Build #81 (Oct 7, 2020, 6:57:05 PM)

1. intel/fs: Use a single untyped surface read for load_num_work_groups (details / githubweb)
2. intel/nir: Lower load_num_work_groups to 32-bit if needed (details / githubweb)
3. iris: Re-emit push constants if we have a varying workgroup size (details / githubweb)
4. docs: add release notes for 20.1.7 (details / githubweb)
5. docs: update calendar and link releases notes for 20.1.7 (details / githubweb)
6. docs: shift 20.2 rc dates by two weeks to match reality (details / githubweb)
7. freedreno: Fix missing rsc->seqno updates (details / githubweb)
8. freedreno: fence_server_sync() fixes (details / githubweb)
9. radeonsi: don't crash if input_usage_mask is 0 for a VS input (details / githubweb)
10. radeonsi: get color interpolation info from shader_info (details / githubweb)
11. radeonsi: clean up code for loading VS inputs (details / githubweb)
12. ac/nir: handle all lowered IO intrinsics (details / githubweb)
13. radeonsi: lower IO intrinsics - complete rewrite of input/output scanning (details / githubweb)
14. radeonsi: remove in/out/uniform variables from NIR after lowering IO (details / githubweb)
15. radeonsi: don't lower indirect IO in GLSL (details / githubweb)
16. radeonsi: don't execute LDS stores for TCS outputs that are never read (details / githubweb)
17. radeonsi: simplify handling color interp modes in si_emit_spi_map (details / githubweb)
18. radeonsi: change PIPE_SHADER to MESA_SHADER (si_shader_selector::type) (details / githubweb)
19. radeonsi: change PIPE_SHADER to MESA_SHADER (si_shader_context::type) (details / githubweb)
20. radeonsi: change PIPE_SHADER to MESA_SHADER (debug flags) (details / githubweb)
21. radeonsi: change PIPE_SHADER to MESA_SHADER (si_compile_llvm) (details / githubweb)
22. radeonsi: change PIPE_SHADER to MESA_SHADER (si_get_shader_part) (details / githubweb)
23. radeonsi: remove unused si_shader_context::type (details / githubweb)
24. radeonsi: change PIPE_SHADER to MESA_SHADER (si_shader_dump_disassembly) (details / githubweb)
25. radeonsi: precompute si_*_descriptors_idx in si_shader_selector (details / githubweb)
26. radeonsi: change PIPE_SHADER to MESA_SHADER (si_dump_descriptors) (details / githubweb)
27. radeonsi: remove si_shader_selector::type (details / githubweb)
28. compiler: add INTERP_MODE_COLOR for radeonsi (details / githubweb)
29. radeonsi: replace TGSI_INTERPOLATE with INTERP_MODE (details / githubweb)
30. radeonsi: replace TGSI_SEMANTIC with VARYING_SLOT and FRAG_RESULT (details / githubweb)
31. radeonsi: optimize out the loop in si_get_ps_input_cntl (details / githubweb)
32. gallium/util: use uint sampler for stencil-reads (details / githubweb)
33. zink: generically handle matrix types (details / githubweb)
34. intel/compiler: Handle all indirect lowering choices in brw_nir.c (details / githubweb)
35. nir/lower_indirect_derefs: Add a threshold (details / githubweb)
36. intel/nir: Stop using nir_lower_vars_to_scratch (details / githubweb)
37. nir: Don't bail too early in lower_mem_constant_vars (details / githubweb)
38. clover: Call nir_lower_mem_constant_vars (details / githubweb)
39. spirv: Don't accept CPacked decoration on struct members (details / githubweb)
40. spirv: Propagate packed information to glsl_type (details / githubweb)
41. glsl: Propagate packed info in get_explicit_type_for_size_align() (details / githubweb)
42. nir/glsl: Consider block interfaces as structs when it comes to size/align calculation (details / githubweb)
43. nir: Expose the packed attribute attached to glsl_type objects (details / githubweb)
44. compiler/types: Make booleans 32-bit for cl_size/align (details / githubweb)
45. nir/glsl: Add an explicit_alignment field to glsl_type (details / githubweb)
46. nir: Add alignment information to cast derefs (details / githubweb)
47. nir: Handle all array stride cases in nir_deref_instr_array_stride (details / githubweb)
48. nir: Add a helper for getting the alignment of a deref (details / githubweb)
49. nir/lower_io: Apply alignments from derefs when available (details / githubweb)
50. nir/opt_deref: Don't remove casts with alignment information (details / githubweb)
51. nir/opt_deref: Remove restrictive alignment information from casts (details / githubweb)
52. spirv: Add pointer helper vars to OpCopyMemory (details / githubweb)
53. spirv: Propagate alignments to deref chains via casts (details / githubweb)
54. nir: Allow var_mem_global in nir_lower_vars_to_explicit_types (details / githubweb)
55. nir: Allow uniform in nir_lower_vars_to_explicit_types (details / githubweb)
56. clover: Use args.size() to compute new var locations (details / githubweb)
57. spirv: Stop counting inputs in entry_point_wrapper (details / githubweb)
58. clover/nir: Use lower_vars_to_explicit for uniform and global (details / githubweb)
59. spirv: Drop the OpenCL type layout code (details / githubweb)
60. anv: Set alignments on UBO/SSBO root derefs (details / githubweb)
61. radv: remove descriptor_indexing fails from expected fails (details / githubweb)
62. aco: fix mad splitting after applying output modifiers (details / githubweb)
63. aco: remove omod_success/clamp_success (details / githubweb)
64. radv: Fix various non-critical integer overflows (details / githubweb)
65. aco: Fix integer overflows when emitting parallel copies during RA (details / githubweb)
66. amd/common: Fix various non-critical integer overflows (details / githubweb)
67. ci/bare-metal: Use re.search() instead re.match() for our line matching. (details / githubweb)
68. ci/bare-metal: Fix detection of "POWER_GOOD not seen in time" fails (details / githubweb)
69. ci/bare-metal: Include a timestamp in our serial reads. (details / githubweb)
70. ci/bare-metal: Log why our run restarts when it does. (details / githubweb)
71. ci/bare-metal: Fix capturing of serial output as job artifacts. (details / githubweb)
72. ci/bare-metal: Use python for handling fastboot booting and parsing (details / githubweb)
73. iris: Fix aux assertion in resource_get_handle (details / githubweb)
74. panfrost: Fix gnu-empty-initializer errors. (details / githubweb)
75. radv,aco: disable opts if VK_PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT (details / githubweb)
76. meson: drop leftover PTHREAD_SETAFFINITY_IN_NP_HEADER (details / githubweb)
77. nir: fix const-cast warning on MSVC (details / githubweb)
78. ac/llvm: fix unaligned VS input loads on gfx10.3 (details / githubweb)
79. intel: add INTEL_DEBUG=shaders (details / githubweb)
80. intel/fs: add hint how to get more info when shader validation fails (details / githubweb)
81. aco: fix byte_align_scalar for 3 dword vectors (details / githubweb)
82. nir/load_store_vectorize: rework alignment calculation (details / githubweb)
83. nir/load_store_vectorizer: Use more imm helpers in the tests. (details / githubweb)
84. nir/load_store_vectorizer: Add unit tests for alignment handling. (details / githubweb)
85. aco: handle unaligned loads on GFX10.3 (details / githubweb)
86. iris: Fold a condition into no_gpu for consistency (details / githubweb)
87. iris: Make iris_has_color_unresolved more generic (details / githubweb)
88. iris: Avoid resolving Z/S reads in transfer_map (details / githubweb)
89. iris: Drop a use of the need_resolve boolean (details / githubweb)
90. iris: Better determine map_would_stall for Z/S (details / githubweb)
91. nir: get ffma support from NIR options for nir_lower_flrp (details / githubweb)
92. nir/algebraic: trivially enable existing 32-bit patterns for all bit sizes (details / githubweb)
93. nir/algebraic: add 16-bit versions of a few 32-bit patterns (details / githubweb)
94. intel/compiler: match brw_compile_* declarations with their definitions (details / githubweb)
95. intel/compiler: use the same name for nir shaders in brw_compile_* functions (details / githubweb)
96. intel/compiler: move extern C functions out of namespace brw (details / githubweb)
97. intel/compiler: print dispatch width when shader fails to compile (details / githubweb)
98. intel/compiler: fix typo in a comment (details / githubweb)
99. gallivm: add InstSimplify pass (details / githubweb)
100. freedreno/regs: add 7nm DSI PHY/PLL regs (details / githubweb)
101. compiler/types: Fix deserializing structs with >= 15 members (details / githubweb)
102. spirv: fix emitting switch cases that directly jump to the merge block (details / githubweb)
103. glsl_to_nir: fix crashes with int16 shifts (details / githubweb)
104. egl/x11: Free memory allocated for reply structures on error (details / githubweb)
105. radeonsi: remove redundant no-signed-zero-fp-math LLVM attribute (details / githubweb)
106. radeonsi: move nir_shader_compiler_options into si_screen (details / githubweb)
107. Revert "ac: generate FMA for inexact instructions for radeonsi" (details / githubweb)
108. ac/llvm: remove stub prototype for fmed3 (details / githubweb)
109. ac/llvm: fix amdgcn.rcp for v2f16 (details / githubweb)
110. ac/llvm: fix amdgcn.fract for v2f16 (details / githubweb)
111. ac/llvm: fix amdgcn.rsq for v2f16 (details / githubweb)
112. ac/llvm: fix bcsel for v2*16 (details / githubweb)
113. ac/llvm: remove dead code handling for fmod (details / githubweb)
114. ac/llvm: add better code for isign (details / githubweb)
115. ac/llvm: add better code for fsign (details / githubweb)
116. ac/llvm: fix b2f for v2f16 (details / githubweb)
117. android: amd/registers: switch to new generated register definitions (details / githubweb)
118. android: util: fix missing include path (details / githubweb)
119. radeonsi: fix syncobj wait timeout (details / githubweb)
120. radeonsi: fix user fence space when MCBP is enabled (details / githubweb)
121. ci: Run deqp-gles3 and deqp-gles31 on RadeonSI (details / githubweb)
122. android: nv50/ir: Add nv50_ir_prog_info_out serialize and deserialize (details / githubweb)
123. vulkan/device_select: don't pick a cpu driver as the default (details / githubweb)
