: Marek Ol=C5=A1=C3=A1k <[email protected]> (cherry picked from commit 2c3c8b7fc267a226ad5d38035b9e86526d6063d2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/0ffb53c2520020f4834ef2c22e48440736c87382">0ffb= 53c2</a></strong> <div> <span> by Peng_Lx </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>turnip/kgsl: close = the dma-buf fd of our own allocations The three internal allocation paths -- dma-heap, ION and legacy ION -- obtain a dma-buf fd from the allocator and hand it straight to tu_bo_init_dmabuf() as the function's return value. tu_bo_init_dmabuf() only borrows that fd: kgsl_bo_init_dmabuf() keeps a dup of its own in bo->shared_fd, and that dup is what kgsl_bo_finish()= closes. Borrowing is the right contract for an imported dma-buf, where the fd belongs to the caller, but these three allocators created the fd themselves, so the reference they hold is never dropped. The result is that every internally allocated shareable BO leaks one dma-buf fd. kgsl_bo_finish() frees the KGSL handle and closes its dup, but the dma-buf refcount never reaches zero, so the memory stays allocated for the lifetime of the process. It is only reachable for exportable BOs, which makes it easy to miss: ordinary offscreen textures are unaffected. It shows up on a Wayland compositor, whose EGL window back buffers are exportable and are reallocated whenever Mesa trims the ones that went idle. Toggling the niri overview on an Adreno 830 (KGSL, 3840x2560 panel) leaked one 37.5 MB buffer per toggle, growing without bound -- 80 toggles took dma-buf usage from 202 MB to 3956 MB. With this patch the same run stays flat at 202-239 MB, and the allocate/close ledger balances exactly. Fixes: 17c12a9924a0 ("turnip/kgsl: Support external memory via ION/DMABUF= buffers") Cc: mesa-stable Signed-off-by: Peng_Lx <[email protected]> (cherry picked from commit 9f955af118d0cdabae776d08e36f99f7aa92317b) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b4e4d4acd2696bbb20b8858137351acc454ea484">b4e4= d4ac</a></strong> <div> <span> by Caio Oliveira </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nir: Account for cm= at memory accesses in copy_prop_vars Cc: mesa-stable Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15932 Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit b277a4925a161592801d711a8258433e2e2cda42) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/71378f32ccffb07d4e7e62b07d5c041af45a4dfe">7137= 8f32</a></strong> <div> <span> by Marek Ol=C5=A1=C3=A1k </span> <i> at 2026-07-29T13:35:27+02:00 = </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv: fix determini= ng the raster prim for guardband No change in correctness, but this is more accurate and faster because we= now have a tighter guardband around the viewport more often. The following commits that are fixes depend on this. Cc: mesa-stable Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit c61fdeb64c4765927708a1426426055266ba8776) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/c9ffe755fe55259b7e469ceadb44704e2d5a345d">c9ff= e755</a></strong> <div> <span> by Marek Ol=C5=A1=C3=A1k </span> <i> at 2026-07-29T13:35:27+02:00 = </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv: fix determini= ng the raster prim for line mode Cc: mesa-stable Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit 6b534309ead43c157b11f467bbdc059dd1816dcc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/6ac27d7f683808f4ad90cd0ff0bebef5c1798f25">6ac2= 7d7f</a></strong> <div> <span> by Marek Ol=C5=A1=C3=A1k </span> <i> at 2026-07-29T13:35:27+02:00 = </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv: fix determini= ng the dynamic raster prim for FS barycentrics Cc: mesa-stable Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit 9a77ee629d75f35a048649462c44fdc3f1e1d676) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/f7b88a2431b761e271ca699c00d1a3e97c234365">f7b8= 8a24</a></strong> <div> <span> by Marek Ol=C5=A1=C3=A1k </span> <i> at 2026-07-29T13:35:27+02:00 = </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv: fix determini= ng the static raster prim for FS barycentrics and front_face This also happens to fix the replacement of front_face with a constant wh= en face culling is enabled, improving performance. No front_face replacement= was done before this commit according to gpu-ratemeter. That may have bee= n because the unknown_rast_prim condition was incorrect and its result was unnecessarily true most of the time. Cc: mesa-stable Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit edeb6dd258b216bc45c9b2e7d260b01cce81e3af) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d0ccc7b803d1cfbcb35383ff639831ed71aeef06">d0cc= c7b8</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>rusticl/memory: ret= urn 0 for CL_IMAGE_SLICE_PITCH also for 2d images Backport-to: * Tested-by: Rob Clark <[email protected]> Reviewed-by: Rob Clark <[email protected]> (cherry picked from commit 1b0f0e0a83c4b20a0a24df41bafc6d0ca88d5fcb) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/bea70de76d1aefd7b666007d21a4c9d3e27c15ef">bea7= 0de7</a></strong> <div> <span> by Ryan Mckeever </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>docs: advertise VK_= KHR_multiview support for Bifrost Fixes: a1376449c80 ("panvk: Expose support for multiview on v7") Reviewed-by: Lorenzo Rossi <[email protected]> Reviewed-by: Erik Faye-Lund <[email protected]> (cherry picked from commit f11071ce865144ca3ff759c9d1a6c46ede06a16e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/8bb6d464039cacf0eb278027e9cea1d0d50120dd">8bb6= d464</a></strong> <div> <span> by Caio Oliveira </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>anv: Include build = and device identity in shader binary UUID In anv_physical_device.c we document that shaderBinaryUUID includes the driver build identity, but anv_shader_init_uuid() only hashed compile= r knobs and workaround settings. Add the build id and device info to the U= UID so driver or device-specific shader binary ABI changes invalidate VK_EXT_shader_object binaries. Fixes: 4314c891f4c ("anv: expose VK_EXT_shader_object") Assisted-by: Pi coding agent (GPT-5.5) Reviewed-by: Lionel Landwerlin <[email protected]> (cherry picked from commit c4039495967db0bff1ffaf2c955bc8586ce9c269) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/06845a1b7e2645680d7b7d5356f36aed1b3c7bba">0684= 5a1b</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: clean up syn= c2 treatment leftovers (cherry picked from commit ebf0bafa7dae458208c4fa2c75f074dd22fe27b5) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d083a51d1a512dddf5ec0fd32b770e02df2af423">d083= a51d</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: simplify syn= c feedback tracking (cherry picked from commit 5c46ee7379e5f3f95d48402b1f7b9560ac3276aa) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/e9464e48894d7e01d051c68c3ec274beff2b0b1f">e946= 4e48</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: update pnext= fix tracking Also move pnext alloc and assign right after main batch. (cherry picked from commit 9dd3d12f2d1ddabdd90992f3a27398d8c5d0d63e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/82c842503d413374f75ef5cc568a68517e0fd402">82c8= 4250</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: explicitly t= rack if need to fix batch This is to prepare for later fixes that don't need a batch: e.g. renderer sync preparations (cherry picked from commit 0406c3b0ae07b98a5842d0f6164ce71f752bdcce) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/60b5537dbf47aac1302549aefca0e81a5b08c4c4">60b5= 537d</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: flatten sync= feedback cmd counting This change: 1. move ffb cmd counting to ffb detection time 2. early return after suspending ffb and sfb if unable to do feedback 3. flatten qfb and later device group submit detection (cherry picked from commit b1f44187bc8ed1b1ab6e91553ac4cf5cd0326c66) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ff18906a32b293b61f94f665e69ceffd938e87e3">ff18= 906a</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: deprecate fe= nce feedback This puts fence back to synchronous driver<->renderer polling temporarily. (cherry picked from commit 4c1938c8adbe0e328f9db69362e79105c1e52d56) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/12632852e96254d3fcd91f15fb0fe12d376e84a2">1263= 2852</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:27+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: split ring s= ubmission to vn_queue_submission_do_submit (cherry picked from commit 2841b9b173c78cbe76d586070315fdfe05d64d2a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/e18e363c53169f12a988de43c38183b7bf902be2">e18e= 363c</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: skip empty b= atch submission With do_submission focusing on ring submission, now we can skip the empty batch after fully resolving the batch. This is to prepare for fence and sempahore being dropped from submission. (cherry picked from commit