xref: /aosp_15_r20/external/mesa3d/docs/relnotes/24.0.2.rst (revision 6104692788411f58d303aa86923a9ff6ecaded22)
1Mesa 24.0.2 Release Notes / 2024-02-28
2======================================
3
4Mesa 24.0.2 is a bug fix release which fixes bugs found since the 24.0.1 release.
5
6Mesa 24.0.2 implements the OpenGL 4.6 API, but the version reported by
7glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
8glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used.
9Some drivers don't support all the features required in OpenGL 4.6. OpenGL
104.6 is **only** available if requested at context creation.
11Compatibility contexts may report a lower version depending on each driver.
12
13Mesa 24.0.2 implements the Vulkan 1.3 API, but the version reported by
14the apiVersion property of the VkPhysicalDeviceProperties struct
15depends on the particular driver being used.
16
17SHA256 checksum
18---------------
19
20::
21
22    94e28a8edad06d8ed2b83eb53f253b9eb5aa62c3080f939702e1b3039b56c9e8  mesa-24.0.2.tar.xz
23
24
25New features
26------------
27
28- None
29
30
31Bug fixes
32---------
33
34- KHR-Single-GL46.arrays_of_arrays_gl.AtomicUsage fails on MTL
35- GTF-GL46.gtf42.GL3Tests.texture_storage.texture_storage_texture_as_framebuffer_attachment fails on MTL
36- [intel][anv][build][regression] - genX_grl.h:27:10: fatal error: grl/grl_cl_kernel.h: No such file or directory
37- RX 6600 VDPAU not recognizing HEVC_MAIN_10 correctly
38- Running an app on another AMD GPU (offload, DRI_PRIME) produces corrupted frames on Wayland.
39- VDPAU declares a texture as "immutable" without also setting its ImmutableLevels attribute.
40- RX6600 hardware HEVC video decode fails for VDPAU but works for VA-API. (Can lock up GPU!)
41- Rusticl panics when getting program build logs using opencl.hpp
42- ue5 game issues lighting Rog Ally 7080u (z1e)
43- Missing textures in RoboCop: Rogue City with mesh shaders enabled
44- radv: Multiview PSO forgets to export layer in some cases.
45- zink: flickering artifacts in Selaco
46
47
48Changes
49-------
50
51Boyuan Zhang (1):
52
53- radeonsi/vcn: only use multi slices reflist when available
54
55Chia-I Wu (1):
56
57- radv: fix pipeline stats mask
58
59Chris Rankin (2):
60
61- vdpau: Declare texture object as immutable using helper function.
62- vdpau: Refactor query for video surface formats.
63
64Connor Abbott (1):
65
66- tu: Follow pipeline compatibility rules for dynamic descriptors
67
68Daniel Schürmann (1):
69
70- spirv: Fix SpvOpExpectKHR
71
72Daniel Stone (2):
73
74- egl/wayland: Add opaque-equivalent FourCCs
75- egl/wayland: Fix EGL_EXT_present_opaque
76
77Dave Airlie (2):
78
79- nouveau/winsys: fix bda heap leak.
80- nvk: fix dri options leak.
81
82David Rosca (1):
83
84- frontends/va: Only set VP9 segmentation fields when segmentation is enabled
85
86Eric Engestrom (10):
87
88- docs: add sha256sum for 24.0.1
89- [24.0-only change] ci: increase the kernel+rootfs builds timeout to 2h
90- .pick_status.json: Update to c6e855b64b9015235462959b2b7f3e9fc34b2f1f
91- .pick_status.json: Update to dce20690542c84ac00509a6db7902dcfc90b25bb
92- .pick_status.json: Update to c12300844d3f084ca011a3f54f0cbaa9807418f0
93- .pick_status.json: Mark 3b927567ac927316eb11901f50ee1573ead44fd2 as denominated
94- .pick_status.json: Update to 423add61e2d5b6ab6b5505d1feec01b93609f8fc
95- .pick_status.json: Update to 4071c399a27932ea9253eb8a65d5725504bac6f3
96- .pick_status.json: Update to 82ff9204abab5267f82a9ce73f9dca1541ef5ee6
97- [24.0 only] disable clang-format
98
99Erik Faye-Lund (1):
100
101- mesa/main: allow GL_BGRA for FBOs
102
103Faith Ekstrand (1):
104
105- nvk: Invalidate the texture cache before MSAA resolves
106
107Hans-Kristian Arntzen (1):
108
109- radv: export multiview in VS/TES/GS for depth-only rendering
110
111Iago Toral Quiroga (1):
112
113- v3d,v3dv: fix BO allocation for shared vars
114
115Ian Romanick (1):
116
117- nir: Mark nir_intrinsic_load_global_block_intel as divergent
118
119Jesse Natalie (1):
120
121- dzn: Don't set view instancing mask until after the PSO
122
123Jordan Justen (1):
124