124. amd/llvm: switch to 3-spaces style (details / githubweb)
125. amd/common: switch to 3-spaces style (details / githubweb)
126. swr: Fix crashes on non-AVX hardware (details / githubweb)
127. lima/ppir: Skip instruction merge when having more than one successor (details / githubweb)
128. radeonsi: stop using TGSI_PROPERTY_NEXT_SHADER (details / githubweb)
129. radeonsi: stop using TGSI_PROPERTY_VS_WINDOW_SPACE_POSITION / VS_BLIT_SGPRS_AMD (details / githubweb)
130. radeonsi: stop using TGSI_PROPERTY_TCS_VERTICES_OUT (details / githubweb)
131. radeonsi: stop using TGSI_PROPERTY_TES_POINT_MODE / TES_PRIM_MODE (details / githubweb)
132. radeonsi: stop using TGSI_PROPERTY_TES_SPACING (details / githubweb)
133. radeonsi: stop using TGSI_PROPERTY_TES_VERTEX_ORDER_CW (details / githubweb)
134. radeonsi: stop using TGSI_PROPERTY_GS_* (details / githubweb)
135. radeonsi: stop using TGSI_PROPERTY_CS_* (details / githubweb)
136. radeonsi: stop using TGSI_PROPERTY_FS_EARLY_DEPTH_STENCIL (details / githubweb)
137. radeonsi: stop using TGSI_PROPERTY_FS_POST_DEPTH_COVERAGE (details / githubweb)
138. radeonsi: stop using TGSI_PROPERTY_FS_COORD_PIXEL_CENTER (details / githubweb)
139. radeonsi: stop using TGSI_PROPERTY_FS_DEPTH_LAYOUT (details / githubweb)
140. radeonsi: stop using TGSI_PROPERTY_CS_LOCAL_SIZE (details / githubweb)
141. radeonsi: stop using TGSI_PROPERTY_FS_COLOR0_WRITES_ALL_CBUFS (details / githubweb)
142. radeonsi: remove info::samplers_declared, image_buffers, msaa_images_declared (details / githubweb)
143. radeonsi: remove redundant si_shader_info::shader_buffers_declared (details / githubweb)
144. radeonsi: remove redundant si_shader_info::images_declared (details / githubweb)
145. radeonsi: remove redundant si_shader_info::const_buffers_declared (details / githubweb)
146. radeonsi: remove redundant si_shader_info:*(clip|cull)* fields (details / githubweb)
147. radeonsi: remove unused si_shader_info::uses_(vertexid|basevertex) (details / githubweb)
148. radeonsi: merge uses_persp_opcode_interp_sample/uses_linear_opcode_interp_sample (details / githubweb)
149. radeonsi: remove redundant si_shader_info::uses_kill (details / githubweb)
150. radeonsi: reduce type sizes in si_shader_selector (details / githubweb)
151. radeonsi: rename num_memory_instructions -> num_memory_stores (details / githubweb)
152. radeonsi: remove redundant si_shader_info::writes_memory (details / githubweb)
153. radeonsi: remove redundant GS variables in si_shader_selector (details / githubweb)
154. radeonsi: remove redundant si_shader_selector::max_gs_stream (details / githubweb)
155. radeonsi: remove redundant si_shader_info::uses_derivatives (details / githubweb)
156. radeonsi: use shader_info::cs::local_size_variable to clean up some code (details / githubweb)
157. radeonsi: deduplicate setting key.mono.u.vs_export_prim_id (details / githubweb)
158. radeonsi: kill point size VS output if it's not used by the rasterizer (details / githubweb)
159. swr: fix build with mingw (details / githubweb)
160. spirv: Improve the "Entry point not found" error message (details / githubweb)
161. spirv2nir: Rework argument handling (details / githubweb)
162. ci: Remove any existing results directory before running piglit (details / githubweb)
163. nir/lower_io: Fix the unknown-array-index case in get_deref_align (details / githubweb)
164. nir/opt_shrink_vectors: shrink image stores using the format (details / githubweb)
165. radeonsi: Work around Wasteland 2 bug. (details / githubweb)
166. android: freedreno: Implement pipe screen's get_device/driver_uuid() (details / githubweb)
167. android: freedreno/common: add libmesa_git_sha1 static dependency (details / githubweb)
168. egl/android: HAVE_DRM_GRALLOC path fixes (v2) (details / githubweb)
169. mesa: refactor floating point texture fbo completeness check on gles (details / githubweb)
170. mesa: add EXT_color_buffer_half_float plumbing (details / githubweb)
171. mesa/st: enable EXT_color_buffer_half_float when formats supported (details / githubweb)
172. tracie: Make tests independent of environment (details / githubweb)
173. tracie: Produce JUnit XML results (details / githubweb)
174. gitlab-ci: Enable unit test reports for normal runner traces jobs (details / githubweb)
175. gitlab-ci: Enable unit test reports for lava traces jobs (details / githubweb)
176. gitlab-ci: Enable unit test report for arm64_a630_traces (details / githubweb)
177. aco: fix one-off error in Operand(uint16_t) (details / githubweb)
178. aco: improve fsign selection (details / githubweb)
179. swr: Use ElemenCount constructor for LLVM 11 (details / githubweb)
180. aco/isel: refactor code and remove unnecessary v_mov (details / githubweb)
181. aco/isel: refactor emit_vop3a_instruction() to handle 2 operand instructions (details / githubweb)
182. radeonsi: set outputs_written_before_ps for geometry shaders too (details / githubweb)
183. radeonsi: eliminate unused shader outputs for separate NGG geometry shaders (details / githubweb)
184. radeonsi: remove swizzle == ~0 dead code in si_llvm_load_input_gs (details / githubweb)
185. ac,radeonsi: lower 64-bit IO to 32 bits and remove all dead code (details / githubweb)
186. radeonsi: inline trivial PS functions (details / githubweb)
187. nir: Update the comment about nir_lower_uniforms_to_ubo()'s multiplier. (details / githubweb)
188. nir: Add a range_base+range to nir_intrinsic_load_ubo(). (details / githubweb)
189. freedreno/ir3: Use the new NIR UBO ranges in UBO analysis. (details / githubweb)
190. freedreno/ir3: Apply the max upload limit to initial range setup (details / githubweb)
191. nir: Use explicit deref information to provide real UBO ranges. (details / githubweb)
192. nir/opt_if: fix opt_if_merge when destination branch has a jump (details / githubweb)
193. nir: Add a dominance validation pass (details / githubweb)
194. spirv: Run repair_ssa if there are discard instructions (details / githubweb)
195. intel/nir: Call validate_ssa_dominance at both ends of the NIR compile (details / githubweb)
196. aco/isel: Turn the function template emit_load into a proper function (details / githubweb)
197. aco/isel: Simplify nested branching code (details / githubweb)
198. aco/isel: Consistently use references for input parameters in emit_load (details / githubweb)
199. aco/isel: Remove unused definitions (details / githubweb)
200. aco/isel: Move context initialization code to a dedicated file (details / githubweb)
201. aco/isel: Move add_startpgm to aco_instruction_selection.cpp (details / githubweb)
202. aco/isel: Compile all helper functions with static linkage (details / githubweb)
203. llvmpipe: include gallivm perf flags in shader cache. (details / githubweb)
204. iris: Add missing range_base/range to our nir_load_ubos. (details / githubweb)
205. radv,gallium: Add driconf option to reduce advertised VRAM size. (details / githubweb)
206. docs/features: Add missing Panfrost extensions (details / githubweb)
207. panfrost: Cleanup panfrost_get_param (details / githubweb)
208. panfrost: Remove old comment on broken depth reload (details / githubweb)
209. panfrost: Correctly set modifier_constant (details / githubweb)
210. panfrost: Seperate resource setup and bo creation (details / githubweb)
211. panfrost: Move tiled-linear conversion checking to a new function (details / githubweb)
212. panfrost: AFBC to linear layout conversion (details / githubweb)
213. nir/opt_loop_unroll: fix is_access_out_of_bounds with vectors (details / githubweb)
214. aco: fix v_writelane_b32 with two sgprs (details / githubweb)
215. aco: workaround disassembler bug of v_writelane_b32 with literal (details / githubweb)
216. aco: don't apply constant to SDWA on GFX8 (details / githubweb)
217. aco: fix value numbering of reductions (details / githubweb)
218. aco: fix validation of sub-dword parallel-copies (details / githubweb)
219. turnip: delete tu_physical_device path field (details / githubweb)
220. turnip: delete unused tu_image fields (details / githubweb)
221. turnip: fix the type of tu_shader_module code field, delete unused sha1 (details / githubweb)
222. turnip: delete unused "tu_cmd_buffer_upload" (details / githubweb)
223. turnip: remove some unnecessary regs init (details / githubweb)
224. turnip: Fix a compiler warning in release builds of the query code. (details / githubweb)
225. freedreno: Make the pack struct have a .qword for wide addresses. (details / githubweb)
226. turnip: Fix truncation of CS shader iovas to 32 bits. (details / githubweb)
227. turnip: Fix truncation of iovas to 32 bits in queries. (details / githubweb)
228. ci/bare-metal: Update the kernel to msm-next-pgtables (details / githubweb)
229. ci/bare-metal: Allow wget of the kernel/dtb for kernel development. (details / githubweb)
230. mesa: Fix GL_CLAMP handling in glSamplerParameter (details / githubweb)
231. mesa: Generate more errors from GetSamplerParameter (details / githubweb)
232. st/nir: Call st_glsl_to_nir_post_opts before interface unification (details / githubweb)
233. nir_dominance: Use uint32_t instead of int16_t for dominance counters (details / githubweb)
234. gallium/dri2: Report I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS num_planes (details / githubweb)
235. gallium/dri2: Support I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS import (details / githubweb)
236. intel/isl: Describe I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS (details / githubweb)
237. intel/isl: Support ISL_AUX_USAGE_MC in surface states (details / githubweb)
238. intel/isl: Add YUV format info for the aux-map (details / githubweb)
239. st/mesa: Don't map all P01X DRM formats to P016 (details / githubweb)
240. intel/common: Add get_aux_map_format_bits() (details / githubweb)
241. iris: Support planar resource imports for MC (details / githubweb)
242. intel/common: Drop unused gen_aux_map_add_image (details / githubweb)
243. iris: Support MC modifier in plane count queries (details / githubweb)
