summaryrefslogtreecommitdiff
BranchCommit messageAuthorAge
11.0-meta-cannot-use-glGenBuffersmeta/TexSubImage: Don't pollute the buffer object namespaceIan Romanick8 years
11.1-meta-cannot-use-glGenBuffersmeta: Don't save or restore the active client textureIan Romanick8 years
6b452c51941fixup! WIP: nir: Copy propagate from bcsel source to phi nodeIan Romanick5 years
ARB_transform_feedback_overflow_querymesa: Track query objects for transform feedback overflow queriesIan Romanick7 years
CID-1429517mesa: Fix tautological compareIan Romanick6 years
EXT_shader_samples_identicali965/vec4: Implement ir_samples_identicalIan Romanick8 years
INTEL_shader_atomic_float_minmaxi965: Enable INTEL_shader_atomic_float_minmax on Gen9+Ian Romanick6 years
NV_shader_atomic_int64haxIan Romanick6 years
NaNu-NaNuWIP: nir/algebraic: Add algebraic opt for fge comparison with identical opera...Ian Romanick3 years
always-ARB_framebuffer_objectmesa: Remove _EXT suffixes from FBO related enumsIan Romanick5 years
arb_gpu_shader_int64st/mesa: add support for enabling ARB_gpu_shader_int64.Dave Airlie7 years
asmir_to_mesa: Don't set component for ir_dereference in ir_quadop_vectorIan Romanick12 years
assume-CMS-in-precompilemisc algabraIan Romanick8 years
ast_to_hir-workglsl: Refactor type checking for redeclarationsIan Romanick6 years
bool-optimizationsi965/fs: Emit smarter code for b2fIan Romanick10 years
bool-optimizations-v2i965/fs: Emit different code for b2fIan Romanick9 years
bool-optimizations-v4i965/vs: Emit AND instead of SEL if the false value is zero on SNB+Ian Romanick9 years
bool-optimizations-v5glsl: Distribute ir_unop_neg over ir_binop_mul of a constantIan Romanick9 years
bool-optimizations-v6glsl: Rewrite ir_triop_csel when condition is ir_unop_logic_notIan Romanick9 years
byt-cpu-optimizationsmesa: Micro-optimize _mesa_is_valid_prim_modeIan Romanick9 years
c1896cdfefcnir/algebraic: Recognize open-coded flrp(a, b, a)Ian Romanick5 years
cmod-barglsl: Optimize certain if-statements to just casts from the conditionIan Romanick9 years
cmod-bazi965: Convert CMP.GE -(abs)reg 0 -> CMP.Z reg 0.Matt Turner9 years
cmod-fooglsl: Rewrite (-abs(x) >= 0) as (x == 0)Ian Romanick9 years
cmod-propagation-worki965/fs: Propagate conditional modifiers from not instructions to compare ins...Ian Romanick6 years
combine-constantsherp derpIan Romanick4 years
commutative-expression-limitsnir/serach: Don't retry a commutative op with duplicate sourcesIan Romanick5 years
compare-with-integer-frobbingi965: nir_opt_constant_folding one more timeIan Romanick6 years
cse-negglsl/cse: Restore the previous ae list after existing a nested blockIan Romanick9 years
cse-neg-v2018.02i965/fs: Optimize a-a to zeroIan Romanick6 years
daniel-schuermann/mesa-bitfieldamd/common: lower bitfield_extract to ubfe/ibfe.Daniel Schürmann5 years
derpnir: implement the GLSL equivalent of if simplication in nir_opt_ifSamuel Pitoiset6 years
double-schedulerremove SIMD16 heuristicIan Romanick6 years
dp-narrowingadd modificationsIan Romanick5 years
ecd3c4bb555fixup! WIP: nir: Copy propagate from bcsel source to phi nodeIan Romanick5 years
emit-spirvglsl/spirv: Emit ir_emit_vertex and ir_end_primitiveIan Romanick6 years
experiment/Gen5-Boolean-resolveWIP: intel/compiler: Don't require resolve when a Boolean is only used by dis...Ian Romanick5 years
experiment/fsat-algebraicWIP: nir/algebraic: fsat(fsat(a)*fsat(b)) -> fsat(a)*fsat(b)Ian Romanick5 years
