Mesa 23.1.5 Release Notes / 2023-08-02

Mesa 23.1.5 is a bug fix release which fixes bugs found since the 23.1.4 release.

Mesa 23.1.5 implements the OpenGL 4.6 API, but the version reported by glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) / glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used. Some drivers don’t support all the features required in OpenGL 4.6. OpenGL 4.6 is only available if requested at context creation. Compatibility contexts may report a lower version depending on each driver.

Mesa 23.1.5 implements the Vulkan 1.3 API, but the version reported by the apiVersion property of the VkPhysicalDeviceProperties struct depends on the particular driver being used.

SHA256 checksum

3cf88576fdebf24fc4047067936131c90cb6541c27365996b79b661dec1fb153  mesa-23.1.5.tar.xz

New features

  • None

Bug fixes

  • Baldurs Gate 3 (DX11) - Graphical corruption on RDNA3 (ACO regression)

  • intel: Deathloop and other DX12 games fail assert(validated) with invalid SEL instruction

Changes

Alex Denes (1):

  • virgl: link VA driver with build-id

Alyssa Rosenzweig (3):

  • panfrost: Lower vertex_id for XFB

  • panfrost: Fix transform feedback on v9 harder

  • nir/lower_blend: Fix 32-bit logicops

Bas Nieuwenhuizen (2):

  • aco: fix nir_op_vec8/16 with 16-bit elements.

  • aco: Fix some constant patterns in 16-bit vec4 construction with s_pack.

Connor Abbott (1):

  • tu: Fix vk2tu_*_stage flag type

Daniel Stone (1):

  • egl/wayland: Always initialise fd_display_gpu

Danylo Piliaiev (3):

  • freedreno/regs: Rename SP_FS_CTRL_REG0.DIFF_FINE into LODPIXMASK

  • ir3: Fix FS quad ops returning wrong values from helper invocations

  • tu,freedreno: Forbid blit event for R8G8_SRGB due to gpu faults

Dave Airlie (1):

  • gallivm: fix atomic global temporary storage.

Eric Engestrom (4):

  • docs: add sha256sum for 23.1.4

  • .pick_status.json: Update to c75bf084d0b0ee4d4ba1d74ba5a7486a07ba8334

  • .pick_status.json: Update to 2b7599dc495669a9f935dcb80e5b62a100a9b788

  • .pick_status.json: Update to 8d7f682bdbaee2cd7185203770bfc3f0e07ee427

Georg Lehmann (1):

  • aco/gfx11: fix get_gfx11_true16_mask with v_cmp_class_f16

Karol Herbst (2):

  • nv50/ir/nir: fix txq emission on MS textures

  • nv50/ir/nir: Fix zero source handling of tex instructions.

Konstantin Seurer (2):

  • gallivm: Fix atomic_global types

  • draw: Do not restart the primitive_id at 0

Lionel Landwerlin (1):

  • intel/fs: fix missing predicate on SEL instruction

Mike Blumenkrantz (7):

  • zink: be even dumber about buffer refs when replacing storage

  • zink: emit SpvCapabilitySampleMaskPostDepthCoverage with SpvExecutionModePostDepthCoverage

  • kopper: handle pixmap creation failure more gracefully

  • glxsw: check geometry of drawables on creation

  • zink: don’t clobber descriptor mode on multiple screen creation

  • nir: fix slot calculations for compact variables with location_frac

  • lavapipe: use the component offset directly for xfb

Nanley Chery (2):

  • anv: Don’t support ASTC images with modifiers

  • intel/blorp: Ambiguate after CCS resolves on gfx7-8

Rhys Perry (4):

  • radv: fix radv_get_ballot_bit_size with CS

  • nir/opt_dead_cf: remove nodes after a jump earlier

  • aco: insert s_nop before VGPR deallocation

  • ac/llvm: fix AC_TM_CHECK_IR

Robert Mader (1):

  • egl/wayland: wait for compositor to release shm buffers

Yiwei Zhang (3):

  • turnip: flush cache for dstBuffer in vkCmdCopyQueryPoolResults

  • venus: fix a cmd builder render_pass state leak across reset

  • venus: fix cmd state leak across implicit reset