244. iris: Support I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS (details / githubweb)
245. nir: More NIR_MAX_VEC_COMPONENTS fixes (details / githubweb)
246. nir/idiv_const: Use the modern nir_src_as_* constant helpers (details / githubweb)
247. nir: More NIR_MAX_VEC_COMPONENTS fixes (details / githubweb)
248. pan/mdg: Fix spilling of non-32-bit types (details / githubweb)
249. freedreno: Fix file descriptor leak. (details / githubweb)
250. svga: Fix unused printf argument. (details / githubweb)
251. spirv: Initialize spirv_test member shader. (details / githubweb)
252. freedreno: Add another new sysmem flake. (details / githubweb)
253. freedreno/cffdec: Fix up texturator parsing scripts for XML changes. (details / githubweb)
254. freedreno/cffdec: Add support for texturator's 2DMS layout setup. (details / githubweb)
255. freedreno/fdl: Add layout test for the Android CTS's MSAA mustpass surface. (details / githubweb)
256. gallivm: disable brilinear for lod bias and explicit lod. (details / githubweb)
257. freedreno: Fix rast state for multisample clear (details / githubweb)
258. freedreno: Don't bypass fd_draw_vbo() in clear fallback (details / githubweb)
259. freedreno/a6xx: Skip empty tile_setup (details / githubweb)
260. freedreno/a6xx: Fix fd6_draw_vbo() return (details / githubweb)
261. freedreno: Clear gs/tcs/tes state for clear blits (details / githubweb)
262. freedreno/a6xx: Fix MSAA clear (details / githubweb)
263. mesa: move u_idalloc from gallium/aux/util to util (details / githubweb)
264. util/idalloc: add util_idalloc_reserve (details / githubweb)
265. util/idalloc: add lowest_free_idx to avoid iterating from 0 (details / githubweb)
266. mesa: add a isGenName parameter to _mesa_HashInsert (details / githubweb)
267. mesa: add GL name reuse support (details / githubweb)
268. mesa: add _mesa_HashFindFreeKeys (details / githubweb)
269. mesa: use _mesa_HashFindFreeKeys for GL functions (details / githubweb)
270. driconf: add option to reuse GL names (details / githubweb)
271. anv: fix minor gen_ioctl(I915_PERF_IOCTL_CONFIG) error handling issue (details / githubweb)
272. intel/compiler: remove unused fs_validator::param_size (details / githubweb)
273. intel/compiler: initialize remaining fields of various classes (details / githubweb)
274. intel/tools: fix possible memory leak in the error path (details / githubweb)
275. intel/tools: handle ftell errors (details / githubweb)
276. intel/compiler: quiet Coverity warnings (details / githubweb)
277. intel/tools: fix possible randomly increased verbosity of error2aub (details / githubweb)
278. turnip: rework vertex buffers draw state handling (details / githubweb)
279. ci: do not build libdrm for vc4, reedreno and etnaviv (details / githubweb)
280. wsi/x11: Hook up VK_PRESENT_MODE_FIFO_RELAXED_KHR (details / githubweb)
281. glx: Use GLX_FUNCTION2 only for actually aliased function names (details / githubweb)
282. glx: Collect all the non-applegl extensions in the GetProcAddress table (details / githubweb)
283. glx: Reject glXSwapIntervalMESA greater than INT_MAX (details / githubweb)
284. glx: Implement GLX_EXT_swap_control for DRI2 and DRI3 (details / githubweb)
285. glx/dri3: Implement GLX_EXT_swap_control_tear (details / githubweb)
286. nir: add mediump flag to IO semantics (details / githubweb)
287. nir: fix lower_mediump_outputs to not require variables (details / githubweb)
288. blorp: Fix alignment test for HIZ_CCS_WT fast-clears (details / githubweb)
289. blorp: Drop trailing whitespace in blorp_clear.c (details / githubweb)
290. pan/mdg: Obey f2fmp size restriction in fuse_io_16 (details / githubweb)
291. nir/algebraic: add flrp patterns for 16 and 64 bits (details / githubweb)
292. nir/algebraic: expand existing 32-bit patterns to all bit sizes using loops (details / githubweb)
293. nir: remove redundant opcode u2ump (details / githubweb)
294. nir: enforce 32-bit src type requirement for f2fmp and i2imp (details / githubweb)
295. nir: add new mediump opcodes f2[ui]mp, i2fmp, u2fmp (details / githubweb)
296. nir/algebraic: collapse conversion opcodes (many patterns) (details / githubweb)
297. nir/algebraic: add late optimizations that optimize out mediump conversions (v3) (details / githubweb)
298. nv50/ir: Add fallthrough statement. (details / githubweb)
299. nv50/ir: Remove duplicate mask assignment. (details / githubweb)
300. disk_cache: move cache dir generation into OS specific helper file (details / githubweb)
301. disk_cache: add disk_cache_enabled() helper (details / githubweb)
302. disk_cache: move index mmap into OS specific helper (details / githubweb)
303. disk_cache: move munmap into an OS specific helper (details / githubweb)
304. disk_cache: move evict_lru_item() to an OS specific helper (details / githubweb)
305. disk_cache: create new helper for writing cache items to disk (details / githubweb)
306. disk_cache: move get_cache_file() to an OS specific helper (details / githubweb)
307. disk_cache: add new OS specific helper disk_cache_evict_item() (details / githubweb)
308. disk_cache: move cache item loading code into disk_cache_load_item() helper (details / githubweb)
309. v3d/compiler: fix V3D double-rounding of .8 fixed-point XY coordinates (details / githubweb)
310. radeonsi: fix max syncobj wait timeout (details / githubweb)
311. freedreno: fix fence-fd leak (details / githubweb)
312. nir/opt_vectorize: don't lose exact and no_*_wrap flags (details / githubweb)
313. anv/image: Disable multi-layer CCS_E on TGL+ (details / githubweb)
314. anv, iris: Set MediaSamplerDOPClockGateEnable for gen12+ (details / githubweb)
315. turnip: device global bo list (details / githubweb)
316. lima: dont split vec3 unaligned load inputs (details / githubweb)
317. glsl: mark some builtins with correct glsl(es) version check (details / githubweb)
318. aco: Fix emit_boolean_exclusive_scan in wave32 mode. (details / githubweb)
319. aco: Clean up emit_mbcnt. (details / githubweb)
320. aco: Add base argument to emit_mbcnt. (details / githubweb)
321. glsl: fix per_vertex_accumulator::fields size (details / githubweb)
322. nir/algebraic: fold some nested comparisons with ball and bany (details / githubweb)
323. turnip: avoid heap allocations in QueueSubmit when semaphores are used (details / githubweb)
324. panfrost: Fix nonzero stencil mask on vertex/compute (details / githubweb)
325. pan/bit: Set d3d=true for CMP tests (details / githubweb)
326. pan/bit: Fix unit tests (details / githubweb)
327. etnaviv: call nir_opt_shrink_vectors(..) in opt loop (details / githubweb)
328. android: aco/isel: Move context initialization code to a dedicated file (details / githubweb)
329. ac/llvm: Fix nonportable sizeof. (details / githubweb)
330. freedreno: Check file descriptor before write. (details / githubweb)
331. nv50/ir: Initialize Converter members. (details / githubweb)
332. intel/genxml: make sure test assert are compiled in (details / githubweb)
333. iris: remove additional pipe control done before hiz for older gens (details / githubweb)
334. aco: pass -fno-exceptions and -fno-rtti (details / githubweb)
335. frontends/va: Add support for NV12/P010/P016 to vaDeriveImage (details / githubweb)
336. frontends/va: Derive image from interlaced buffers (details / githubweb)
337. frontends/va: Derive image from interlaced buffers in some cases (details / githubweb)
338. freedreno/ir3: allow layer/viewport output for VS/GS/DS (details / githubweb)
339. freedreno/ir3: add view_zero to shader key (details / githubweb)
340. turnip: multiViewport and VK_EXT_shader_viewport_index_layer (details / githubweb)
341. spirv: Handle OpTypeOpaque (details / githubweb)
342. libgl-gdi: Fix unused-variable warnings. (details / githubweb)
343. st/mesa: don't enable NV_copy_depth_to_color if NIR doesn't support FP64 (details / githubweb)
344. nir,radeonsi: move ffma fusing to late optimizations for better codegen (details / githubweb)
345. radeonsi: clean up ffma handling (details / githubweb)
346. Revert "radeonsi: set BIG_PAGE fields on gfx10.3" (details / githubweb)
347. Revert "radeonsi: move L2_CACHE_CONTROL registers into si_emit_framebuffer_state" (details / githubweb)
348. docs/download: mention tarball GPG signatures and link to the keys (details / githubweb)
349. docs: add another 20.1.x release to allow for more overlap with 20.2 (details / githubweb)
350. docs/release-calendar: update 20.2 (details / githubweb)
351. r600/uvd: set dec->bs_ptr = NULL on unmap (details / githubweb)
352. radeon/vcn: set dec->bs_ptr = NULL on unmap (details / githubweb)
353. radeonsi: fix quant_mode selection for large negative values (details / githubweb)
354. radeonsi: fix guardband handling for large values (details / githubweb)
355. r600: revert disabling llvm draw (details / githubweb)
356. r600/nir: fetch sources and split uniforms before emittting alu instructions (details / githubweb)
357. r600/sfn: correct ring op patching (details / githubweb)
358. r600/sfn: Fix loading vertex attributes (details / githubweb)
359. r600/sfn: clone shader before lowering to registers and src/dest modifiers (details / githubweb)
360. r600/sfn: Fix ordering of tex param moves (details / githubweb)
361. r600/sfn: avoid some copies (details / githubweb)
362. r600/sfn: Lower *sign opcodes in nir (details / githubweb)
363. r600/sfn: Fix split_alu_modifiers (details / githubweb)
364. r600/sfn: Fix bitfield ops and 2x16 split_y (details / githubweb)
365. r600/sfn: Fix source swizzle for gradient queries (details / githubweb)
366. r600/sfn: more fixing of vec4 fetching (details / githubweb)
367. r600/sfn: Fix comparison with different signedness (details / githubweb)
368. v3d: remove unused header (details / githubweb)
369. vc4: remove unused header (details / githubweb)
370. gallium/aux: remove unused u_blit.[ch] (details / githubweb)
371. nir: Add option lower_uniforms_to_ubo (details / githubweb)