ca69ef0660963b9023dbe0265d1af9d9544de764) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/82d4f2801eae9153b2dc7cce2676550ec27f0dee">82d4= f280</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: drop vn_sync= _payload_external from submission tracking struct vn_sync_payload_external will be deprecated later after migrating external fence and semaphore over to syncobj. (cherry picked from commit 044c099959c22a949fa9866de51060c30c35c5e7) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/36205ba4e71c176b4596cf2ef87d6a629309cccf">3620= 5ba4</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: extract vn_t= imeout_to_poll_timeout (cherry picked from commit c90f8dfd01d45e6efbc3878492bba08d5ec1ebcc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/5ab4de36146eaaabbf18cda6af3a3a411735da3c">5ab4= de36</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus/virtgpu: only= signal non-zero initial value (cherry picked from commit 29abb028d912130bc4f57fef19d84a1f1cb4cc18) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/a34ca6868bbd13366475d9691d75988352f61ba0">a34c= a686</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus/virtgpu/vtest= : drop initial value from sync reset No use cases to reset to non-zero. (cherry picked from commit cfe2a5987f4e9009a2ea8dcf45c5df3febff58e6) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/00b415ff6c99d41d5533213a213fbe4693b5ce89">00b4= 15ff</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: prepare for = VN_SYNC_TYPE_SYNC This submits the syncs to renderer with a dependency on the last queue submit ring seqno. (cherry picked from commit d921c271c8b6ef53f9b0a52e9021403739ad3c0f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/275c59164c6e7b8ee8182d9da6dc1d33affb3ebf">275c= 5916</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: migrate fenc= e over to VN_SYNC_TYPE_SYNC This change: - stops creating renderer fence object - drops fence->external_payload trick for SYNC_FD fence export - adopts vn_renderer_sync for fence support in: - vn_CreateFence/vn_DestroyFence/vn_ResetFences - vn_GetFenceStatus/vn_WaitForFences - vn_ImportFenceFdKHR/vn_GetFenceFdKHR - skips ring submission if submitting just a fence with empty batch To be noted, VN_SYNC_TYPE_IMPORTED_SYNC_FD is preserved in favor of: - handling of -1 across drm_syncobj and sim_syncobj - vtest backend compatibility - avoiding unnecessary virtgpu ioctls (cherry picked from commit 6aded3b9d65b8af85f1684d042249f1a0b32a9c9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/c073f335d7a982fa9da07a83935f60c44cc28018">c073= f335</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: relax SYNC_F= D fence export requirement No longer need such requirement after fence being migrated to driver side syncobj. (cherry picked from commit 772cb76cc80ed5bde9c8a3972f0852d68fa54f9d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d32fa1104063c13b4d1c916614b7fcf5edfa63af">d32f= a110</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: deprecate qu= eue idle wait workaround Now that fence payload is submitted via the renderer, it has fenced prior submissions on the same ring_idx. We can safely drop this semi-working roundtrip. (cherry picked from commit e966e8293b6d5ae0519c6cae00a606b48de9a2ea) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/20a0c78512a45814471e884d08fb3e52417c2c6b">20a0= c785</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: rename to be= explicit about sync fd semaphore Let's call out sync fd explicitly. The special treatment isn't specific to imported sync fd but will also apply to exportable sync fd semaphore. We can't preserve the renderer object without requirement more on the renderer side due to the copy transference nature of sync fd. (cherry picked from commit f69094c19828feba0e65ba5c2530c00e5a2c8f2d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/34af8e275568e0fef3ee59b20953d5aed536b837">34af= 8e27</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: vn_semaphore= _(is|wait)_sync_fd to support VN_SYNC_TYPE_SYNC (cherry picked from commit 919b44d331305461cde32a6646df479e9fdf615c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/91fa1d93afb1b06dd5a19505229a0853dd438ec3">91fa= 1d93</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: rename exist= ing queue submission wait semaphore tracking Make it specific to wait semaphores. (cherry picked from commit 25fdd7a27ef446ee052b9861a8b7f0d460f27323) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/840221a3b4e92eaac6634f6ffd7b72b3fafdd624">8402= 21a3</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: count and pr= epare storage to scrub SYNC_FD signal semaphore (cherry picked from commit 90f58caf8cde8853499eaa1163f96321696be2be) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/c0ceec78eb8d32abc11175197f632212ec606429">c0ce= ec78</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: extract sync= s and scrub SYNC_FD signal semaphores 1. properly fix the pNext chain for those to be dropped 2. extract the syncs from the SYNC_FD semaphores 3. drop those semaphores from the ring submission batch (cherry picked from commit 6f3a570d4181e4c762388afed8a12a2f6f92c5de) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d5c11ff3976a7db3f010461e5fef09dfe2d4779b">d5c1= 1ff3</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: ensure rende= rer sync fence is submitted between queue batches Previously, the later queue batches might end up submitted before the sync fence submission on the renderer side. Now we submit roundtrip and have the followed queue batch submission via the ring to properly wait for the submitted roundtrip, which ensures the completion of sync fence submission. (cherry picked from commit 36f79c3c6459bd3442519b223be071e15582b15c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/c23bf808c0bc9b698a0e3cb4fd3d9b691e5fb0c4">c23b= f808</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: migrate SYNC= _FD semaphore over to VN_SYNC_TYPE_SYNC This change: - stops creating renderer semaphore object for SYNC_FD semaphore - drops sem->external_payload trick for SYNC_FD semaphore export To be noted, VN_SYNC_TYPE_IMPORTED_SYNC_FD is preserved in favor of: - handling of -1 across drm_syncobj and sim_syncobj - vtest backend compatibility - avoiding unnecessary virtgpu ioctls (cherry picked from commit c5eda762851d1ec8226e10ecd20795fe2511ea6d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/88114a51baa9dd606d3d290d88e4febb32814b9d">8811= 4a51</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: relax SYNC_F= D semaphore export requirement (cherry picked from commit abbaca283bbde0d208cc1b5ff5c39ea609799a34) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/885c401d0bec2a85ec5548b4ff6939df4a795130">885c= 401d</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus/virtgpu: hide= has_timeline_sync behind a new perf option This change: 1. adds a new VN_PERF=3Dno_timeline_sync perf option 2. refactors to use renderer instead of gpu->base indirection (cherry picked from commit b17c963ec861abc74900b245915d680d8fe69408) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/dbdd5b7c9f1623b70c3b47c1b2b3baab0f23bb28">dbdd= 5b7c</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus/vtest: advert= ise timeline syncobj support The vtest interface does support timeline sync. Similarly, hide behind the new perf option. (cherry picked from commit b64c5743837d0564131ef901e5b073e3dcfdbad8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/c9fd788b9efa62a105d41cd4cae00c5d47c47ca3">c9fd= 788b</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: drop vn_rend= erer_sync_flags At this point, I'm pretty sure we don't need this flag at all. Meanwhile, drop some obsolete TODOs. (cherry picked from commit 23c8ebd60c79c43599675042dc8d613095df34eb) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/6b1a82e89095ddee5e0aa55469712b83645cd1a6">6b1a= 82e8</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: add VN_SYNC_= TYPE_TIMELINE_SYNC (cherry picked from commit 0ed627fed42316bc3e4ff0c962507b632c8a9d2c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/975a86f64179e2dca51a6336623e6eb9c33e675f">975a= 86f6</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: track queue = internal array index within vn_device::queues This will be used to map queue timeline to timeline semaphore syncs. (cherry picked from commit 71732947f30be3d9def4a43884394b38ca86e580) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/9dccd7a6f31a4e3cd416c517d94e83124d5a4adc">9dcc= d7a6</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: count and in= it syncs from timeline semaphores For timeline semaphore, the wait and signal semaphores are both forwarded to the renderer side for device operation. So we only count the signal timeline semaphores and submit those extracted syncs. (cherry picked from commit fe293af9f18a5bb87b15bb85cf96bfb5c80e0961) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/f707650bb08ca1d526e9c1dfb5f1b4e9cdf78a11">f707= 