experiment/lower-flrpWIP: nir/algebraic: Reassociate open-coded flrp(-1, b, c)Ian Romanick5 years
experiment/nan-compare-logic-opsWIP: intel/fs: Optimize some SEL instructions to predicate MOVIan Romanick4 years
experiment/optimize-bfesquash! intel/fs: Emit better code for addition of extractIan Romanick4 years
experiment/optimize-shiftsWIP: nir/algebraic: Reassociate shift-by-constant of shift-of-constantIan Romanick5 years
experiment/rematerialize-comparesnir: Rematerialize comparisons for nir_intrinsic_discard_if tooIan Romanick5 years
experiment/simple-range-analysisWIP: nir/algebraic: Let the open-coded-{fmax,fmin} optimizations have precedenceIan Romanick5 years
experimentsi965/fs: Invert if-condition for nir_op_inotIan Romanick6 years
expose-fewer-visualsWIP: gallium/dri2: Don't expose accumulation buffer with RGBA visualIan Romanick4 years
extension-refactormesa: Always enable GL_ARB_vertex_programIan Romanick10 years
fix-GL45-CTS.program_interface_query.uniform-block-typesnir: Trivial clean ups in the generated nir_constant_expressions.cIan Romanick7 years
flexarrayglsl: flexarray.Kenneth Graunke12 years
flexarray-funcstashKenneth Graunke12 years
for-bsinir/algebraic: Reassociate some address arithmeticIan Romanick4 years
fortran965 haxKenneth Graunke13 years
generate-code-generatorWIP: nir/algebraic: weird flrp to bcsel stuffIan Romanick4 years
glsl-diet-v3mesa: Free the compiled shader IR after it has been linked.Eric Anholt10 years
glsl-diet-v3-extra-loggingglsl: Log information about variable names that exist after linkingIan Romanick10 years
glsl-diet-v4glsl: Add Bloom filter to get_static_nameIan Romanick10 years
i915-fixesir_to_mesa: Print Mesa IR again after running _mesa_optimize_programIan Romanick7 years
i965-license-headeri965: Re-wrap lines in the license, part 2Ian Romanick9 years
inverted-conditionsnir/algebraic: Remove comparison simplifications with different NaN behaviorIan Romanick5 years
ir_variable-dietglsl: Dynamically allocate ir_variable interface data storageIan Romanick10 years
ir_variable-diet-v2glsl: Rebuild the symbol table without unreachable symbolsIan Romanick10 years
iris-threaded-shadersWIP: iris: Enable threaded shader compilationIan Romanick3 years
jenkinsnir/algebraic: Distribute source modifiers into instructionsIan Romanick5 years
jenkins-vkspirv: Generate spirv_info.cIan Romanick7 years
kill-nv04nouveau: Move common extension enables to nouveau_context_initIan Romanick8 years
lock_surfaceegl: Some support for various config flags used by EGL_KHR_lock_surfaceIan Romanick11 years
logic-expression-frobbingBoolean simplifications, part 2Ian Romanick6 years
loopglsl: Skip type conversions in loop-size estimates on float-only arches.Kenneth Graunke12 years
lower_bool_to_int32-earlierWIP: intel/compiler: Move nir_lower_bool_to_int32 earlierIan Romanick5 years
masterintel/compiler: Fix nir_op_b2[fi] with 64-bit result on Gen8 LP and Gen9 LPJason Ekstrand6 years
meta-cannot-use-Gen-v2017.12more debug infoIan Romanick6 years
meta-fixesi965: Enable EWA anisotropic filtering algorithmIan Romanick10 years
meta-msaa-blitmeta/blit: Use GL_EXT_shader_samples_identical in MSAA-SS resolve blitIan Romanick8 years
meta-work-20170628meta: Make meta_alloc_texture actually allocate a textureIan Romanick7 years
misc-algebraWIP: nir/algebraic: Replace signed range [0, constant b] checks with unsigned...Ian Romanick4 years