372. radeonsi: set compiler flag lower_uniforms_to_ubo (details / githubweb)
373. freedreno/ir3: set lower_uniforms_to_ubo compiler flag (details / githubweb)
374. intel/compiler: Set lower_uniform_to_ubo compiler flag (details / githubweb)
375. llvmpipe: set lower_uniform_to_ubo compiler flag (details / githubweb)
376. gallium+mesa/st: lower uniforms based on compiler flag instead of packed uniforms cap (details / githubweb)
377. docs: add release notes for 20.1.8 (details / githubweb)
378. docs: update calendar and link releases notes for 20.1.8 (details / githubweb)
379. anv: improve error message when failing to open device path (details / githubweb)
380. anv: assert that the target bo is valid when adding a reloc list (details / githubweb)
381. pan/bi: Lower flrp16 (details / githubweb)
382. pan/bi: Add XML describing the instruction set (details / githubweb)
383. pan/bi: Add ISA parser (details / githubweb)
384. pan/bi: Add packing generator (details / githubweb)
385. pan/bi: Add disassembler generator (details / githubweb)
386. pan/bi: Add disassembly prototypes (details / githubweb)
387. pan/bi: Add bi_disasm_dest_* helpers (details / githubweb)
388. pan/bi: Export dump_src (details / githubweb)
389. pan/bi: Use new disassembler (details / githubweb)
390. pan/bi: Use canonical syntax for registers/uniforms/imms (details / githubweb)
391. pan/bi: Use canonical syntax for special constants (details / githubweb)
392. pan/bi: Add dummy carry/borrow argument for iadd/isub (details / githubweb)
393. pan/bi: Introduce segments into the IR (details / githubweb)
394. pan/bi: Add format field to IR (details / githubweb)
395. pan/bi: Track compute_lod in IR (details / githubweb)
396. pan/bi: Pass blend descriptor explicitly in IR (details / githubweb)
397. pan/bi: Use 8-bit shifts (details / githubweb)
398. pan/bi: Use src1/dest_invert instead of src_invert[] (details / githubweb)
399. pan/bi: Move packing helpers to dedicated file (details / githubweb)
400. pan/bi: Use new packing (details / githubweb)
401. pan/bi: Remove unused prints (details / githubweb)
402. pan/bi: Remove unused packing data structures (details / githubweb)
403. pan/bi: Drop *FMIN reference (details / githubweb)
404. nir/lower_io: don't reduce range if parent length is zero (details / githubweb)
405. util: Makefile.sources: Add disk_cache_os.{c,h} (details / githubweb)
406. turnip: Add support for a615. (details / githubweb)
407. aco: fix incorrect assertion in emit_vop3a_instruction() (details / githubweb)
408. r600: enable lowering uniforms to UBO (details / githubweb)
409. r600/sfn: Use load_ubo_vec4 lowering pass (details / githubweb)
410. nir: remove ubo_r600 instrinsic since ubo_vec4 is used now (details / githubweb)
411. radv: initialize with expanded cmask if the destination layout needs it (details / githubweb)
412. anv: Fix the target_bo assertion in anv_reloc_list_add (details / githubweb)
413. glx: Allow depth-30 pbuffers to work without a depth-30 pixmap format (details / githubweb)
414. android: pan/bi: Use new disassembler (details / githubweb)
415. android: pan/bi: Use new packing (details / githubweb)
416. disk_cache: Fix filename leak on error path. (details / githubweb)
417. radesonsi: Remove unsigned comparison to zero. (details / githubweb)
418. panfrost: Delete debug allocated syncobj. (details / githubweb)
419. turnip: Release bo_mutex lock before potential error path. (details / githubweb)
420. glsl: take EXT_gpu_shader4 in to account when adding round (details / githubweb)
421. swr: missing _BitScanForward64 on 32 bits win (details / githubweb)
422. swr: fix _BitScanForward64 on unix (details / githubweb)
423. etnaviv: shuffle some variant fields (details / githubweb)
424. etnaviv: add disk cache (details / githubweb)
425. mesa: fix glUniform* when a struct contains a bindless sampler (details / githubweb)
426. radv: fix null descriptor for dynamic buffers (details / githubweb)
427. radv: fix vertex buffer null descriptors (details / githubweb)
428. glsl_type: Add packed to structure type comparison for hash map (details / githubweb)
429. radeonsi: fix user fence GPU address (details / githubweb)
430. amd/common: Store non-displayable DCC pitch. (details / githubweb)
431. radeonsi: Put retile map in separate buffers. (details / githubweb)
432. radeonsi: Move display dcc dirty tracking to framebuffer emission. (details / githubweb)
433. lima: allocate new bo for stream draw (details / githubweb)
434. lima: fix vertex shader uniform buffer size (details / githubweb)
435. panfrost: Set modifier_constant to true for exported resources (details / githubweb)
436. pan/mdg: Return a bool from midgard_nir_lod_errata (details / githubweb)
437. pan/mdg: Use nir_shader_instructions_pass for nir_lod_errata (details / githubweb)
438. pan/mdg: Use nir_shader_instructions_pass for fdot2 lowering (details / githubweb)
439. clover: Pull the stride from pipe_transfer for image maps (details / githubweb)
440. spirv: Access qualifiers are not a bitfield (details / githubweb)
441. spirv: Plumb access qualifiers through from image types (details / githubweb)
442. nir: Add a pass for lowering CL-style image ops to texture ops (details / githubweb)
443. intel/fs/swsb: SCHEDULING_FENCE only emits SYNC_NOP (details / githubweb)
444. ac/surface: Fix depth import on GFX6-GFX8. (details / githubweb)
445. radv,radeonsi: Disable compression on interop depth images (details / githubweb)
446. Revert "radv: set BIG_PAGE to improve performance on GFX10.3" (details / githubweb)
447. Revert "radv: emit {CB,DB}_RMI_L2_CACHE_CONTROL at framebuffer time" (details / githubweb)
448. Revert "panfrost: Drop implicit blend pooling" (details / githubweb)
449. panfrost: gen_pack: Minor formatting improvement (details / githubweb)
450. panfrost: gen_pack: Fix __gen_unpack_uint() (details / githubweb)
451. panfrost: gen_pack: Add pan_{unpack,print}() helpers (details / githubweb)
452. panfrost: gen_pack: Move the group get_length() logic to its own method (details / githubweb)
453. panfrost: gen_pack: Add the aggregate concept (details / githubweb)
454. panfrost: gen_pack: Allow empty structs (details / githubweb)
455. panfrost: gen_pack: Add an align() modifier (details / githubweb)
456. panfrost: gen_pack: Add a log2 modifier (details / githubweb)
457. panfrost: gen_pack: Allow enum/define values expressed in hexadecimal (details / githubweb)
458. panfrost: decode: Make the indentation consistent with auto-generated print helpers (details / githubweb)
459. panfrost: decode: Rework the DUMP_{CL,ADDR}() macros (details / githubweb)
460. panfrost: decode: Add a macro to dump unpacked descriptors (details / githubweb)
461. panfrost: decode: Use pan_{unpack,print}() when applicable (details / githubweb)
462. panfrost: XML-ify the local storage descriptor (details / githubweb)
463. panfrost: Clarify what TILED mode is (details / githubweb)
464. panfrost: Add Tiled linear mode to the Block Format enum (details / githubweb)
465. panfrost: XML-ify the midgard tiler descriptor (details / githubweb)
466. panfrost: XML-ify the single target framebuffer descriptor (details / githubweb)
467. panfrost: XML-ify the bifrost tiler descriptors (details / githubweb)
468. panfrost: XML-ify the multi-target framebuffer descriptors (details / githubweb)
469. panfrost: XML-ify the job header descriptor (details / githubweb)
470. panfrost: XML-ify the write value job descriptor (details / githubweb)
471. panfrost: XML-ify the fragment job descriptor (details / githubweb)
472. panfrost: Rename the Blend dither disable flag (details / githubweb)
473. panfrost: XML-ify the compute job descriptor (details / githubweb)
474. panfrost: Avoid copying job descriptors around when we can (details / githubweb)
475. panfrost: decode: Misc formatting improvements (details / githubweb)
476. radv,aco: fix reading primitive ID in FS after TES (details / githubweb)
477. gallium: add PIPE_CAP_MAX_TEXTURE_MB (details / githubweb)
478. radeonsi: move GL vendor workaround to drirc (details / githubweb)
479. radeonsi: reduce PIPE_SHADER_CAP_MAX_CONST_BUFFER_SIZE value (details / githubweb)
480. radeonsi: change vendor name to AMD (details / githubweb)
481. aco: keep track of temporaries' regclasses in the Program (details / githubweb)
482. aco: use bit vectors for liveness sets (details / githubweb)
483. radv: fix transform feedback crashes if pCounterBufferOffsets is NULL (details / githubweb)
484. ci: Add "is scheduled pipeline" YAML anchor (details / githubweb)
485. ci: Add "is master branch of main project" YAML anchor (details / githubweb)
486. ci: Add "is pre-merge pipeline for Marge Bot" YAML anchor (details / githubweb)
487. ci: Add "is post-merge pipeline, not for Marge Bot" YAML anchor (details / githubweb)
488. ci: Add "is forked branch or pre-merge pipeline" YAML anchor (details / githubweb)
489. ci: Add "is forked branch" YAML anchor (details / githubweb)
490. ci: Add "is post-merge pipeline" YAML anchor (details / githubweb)
491. ci: Add "is pre-merge pipeline" YAML anchor (details / githubweb)
492. ci: Add "is for Marge Bot" YAML anchor (details / githubweb)
493. ci: Always use CI_PROJECT_NAMESPACE instead of CI_PROJECT_PATH (details / githubweb)
494. ci: Prevent pages job from running in pre-merge pipelines (details / githubweb)
495. ci: Don't create test-docs job if the pages one exists in the pipeline (details / githubweb)
496. radv: add a helper for loading meta descriptors (details / githubweb)
497. radv: do not lower UBO/SSBO access to offsets (details / githubweb)
498. gallium/auxiliary/vl: Odd Dimensions are failing (details / githubweb)
499. r600/sfn: make number of source components a local variable (details / githubweb)
500. r600/sfn: Fix component count for fdph (details / githubweb)
501. r600/sfn: Fix typo in comment (details / githubweb)
502. gallium: Parse packed HEVC SPS encode header for crop parameters (details / githubweb)
503. radeon: Pass HEVC encode crop parameters to the encoder (details / githubweb)