650b</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: implement ho= st signal for TIMELINE_SYNC Only need to signal the cpu timeline sync (back of the payload syncs) to the requested counter value. (cherry picked from commit 4219bb97c5ae5fbcdfe84d122017bfec74b99e92) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/8c2b91beec0910a7c09ffe6e5ebffe9a1c3a9054">8c2b= 91be</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: implement co= unter query for TIMELINE_SYNC One of the TIMELINE_SYNC payload syncs has the current point of the timeline semaphore, so we query each of them to return the greatest. (cherry picked from commit cf06a29d206ae140a5dd15c8a2bfea7f51767726) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b34e5b2d71631e369c495be33ad36471a62b2f82">b34e= 5b2d</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: add vn_wait_= semaphores_legacy for legacy wait Add a simple sem payload type test to prepare for timeline sync. (cherry picked from commit 85543981c2f002555f1ae42cf4d77b99dbfbcc87) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/5fce4ae95ecfbc1aac6a2a84691a38d0662ccdd2">5fce= 4ae9</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: implement se= maphore wait for TIMELINE_SYNC For WAIT_ANY, collect all syncs from all semaphores and wait any one of them to signal with the passed timeout in one pass. Otherwise, wait for individual semaphores one by one. For each of the semaphores, wait for any one of the semaphore timeline syncs to signal. (cherry picked from commit f72c0018463eaed412eee984537d2f94d3511aac) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/284dca2d5ea1ddd539923253c29f514a75aceeca">284d= ca2d</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: migrate time= line semaphore to VN_SYNC_TYPE_TIMELINE_SYNC (cherry picked from commit 0959c8ec29776c054bb07d48609d71093e37f2b1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/be915e54e9f7cee91593736afa83d23ab931f885">be91= 5e54</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: document tim= eline semaphore implementation (cherry picked from commit 8b794a5167a729758cac9bde45fbe1886eb89c91) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d1a613998f7de05955d34b204d4eec3299ddf1ad">d1a6= 1399</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>pick-ui: turn commi= t.date into a (cached) property This is a property of the commit, not an action performed on the commit, but it's expensive to compute so let's cache it. (cherry picked from commit b6386e96e5a81e074a90dff724ef3c35b1a9d1e7) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/dc260c14dd129881ebcde613dc18b836cd500427">dc26= 0c14</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>pick-ui: show MR nu= mber for additional context Shows which commits belong together, and shows context for problematic commits (doesn't backport, doesn't build, fails the unit tests, breaks a CI test). (cherry picked from commit 2086df71501a2e2934d36a33d5ea9087057423e9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/e7b165db96d3ff23f0c5b69b8ba6e52690d5e5c7">e7b1= 65db</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>rusticl/kernel: add= libclc source hash to kernel shader keys Backport-to: * (cherry picked from commit 780727e68adc62f446dc1df7bcaf86f20b6c03d1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/882dcedd880f71fd492bb97bef42c5b844e5b0d7">882d= cedd</a></strong> <div> <span> by Christoph Pillmayer </span> <i> at 2026-07-29T13:35:28+02:00 </= i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>pan/kraid: Fix 32bi= t hw_runner builds On armhf, unsigned long is uint32_t, so passing a u64 into drmIoctl doesnt work. Also don't pull in everything into bindgen. Fixes: 1d171a41 ("kraid: Add hw_runner") Reviewed-by: Lars-Ivar Hesselberg Simonsen <[email protected]= > (cherry picked from commit 6cc038ecd1fdf6b654d7b976752a68bc60ba7b50) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/80747baec7ea10d03cc5a0f39e5380a11f8e0272">8074= 7bae</a></strong> <div> <span> by Christoph Pillmayer </span> <i> at 2026-07-29T13:35:28+02:00 </= i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>pan/perf: Fix 32bit= build of panquick Fixes: 5c15ef50 ("pan/perf: Transition to auto-generated derived counters= ") Reviewed-by: Erik Faye-Lund <[email protected]> (cherry picked from commit 6015d5e538d531f230bb102e0af10eed566e6900) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b7dc7ca1e040a8cf63436df9ee349707a2cbce0c">b7dc= 7ca1</a></strong> <div> <span> by Rhys Perry </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>ac/nir/ngg: don't s= hrink device-scope memory barriers Signed-off-by: Rhys Perry <[email protected]> Backport-to: * Reviewed-by: Marek Ol=C5=A1=C3=A1k <[email protected]> (cherry picked from commit 027a2b45fcb69ed5e421e6a3f829ea1fd7c65787) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/26eb9ed2f42c9a6e8425e55a3dadacb300a7378b">26eb= 9ed2</a></strong> <div> <span> by Rhys Perry </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>ac/nir/ngg: track u= niformity with multiple set_vertex_and_primitive_count This can be in uniform control flow. Signed-off-by: Rhys Perry <[email protected]> Fixes: 863fdcf5196 ("ac/nir/ngg: Use workgroup divergence analysis for me= sh output counts.") Reviewed-by: Marek Ol=C5=A1=C3=A1k <[email protected]> (cherry picked from commit ba8dfc5aa8ac3af757d7330f69e1cd8573f5772e) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/bd94a96c9994c7108424b1ba1dce0e469d401ae9">bd94= a96c</a></strong> <div> <span> by Yiwei Zhang </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>venus: ensure cache= d vn_ring_submit batches are bounded The prior fix can still end up with unbounded leak since the actual shmem count in the storage isn't tracked. This change reverts the prior fix while avoiding caching over-sized batch. Practically, there's no need to cache more since the malloc overhead is trivial as compared to indirect uploads. Fixes: 2cf1f6cb508 ("venus: fix unbound malloc leak in vn_ring_get_submit= s") (cherry picked from commit 09fb7ca8d824616712e8d3b2639240087800e3d5) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/b6130bce0ab3a9e4d0f61bcc9433b0f4555088e1">b613= 0bce</a></strong> <div> <span> by Danylo Piliaiev </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Fix gfx_write_a= ccess checking for TRANSFORM_FEEDBACK_COUNTER_READ_BIT It should check for WRITE_BIT. Fixes: 59259a01671 ("tu: Convert to sync2 entrypoints") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit 221d58dc47746e344d10b3149322131b4e0f2b27) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/3a789ca9aaada161341429dee1da878add337705">3a78= 9ca9</a></strong> <div> <span> by Lionel Landwerlin </span> <i> at 2026-07-29T13:35:28+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>anv: fix push buffe= r descriptor address relocation Fixes a few tests on < Gfx12.5 like : ANV_DEBUG=3Dexperimental ./deqp-vk -n dEQP-VK.dgc.ext.compute.misc.desc= riptor_buffer_push_descriptor_with_ies Signed-off-by: Lionel Landwerlin <[email protected]> Fixes: 1ffdbe276d ("anv: Replace va.indirect_descriptor_pool access with = a function") Reviewed-by: Tapani P=C3=A4lli <[email protected]> (cherry picked from commit 849eb7067fbaddf07c350a5d00760d700e27de86) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/fbed83c8a4b1fb9c82a4c6ec8cda061b060961a4">fbed= 83c8</a></strong> <div> <span> by Janne Grunau </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>asahi: nir: lower f= lrp64 Fixes unhandled ALU instruction asserts | Unhandled ALU op flrp | src/asahi/compiler/agx_compile.c:2160: agx_emit_alu: Assertion `!"Unhan= dled ALU instruction"' failed. observed since commit ac679a85049 ("glsl: translate fma as fma_weak") in following OpenGL CTS tests: - KHR-GL46.gpu_shader_fp64.builtin.mix_double - KHR-GL46.gpu_shader_fp64.builtin.mix_dvec2 - KHR-GL46.gpu_shader_fp64.builtin.mix_dvec3 - KHR-GL46.gpu_shader_fp64.builtin.mix_dvec4 Backport-to: 26.2 Signed-off-by: Janne Grunau <[email protected]> (cherry picked from commit ed30ed7c39d2fae442c4a353144ba1112e9f0d3d) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/6ebbe5d71a6f2c03c72e81e4fbefe4f00cc76191">6ebb= e5d7</a></strong> <div> <span> by Danylo Piliaiev </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Fix blit_cache_= cleaned never being set to true tu6_emit_flushes clears cache's flush_bits, resulting in the follow up `flush_bits & TU_CMD_FLAG_BLIT_CACHE_CLEAN` never being true, leading to excessive flushing when dynamic input attachments were used. Fixes: a99600322c1 ("tu: Track possible feedback loops for dynamic render= passes") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit ba8de990e36474c9b46a103ca274c52fc15c87b8) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/409d6f1e91a74a56fcce2208c42afd2baea27d03">409d= 6f1e</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radeonsi: Stop forc= ing GTT with no gfx/compute This is no longer needed as staging texture is now never used without gfx/compute. Backport-to: 26.2 Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@am= d.com> (cherry picked from commit 1e70c9859d16a9fcca3d9d70c545c8e10c81c8e2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/119a930a2f45dac848e4d6f6a477e7da5b72a97f">119a= 930a</a></strong> <div> <span> by David Rosca </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>ac/video: Fix numbe= r of AV1 single refs 2 refs are only supported with compound. Fixes: 914b4a700cc ("ac/info: Add video codec caps") Reviewed-by: Ruijing Dong <[email protected]> (cherry picked from commit 6f0f36e3b1f9462335d717a9363d0880ae5726be) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/0b940b05317375d1c37c37c8eb11c067803f12f3">0b94= 0b05</a></strong> <div> <span> by Jaakko Jokinen </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nir: Add cases to n= ir_get_io_offset_src_number() Added cases nir_intrinsic_store_shared_unlock_nv and nir_intrinsic_load_shared_lock_nv. This fixes segmentation faults caused by the unexpected default return value -1. Fixes: 47f6c74b71c ("nir,nak: Add KeplerB shared atomics intrinsics and l= owering") Reviewed-by: Mary Guillemard <[email protected]> (cherry picked from commit 61f6c264bcf917468854a82b2a0c9a0912bc6c1a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/f56d024f523e886a7cda50e65cf23b262af7d1fb">f56d= 024f</a></strong> <div> <span> by Danylo Piliaiev </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Fix LRZ handlin= g for VK_EXT_custom_resolve Custom resolve can resolve depth attachments, this case is a bit narrower that general case of RP using several depth attachment. Here we should have LRZ enabled for the main part of RP and disabling it only for custom resolve. Note, if custom resolve resolves depth we still disable LRZ for the main depth attachment _after_ the RP, which is not strictly necessary but I don't see a point in handling this case. Fixes: 520e3f3a473 ("tu: Implement VK_EXT_custom_resolve") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit adca1f255fd5cb9702db1e0a2bc25d13a521ae5f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/a60580261aaf606e29e868b30d646b8f90ec77b3">a605= 8026</a></strong> <div> <span> by Danylo Piliaiev </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Dirty LRZ after= changing attachment locations disable LRZ writes We called tu_lrz_disable_write_for_rp, but didn't trigger LRZ state reemission. Fixes: 7ee9835475c ("tu/lrz: Disable LRZ when CmdSetRenderingAttachmentLo= cations is used") Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit 3c0803f5a17f13bad5c2f2da7cdeaee4f792eeb2) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ac675d7e8c1c0c7ea9d3cf364b6b6ab45d9d506c">ac67= 5d7e</a></strong> <div> <span> by Lionel Landwerlin </span> <i> at 2026-07-29T13:35:28+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>brw: add missing IN= TEL_FS_CONFIG_PER_PRIMITIVE_REMAPPING handling Fixes tests in dEQP-VK.mesh_shader.ext.pipeline.builtin.fast_lib.* in DG2= . Signed-off-by: Lionel Landwerlin <[email protected]> Fixes: df5a6d7b87 ("brw/jay: move some coarse lowering to NIR") Acked-by: Alyssa Rosenzweig <[email protected]> (cherry picked from commit 3cd07aacb9ebe839bd73c9c5e07afff4db0e0d00) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/1c7425fa44b00862df3e97623836fb95664a57d3">1c74= 25fa</a></strong> <div> <span> by Jan Meisel </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv: fix RADV_PERF= TEST=3Dnircache enablement RADV_PERFTEST_NIR_CACHE is parsed from RADV_PERFTEST, but enable_nir_cache reads debug_flags after radv_compiler_info was introduced. This makes RADV_PERFTEST=3Dnircache a no-op. It also accidentally ties the feature to RADV_DEBUG=3Dinfo because both enums use bit 13. Use perftest_flags so the documented perftest flag enables per-stage graphics NIR caching again. On an RX 9070 XT (GFX1201), a local graphics pipeline benchmark with 1000 vertex-input permutations improved from 368.345 ms to 255.500 ms with RADV_PERFTEST=3Dnircache (-30.636%, n=3D5, interleaved cold-cache runs). Fixes: 4a91fd8babce ("radv: add a radv_compiler_info object") Reviewed-by: Konstantin Seurer <[email protected]> Assisted-by: Codex Signed-off-by: Jan Meisel <[email protected]> (cherry picked from commit ad0cdded0e439515155677941cada1cb4bb4d388) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/261a270c2cc5e8750d32a14df1f1274adc631a72">261a= 270c</a></strong> <div> <span> by utzcoz </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>virtio: magma-gpu-r= s: accept a null device in virtgpu_kumquat_finish virtgpu_kumquat_init leaves the caller's pointer untouched when it fails,= and VirtGpuKumquatDevice releases the device unconditionally from its destructor, so virtgpu_kumquat_finish is reached with a null pointer on every failed initialization. Box::from_raw on a null pointer raises a non-unwinding panic, which aborts the process and takes with it the error= that explained why initialization failed. The surrounding catch_unwind cannot intercept that. Make the release side tolerate null instead of fixing the one caller. This is a C init/finish pair, and C's canonical release function treats null as a no-op, so C callers reasonably expect that; it also covers callers we do not control. Fixes: f2b07903abaf ("mesa: import virtgpu_kumquat_ffi") (cherry picked from commit 0989349030b36e6ce587cf70f2a5c7615465f04f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/dcfeffe0b89eddbe5627da1f840328d919c82807">dcfe= ffe0</a></strong> <div> <span> by Caio Oliveira </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>brw: Match fill/spi= ll optimization scratch accesses by logical offset brw_opt_fill_and_spill matches and invalidates scratch accesses by physical offset but removes dead spills by logical offset. With scratch slot reuse these two views disagree. For example, values A, B and C take turns using the same physical slot, the first operations on B and C are partial writes (which trigger read-modify-write sequences), and C is defined by the same `mad` that last reads B. ``` fill A ; A's last use mad X, A, ... fill B ; B's RMW fill mad B:hf, ... spill B ... fill C ; C's RMW fill (C was spilled before B) fill B ; B's last use mad C:hf, B, ... spill C ``` The duplicate fill elimination matches the first `fill B` with `fill A` and the second one with `fill C`. Same physical offsets, even though all their logical offsets differ: ``` fill A mad X, A, ... mov B, A ; was `fill B`, matched by physical offset mad B:hf, ... spill B ... fill C mov B, C ; was `fill B`, matched by physical offset mad C:hf, B, ... spill C ``` Then, with no fills of B left, `spill B` looks like a dead write based on logical offsets and is removed: ``` fill A mad X, A, ... mov B, A mad B:hf, ... ; `spill B` was removed ... fill C mov B, C mad C:hf, B, ... spill C ``` But `spill B` was not dead: its value is still used by the last `mad`. The `mov`s hid that use. B's value now flows through `fill C` and `mov B, C`, so without the spill, `fill C` reads A's old bytes and the last `mad` gets a stale B. The right thing here would be to not change those `fill B`s; they should only match other accesses of B. So change the rest of brw_opt_fill_and_spill subpasses to also use logical offsets. This only narrows the matches (same logical offset implies same physical offset), and the invalidation checks remain correct since only one value at a time can access its slot. Found by the llama.cpp unit test on BMG: ``` test-backend-ops test -o FLASH_ATTN_EXT -p hsk=3D512 ``` No changes to fossil-db on BMG. Fixes: a9b67ad0954b ("brw: Reuse scratch slots between non-interfering sp= illed VGRFs") Assisted-by: Claude Code (Fable 5) Reviewed-by: Ian Romanick <[email protected]> (cherry picked from commit 28f0804ea039c333d2294bdd6565332dbdfe7979) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ad94120ba2bc69843141edc75e33721eb73df981">ad94= 120b</a></strong> <div> <span> by Janne Grunau </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>asahi: ci: Drop no = longer failing VK.wsi.xcb.present_timing test "dEQP-VK.wsi.xcb.present_timing.basic.timing_queue" no longer fails after implementing present timing for X11 WSI in https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39551 Fixes: 1546dabd4a9 ("wsi/x11: Implement main body of present timing.") Reviewed-by: Eric Engestrom <[email protected]> Signed-off-by: Janne Grunau <[email protected]> (cherry picked from commit c523afe2b0b258a701ad0e0cf2851524be93c7c1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/e151dd11b44f4894fe40725b71dab686e41d90ea">e151= dd11</a></strong> <div> <span> by Janne Grunau </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>panfrost: ci: Drop = no longer failing VK.wsi.xcb.present_timing test "dEQP-VK.wsi.xcb.present_timing.basic.timing_queue" no longer fails after implementing present timing for X11 WSI in https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39551 Fixes: 1546dabd4a9 ("wsi/x11: Implement main body of present timing.") Reviewed-by: Eric Engestrom <[email