nir_lower_clamp_color_outputs-vs-b2fnir/algebraic: Optimize away an fsat of a b2fIan Romanick5 years
opt-minimize-BooleanWIP: nir/opt_bool: Factor out common terms differentlyIan Romanick6 years
optimize-bfeWIP: nir/algebraic: Combine right shift with bitfield extractIan Romanick4 years
optimize-dph-and-othersreassociate ffma of fmul, lateIan Romanick5 years
optimize-flrp-and-ffmaWIP: nir/algebraic: Evaluate [iu]{min,max} with maximum or minimum constantsIan Romanick5 years
paulimeta: texture rectangle textures may not have mipmapsPauli Nieminen12 years
peephole_selectWIP: intel/fs: Optimize some SEL instructions to predicate MOVIan Romanick4 years
pragma-warningglsl: Add pragma to disable all warningsIan Romanick5 years
ptn-narrowingir_to_mesa: Emit smarter ir_binop_logic_or for vertex programsIan Romanick8 years
query-rendererglx/tests: Add unit tests for the DRI2 part of GLX_MESA_query_rendererIan Romanick11 years
range-analysis-cachenir/range-analysis: Use types to provide better ranges from bcsel and movIan Romanick5 years
reassociate-vector-scalar-mathnir: Mark fmax and fmin as associativeIan Romanick5 years
remove-ir_assignment-conditionglsl: Simplify ir_assignment::cloneIan Romanick7 years
reverse-code-geni965: Use De Morgan's laws to avoid logical-not of a logic result on Gen8+Ian Romanick7 years
review/issue-6365WIP: Use NIR lowering for tg4 offset arrays instead of GLSL loweringIan Romanick23 months
review/lower-flrpRFC: intel/compiler: Don't always require precise lowering of flrpIan Romanick5 years
robustnessdri: Allow __DRI_CTX_FLAG_ROBUST_BUFFER_ACCESS in driCreateContextAttribsIan Romanick10 years
shader-workglsl/loop_unroll: unroll loops with cond breaks anywhere, not just the endLuca Barbieri13 years
source-modifiersWIP: nir/algebraic: fmin / fmax of fabs to fsatIan Romanick5 years
sso4wip2Ian Romanick11 years
sso5mesa: Eliminate gl_shader_program::InternalSeparateShaderIan Romanick10 years
sso6docs: update ARB_buffer_storage for nouveauIlia Mirkin10 years
standalone-compilerglsl_compiler: Enable any extension that any Mesa driver enablesIan Romanick11 years
standalone-ir_builderglsl/standalone: Add the ability to generate ir_builder codeIan Romanick8 years
standalone-ir_builder-v2glsl/standalone: Add the ability to generate ir_builder codeIan Romanick8 years
t_dd_dmatmpglsl: In later GLSL versions, sequence operator is cannot be a constant expre...Ian Romanick9 years
ud-chainsglsl: Add optimization pass to trim unreachable portions of the CFGIan Romanick13 years
uniform-settersmarter blendIan Romanick9 years
uniform-setter-20190513mesa: Add SSE2 paths for _mesa_Uniform*fIan Romanick5 years
uniform-setter-fastno loggingIan Romanick10 years
uniform-setter-v2mesa: Add SSE2 paths for _mesa_Uniform*fIan Romanick9 years
uniform-setter-wipsmarter blendIan Romanick9 years
vertex-shader-gcmintel/compiler: Enable GCM for scalar vertex stageIan Romanick4 years
viewport-arrayset all scissor rectsIan Romanick10 years
wip/combine-constantsfixup! intel/fs: Combine constants for integer instructions tooIan Romanick2 years
wip/fix-shift-countsWIP: intel/fs: Fix shift counts for 8- and 16-bit typesIan Romanick12 months
wip/improve-range-analysisnir/range-analysis: Improve "is a number" range analysis for fexp2 and fpowIan Romanick3 years
yuv-formatsanv: Packed, subsampled formats are not disjointIan Romanick3 years
 
generated by cgit v1.2.3 (git 2.39.1) at 2024-04-26 17:40:21 +0000