504. frontends/va: Enabled packed headers for H.264 encoder (details / githubweb)
505. gallium/auxiliary/vl: Include src region in scale_y calculation (details / githubweb)
506. wsi/x11: Create a present queue for VK_PRESENT_MODE_FIFO_RELAXED_KHR (details / githubweb)
507. intel/compiler: fixup Gen12 workaround for array sizes (details / githubweb)
508. turnip: Add kgsl backend (details / githubweb)
509. turnip/kgsl: Associate fences with submits. (details / githubweb)
510. st/mesa: Deal with empty textures/buffers in semaphore wait/signal. (details / githubweb)
511. pan/bi: Fix typo. (details / githubweb)
512. glsl: Initialize ir_constant member const_elements in all constructors. (details / githubweb)
513. ac/nir: implement nir_op_[un]pack_[64/32]_* (details / githubweb)
514. radeonsi: don't lower pack for better 16-bit vectorization (details / githubweb)
515. radeonsi: set flags for FP16 in shaders (details / githubweb)
516. radeonsi: implement 16-bit FS color outputs (details / githubweb)
517. radeonsi: vectorize IO for better ALU vectorization (details / githubweb)
518. radeonsi: don't scalarize 16-bit vec2 ALU opcodes (details / githubweb)
519. radeonsi: add 16-bit ALU vectorization (details / githubweb)
520. gallium: rename PIPE_TRANSFER_* -> PIPE_MAP_* (details / githubweb)
521. gallium: rename pipe_transfer_usage -> pipe_map_flags (details / githubweb)
522. gallium: rename transfer flags -> map flags in comments (details / githubweb)
523. radeon: rename RADEON_TRANSFER_* -> RADEON_MAP_* (details / githubweb)
524. nir/lower_io: Eliminate oob writes and return zero for oob reads (details / githubweb)
525. nir/large_constants: Eliminate out-of-bounds writes to large constants (details / githubweb)
526. nir/lower_samplers: Clamp out-of-bounds access to array of samplers (details / githubweb)
527. vulkan: bump headers/registry to 1.2.154 (details / githubweb)
528. anv: implement VK_KHR_copy_commands2 (details / githubweb)
529. aco: Use NIR IO semantics for tess factor IO locations. (details / githubweb)
530. aco: use io semantics to get an intrinsic's slot (details / githubweb)
531. aco: use nir_get_io_offset_src() in visit_load_input() (details / githubweb)
532. aco: use nir's constant source helpers more (details / githubweb)
533. aco: remove dead indirect fs input loading (details / githubweb)
534. aco: stop multiplying driver_location by 4 (details / githubweb)
535. nir: Rename get_buffer_size to get_ssbo_size (details / githubweb)
536. etnaviv: fix comment for source of etna_mesa_debug (details / githubweb)
537. r600/sfn: use cnde instead of cnde_int (details / githubweb)
538. r600/sfn: run late algebraic optimizations (details / githubweb)
539. r600/sfn: remove a useless if-condition (details / githubweb)
540. r600: Add flag for dual-source blending to shader key (details / githubweb)
541. r600/sfn: Sort the outputs of the FS according to data index (details / githubweb)
542. r600/sfn: Keep info about dual-source blend in FS (details / githubweb)
543. r600/sfn: Handle number of color outputs taking dual source blending into account (details / githubweb)
544. r600/sfn: Take dual source blending output indices into account (details / githubweb)
545. r600/sfn: Acquire the number of FS outputs and the write_all info early (details / githubweb)
546. r600/sfn: Be a bit more verbose when logging skipped FS outputs (details / githubweb)
547. tu: Expose shaderStorageImageExtendedFormats (details / githubweb)
548. tu: Expose shaderImageGatherExtended (details / githubweb)
549. ir3: Don't use the format to get the image type (details / githubweb)
550. tu: Expose shaderStorageImage*WithoutFormat (details / githubweb)
551. ci: Use ignore_scheduled_pipelines anchor in .radeonsi-rules (details / githubweb)
552. radeonsi: set TRUNC_COORD=0 for Total War: WARHAMMER to fix it (details / githubweb)
553. radeonsi: move debug options from si_disk_cache_create to si_get_ir_cache_key (details / githubweb)
554. radeonsi: remove KILL_PS_INF_INTERP/CLAMP_DIV_BY_ZERO, use screen::options (details / githubweb)
555. amd: add Dimgrey Cavefish support (details / githubweb)
556. amd: add VanGogh support (details / githubweb)
557. mesa: Make the android_stub be a set of non-installed shared libraries. (details / githubweb)
558. android: Disable trying to read/write to the disk cache. (details / githubweb)
559. nir_lower_system_values: Fix load_global_invocation_id to use base_work_group_id even with no base_global id (details / githubweb)
560. gallium/util: add shader for stencil-blits (details / githubweb)
561. gallium/util: add blitter-support for stencil-fallback (details / githubweb)
562. radv: remove useless assignment of MAX_API_VERSION (details / githubweb)
563. radv: bump the advertised patch version to 145 (details / githubweb)
564. radv: add VK_KHR_copy_commands2 but leave it disabled (details / githubweb)
565. radv: add support for CmdBlitImage2KHR() (details / githubweb)
566. radv: add support for CmdCopyBuffer2KHR() (details / githubweb)
567. radv: add support for CmdCopyBufferToImage2KHR() (details / githubweb)
568. radv: add support for CmdCopyImage2KHR() (details / githubweb)
569. radv: add support for CmdCopyImageToBuffer2KHR() (details / githubweb)
570. radv: cleanup selecting the hardware resolve path (details / githubweb)
571. radv: add support for CmdResolveImage2KHR() (details / githubweb)
572. radv: advertise VK_KHR_copy_commands2 (details / githubweb)
573. lima: add natively supported vertex buffer formats (details / githubweb)
574. radv: Disable NGG on APUs. (details / githubweb)
575. radv: Simplify radv_is_hw_resolve_pipeline. (details / githubweb)
576. radv: Add VK_FORMAT_E5B9G9R9_UFLOAT_PACK32 rendering support. (details / githubweb)
577. st/nir: call nir_opt_access before gl_nir_lower_buffers (details / githubweb)
578. radeonsi: don't use nir_opt_access (details / githubweb)
579. radv: set KEEP_TOGETHER_ENABLE if necessary on GFX10+ (details / githubweb)
580. radv: add a tweak for PS wave CU utilization for gfx10.3 (details / githubweb)
581. pan/bi: Annotate stop bit (canonically "Z-bit") (details / githubweb)
582. pan/bi: Annotate disassemble with format names (details / githubweb)
583. pan/bi: Inline dump_instr (details / githubweb)
584. pan/bi: Track M values of disassembled constants (details / githubweb)
585. pan/bi: Decode M values in disasm (details / githubweb)
586. pan/bi: Disassemble PC-relative addresses (details / githubweb)
587. pan/bi: Add bifrost_reg_mode enum (details / githubweb)
588. pan/bi: Pass 'first' through disassembler (details / githubweb)
589. pan/bi: Decode all 32-bit register modes (details / githubweb)
590. pan/bi: Rename port -> slot (details / githubweb)
591. pan/bi: Use canonical register packing (details / githubweb)
592. pan/bi: Remove old register mode definitions (details / githubweb)
593. radeonsi: set KEEP_TOGETHER_ENABLE if needed (details / githubweb)
594. radeonsi: move binning parameters into si_screen (details / githubweb)
595. radeonsi: break a binning batch on a new PS if bins can use multiple state sets (details / githubweb)
596. radeonsi: add a tweak for PS wave CU utilization for gfx10.3 (details / githubweb)
597. radeonsi: force linear for textures with height=1 (gfx6-8) (details / githubweb)
598. nir: split fuse_ffma into fuse_ffma16/32/64 (details / githubweb)
599. nir: split lower_ffma into lower_ffma16/32/64 (details / githubweb)
600. radeonsi: fuse or lower ffma optimally on all chips (details / githubweb)
601. nir: replace lower_ffma and fuse_ffma with has_ffma (details / githubweb)
602. panfrost: Clamp uniform buffer size (details / githubweb)
603. panfrost: Handle non-positive viewport positions (details / githubweb)
604. panfrost: Remove redundant casts of viewport position (details / githubweb)
605. radeonsi: use optimal order of operations when setting up a compute dispatch (details / githubweb)
606. radeonsi: call si_upload_graphics_shader_descriptors before the big conditional (details / githubweb)
607. radeonsi: move a displaced comment in si_draw_vbo (details / githubweb)
608. radeonsi: don't call emit_cache_flush after uploading bindless descriptors (details / githubweb)
609. radeonsi: reorganize the code around the gfx9 scissor bug (details / githubweb)
610. radeonsi: move si_upload_vertex_buffer_descriptors into si_state_draw.c (details / githubweb)
611. radeonsi: add unlikely statements into si_draw_vbo (details / githubweb)
612. radeonsi: lift the conditional for skipping si_upload_vertex_buffer_descriptors (details / githubweb)
613. radeonsi: always inline draw-related functions that have only one use (details / githubweb)
614. vulkan/wsi/display: add option for display fence to signal syncobj (details / githubweb)
615. ci: adjust RadeonSI rules (details / githubweb)
616. mesa: handle GL_FRONT after translating to it (details / githubweb)
617. turnip: delete unused tu_fence_signal function (details / githubweb)
618. turnip: add a fd field to tu_device (details / githubweb)
619. turnip: require syncobj support (details / githubweb)
620. turnip: rework fences to use syncobjs (details / githubweb)
621. radeonsi/tmz: use secure job if framebuffer has dcc (details / githubweb)
622. radeonsi/tmz: use secure job if using an encrypted z/s buffer (details / githubweb)
623. radeonsi/tmz: add safety assert when tmz is enabled (details / githubweb)
624. radeonsi/tmz: allocate depth/stencil buffers as encrypted (details / githubweb)
625. radeonsi: introduce SI_RESOURCE_FLAG_INTERNAL / RADEON_FLAG_DRIVER_INTERNAL (details / githubweb)
626. amd: add AMDGPU_IDS_FLAGS_TMZ definition to amdgpu_drm.h (details / githubweb)
627. ac/gpu_info: add detection of TMZ support (details / githubweb)
628. radeonsi/tmz: allow secure job if the app made a tmz allocation (details / githubweb)
629. amd/winsys: add RADEON_FLUSH_TOGGLE_SECURE_SUBMISSION (details / githubweb)