protected]> Signed-off-by: Janne Grunau <[email protected]> (cherry picked from commit aa4a378f301079faec3a042081ba19ba6588c7fc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/cfdb81f41764e060e56ca522fb82bccb346dfa8e">cfdb= 81f4</a></strong> <div> <span> by Janne Grunau </span> <i> at 2026-07-29T13:35:28+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>asahi: ci: Add fail= ing b10g11r11 and e5b9g9r9 copy tests These tests are new in VKCTS 1.4.5.3 and the tested functionality was likely always broken. It is not immediately clear why the tests fail so mark them as fails for now until they can be fixed in one way or another. This simplifies chasing other VK CTS regressions for honeycomb. Backport-to: 26.2 Acked-by: Eric Engestrom <[email protected]> Signed-off-by: Janne Grunau <[email protected]> (cherry picked from commit c276444b7190c2feaa722168253baaf9f3c737e9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/ae84375803121e91495eb963a35401a2d2fd5af4">ae84= 3758</a></strong> <div> <span> by Karol Herbst </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>vtn/opencl: fix lib= clc needing fp16 lowering to fp32 fp16 builtins inside libclc might be implemented on top of other fp16 builtins. However some of those might not have a fp16 version in which case we have to fall back to the fp32 one. This might be a libclc build system configuration issue, but we can easily workaround this by looking up in the current shaders whether we find a fp32 version of a builtin. Backport-to: * Reviewed-by: Zolt=C3=A1n B=C3=B6sz=C3=B6rm=C3=A9nyi <[email protected]= > (cherry picked from commit 60654ceb9139ab8f289c6252bc12659fcf655e1c) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/2bce69ca3f88ecc6a458d435883462ce81cf6a90">2bce= 69ca</a></strong> <div> <span> by Christian Gmeiner </span> <i> at 2026-07-29T13:35:29+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nir/opt_vectorize: = Remove the phis that were combined Combining two phis rewrote their uses but left the phis themselves in the block. They still hash equal and still dominate, so the next run of the pass combined them again, reported progress again and added another dead phi. The ALU path already drops its inputs, in rewrite_uses(). Fixes: 850768bacbf ("nir/opt_vectorize: prefer to swizzle vector phis at = the destination, not the source") Signed-off-by: Christian Gmeiner <[email protected]> Reviewed-by: Georg Lehmann <[email protected]> (cherry picked from commit 829e6d16301d3daf83743f4edc777567cf76d8f4) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d5c0266b37b5179881c86f62db9a5f76bbfd4530">d5c0= 266b</a></strong> <div> <span> by Jeremy Huddleston </span> <i> at 2026-07-29T13:35:29+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>glx/apple: only ski= p AppleGL election on macOS 26.0 through 26.5 The OpenGL.framework breakage that made AppleGL unusable on macOS 26 was = addressed in macOS 26.6, so limit the fallback to Gallium to the affected releases and elect AppleGL as usu= al on newer systems. Fixes: fa0974ee1cbf ("glx/apple: skip AppleGL election on macOS 26 and ne= wer") Signed-off-by: Jeremy Huddleston Sequoia <[email protected]> (cherry picked from commit c810f97ad70db582277b74326d610ec62528eeb9) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/a2642c7869e9634d6c765549e7fb2ab1d1dbdb53">a264= 2c78</a></strong> <div> <span> by Frank Binns </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>pvr: rearrange some= functions in pvr_arch_border.c This is in preparation for a follow on commit. Backport-to: 26.1 Signed-off-by: Frank Binns <[email protected]> Reviewed-by: Luigi Santivetti <[email protected]> (cherry picked from commit 311207091a7d9eecc5a0be6f9828e942709fa351) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/1beb874992f1a48ab9521645b80a39cb6f484dd4">1beb= 8749</a></strong> <div> <span> by Frank Binns </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>pvr: setup all form= at fields for custom border color entries This fixes an issue seen with depth/stencil formats where Zink creates a = sampler with either a depth only or stencil only custom border color, but then us= es this with an image with a combined depth/stencil format, taking either th= e depth or stencil aspect. This results in a mismatch between the format field th= at gets written in the border color entry and the one that gets read from the bor= der color entry. For example, if a custom border color is setup for VK_FORMAT_S8_UINT and = used in combination with a VK_FORMAT_D32_SFLOAT_S8_UINT image, taking the stencil= aspect, then the format in the image state words will be X24G8X32. This r= esults in the value in the X24G8X32 format field being looked up in the border c= olor entry rather than the S8 field. One way to resolve this would be to add additional handling for depth/ste= ncil formats to write values to all relevant format fields in a border color entry. However, a simpler approach is to setup all format fields for cust= om border color entries, as is already done for built-in border colors. This= allows the code to be simplified, since the same function can be used for both c= ases. Fixes tests in dEQP-GLES31.functional.texture.border_clamp.*. Fixes: bf9c84991fc ("pvr: Add support for custom border colors") Signed-off-by: Frank Binns <[email protected]> Reviewed-by: Luigi Santivetti <[email protected]> (cherry picked from commit ea6e5b6b38322599bd36890ee1a4dec08771faf1) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/8d5f395286d2938e9250021377a8a1f60f334713">8d5f= 3952</a></strong> <div> <span> by Lionel Landwerlin </span> <i> at 2026-07-29T13:35:29+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>brw: fix wa_1801911= 0168 lowering If the linked mesh shader to the fragment shader doesn't have the workaround active, there is nothing to do. Signed-off-by: Lionel Landwerlin <[email protected]> Fixes: fcf4401824 ("brw: handle wa_18019110168 with independent shader co= mpilation") Reviewed-by: Ivan Briano <[email protected]> (cherry picked from commit f8801d555b58d1965edab147018250830a3483a3) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/637a477386e913b7e954c625bab81d23e553d00e">637a= 4773</a></strong> <div> <span> by Natalie Vock </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv/nir: Clean up = descriptor index lowering This switches to a forward iteration over descriptor-related intrinsics. The biggest advantage of this is that we can now change the result types of vulkan_resource_(re)index, which also allows us to nuke all the special-casing for raytracing descriptors. Aside from just looking better, this will also allow us to implement mutable acceleration structure descriptors. Since this is needed to fix Unreal Engine 5.8, it should be backported even though it's a cleanup. Cc: mesa-stable (cherry picked from commit 0be03076dcdc2131f8084a67f5db4c07579a76fb) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/504d077c7c49b20146e3b3a769fe55e1d1ad523d">504d= 077c</a></strong> <div> <span> by Natalie Vock </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>radv: Expose mutabl= e acceleration structure descriptors These Just Work(tm) now. Unreal Engine 5.8 needs it. Cc: mesa-stable (cherry picked from commit dcb6045ac5b45a0d086d37904588a4fc05262233) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/70b85dc2bce74b6bb287a470361affe44aa9fffa">70b8= 5dc2</a></strong> <div> <span> by Danylo Piliaiev </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>nir: Include scalar= ized component offsets in UBO ranges When scalarizing UBO loads, we should take into account per-component offset when calculating base and range. This is important for dynamically index row major matrices, where without this fix each row's base/range is calculated as if it starts from offset 0. ``` layout(std140, row_major, offset =3D 512) mat4 mat_arr[2]; ... =3D mat_arr[gl_ViewIndex] ``` Without the fix it was scalarized, then vectorized back into: ``` 32x4 %1266 =3D @load_ubo (%2638, %1297) (access=3Dnone, align_mul=3D= 64, align_offset=3D0, range_base=3D512, range=3D92) 32 %2725 =3D load_const (0x00000210 =3D 528) 32 %2726 =3D iadd %2725 (0x210), %1295 32x4 %1273 =3D @load_ubo (%2638, %2726) (access=3Dnone, align_mul=3D= 64, align_offset=3D16, range_base=3D512, range=3D92) 32 %2727 =3D load_const (0x00000220 =3D 544) 32 %2728 =3D iadd %2727 (0x220), %1295 32x4 %1280 =3D @load_ubo (%2638, %2728) (access=3Dnone, align_mul=3D= 64, align_offset=3D32, range_base=3D512, range=3D92) 32 %2729 =3D load_const (0x00000230 =3D 560) 32 %2730 =3D iadd %2729 (0x230), %1295 32x4 %1287 =3D @load_ubo (%2638, %2730) (access=3Dnone, align_mul=3D= 64, align_offset=3D48, range_base=3D512, range=3D92) ``` Notice the same base, which caused ir3 to not preload the whole 128b. After the fix we get: ``` 32x4 %1266 =3D @load_ubo (%2638, %1297) (access=3Dnone, align_mul=3D= 64, align_offset=3D0, range_base=3D512, range=3D80) 32 %2725 =3D load_const (0x00000210 =3D 528) 32 %2726 =3D iadd %2725 (0x210), %1295 