[ANNOUNCE] mesa 24.0.5

Eric Engestrom eric at engestrom.ch
Wed Apr 10 20:33:30 UTC 2024

Hello everyone,

The bugfix release 24.0.5 is now available.

If you find any issues, please report them here:

The next bugfix release is due in two weeks, on April 24th.



Axel Davy (5):
      frontend/nine: Fix ff ps key
      frontend/nine: Fix programmable vs check
      frontend/nine: Fix missing light flag check
      frontend/nine: Fix destruction race
      frontend/nine: Reset should EndScene

Connor Abbott (2):
      freedreno/a7xx: Add CP_CCHE_INVALIDATE
      tu: Implement CCHE invalidation

Dave Airlie (1):
      mesa: reorder st context teardown

David Heidelberg (7):
      r300: add missing licence to the r300_public.h
      r300: add missing copyright header
      docs: we support EGL 1.5 for a long time
      r600: add license header to r600_formats.h
      r600: add license info to the r600_opcodes.h
      r600: add license information to the sfn_shader_gs.h

David Stern (1):
      vulkan/wsi/x11: Explicitly discard errors from xcb_present_pixmap.

Eric Engestrom (7):
      docs: add sha256sum for 24.0.4
      .pick_status.json: Update to 3d68dd78d07b30cefe90d76af681075f4ed6b33d
      .pick_status.json: Update to fcb568a5d5a52db75fa2f6d04579bb404ca7f597
      .pick_status.json: Update to 078fe5454e97d073feb18bcdcf7ed1874e8b4835
      .pick_status.json: Update to 2c1cb65949933a05eedb2eacc15cd893ecaef8aa
      docs: add release notes for 24.0.5
      VERSION: bump for 24.0.5

Eric R. Smith (2):
      panfrost: mark indirect compute buffer as read
      gallium: handle copy_image of depth textures

Faith Ekstrand (2):
      nvk: Add a _pad field to nvk_cbuf
      nvk: Add a _pad field to nvk_fs_key

Georg Lehmann (2):
      aco: don't combine mul+add_clamp to mad_clamp
      aco/ra: use SDWA for 16bit instructions when the second byte is blocked

Iago Toral Quiroga (2):
      v3d: implement fix for GFXH-1602
      broadcom/compiler: fix workaround for GFXH-1602

Ian Romanick (3):
      intel/brw: Clear write_accumulator flag when changing the destination
      intel/brw: Use enums for DPAS source regioning
      nir: intel/brw: Change the order of sources for nir_dpas_intel

Jesse Natalie (1):
      glsl: Use a stable attr sort for VS in / FS out

Jordan Justen (1):
      intel/dev: Add 0x56be and 0x56bf DG2 PCI IDs

José Roberto de Souza (4):
      anv: Fix calculation of syncs required in Xe KMD
      iris: Wait for drm_xe_exec_queue to be idle before destroying it
      anv: Create protected engine context when i915 supports vm control
      intel: Enable Xe KMD support by default

Juston Li (1):
      Revert "zink: store last pipeline directly for zink_gfx_program::last_pipeline"

Karol Herbst (1):
      meson: fix link failure with llvm-18

Kenneth Graunke (2):
      intel/brw: Fix generate_mov_indirect to check has_64bit_int not float
      intel/brw: Fix lower_regioning for BROADCAST, MOV_INDIRECT on Q types

Konstantin Seurer (1):
      nir/serialize: Encode data for temporaries

Lionel Landwerlin (7):
      anv: fix protected memory allocations
      anv: disable protected content around surface state copies
      anv: disable generated draws in protected command buffers
      anv: update protection fault property
      anv: add missing data flush out of L3 for transform feedback writes
      anv: mark descriptors & pipeline dirty after blorp compute
      isl: set NullPageCoherencyEnable for depth/stencil sparse surfaces

Lucas Stach (2):
      etnaviv: fix depth writes without testing
      etnaviv: rs: take src dimensions into account when increasing height alignment

Mike Blumenkrantz (12):
      zink: only check that CUBE_COMPATIBLE for images doesn't subtract flags
      zink: don't use set_foreach_remove with dmabuf_exports
      zink: make descriptor pool creation more robust
      zink: fix shaderdb pipeline compile
      zink: don't clobber indirect array reads with missing components
      zink: fix add_derefs case for compact arrays
      llvmpipe: fix DRAW_USE_LLVM=0
      glsl: handle xfb resources for spirv before running varying opts
      mesa: clamp binary pointer in ShaderBinary if length==0
      glsl: set PSIZ bit in outputs_written when injecting a 1.0 psiz write
      nir/lower_clamp_color_outputs: fix use with lowered io
      nir/texcoord_replace: fix scalarized io handling

Nikita Popov (1):
      Pass no-verify-fixpoint option to instcombine in LLVM 18

Patrick Lerda (1):
      r300: fix constants_remap_table memory leak related to the dummy shader path

Paul Gofman (3):
      glsl: allow out arrays in #110 with allow_glsl_120_subset_in_110
      driconf: add a workaround for Joe Danger 2
      driconf: add a workaround for Joe Danger

Paulo Zanoni (2):
      anv/xe: don't leak xe_syncs during trtt submission
      anv, iris: add missing CS_STALL bit for GPGPU texture invalidation

Samuel Pitoiset (3):
      radv: fix conditional rendering with mesh+task and multiview (again)
      radv: enable radv_zero_vram for Red Dead Redemption 2
      radv: make sure the heap budget is less than or equal to the heap size

Tapani Pälli (1):
      anv: disable fcv optimization on >= gfx125

Yonggang Luo (1):
      util: Fixes futex_wait on win32

Zack Rusin (1):
      svga: Fix instanced draw detection

Zan Dobersek (1):
      tu: fix memory leaks in tu_shader

git tag: mesa-24.0.5

SHA256: 38cc245ca8faa3c69da6d2687f8906377001f63365348a62cc6f7fafb1e8c018  mesa-24.0.5.tar.xz
SHA512: 9476af4b9ac8db5dce397084ef169927d10b28adea7e74aa5b3136810b499ac98ddf7ab564e7d1ff81c887208c8ebab3ad2d4e27e7f46136609b5c67527018eb  mesa-24.0.5.tar.xz
PGP:  https://mesa.freedesktop.org/archive/mesa-24.0.5.tar.xz.sig

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20240410/9f0c340e/attachment.sig>

More information about the mesa-dev mailing list