630. radeonsi/tmz: fail si_texture_transfer_map if tex is encrypted (details / githubweb)
631. radeonsi/tmz: add tmz variant of sctx::wait_mem_scratch (details / githubweb)
632. radeonsi/tmz: add tmz variant for sctx::tess_rings (details / githubweb)
633. radeonsi: disable primitive discard if tmz is in use (details / githubweb)
634. radeonsi/tmz: add a tmz variant for sctx::eop_bug_scratch (details / githubweb)
635. radeonsi/tmz: add workaround for mpv/vaapi subtitles (details / githubweb)
636. amd/tmz: move uses_secure_bos to radeon_winsys (details / githubweb)
637. radv: fix incorrect ResetFences path for WSI fence (details / githubweb)
638. radv: use syncobj for wsi fence (details / githubweb)
639. Revert "nir: replace lower_ffma and fuse_ffma with has_ffma" (details / githubweb)
640. zink: use correct value for color buffer sample count when creating renderpass (details / githubweb)
641. zink: use correct number of samples on framebuffer in set_framebuffer_state (details / githubweb)
642. zink: use correct layer count when creating framebuffer (details / githubweb)
643. zink: clamp min created fb size to 1x1 (details / githubweb)
644. zink: verify that src and dst aspects are the same in resource_copy_region hook (details / githubweb)
645. zink: implement ARB_instanced_arrays (details / githubweb)
646. zink: move viewport count to zink_gfx_pipeline_state (details / githubweb)
647. zink: set multiviewport cap in ntv when gl_ViewportIndex is a written output (details / githubweb)
648. zink: correctly set up fb-sized scissors for each viewport (details / githubweb)
649. zink: apply viewport count when creating pipelines (details / githubweb)
650. zink: reorder create_stream_output_target to fix failure case leak (details / githubweb)
651. radeonsi: Only call nir_lower_var_copies at the end of the opt loop (details / githubweb)
652. gallium/drm: Deduplicate screen creation for the dynamic (clover) pipe loader. (details / githubweb)
653. gallium/drm: Refactor the stub screen create functions. (details / githubweb)
654. gallium/drm: Define the DRM entrypoints in drm_helper.h (details / githubweb)
655. gallium/drm: Make the pipe loader handle the driconf merging. (details / githubweb)
656. r600/sfn: Initialize GPRValue member m_pin_to_channel. (details / githubweb)
657. gallium/dri2: Move image->texture assignment after image NULL check. (details / githubweb)
658. spirv: vtn_fail with a nice message on unsupported rounding modes (details / githubweb)
659. nir: gather indirect info from lowered IO intrinsics (details / githubweb)
660. nir: gather tess.tcs_cross_invocation info from lowered IO intrinsics (details / githubweb)
661. nir: set system_values_read for all intrinsics (details / githubweb)
662. nir: gather fs.uses_sample_qualifier from lowered IO (details / githubweb)
663. nir: fix input/output info gathering for lowered IO (details / githubweb)
664. nir: gather information about fbfetch and dual source color (details / githubweb)
665. panfrost: gen_pack: Fix gnu-empty-initializer errors (details / githubweb)
666. ci: Extend meson-clang coverage by compiling all gallium drivers (details / githubweb)
667. etnaviv: simplify linear stride implementation (details / githubweb)
668. radeonsi: fix indirect dispatches with variable block sizes (details / githubweb)
669. radeonsi: call nir_shader_gather_info after lowering and optimizing NIR (details / githubweb)
670. radeonsi: use info.system_values_read (details / githubweb)
671. radeonsi: get information about FS color outputs from shader_info directly (details / githubweb)
672. radeonsi: get input/output usage flags from shader_info directly (details / githubweb)
673. radeonsi: run NIR optimizations that glsl_to_nir runs but other places might not (details / githubweb)
674. radeonsi: assume that constant load_local_group_size has been optimized out (details / githubweb)
675. radeonsi: remove redundant variables from struct si_compute (details / githubweb)
676. radeonsi: remove redundant info.uses_fbfetch (details / githubweb)
677. zink: correct typo in stencil-setup (details / githubweb)
678. zink: store base-object of DSA-state (details / githubweb)
679. zink: only set stencil-ref for back if two-sided (details / githubweb)
680. intel/fs: Disable sample mask predication for scratch stores (details / githubweb)
681. kmsro: Extend to include imx-dcss (details / githubweb)
682. lima: fix glCopyTexSubImage2D (details / githubweb)
683. nir/instr_set: hash intrinsic sources (details / githubweb)
684. turnip: fix wrong indentation in tu6_draw_common (details / githubweb)
685. turnip: move A6XX_RB_ALPHA_CONTROL write to init_hw (details / githubweb)
686. turnip: implement VK_EXT_extended_dynamic_state (details / githubweb)
687. nir/liveness: Consider if uses in nir_ssa_defs_interfere (details / githubweb)
688. freedreno/layout: add tile_all flag to the layout (details / githubweb)
689. meson: add support for loongson's mips/mips64 arch. (details / githubweb)
690. blorp: Ensure aligned HIZ_CCS_WT partial clears (details / githubweb)
691. util/xmlconfig: Add a unit test of the code. (details / githubweb)
692. virgl: Clean up the driconf definition of GLES_SAMPLES_PASSED_VALUE. (details / githubweb)
693. driconf: Use nesting macros for defining options. (details / githubweb)
694. nir: Add an internal flag to shader_info (details / githubweb)
695. nir: Add glsl_base_type unsigned -> signed version helper (details / githubweb)
696. vtn: add an option to create a nir library from spirv (details / githubweb)
697. nir/vtn: Add handling for SPIR-V event variables (details / githubweb)
698. vtn/opencl: Rework type handling for CL extension opcodes (details / githubweb)
699. clover/nir: add libclc lowering pass (details / githubweb)
700. vtn/opencl: Add infrastructure for calling out to libclc (details / githubweb)
701. util: add missing extern C (details / githubweb)
702. clover: handle libclc shader (v3) (details / githubweb)
703. vtn/opencl: Implement a lot of opcodes via libclc (details / githubweb)
704. vtn/opencl: Rework handle_instr to be able to handle core SPIR-V opcodes via libclc (details / githubweb)
705. vtn/opencl: Hook up OpenCL async copy and group wait opcodes via libclc (details / githubweb)
706. vtn/opencl: Switch non-native trig to use libclc (details / githubweb)
707. vtn/opencl: Switch exp/pow/log to use libclc (details / githubweb)
708. vtn/opencl: Switch division-related ops to use libclc (details / githubweb)
709. vtn/opencl: Switch some nir-sequence ops to use libclc (details / githubweb)
710. vtn/opencl: Only use libclc ldexp when lower_ldexp is set (details / githubweb)
711. vtn/opencl: Switch fma to conditionally use libclc for 32bit floats (details / githubweb)
712. meson/anv: Use variable that checks for --build-id (details / githubweb)
713. gallivm: add support for lowered IO in vertex shaders (details / githubweb)
714. compiler/types: Add glsl_baseN_t_type(bit_size) helpers (details / githubweb)
715. spirv: Use the new types helpers (details / githubweb)
716. nir: Add a new memcpy intrinsic (details / githubweb)
717. nir: Add a lowering pass to lower memcpy (details / githubweb)
718. spirv: Add support for OpCopyMemorySized (details / githubweb)
719. clover/nir: Call the memcpy lowering pass (details / githubweb)
720. clover: set LLVM min version to 8.0.1 (details / githubweb)
721. ci: add dEQP-VK.info.device_extensions to the list of skipped tests (details / githubweb)
722. panfrost: Fix bifrost tiler descriptor definition (details / githubweb)
723. panfrost: Fix bifrost tiler job emission (details / githubweb)
724. android: pan/bi: fix typo in bifrost_gen_disasm.c gen rules (details / githubweb)
725. lima: set clear depth value to 0x00ffffff as default (details / githubweb)
726. docs: escape backquote character (details / githubweb)
727. docs: show 'Edit on GitLab'-link (details / githubweb)
728. docs: store prefixes in redirects (details / githubweb)
729. docs: remove webmaster article (details / githubweb)
730. docs: everytime -> every time (details / githubweb)
731. docs: apis -> APIs (details / githubweb)
732. docs: scons -> SCons (details / githubweb)
733. docs: frambuffer -> framebuffer (details / githubweb)
734. docs: make two acronyms upper-case (details / githubweb)
735. docs: unecessarily -> unnecessarily (details / githubweb)
736. docs: behaviour -> behavior (details / githubweb)
737. docs: timeplan -> time plan (details / githubweb)
738. docs: initialisation -> initialization (details / githubweb)
739. docs: gitlab -> GitLab (details / githubweb)
740. docs: url -> URL (details / githubweb)
741. docs: recognisable -> recognizable (details / githubweb)
742. glsl/linker: Add support for XFB varying lowering in geometry shader (details / githubweb)
743. panfrost: Remove extra printf arguments. (details / githubweb)
744. nir/load_store_vectorize: improve vectorization with identical operations (details / githubweb)
745. panfrost: Mark blit shaders as internal (details / githubweb)
746. pan/mdg: Infer whether to disassemble shaders from info.internal (details / githubweb)
747. nir/lower_memory_model: return progress when visiting instructions (details / githubweb)
748. nir/lower_memory_model: do not break with global atomic operations (details / githubweb)
749. zink: generate extension infrastructure using a python script (details / githubweb)
750. zink: hook zink_device_info.py to build system (details / githubweb)
751. zink: use the new extension infrastructure in device creation (details / githubweb)
752. zink: use the new, generated extension infrastructure (details / githubweb)
753. zink: remove old extension infrastructure (details / githubweb)
754. radv: Fix emitting SQTT userdata. (details / githubweb)
755. radv: Use correct alignment for SQTT buffer sizes. (details / githubweb)
756. radv: Fix RGP Asic CU info for GFX10+. (details / githubweb)
757. radv: Include flushes in the barrier. (details / githubweb)