32x4 %1273 =3D @load_ubo (%2638, %2726) (access=3Dnone, align_mul=3D= 64, align_offset=3D16, range_base=3D528, range=3D80) 32 %2727 =3D load_const (0x00000220 =3D 544) 32 %2728 =3D iadd %2727 (0x220), %1295 32x4 %1280 =3D @load_ubo (%2638, %2728) (access=3Dnone, align_mul=3D= 64, align_offset=3D32, range_base=3D544, range=3D80) 32 %2729 =3D load_const (0x00000230 =3D 560) 32 %2730 =3D iadd %2729 (0x230), %1295 32x4 %1287 =3D @load_ubo (%2638, %2730) (access=3Dnone, align_mul=3D= 64, align_offset=3D48, range_base=3D560, range=3D80) ``` Which results in load of the correct UBO range. Cc: mesa-stable Signed-off-by: Danylo Piliaiev <[email protected]> (cherry picked from commit d9d106a5295dd4a220eb7918e1d0f16524ebb3d4) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/cc44702a1d13762282a6055f380c82b02e167e24">cc44= 702a</a></strong> <div> <span> by Emma Anholt </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>freedreno: Don't fo= rce image component A=3D1 substitution on R/RG textures. This is a no-op for normal texture sampling, but with QCOM_image_processing the image component substitution happens before filtering (successfully, due to the underlying format), so re-substitutio= n during the texture component swizzle stage causes incorrect A=3D1 when yo= u should be getting the filtered value. This also means we need to refuse weight image sampling for BC1 RGB, where we have no choice but to do the A=3D1 substitution in the swizzle. Cc: mesa-stable (cherry picked from commit 3bdfd556b1ca602bef7936dc548f624708d1a5d7) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/3c47d782071408f4d26dcc29faacae7f90e53514">3c47= d782</a></strong> <div> <span> by Pohsiang (John) Hsu </span> <i> at 2026-07-29T13:35:29+02:00 </= i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>d3d12: fix msvc bui= ld warning C4819 Fixes: 42f636bee95 ("d3d12/video: Don't reset batches in fence_wait") Reviewed-by: Sil Vilerino <[email protected]> (cherry picked from commit a25e41df17602b3c3962805be6fe748ad226abbc) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/76589dcd2d633a22c77e2eca6e8ad7b9e96bbfa6">7658= 9dcd</a></strong> <div> <span> by Caio Oliveira </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>intel/gen: Fix Gfx1= 1 3-src accumulator file encoding The decoder was correct, but encoder was wrongly using IMM instead of ARF in two fields. Fixes: 03c32e90a98f ("intel/gen: Add gen encoding module") Reviewed-by: Ian Romanick <[email protected]> Reviewed-by: Jordan Justen <[email protected]> (cherry picked from commit 1c0e6ad8cbf1a36b5eb22bfbd0389264955b8b60) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/5d440d08ae3085e373d4fecdf94572cb2641efa6">5d44= 0d08</a></strong> <div> <span> by Lionel Landwerlin </span> <i> at 2026-07-29T13:35:29+02:00 </i>= </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>anv: fix leak in RT= binding point Signed-off-by: Lionel Landwerlin <[email protected]> Cc: mesa-stable Reviewed-by: Michael Cheng <[email protected]> (cherry picked from commit b287c8fd44bfae03b9e7ef2585bf28c30d5bc67f) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/d6d8b826191f025f095ea77b6871a821d4be452c">d6d8= b826</a></strong> <div> <span> by Georg Lehmann </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>aco/live_var_analys= is: make sure shared vgprs are within the encodable vgprs If we are fine with a low number of waves, vgprs will be greater than 256= before we clamp to the vgpr limit. Cc: mesa-stable Reviewed-by: Daniel Sch=C3=BCrmann <[email protected]> (cherry picked from commit da3e65d22d7a678160bfa68c4eee106282e1eea4) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/a293c85af63c74b523e8b8d3ecc3ba5b8165a5a2">a293= c85a</a></strong> <div> <span> by Valentine Burley </span> <i> at 2026-07-29T13:35:29+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>tu: Report correct = maxFragmentInputComponents limits maxFragmentInputComponents doesn't need to reserve space for builtin variables. Remove the conservative adjustment and report the actual hardware limits. Cc: mesa-stable Signed-off-by: Valentine Burley <[email protected]> (cherry picked from commit 42694cb11015d3efea30abb3358b8a0b3875f71a) Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/43= 295> </pre> </li> <li> <strong style=3D"font-weight: 600;"><a href=3D"https://salsa.debian.org/x= org-team/lib/mesa/-/commit/5db7111ffb9f932815baf7b299cd4db772e6d301">5db7= 111f</a></strong> <div> <span> by Eric Engestrom </span> <i> at 2026-07-29T18:32:23+02:00 </i> </div> <pre class=3D"commit-message" style=3D'white-space: pre-wrap; display: bl= ock; font-size: 14px; color: #3a383f; position: relative; font-family: "G= itLab Mono", "JetBrains Mono", "Menlo", "DejaVu Sans Mono", "Liberation M= ono", "Consolas", "Ubuntu Mono", "Courier New", "andale mono", "lucida co= nsole", monospace; font-variant-ligatures: none; word-break: break-all; w= ord-wrap: break-word; background-color: #fbfafd; border-radius: 2px; marg= in: 0; padding: 8px 12px; border: 1px solid #dcdcde;'>VERSION: bump for 2= 6.2.0-rc3 </pre> </li> </ul> <h4 style=3D"margin-top: 10px; margin-bottom: 10px;"> 158 changed files: </h4> <ul> <li class=3D"file-stats"> <a href=3D"#c9b148249d6d20e30e29f5037a9fae3f60b9edc6"> <span class=3D"new-file"> + .pick_status.json </span> </a> </li> <li class=3D"file-stats"> <a href=3D"#b33892fbddeffc3368077b4904ad0b3d562a7312"> VERSION </a> </li> <li class=3D"file-stats"> <a href=3D"#30a9a7b13e862b56ca5ce6226b94d4adf2799d00"> bin/pick/core.py </a> </li> <li class=3D"file-stats"> <a href=3D"#56565a5b717a320525bada24fed720139241b252"> bin/pick/ui.py </a> </li> <li class=3D"file-stats"> <a href=3D"#835e13c1d5068caa0da837d6f4ba70e697ad6afd"> docs/features.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#c3cf206d71203e77a4252c3915daf913c9251dc3"> src/amd/common/ac_gpu_info.c </a> </li> <li class=3D"file-stats"> <a href=3D"#f9639eb777a93e018809f0f73e85c1b71c528403"> src/amd/common/ac_surface.c </a> </li> <li class=3D"file-stats"> <a href=3D"#4e423b746dee4d56f2591bf72f38971ff9f06ff1"> src/amd/common/ac_vcn_dec.c </a> </li> <li class=3D"file-stats"> <a href=3D"#133ebf24bbc29cdf91294a4d988ea71f57256dfa"> src/amd/common/ac_vcn_enc.c </a> </li> <li class=3D"file-stats"> <a href=3D"#3c202f52c7dc216f866acb2d0e853632079c1f31"> src/amd/common/ac_video.c </a> </li> <li class=3D"file-stats"> <a href=3D"#91f9aa804da7e608d079db92de1963580a148aa2"> src/amd/common/amd_family.h </a> </li> <li class=3D"file-stats"> <a href=3D"#f9a4cc86822c48705c64d5ec43a7b89ee0ea699a"> src/amd/common/nir/ac_nir_lower_ngg_mesh.c </a> </li> <li class=3D"file-stats"> <a href=3D"#cce7b9a32ecff216bdb51c0d54cf252372a9fe62"> src/amd/compiler/aco_live_var_analysis.cpp </a> </li> <li class=3D"file-stats"> <a href=3D"#7d12b8df6842a3526888529d01d9e799443895dc"> src/amd/vulkan/bvh/update.comp </a> </li> <li class=3D"file-stats"> <a href=3D"#690fac1e4b74485e7ab8ddeb24cadc53f025e0da"> src/amd/vulkan/nir/radv_nir.h </a> </li> <li class=3D"file-stats"> <a href=3D"#5df9b3fccd9075be57a3bbb8a21584a6777a0b2a"> src/amd/vulkan/nir/radv_nir_lower_abi.c </a> </li> <li class=3D"file-stats"> <a href=3D"#bf47bc13a0ba3d9e3dbbf382d89d052b32beb16b"> src/amd/vulkan/nir/radv_nir_lower_descriptors.c </a> </li> <li class=3D"file-stats"> <a href=3D"#cbfe76490e742e945fde031fe49963b401c5acb4"> src/amd/vulkan/nir/radv_nir_lower_fs_barycentric.c </a> </li> <li class=3D"file-stats"> <a href=3D"#3d515078094413b641294a531f408bffe3be33ac"> src/amd/vulkan/nir/radv_nir_opt_fs_builtins.c </a> </li> <li class=3D"file-stats"> <a href=3D"#64237df6922fd4a9815d75be66ce4d02629ffdb8"> src/amd/vulkan/radv_cmd_buffer.c </a> </li> <li class=3D"file-stats"> <a href=3D"#709e19a703af64174614c44b4627978a829dc8b1"> src/amd/vulkan/radv_cmd_buffer.h </a> </li> <li class=3D"file-stats"> <a href=3D"#50dcbf8c7cfccfbf55aa91e69956cbca1e0d4ea9"> src/amd/vulkan/radv_descriptors.c </a> </li> <li class=3D"file-stats"> <a href=3D"#08ef5c5e61fcf426775b521bc17777985842d4ac"> src/amd/vulkan/radv_device.c </a> </li> <li class=3D"file-stats"> <a href=3D"#e572c777afaf9d0f7b12156d15260c09d391d12e"> src/amd/vulkan/radv_physical_device.c </a> </li> <li class=3D"file-stats"> <a href=3D"#4637426b46e769efd45fb852759628a2b7bf7966"> src/amd/vulkan/radv_pipeline_graphics.c </a> </li> <li class=3D"file-stats"> <a href=3D"#6286d40aeffd5d5b8cdca5a8629a40c1e54f20ed"> src/amd/vulkan/radv_pipeline_graphics.h </a> </li> <li class=3D"file-stats"> <a href=3D"#0f806abc817845d744313f92a86eb33ef0a11771"> src/amd/vulkan/radv_shader.h </a> </li> <li class=3D"file-stats"> <a href=3D"#dd2688a2313251d84a540c53bdcb09764bfd5359"> src/amd/vulkan/radv_shader_args.c </a> </li> <li class=3D"file-stats"> <a href=3D"#c42c9cee35b5ea279662d2ff1237b29f3c135771"> src/amd/vulkan/radv_shader_object.c </a> </li> <li class=3D"file-stats"> <a href=3D"#b7ab6f61deeb790b79afc25b90a99eaaee1e57cb"> src/asahi/ci/asahi-agx2-fails.