125- intel/dev: Add 2 additional ADL-N PCI ids
126
127Juston Li (1):
128
129- venus: fix image reqs cache store locking
130
131Karol Herbst (3):
132
133- zink: lower unaligned memory accesses
134- rusticl/program: fix CL_PROGRAM_BINARIES for devs with no builds
135- meson: do not pull in clc for clover
136
137Konstantin Seurer (5):
138
139- zink: Always set mfence->submit_count to the fence submit_count
140- Revert "zink: always force flushes when originating from api frontend"
141- llvmpipe: Use full subgroups when possible
142- gallivm: Consider the initial mask when terminating loops
143- ci: Update llvmpipe trace checksums
144
145Lionel Landwerlin (8):
146
147- vulkan/runtime: add helper to query attachment layout
148- anv: fixup push descriptor shader analysis
149- anv: reenable ANV_ALWAYS_BINDLESS
150- anv: fix Wa_16013994831 macros
151- anv: disable Wa_16013994831
152- intel/nir: only consider ray query variables in lowering
153- anv: limit depth flush on dynamic render pass suspend
154- anv: add missing generated file dep
155
156Martin Roukala (né Peres) (1):
157
158- radv/ci: switch vkcts-polaris10 from mupuf to KWS' farm
159
160Michel Dänzer (1):
161
162- egl/wayland: Flush after blitting to linear copy
163
164Mike Blumenkrantz (25):
165
166- zink: prune dmabuf export tracking when adding resource binds
167- zink: fix sparse bo placement
168- zink: zero allocate resident_defs array in ntv
169- zink: move sparse lowering up in file
170- zink: run sparse lowering after all optimization passes
171- zink: adjust swizzled deref loads by the variable component offset
172- zink: clamp zink_gfx_lib_cache::stages_present for generated tcs
173- zink: promote gpl libs freeing during shader destroy out of prog loop
174- zink: don't add VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT for sparse textures
175- zink: delete maxDescriptorBufferBindings checks
176- zink: avoid infinite recursion on (very) small BAR systems in bo alloc
177- zink: add checks/compat for low-spec descriptor buffer implementations
178- zink: add a second fence disambiguation case
179- zink: force host-visible allocations for MAP_COHERENT resources
180- zink: handle stencil_fallback in zink_clear_depth_stencil
181- zink: don't destroy the current batch state on context destroy
182- mesa: check driver format support for certain GetInternalformativ queries
183- vk/wsi/x11/sw: use swapchain depth for putimage
184- zink: only scan active batch states for free states if > 1 exist
185- zink: fix longstanding issue with active batch state recycling
186- zink: assert that batch_id is valid in zink_screen_check_last_finished()
187- zink: clamp in_rp clears to fb size
188- zink: fix (dynamic rendering) execution of scissored clears during flush
189- zink: lock buffer age when chundering swapchain for readback
190- zink: flag acquired swapchain image as readback target on acquire, not present
191
192Patrick Lerda (3):
193
194- r300: fix vertex_buffer related refcnt imbalance
195- r300: fix r300_destroy_context() related memory leaks
196- r300: fix memory leaks when register allocation fails
197
198Pavel Ondračka (1):
199
200- r300: add explicit flrp lowering
201
202Rhys Perry (2):
203
204- aco/ra: don't initialize assigned in initializer list
205- aco/ra: fix GFX9- writelane
206
207Sagar Ghuge (1):
208
209- nir: Allow nir_texop_tg4 in implicit derivative
210
211Samuel Pitoiset (4):
212
213- radv: fix RGP barrier reason for RP barriers inserted by the runtime
214- radv: enable GS_FAST_LAUNCH=2 by default for RDNA3 APUs (Phoenix)
215- spirv: only consider IO variables when adjusting patch locations for TES
216- radv: fix indirect dispatches on compute queue with conditional rendering on GFX7
217
218Tapani Pälli (2):
219
220- intel/blorp: disable use of REP16 independent of format
221- iris: make sure DS and TE are sent in pairs on >= gfx125
222
223Yiwei Zhang (2):
224
225- venus: force async pipeline create on threads creating descriptor pools
226- venus: fix the cmd stride used for qfb recording
227
228thfrwn (1):
229
230- mesa: fix off-by-one for newblock allocation in dlist_alloc
231