758. radv: Record cache flushes for RGP. (details / githubweb)
759. mesa: Promote Intel's simple logging façade for Android to util/ (details / githubweb)
760. turnip: Replace tu_log*() with mesa_log*() (details / githubweb)
761. ci/freedreno: Sort the traces in the .yml of expectations (details / githubweb)
762. ci/freedreno: Add trace tests for glxgears, 0 A.D., and xonotic. (details / githubweb)
763. nir/lower_clip: Add i/o semantics for load/store intrinsics. (details / githubweb)
764. r600/sfn: Fix emitting shared atomics with constant sources (details / githubweb)
765. r600/sfn: Handle nir_op_b2b32 (details / githubweb)
766. r600/sfn: lower to scalar for some optimizations and vectortize later (details / githubweb)
767. r600/sfn: Support group memory barrier (details / githubweb)
768. r600/sfn: save some instructions when doing multisample on sample 0 (details / githubweb)
769. r600/sfn: use fine gradient evaluation for interpolate_at_offset (details / githubweb)
770. r600/sfn: Fix interpolate at sample (details / githubweb)
771. r600/sfn: Fix indirect const buffer access (details / githubweb)
772. r600/sfn; go back to not lowering uniforms to UBOs (details / githubweb)
773. r600/sfn: replace hand-backed literal check by NIR function (details / githubweb)
774. r600/sfn: remove old code to track uniforms as it is no longer needed (details / githubweb)
775. r600/sfn: Add support for helper invocations (details / githubweb)
776. intel/vec4: Silence unused paramter warnings in brw_vec4_generator.cpp (details / githubweb)
777. intel/compiler: Silence unused parameter warning in brw_surface_payload_size (details / githubweb)
778. intel/compiler: Don't fallback to vec4 when scalar GS compile fails [v2] (details / githubweb)
779. intel/vec4: Remove inline lowering of LRP (details / githubweb)
780. intel/compiler: Remove INTEL_SCALAR_... env variables (details / githubweb)
781. intel/vec4: Remove all support for Gen8+ [v2] (details / githubweb)
782. intel/vec4: Remove everything related to VS_OPCODE_SET_SIMD4X2_HEADER_GEN9 (details / githubweb)
783. anv: Check file descriptor before closing. (details / githubweb)
784. aco: Initialize mad_info member literal_idx. (details / githubweb)
785. gallium/swr: Remove unreachable code. (details / githubweb)
786. pan/mdg: Fix memory leak on error path. (details / githubweb)
787. CI: Temporarily disable Panfrost T7xx (details / githubweb)
788. ac/nir: implement nir_intrinsic_{load,store}_global (details / githubweb)
789. ac/nir: implement nir_intrinsic_global_atomic_* (details / githubweb)
790. radv: lower deref operations for global memory for both backends (details / githubweb)
791. ac/llvm: fix invalid IR if image stores are shrinked using the format (details / githubweb)
792. mesa: bptc fixes for decompress rgba_unorm and rgb_float (details / githubweb)
793. panfrost: Adjust quirks for bifrost v6 (details / githubweb)
794. panfrost: Add preliminary support for Mali G72 (details / githubweb)
795. kmsro: Add mediatek entry point (details / githubweb)
796. docs: drop outdated gallium-docs comment (details / githubweb)
797. docs: clippping -> clipping (details / githubweb)
798. docs: consistantly -> consistently (details / githubweb)
799. docs: stabilisation -> stabilization (details / githubweb)
800. docs: flavours -> flavors (details / githubweb)
801. docs: debian -> Debian (details / githubweb)
802. docs: docker -> Docker (details / githubweb)
803. docs: gallium -> Gallium (details / githubweb)
804. st/mesa: use roundf instead of floorf for lod-bias rounding (details / githubweb)
805. nir/lower_io: change nir_io_add_const_offset_to_base to use bitfield modes (details / githubweb)
806. radeonsi: call nir_io_add_const_offset_to_base only once per shader (details / githubweb)
807. docs: add INTEL_SHADER_ASM_READ_PATH description (details / githubweb)
808. radv/llvm: call nir_lower_io_to_vector with FS to fix array tests (details / githubweb)
809. gallium/util: set right dst-dimensions (details / githubweb)
810. gallium/util: fix texture-coordinates for stencil-fallback (details / githubweb)
811. gallium/util: allow scaling blits for stencil-fallback (details / githubweb)
812. nir: Add nir_lower_multiview pass (details / githubweb)
813. anv: Use nir_lower_multiview pass (details / githubweb)
814. nir: Count i/o slots correctly for per-view variables (details / githubweb)
815. nir/lower_io_arrays: Fix xfb_offset bug (details / githubweb)
816. nir: Add per_view to IO semantics (details / githubweb)
817. nir: Handle per-view io in nir_io_add_const_offset_to_base() (details / githubweb)
818. CI: Re-enable Panfrost T7xx (details / githubweb)
819. panfrost: Add support for rbg16 formats (details / githubweb)
820. turnip: remove unused cmd_buffer/device arguments in descriptor sets (details / githubweb)
821. turnip: delete unused/broken pipeline layout hashing code (details / githubweb)
822. turnip: initial implementation of VK_KHR_push_descriptor (details / githubweb)
823. gallium: prepare framework for supporting GL_NV_shader_atomic_int64 (details / githubweb)
824. mesa,glsl: add support for GL_NV_shader_atomic_int64 (details / githubweb)
825. radeonsi: enable support for GL_NV_shader_atomic_int64 (details / githubweb)
826. radv/aco: Set I/O variable locations outside ACO. (details / githubweb)
827. radv: call nir_io_add_const_offset_to_base for FS outputs (details / githubweb)
828. radv: move lowering of FS outputs outside of ACO (details / githubweb)
829. gallium: Make pipe_viewport_state swizzle_x/y/z/w bit-fields 8 bits wide (details / githubweb)
830. tu: Write multiview control registers in binning pass (details / githubweb)
831. tu: Refactor shader compilation flow (details / githubweb)
832. ir3, tu: Run optimization loop twice (details / githubweb)
833. ir3, tu: Link per-view position correctly (details / githubweb)
834. tu: Enable multi-position output (details / githubweb)
835. gallium/vl: do not call transfer_unmap if transfer is NULL (details / githubweb)
836. gallium/vl: add chroma_format arg to vl_video_buffer functions (details / githubweb)
837. Add driver override to envvars.rst (details / githubweb)
838. intel: Add support for i945g to intel_stub_gpu. (details / githubweb)
839. ci/deqp-runner: Allow overriding width/height/config (details / githubweb)
840. ci: cherry-pick deqp fix for config choosing (details / githubweb)
841. ci: Enable remaining (non-rotate) mustpass CTS tests (details / githubweb)
842. radv: Write correct dispatch size for RGP. (details / githubweb)
843. zink: combine all surface layout-setting for src/dst into util function (details / githubweb)
844. zink: unify all occurrences of waiting on a fence (details / githubweb)
845. freedreno/drm: drop bo's dev reference (details / githubweb)
846. freedreno: Don't leak border_color_buf reference (details / githubweb)
847. freedreno/a6xx: Small cleanup (details / githubweb)
848. freedreno/drm: Also clean ring_cache (details / githubweb)
849. freedreno/registers: Add a couple things used on kernel side (details / githubweb)
850. freedreno: Don't leak LRZ bo's (details / githubweb)
851. r600/sfn: Fix enabling the right interpolator for inerpolate_at_sample (details / githubweb)
852. r600/sfn: Fix IDX register ID (details / githubweb)
853. r600/sfn: Add support for more barrier instructions (details / githubweb)
854. r600/sfn: extend life range of all variables by one (details / githubweb)
855. gallivm: fix pow(0, y) to be 0 (details / githubweb)
856. turnip: clean up tu_device_memory (details / githubweb)
857. turnip: always create permanent syncobj for semaphore (details / githubweb)
858. turnip: set MSM_SUBMIT_SYNCOBJ_RESET for submit pWaitSemaphores (details / githubweb)
859. turnip: semaphores simplification (only syncobj semaphores supported) (details / githubweb)
860. turnip: rework GetSemaphoreFdKHR (details / githubweb)
861. turnip: rework ImportSemaphoreFdKHR (details / githubweb)
862. turnip: remove remaining uses of drmSyncobj helpers (details / githubweb)
863. turnip: share code between semaphores/fences + fence import/export (details / githubweb)
864. turnip: signal fence and semaphore in AcquireNextImage2KHR (details / githubweb)
865. gallivm: fix 64-bit CL intrinsics. (details / githubweb)
866. gallivm/nir: fix up non 32-bit load stores (details / githubweb)
867. gallivm/nir: handle non-32-bit mul high (details / githubweb)
868. llvmpipe: use an alternate env var to enable clover. (details / githubweb)
869. nir: Allow creating variables with nir_var_mem_push_const. (details / githubweb)
870. nir/lower_io: Add support for push constants (details / githubweb)
871. anv,radv,tu,val: Call nir_lower_io for push constants (details / githubweb)
872. spirv: Use derefs for push constants (details / githubweb)
873. vallium: Stop using lower_ubo_ssbo_access_to_offsets (details / githubweb)
874. spirv: Delete the legacy offset/index UBO/SSBO lowering (details / githubweb)
875. docs: softwara -> software (details / githubweb)
876. docs: existant -> existent (details / githubweb)
877. docs: webservice -> web service (details / githubweb)
878. docs: bpp -> BPP (details / githubweb)
879. docs: llvm -> LLVM (details / githubweb)
880. docs: correct reference to meson.build (details / githubweb)
881. docs: meson -> Meson (details / githubweb)
882. docs: python3 -> Python 3 (details / githubweb)
883. docs: flex -> Flex (details / githubweb)
884. docs: bison -> Bison (details / githubweb)
885. docs: mako -> Mako (details / githubweb)
886. docs: chocolatey -> Chocolatey (details / githubweb)
887. docs: ninja -> Ninja (details / githubweb)
888. docs: mingw -> MinGW (details / githubweb)
889. docs: microsoft -> Microsoft (details / githubweb)
890. docs: linux -> Linux (details / githubweb)
891. docs: windows -> Windows (details / githubweb)
892. docs: visual studio -> Visual Studio (details / githubweb)
893. docs: gpu -> GPU (details / githubweb)