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#7ae92cd379acaef2b0bb11cd2317112f6c7a901d"> src/asahi/compiler/agx_compile.h </a> </li> <li class=3D"file-stats"> <a href=3D"#24a5172e0fa8b4ed243fddf9440a5aec4ceb3d84"> src/broadcom/vulkan/v3dv_cmd_buffer.c </a> </li> <li class=3D"file-stats"> <a href=3D"#9f73fa64274ca62ff0de8f8ca278aafd41c6b8f8"> src/broadcom/vulkan/v3dv_descriptor_set.h </a> </li> <li class=3D"file-stats"> <a href=3D"#8cdba06709bcd69c6410fc7ef3056110529efd5e"> src/broadcom/vulkan/v3dv_pipeline.c </a> </li> <li class=3D"file-stats"> <a href=3D"#fa866e885ef15f9d66a5f5294962daa430bf52d4"> src/compiler/clc/mesa_clc.c </a> </li> <li class=3D"file-stats"> <a href=3D"#492a808ed079106e690bd104125c95426c652a04"> src/compiler/nir/nir_divergence_analysis.c </a> </li> <li class=3D"file-stats"> <a href=3D"#0773718e597aec9c0dcb3c03eeac450aa43a403f"> src/compiler/nir/nir_lower_explicit_io.c </a> </li> <li class=3D"file-stats"> <a href=3D"#736b255bfc45bba23e92c36f2653bb19628faffc"> src/compiler/nir/nir_lower_io.c </a> </li> <li class=3D"file-stats"> <a href=3D"#4c0bde7db8a893f9760a9bbe97270dc10b9011e7"> src/compiler/nir/nir_opt_copy_prop_vars.c </a> </li> <li class=3D"file-stats"> <a href=3D"#a0a3ad79e369c5793b1cc03b42f47f086d1f19e1"> src/compiler/nir/nir_opt_dead_write_vars.c </a> </li> <li class=3D"file-stats"> <a href=3D"#f18f7570ee394aa1f061761dfeff78b3df614f5b"> src/compiler/nir/nir_opt_vectorize.c </a> </li> <li class=3D"file-stats"> <a href=3D"#4106ea729ce729be1dcfb2fbb6d9e3240cc8190e"> src/compiler/nir/nir_range_analysis.c </a> </li> <li class=3D"file-stats"> <a href=3D"#62175dfb5f26326d7683e479df5a14be034bc3a3"> src/compiler/spirv/vtn_opencl.c </a> </li> <li class=3D"file-stats"> <a href=3D"#154c0ee5e687bd8053b6a89a54a15a6d7faea986"> src/egl/drivers/dri2/egl_dri2.c </a> </li> <li class=3D"file-stats"> <a href=3D"#a7239aa678e1facf0575f8daab9e543d30065e6a"> src/freedreno/fdl/fd6_view.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#974e0f2cd95b59095477dffd72aa23db14c72cd9"> src/freedreno/ir3/ir3.c </a> </li> <li class=3D"file-stats"> <a href=3D"#44e48c438a091ea64c2454f06df33d8bd927b4de"> src/freedreno/ir3/ir3_compiler.c </a> </li> <li class=3D"file-stats"> <a href=3D"#15078fc3d9d553461ea841bb6437e02cd7296beb"> src/freedreno/ir3/ir3_nir.c </a> </li> <li class=3D"file-stats"> <a href=3D"#9651294053804e144323d75e1aeacad0aa70bc03"> src/freedreno/vulkan/tu_acceleration_structure.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#3b83f563f8016965f6289b927ed3a1b55b7e12ae"> src/freedreno/vulkan/tu_clear_blit.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#e15988c46d76555df1fa0582dfbcde3591aa05e7"> src/freedreno/vulkan/tu_clear_blit.h </a> </li> <li class=3D"file-stats"> <a href=3D"#b503e7682197bd8e5d6e2b60204e95dce70774fa"> src/freedreno/vulkan/tu_cmd_buffer.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#b6cff71259eda0152919b9259cd9a0e440b91ea5"> src/freedreno/vulkan/tu_cs.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#097dac82a18487c0cf545cd294b0856cbc725fd8"> src/freedreno/vulkan/tu_device.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#28268b780fd5caf366c1864e267d62e08f4a4748"> src/freedreno/vulkan/tu_formats.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#b3cee820dacba42a4b75e4a6826f30bb487d326f"> src/freedreno/vulkan/tu_knl_kgsl.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#16c3a62f321ce2545302e0081fc59bfa7d234c91"> src/freedreno/vulkan/tu_lrz.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#40cfc9eed299c56f91bc5895fabe3023526afbfc"> src/freedreno/vulkan/tu_shader.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#617d82ec418416754d340f1a24672753c10ca08f"> src/freedreno/vulkan/tu_subsampled_image.cc </a> </li> <li class=3D"file-stats"> <a href=3D"#8c1b702d68d851220869c7c576bd0953399e8ce0"> src/gallium/auxiliary/pipe-loader/driinfo_gallium.h </a> </li> <li class=3D"file-stats"> <a href=3D"#0a6d5812259d6839a4ca3ba298db78f5914d1d91"> src/gallium/auxiliary/util/u_driconf.c </a> </li> <li class=3D"file-stats"> <a href=3D"#ea17dbf82f715ee7ba9ed3c7ca925473db110754"> src/gallium/auxiliary/vl/vl_proc.c </a> </li> <li class=3D"file-stats"> <a href=3D"#c05767e9826b2e16b9e119647e54028b434362b3"> src/gallium/drivers/d3d12/d3d12_video_dec.cpp </a> </li> <li class=3D"file-stats"> <a href=3D"#9053b6db72e9c78ba081107890d1b68972d6a5fe"> src/gallium/drivers/d3d12/d3d12_video_enc.cpp </a> </li> <li class=3D"file-stats"> <a href=3D"#734f7df739164e3800173c5db17e25d14c7552db"> src/gallium/drivers/llvmpipe/ci/llvmpipe-fails.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#7499645dbcbfcc959d61cdefabf59b9f2cfbcf11"> src/gallium/drivers/llvmpipe/lp_state_derived.c </a> </li> <li class=3D"file-stats"> <a href=3D"#d5d9501aebe2bfd18e58beaf13086af0045a79fa"> src/gallium/drivers/nouveau/codegen/nv50_ir_ra.cpp </a> </li> <li class=3D"file-stats"> <a href=3D"#1b7f59f0c132484ed9f8c5d8f287668a0bc4e6c1"> src/gallium/drivers/nouveau/nouveau_mm.c </a> </li> <li class=3D"file-stats"> <a href=3D"#f1d971aa48b7aeefbc5871dc0936eb04ad9a3a23"> src/gallium/drivers/nouveau/nv30/nv30_context.c </a> </li> <li class=3D"file-stats"> <a href=3D"#7886803ab1e07ce72cfcd13e58d8026b8274d82a"> src/gallium/drivers/nouveau/nv30/nv30_state_validate.c </a> </li> <li class=3D"file-stats"> <a href=3D"#d44ba76864a16a5c6a0e3ae6b919a42423c8f404"> src/gallium/drivers/nouveau/nv30/nvfx_fragprog.c </a> </li> <li class=3D"file-stats"> <a href=3D"#13c248a431dce51cd5eab407387eb8f42ee9f3ed"> src/gallium/drivers/nouveau/nv30/nvfx_shader.h </a> </li> <li class=3D"file-stats"> <a href=3D"#af8b4b2403049882711a8488d37cb6ae71749717"> src/gallium/drivers/panfrost/pan_shader.c </a> </li> <li class=3D"file-stats"> <a href=3D"#29f06220b3e341c3b1b5e29d287875b99febb3d9"> src/gallium/drivers/radeonsi/mm/radeon_vcn_enc.c </a> </li> <li class=3D"file-stats"> <a href=3D"#0a15b0b92bf44e267989b19ad0c527543a016b24"> src/gallium/drivers/radeonsi/mm/radeon_vcn_enc.h </a> </li> <li class=3D"file-stats"> <a href=3D"#b9d1d379e4c544f7b8c762998f43535b15724fce"> src/gallium/drivers/radeonsi/si_buffer.c </a> </li> <li class=3D"file-stats"> <a href=3D"#c5f2982df5fe472882e2234484ac9885cc80b82f"> src/gallium/drivers/radeonsi/si_pipe.c </a> </li> <li class=3D"file-stats"> <a href=3D"#68c82a61fc1db2215a1303f63113e0a50250929c"> src/gallium/drivers/radeonsi/si_state.c </a> </li> <li class=3D"file-stats"> <a href=3D"#57324d0bba7696f5c542553dc755eda7663f65b7"> src/gallium/drivers/radeonsi/si_state.h </a> </li> <li class=3D"file-stats"> <a href=3D"#467ad1c12580db62b3edbaecbbe3b8633686e80f"> src/gallium/drivers/v3d/v3d_blit.c </a> </li> <li class=3D"file-stats"> <a href=3D"#99cc7fd5e7c71b435d71cf405bdf8d2137340b8f"> src/gallium/drivers/zink/ci/zink-tu-a750-fails.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#16063c72e3dcf7b37b23a8c987fb3338e36b5f47"> src/gallium/drivers/zink/zink_kopper.c </a> </li> <li class=3D"file-stats"> <a href=3D"#ff3c2bfe77577206ed4af9a8d1407bb22fbfa3c7"> src/gallium/frontends/rusticl/api/memory.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#d872df3d893680588e1f02adf3e51019aeff1561"> src/gallium/frontends/rusticl/builtin_kernels/copy.cl </a> </li> <li class=3D"file-stats"> <a href=3D"#62e9721605cdc567036476d545400c45be8e8c6d"> <span class=3D"new-file"> + src/gallium/frontends/rusticl/builtin_kernels/copy_ulong.cl </span> </a> </li> <li class=3D"file-stats"> <a href=3D"#80c00fea6fdb2b7822884a0cdbe6336bbbb80241"> src/gallium/frontends/rusticl/builtin_kernels/lib.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#2052939fac30c0fdd5fb90f2e38fdfc200bd91cf"> src/gallium/frontends/rusticl/builtin_kernels/meson.build </a> </li> <li class=3D"file-stats"> <a href=3D"#a1078540e2f03c8d7e84020baca833c35733e7c3"> src/gallium/frontends/rusticl/core/device.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#6cd6aebfba37b26d8d235c9d2e0d3569b634849e"> src/gallium/frontends/rusticl/core/kernel.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#329d84d0a0e828ec97159656281d8f42855e52d1"> src/gallium/frontends/rusticl/core/meta.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#7b851b9b9f366f6b8ef28ac351b4a5eeee89993e"> src/gallium/frontends/rusticl/core/program.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#9a64d7388f55cb0ac2ab2d2a3ae2ce75aae2f3fc"> src/gallium/frontends/rusticl/mesa/compiler/nir.