aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/Target/AMDGPU
AgeCommit message (Expand)AuthorFilesLines
8 hours[AMDGPU][NFC] Correct description (#112847)Mariusz Sikora1-2/+2
11 hours[llvm] Consistently respect `naked` fn attribute in `TargetFrameLowering::has...Alex Rønne Petersen3-6/+6
25 hours[SimplifyLibCall][Attribute] Fix bug where we may keep `range` attr with inco...goldsteinn1-1/+2
26 hours[LLVM] Make more use of IRBuilder::CreateIntrinsic. NFC. (#112706)Jay Foad7-54/+28
33 hours[AMDGPU] Factor out getNumUsedPhysRegs(). NFC. (#112624)Stanislav Mekhanoshin3-38/+18
34 hours[APInt] Fix APInt constructions where value does not fit bitwidth (NFCI) (#80...Nikita Popov2-3/+4
2 days[AMDGPU][True16][CodeGen] fp conversion in true/fake16 format (#101678)Brox Chen3-39/+95
2 days[LLVM] Make more use of IRBuilder::CreateIntrinsic. NFC. (#112546)Jay Foad1-3/+2
2 days[AMDGPU][True16][MC] VOP3 profile in True16 format (#109031)Brox Chen4-78/+330
2 days[LLVM] Add `Intrinsic::getDeclarationIfExists` (#112428)Rahul Joshi3-5/+4
2 days[AMDGPU][NewPM] Fill out addILPOpts. (#108514)Christudasan Devadasan2-0/+8
2 days[CodeGen][NewPM] Port EarlyIfConversion pass to NPM. (#108508)Christudasan Devadasan1-1/+1
2 daysAMDGPU/GlobalISel: Run redundant_and combine in RegBankCombiner (#112353)Petar Avramovic1-1/+2
3 daysRevert "[AMDGPU] Serialize WWM_REG vreg flag (#110229)"Peter Collingbourne3-29/+0
3 days[AMDGPU] Avoid repeated hash lookups (NFC) (#112309)Kazu Hirata1-4/+2
3 days(reland) [AMDGPU][SplitModule] Handle !callees metadata (#108802)Pierre van Houtryve1-15/+48
4 days[AMDGPU] Tidy SIPreAllocateWWMRegs after recent changes (NFCI) (#111967)Carl Ritson1-6/+3
4 daysRevert "[AMDGPU][SplitModule] Handle !callees metadata (#108802)"Nico Weber1-48/+15
4 days[AMDGPU] Skip terminators when forcing emit zero flag (#112116)Shilei Tian1-1/+3
4 days[AMDGPU] Use MachineInstr::mayLoadOrStore. NFC.Jay Foad1-1/+1
4 days[AMDGPU] Serialize WWM_REG vreg flag (#110229)Akshat Oke3-0/+29
4 days[AMDGPU][SplitModule] Handle !callees metadata (#108802)Pierre van Houtryve1-15/+48
4 days[NFC] clang-format llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cppShilei Tian1-10/+9
4 days[NFC][AMDGPU] Remove unnecessary member `ForceEmitZeroWaitcnts` (#112114)Shilei Tian1-5/+1
5 days[AMDGPU] Avoid repeated hash lookups (NFC) (#112115)Kazu Hirata1-4/+2
7 daysRevert "[AMDGPU] Avoid resource propagation for recursion through multiple fu...Janek van Oirschot1-83/+7
7 days[AMDGPU] Fix -Wunused-variable in non-assertion builds after #109818Fangrui Song1-4/+1
7 days[AMDGPU][SIPreEmitPeephole] mustRetainExeczBranch: use BranchProbability and ...Juan Manuel Martinez Caamaño1-21/+68
7 days[AMDGPU] Avoid resource propagation for recursion through multiple functions ...Janek van Oirschot1-7/+83
7 days[TTI][AMDGPU] Allow targets to adjust `LastCallToStaticBonus` via `getInlinin...Shilei Tian2-0/+6
7 days[AMDGPU] Rewrite RegSeqNames using !foreach. NFC. (#111994)Jay Foad1-9/+4
7 days[NFC] Rename `Intrinsic::getDeclaration` to `getOrInsertDeclaration` (#111752)Rahul Joshi12-63/+69
7 daysAMDGPU/GlobalISel: Fix inst-selection of ballot (#109986)Petar Avramovic1-25/+76
7 days[AMDGPU] Enable unaligned scratch accesses (#110219)Fabian Ritter4-27/+21
7 days[AMDGPU][SplitModule] Cleanup CallsExternal Handling (#106528)Pierre van Houtryve1-22/+63
8 daysAMDGPU: Minor improvement and cleanup for waterfall loop generation (#111886)Changpeng Fang1-8/+10
8 days[AMDGPU] Improve codegen for s_barrier_init (#111866)Jay Foad1-3/+1
8 days[AMDGPU] Avoid repeated hash lookups (NFC) (#111787)Kazu Hirata1-4/+6
8 daysRevert "Reapply "[AMDGPU][GlobalISel] Fix load/store of pointer vectors, buff...Mikhail Goncharov3-48/+27
8 days[AMDGPU][NewPM] Fill out AMDGPU addMachineSSAOptimizations (#111658)Akshat Oke2-0/+22
8 daysAMDGPU/GlobalISel: Insert m0 initialization before sextload/zextload (#111720)Matt Arsenault1-0/+2
9 daysReapply "[AMDGPU][GlobalISel] Fix load/store of pointer vectors, buffer.*.pN ...Krzysztof Drewniak3-27/+48
9 days[TTI] NFC: Port TLI.shouldSinkOperands to TTI (#110564)Jeffrey Byrnes4-22/+22
9 daysAMDGPU: Fix incorrectly selecting fp8/bf8 conversion intrinsics (#107291)Matt Arsenault3-5/+17
9 days[AMDGPU] Fix expensive checkJeffrey Byrnes1-1/+1
9 days[AMDGPU] Optionally Use GCNRPTrackers during scheduling (#93090)Jeffrey Byrnes5-74/+431
9 daysAMDGPU: Add instruction flags when lowering ctor/dtor (#111652)Matt Arsenault1-3/+7
9 daysAMDGPU: Use pointer types more consistently (#111651)Matt Arsenault1-9/+7
9 daysAMDGPU: Avoid using hardcoded address space numberMatt Arsenault1-2/+2
9 daysAMDGPU: Remove flat/global fmin/fmax intrinsics (#105642)Matt Arsenault6-43/+0
generated by cgit v1.1 at 2024-10-18 16:52:01 +0000