894. docs: cmake -> CMake (details / githubweb)
895. docs: x11 -> X11 (details / githubweb)
896. docs: wayland -> Wayland (details / githubweb)
897. docs: drm -> DRM (details / githubweb)
898. docs: android -> Android (details / githubweb)
899. docs: git -> Git (details / githubweb)
900. docs: quote "git log" (details / githubweb)
901. docs: scons -> SCons (details / githubweb)
902. docs: ubuntu -> Ubuntu (details / githubweb)
903. docs: vmware -> VMWare (details / githubweb)
904. docs: Sandybridge -> Sandy Bridge (details / githubweb)
905. docs: cpu -> CPU (details / githubweb)
906. bin/gen_release_notes.py: escape special rST characters (details / githubweb)
907. radeonsi: Fix imports with displayable DCC. (details / githubweb)
908. intel/nir: Use nir control flow helpers (details / githubweb)
909. radv: Use nir control flow insertion helpers (details / githubweb)
910. ttn: Use nir control flow insertion helpers (details / githubweb)
911. nir/lower_returns: Use nir control flow insertion helpers (details / githubweb)
912. nir/opt_if: Remove open-coded nir_ssa_def_rewrite_uses() (details / githubweb)
913. nir/opt_if: Use early returns in opt_if_merge() (details / githubweb)
914. nir/copy_propagate: Copy-prop into jump conditions (details / githubweb)
915. nir: Disallow goto and goto_if in clone and [de]serialize (details / githubweb)
916. nir/cf: Better handle intra-block splits (details / githubweb)
917. util: implement f16c - fast half<->float conversions (details / githubweb)
918. util: move util_half_to_float code into _mesa_half_to_float_slow (details / githubweb)
919. util: remove util_float_to_half and util_half_to_float wrappers (details / githubweb)
920. gallium/util: remove redundant util_float_to_half_rtz (details / githubweb)
921. gallium/util: remove empty file u_half.h (details / githubweb)
922. nir/validate: Improve the validation of blocks (details / githubweb)
923. nir/lower_goto_ifs: Don't destroy SSA form in the process (details / githubweb)
924. nir/dominance: Use _mesa_set_clear instead ofhand-rolling it (details / githubweb)
925. spirv: Only run repair_ssa if structured (details / githubweb)
926. nir/lower_goto_ifs: Use rzalloc (details / githubweb)
927. nir/lower_goto_ifs: Add asserts for SSA forks (details / githubweb)
928. nir/lower_goto_ifs: Always include level dom_frontiers in prev_frontier (details / githubweb)
929. Revert "nir/lower_goto_if: Add a route::outside set" (details / githubweb)
930. turnip: implement legacy API functions separately (details / githubweb)
931. freedreno/ir3: Make sure we run the opt loop after lowering UBOs to vec4. (details / githubweb)
932. nir: Document a bit about how align_mul/offset work. (details / githubweb)
933. nir: Print the alignment information on casts. (details / githubweb)
934. nir/nir_lower_uniforms_to_ubo: Set better alignments on our new instructions. (details / githubweb)
935. nir/gl_nir_lower_buffers: Set up align_mul/offset on UBOs. (details / githubweb)
936. nir: Make the load_store_vectorizer provide align_mul + align_offset. (details / githubweb)
937. nir: Drop the high_offset argument to the load_store_vectorizer filter. (details / githubweb)
938. nir: Make nir_lower_ubo_vec4() handle non-vec4-aligned loads. (details / githubweb)
939. freedreno/ir3: Enable the i/o vectorizer on UBOs. (details / githubweb)
940. docs: add release notes for 20.1.9 (details / githubweb)
941. docs: update calendar and link releases notes for 20.1.9 (details / githubweb)
942. add one last 20.1 release to coincide with expected 20.2.1 (details / githubweb)
943. radeonsi: Fix dead lock with aux_context_lock in si_screen_clear_buffer. (details / githubweb)
944. lavapipe: rename vallium to lavapipe (details / githubweb)
945. radv: fix gathering writes_memory for global store/atomic operations (details / githubweb)
946. ac/llvm: fix invalid use of unreachable in ac_build_atomic_rmw() (details / githubweb)
947. ac/nir: fix nir_intrinsic_shared_atomic_fadd (details / githubweb)
948. ttn: Fix number of components for IF/UIF (details / githubweb)
949. radv: gather output usage mask from store_output for VS, TES and GS (details / githubweb)
950. radv/aco: lower IO for all stages outside of ACO (details / githubweb)
951. radeonsi: support for external buffers (ext_external_objects) (details / githubweb)
952. anv: Allow HiZ clears for multi-view (details / githubweb)
953. anv: Use more temp vars in cmd_buffer_begin_subpass (details / githubweb)
954. anv: Skip HiZ and CCS ambiguates which preceed fast-clears (details / githubweb)
955. intel/blorp: Conditionally clear full surface depth and stencil (details / githubweb)
956. radeonsi: simplify NGG culling enablement and add radeonsi_shader_culling option (details / githubweb)
957. radeonsi: kill disabled clip distances and planes at per-channel granularity (details / githubweb)
958. radeonsi: move si_set_active_descriptors_for_shader into si_update_common_shader_state (details / githubweb)
959. radeonsi: use staging buffer uploads for most VRAM buffers (details / githubweb)
960. radeonsi: call nir_lower_bool_to_int32 last because it breaks nir_opt_if (details / githubweb)
961. radeonsi: restructure si_pipe_set_constant_buffer (details / githubweb)
962. intel/gen9: Enable MSC RAW Hazard Avoidance (details / githubweb)
963. gallivm/nir: make sure to mask global reads. (details / githubweb)
964. llvmpipe/cs: add in shader shared size. (details / githubweb)
965. gallivm/nir: fix non-32 bit find lsb/msb (details / githubweb)
966. nir: Split NIR_INTRINSIC_TYPE into separate src/dest indices (details / githubweb)
967. nir: Add a conversion and rounding intrinsic (details / githubweb)
968. nir: Add builder helpers for OpenCL type conversions (details / githubweb)
969. nir: Add a passes for nir_intrinsic_convert_alu_types (details / githubweb)
970. spirv: Add some conversion handling helpers (details / githubweb)
971. spirv: Handle all OpenCL conversion ops with full rounding (details / githubweb)
972. spirv/opencl: Drop dest_type from handle_v_load_store (details / githubweb)
973. spirv: Implement vload[a]_half[n] and vstore[a]_half[n][_r] (details / githubweb)
974. clover/nir: Call nir_lower_convert_alu_types (details / githubweb)
975. nir: Add lowering from regular ALU conversions to the intrinsic (details / githubweb)
976. android: gallium/iris: cleanup iris_driinfo.h gen rules (details / githubweb)
977. android: gallium/radeonsi: cleanup si_driinfo.h gen rules (details / githubweb)
978. android: gallium/virgl: cleanup virgl_driinfo.h gen rules (details / githubweb)
979. Revert F16C series (MR 6774) (details / githubweb)
980. virgl: add flags to (*resource_create) callback (details / githubweb)
981. drm-uapi: virtgpu_drm.h: resource create blob + host visible memory region (details / githubweb)
982. virgl/drm: query for resource blob and host visible memory region (details / githubweb)
983. virgl/drm: add resource create blob function (details / githubweb)
984. virgl: support PIPE_CAP_BUFFER_MAP_PERSISTENT_COHERENT (details / githubweb)
985. virgl: query blob mem (details / githubweb)
986. virgl: fix stride + layer_stride inconsistency (details / githubweb)
987. intel/fs: NoMask initialize the address register for shuffles (details / githubweb)
988. freedreno/cffdec: fix decoding of bindless descriptors (details / githubweb)
989. glsl: don't duplicate state vars as uniforms in the NIR linker (details / githubweb)
990. nir: Fix a misspelling (details / githubweb)
991. ci/bare-metal: Move the "POWER_GOOD not seen in time" check to the right time. (details / githubweb)
992. turnip: add environment variable to disable LRZ (details / githubweb)
993. turnip: create LRZ buffer (details / githubweb)
994. turnip: disable LRZ on specific cases (details / githubweb)
995. turnip: disable LRZ writes when blend is enabled (details / githubweb)
996. turnip: disable LRZ depending on fragment changes (details / githubweb)
997. turnip: add LRZ tracking to command buffer state (details / githubweb)
998. turnip: add LRZ valid tracking for secondary command buffers (details / githubweb)
999. turnip: add support to clear LRZ (details / githubweb)
1000. turnip: emit correct LRZ fast clear setup (details / githubweb)
1001. turnip: disable LRZ on vkCmdClearAttachments() (details / githubweb)
1002. turnip: disable LRZ on vkCmdClearattachments() 3D fallback path (details / githubweb)
1003. turnip: enable LRZ (details / githubweb)
1004. turnip: don't initialize GRAS_LRZ_CNTL/RB_LRZ_CNTL tu6_init_hw() (details / githubweb)
1005. meson: fix power8 option (details / githubweb)
1006. gallium/util: fix memory-leak (details / githubweb)
1007. gallium/util: allow scissored blits for stencil-fallback (details / githubweb)
1008. nir/find_array_copies: Properly discard copies for casts (details / githubweb)
1009. nir: Handle memcpy in copy_prop_vars and combine_stores (details / githubweb)
1010. nir: Add a memcpy optimization pass (details / githubweb)
1011. nir/opt_load_store_vectorize: Use bit sizes when checking mask compatibility (details / githubweb)
1012. nir: Add component mask re-interpret helpers (details / githubweb)
1013. nir/opt_deref: Add an instruction type switch (details / githubweb)
1014. nir/opt_deref: Add an optimization for bitcasts (details / githubweb)
1015. nir: Add a pass to lower vec3s to vec4s (details / githubweb)
1016. aco: fix get_buffer_resource_flags() (details / githubweb)
1017. Update debian/changelog (details / githubweb)
1018. Rebase llvm7 patch. (details / githubweb)
1019. Sync debian directory with upstream experimental. (details / githubweb)
1020. Update debian/changelog. (details / githubweb)
1021. Lover libglvnd-dev version dependency. (details / githubweb)
1022. Fix patch name. (details / githubweb)
1023. Rebase llvm7 patch for latest upstream. (details / githubweb)
1024. Add missing libxcb-shm0-dev dependency. (details / githubweb)

Started by user Ivan J.

• refs/remotes/origin/maemo/beowulf