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#dcb7537fa23c524a4185f647079ad73723f743e2"> src/gallium/frontends/va/image.c </a> </li> <li class=3D"file-stats"> <a href=3D"#50334bfbacbb7eba2b7bedaf06c36473971838fc"> src/gallium/include/frontend/api.h </a> </li> <li class=3D"file-stats"> <a href=3D"#cb0a755e0ba29bda66451c9e1804a10f145cc27d"> src/glx/glxext.c </a> </li> <li class=3D"file-stats"> <a href=3D"#2556094f30e34ce31abace0d07de249214beec98"> src/imagination/vulkan/pvr_arch_border.c </a> </li> <li class=3D"file-stats"> <a href=3D"#7762149dec4a5872b267adca019552413adafae5"> src/imagination/vulkan/pvr_arch_cmd_buffer.c </a> </li> <li class=3D"file-stats"> <a href=3D"#d893e17661870d200ce384b3a662b797a9124d78"> src/imagination/vulkan/pvr_arch_queue.c </a> </li> <li class=3D"file-stats"> <a href=3D"#aabd329ffe24a178e05d98a3fc6a3e9e7238d29a"> src/imagination/vulkan/pvr_job_render.h </a> </li> <li class=3D"file-stats"> <a href=3D"#3299ee10437635a4e82c5201b43cddce1193e5bb"> src/intel/compiler/brw/brw_compile_fs.cpp </a> </li> <li class=3D"file-stats"> <a href=3D"#1547a1debc5ac8af7124ff8cc1549531c059a569"> src/intel/compiler/brw/brw_nir.c </a> </li> <li class=3D"file-stats"> <a href=3D"#6dfa8d4797d65672ca922019127c0ab1ae983e2d"> src/intel/compiler/brw/brw_opt_fill_spill.cpp </a> </li> <li class=3D"file-stats"> <a href=3D"#81817b80d9a6ad8e2e8ebb60ac6e854e87f9d521"> src/intel/compiler/gen/gen_encoding_pre_xe.cpp </a> </li> <li class=3D"file-stats"> <a href=3D"#e4da400ad60b1d7690d71fb29048f4e71edb67ee"> src/intel/dev/intel_debug.c </a> </li> <li class=3D"file-stats"> <a href=3D"#549725679535e2023f7ec12033ea49a4eb3ce618"> src/intel/dev/intel_device_info.c </a> </li> <li class=3D"file-stats"> <a href=3D"#f9e4bdb6dd2c345a66b81daccc6af2c5cef5c730"> src/intel/vulkan/00-anv-defaults.conf </a> </li> <li class=3D"file-stats"> <a href=3D"#bf772703006afde56a422e65daa0f3bae13991cc"> src/intel/vulkan/anv_blorp.c </a> </li> <li class=3D"file-stats"> <a href=3D"#4dd705504acb63c3656e460113031eac02ac08c9"> src/intel/vulkan/anv_cmd_buffer.c </a> </li> <li class=3D"file-stats"> <a href=3D"#4e2cbd75e6c0a1b3b14532fd15a8c66c91baa342"> src/intel/vulkan/anv_shader.c </a> </li> <li class=3D"file-stats"> <a href=3D"#3ee26a23f68c63cb439fa38853753e0d22f497ee"> src/intel/vulkan/anv_shader_compile.c </a> </li> <li class=3D"file-stats"> <a href=3D"#ea45fe239223e26e3554af9807478538f9e8b1df"> src/intel/vulkan/genX_cmd_buffer.c </a> </li> <li class=3D"file-stats"> <a href=3D"#5c096f19c7d00cb7e7acdb9146a6712afad09074"> src/intel/vulkan/genX_query.c </a> </li> <li class=3D"file-stats"> <a href=3D"#3835366103ffe7cf94bef8bf4827445051732f97"> src/kosmickrisp/vulkan/kk_query_pool.c </a> </li> <li class=3D"file-stats"> <a href=3D"#41a96cd943349cee87f836f6db1b21ca0cf047aa"> src/mesa/main/bufferobj.c </a> </li> <li class=3D"file-stats"> <a href=3D"#4a7ec9c9d0d42e6f87af57f030b5a6efe3e3af51"> src/mesa/state_tracker/st_cb_texture.c </a> </li> <li class=3D"file-stats"> <a href=3D"#a558c409f3ed53f514d1c46703baac2c4dd2d150"> src/nouveau/compiler/nak_nir_lower_mesh_intrinsics.c </a> </li> <li class=3D"file-stats"> <a href=3D"#bb34f922e3603e4d81064fefe1a45d4aceb919e6"> src/nouveau/vulkan/nvk_cmd_buffer.c </a> </li> <li class=3D"file-stats"> <a href=3D"#b5c1baf2f895db9b98e2dc1e593ba491fd913edc"> src/nouveau/vulkan/nvk_shader.c </a> </li> <li class=3D"file-stats"> <a href=3D"#519d819ef3fda1c7af7ce3bd281a1a79cb51f656"> src/panfrost/ci/panfrost-g52-fails.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#1ba6c005f63e5e961bf800f0a66e0043dc3418de"> src/panfrost/ci/panfrost-g57-fails.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#f6b32520d74b71d1d2f1a1ae3058c6f5db1d9225"> src/panfrost/ci/panfrost-g57-flakes.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#1380b85fa1188f6dffc93937f45ad16c1d5e3dbd"> src/panfrost/ci/panfrost-g610-fails.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#d3ef7915d6c6398339e2a2889b0e72f7ebd95903"> src/panfrost/ci/panfrost-g610-flakes.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#28cb047cec71c42c6916412c713b428eae5705f7"> src/panfrost/ci/panfrost-g610-skips.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#098363f62adf0f879228572b5026468f712782e4"> src/panfrost/ci/panfrost-g925-fails.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#c1a585cd4389ab2bf92ed4d64aca0498a7828197"> src/panfrost/ci/panfrost-g925-flakes.txt </a> </li> <li class=3D"file-stats"> <a href=3D"#dfb17254c18687fe8fc20fa05a729c4d196edbac"> src/panfrost/compiler/bifrost/bifrost_compile.c </a> </li> <li class=3D"file-stats"> <a href=3D"#bd885bcfa5c791fcd69687a54144b8fd367adb57"> src/panfrost/compiler/kraid/hw_runner/device.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#e654d45672ed862dadcab8973936d6dddce30ae5"> src/panfrost/compiler/kraid/hw_runner/meson.build </a> </li> <li class=3D"file-stats"> <a href=3D"#9ce736bd2177824650db50d57bb09187a3955054"> src/panfrost/compiler/midgard/midgard_compile.c </a> </li> <li class=3D"file-stats"> <a href=3D"#5b389492ca6ec11b171eb06813666cb09c79af31"> src/panfrost/perf/quick.c </a> </li> <li class=3D"file-stats"> <a href=3D"#6f5e1f87d90a96c0197a977d759d390e10146c28"> src/panfrost/vulkan/panvk_vX_physical_device.c </a> </li> <li class=3D"file-stats"> <a href=3D"#9489c9ba7c9c6e53fb67125374e5c3ff3d141270"> src/util/00-mesa-defaults.conf </a> </li> <li class=3D"file-stats"> <a href=3D"#fc5bac2c2e46afe47410836e1a5ae72be4a95ccd"> src/util/driconf.h </a> </li> <li class=3D"file-stats"> <a href=3D"#48ac0cf1296119f074a799a825f5e71360c48fc8"> src/virtio/magma-gpu-rs/ffi/virtgpu_kumquat_ffi/lib.rs </a> </li> <li class=3D"file-stats"> <a href=3D"#99e805ee5fff1a768a555d940ed05f06da0b8a63"> src/virtio/vulkan/vn_common.c </a> </li> <li class=3D"file-stats"> <a href=3D"#32545674dabf8557fdd8b60042fc6ee974b44adf"> src/virtio/vulkan/vn_common.h </a> </li> <li class=3D"file-stats"> <a href=3D"#3890ed50e774a0226795b2c865c51b70c812b088"> src/virtio/vulkan/vn_device.c </a> </li> <li class=3D"file-stats"> <a href=3D"#5b8b27fd97d24a0159dc99b6181996cacb4daa5a"> src/virtio/vulkan/vn_device.h </a> </li> <li class=3D"file-stats"> <a href=3D"#dbb9dd7a5d8f35f6c220023d020e37d33fb35eb1"> src/virtio/vulkan/vn_feedback.c </a> </li> <li class=3D"file-stats"> <a href=3D"#a9f511f5f6caac1f6c79729be5ad71c753359481"> src/virtio/vulkan/vn_feedback.h </a> </li> <li class=3D"file-stats"> <a href=3D"#1d3bfd1460c02699dc488306c9fecc22524510f0"> src/virtio/vulkan/vn_host_copy.c </a> </li> <li class=3D"file-stats"> <a href=3D"#24b35b9bb1b7ae97fce5c0afb49fe0d4f247900a"> src/virtio/vulkan/vn_physical_device.c </a> </li> <li class=3D"file-stats"> <a href=3D"#ce146219441ae39e82d451a2d5e94e12f2b3d569"> src/virtio/vulkan/vn_physical_device.h </a> </li> <li class=3D"file-stats"> <a href=3D"#7dfe394ffde0617fb0fe7ab7b2e0feec0ffe9adb"> src/virtio/vulkan/vn_queue.c </a> </li> <li class=3D"file-stats"> <a href=3D"#399b36afcc7b4ea6ba45f8942cd50f7726f7e9f6"> src/virtio/vulkan/vn_queue.h </a> </li> <li class=3D"file-stats"> <a href=3D"#5f8bf1021c66e5993b67ff3938d12876fa788ff2"> src/virtio/vulkan/vn_renderer.h </a> </li> <li class=3D"file-stats"> <a href=3D"#31fe7ebb0c68cbc2c4be3922f12c564da1823159"> src/virtio/vulkan/vn_renderer_sim_syncobj.c </a> </li> <li class=3D"file-stats"> <a href=3D"#6613153782afd6a10fb4d5bc4adef96af5acfa7c"> src/virtio/vulkan/vn_renderer_sim_syncobj.h </a> </li> <li class=3D"file-stats"> <a href=3D"#c183ce47c0f29e8658a60eda13a65bcdcaf4fa1a"> src/virtio/vulkan/vn_renderer_util.c </a> </li> <li class=3D"file-stats"> <a href=3D"#eecbd99c0d4e0fbfc97ff56d99dcde6202fbbb1b"> src/virtio/vulkan/vn_renderer_virtgpu.c </a> </li> <li class=3D"file-stats"> <a href=3D"#c4c93f41c4df8b94b04e9329c8406b2e5529cf44"> src/virtio/vulkan/vn_renderer_vtest.c </a> </li> <li class=3D"file-stats"> <a href=3D"#250f391f20dddff910ee9cbe57feca14d0cd5c14"> src/virtio/vulkan/vn_ring.c </a> </li> <li class=3D"file-stats"> <a href=3D"#f8a2773779f012a2e5c7070d3b54c3a79eb6c6fe"> src/virtio/vulkan/vn_sync.c </a> </li> <li class=3D"file-stats"> <a href=3D"#91be432f2e87d029d0743031f9cbac20944ea78f"> src/virtio/vulkan/vn_sync.h </a> </li> <li class=3D"file-stats"> <a href=3D"#716e94477865f663d8648ee93e49310dd6802ab7"> src/vulkan/runtime/vk_pipeline.c </a> </li> <li class=3D"file-stats"> <a href=3D"#5801d2e810f777506e404cdb1f968a760784a836"> src/vulkan/runtime/vk_video.c </a> </li> <li class=3D"file-stats"> <a href=3D"#324282c78dfdb597289142f354638cffe048e32a"> src/vulkan/runtime/vk_video.h </a> </li> </ul> <h5 style=3D"margin-top: 10px; margin-bottom: 10px; font-size: .875rem;">= The diff was not included because it is too large. </h5> </div> <div class=3D"footer" style=3D"margin-top: 10px;"> <p style=3D"font-size: small; color: #626168;"> =E2=80=94 <br> <a href=3D"https://salsa.debian.org/xorg-team/lib/mesa/-/compare/57017725= 151dc7e96ba82eeedc35baf1fc13911f...5db7111ffb9f932815baf7b299cd4db772e6d3= 01">View it on GitLab</a>. <br> You're receiving this email because of your account on <a target=3D"_blan= k" rel=3D"noopener noreferrer" href=3D"https://salsa.debian.org">salsa.de= bian.org</a>. <a href=3D"https://salsa.debian.org/-/profile/notifications= " target=3D"_blank" rel=3D"noopener noreferrer" class=3D"mng-notif-link">= Manage all notifications</a> =C2=B7 <a href=3D"https://salsa.debian.org/h= elp" target=3D"_blank" rel=3D"noopener noreferrer" class=3D"help-link">He= lp</a> <span style=3D"color: transparent; font-size: 0; display: none; overflow:= hidden; opacity: 0; width: 0; height: 0; max-width: 0; max-height: 0;"> Notification message regarding https://salsa.debian.org/xorg-team/lib/mes= a/-/compare/57017725151dc7e96ba82eeedc35baf1fc13911f...5db7111ffb9f932815= baf7b299cd4db772e6d301 at 1785770298 </span> </p> </div> </body> </html> ----==_mimepart_6a70b13ab73b_5515b98ebc885935--