aboutsummaryrefslogtreecommitdiff
path: root/llvm/test/Transforms
AgeCommit message (Expand)AuthorFilesLines
6 hoursReland "[DebugMetadata][DwarfDebug] Support function-local types in lexical b...Vladislav Dzhidzhoev2-0/+150
7 hours[ArgPromotion] Add DW_CC_nocall to DISubprogram (#178973)yonghong-song1-1/+16
8 hours[VPlan] Refine exit select check in transformtoPartialReduction.Florian Hahn1-0/+54
11 hoursReapply "[InstCombine] Always fold alignment assumptions into operand bundles...Nikolas Klauser3-61/+47
11 hours[InstCombine] fold icmp ne (and X, 1), 0 --> trunc X to i1 (#178977)Andreas Jonson19-242/+209
12 hoursInstCombine: Handle minnum/maxnum in SimplifyDemandedFPClass (#179299)Matt Arsenault3-351/+222
12 hours[InstCombine] Fold select of intrinsic into intrinsic of select (#178002)Gauravsingh Sisodia1-0/+153
13 hours[InstCombine] Extend canonicalization of addition to positive numbers (#179343)SiliconA-Z1-0/+135
15 hours[SLP]Disable modeling disjoint reduction or as bitcast for big endianAlexey Bataev1-0/+261
16 hoursReland "[CoroCleanup] Noop coroutine elision for load-and-call pattern (#1791...Weibo He2-24/+51
17 hours[SLP][NFC]Add another test for shl-to-add transformation, NFCAlexey Bataev1-0/+34
18 hoursRevert "[SeparateConstOffsetFromGEP] Decompose constant xor operand if possib...Eli Friedman2-438/+2
21 hours[VPlan] Sink recipes from the vector loop region in licm. (#168031)Mel Chen49-432/+439
21 hoursAttributor: Add denormal-fp-math to attributor-light (#79576)Matt Arsenault1-0/+1
29 hours[InstCombine] Preserve fdiv metadata on fneg folds (#179157)Wenju He1-0/+33
31 hours[Matrix] Add test where pointer phi currently blocks tiling.Florian Hahn1-0/+49
35 hours[SLP][NFC]Add a case for missed shl-to-add transformation, NFCAlexey Bataev1-0/+76
35 hoursInstCombine: Fix SimplifyDemandedFPClass bug with known-snan sources (#179244)Matt Arsenault13-28/+55
35 hoursRevert "[VectorCombine] Trim low end of loads used in shufflevector rebroadca...Hans Wennborg1-46/+23
35 hoursInstCombine: Stop using nsz in multi-use min/max fold (#176579)Matt Arsenault4-8/+8
36 hoursValueTracking: Use computeKnownBits for ldexp integer handling (#179234)Matt Arsenault1-2/+2
36 hours[LoopUnroll] Remove preceding whitespace in loop peeling optimization remark ...Justin Fargnoli1-5/+5
36 hoursInstCombine: Add baseline tests for broken snan handling (#179243)Matt Arsenault20-2/+579
36 hoursRevert "[CoroCleanup] Noop coroutine elision for load-and-call pattern" (#179...Weibo He2-51/+24
37 hours[NFC][LoopUnroll] Add `-unroll-runtime-other-exit-predictable=false` to `unro...Marek Sedláček1-2/+2
39 hours[CoroCleanup] Noop coroutine elision for load-and-call pattern (#179154)Weibo He2-24/+51
2 days[IR] Add `fpmath` to keep list of dropUBImplyingAttrsAndMetadata (#179019)Wenju He1-0/+9
2 days[Matrix] Update test to make sure tiled loops can be used. (NFC)Florian Hahn1-1702/+2816
2 days[VPlan] Fold (x | !x) -> true. (#177887)Florian Hahn3-7/+42
2 days[SLP][NFC]Add a test to check modeling shl x, 1 as add x,x, NFCAlexey Bataev1-0/+64
3 daysReapply "[VPlan] Detect and create partial reductions in VPlan. (NFCI) (#1678...Florian Hahn2-0/+48
3 daysRevert "[VPlan] Detect and create partial reductions in VPlan. (NFCI) (#167851)"Martin Storsjö1-1/+0
3 days[VPlan] Detect and create partial reductions in VPlan. (NFCI) (#167851)Florian Hahn1-0/+1
4 days[VectorCombine] Fix the PtrAdd offset in shrinkLoadForShuffles to account for...Deric C.1-10/+10
4 daysVectorCombine: Mark the test `+asserts` (fixup for #178072)NAKAMURA Takumi1-0/+1
4 daysAttributor: Use anchor scope for SimplifyQuery context (#178958)Matt Arsenault1-38/+80
4 days[LV] Add additional partial reduction test coverage for #167851.Florian Hahn3-113/+228
4 daysReapply "[VPlan] Add hidden `-vplan-print-after-all` option" (#178547)Andrei Elovikov1-0/+67
4 days[VectorCombine] Fix crash with poison mask elements in shrinkLoadForShuffles ...puneeth_aditya_56561-0/+14
4 days[Matrix] Add test showing excessive matrix codegen.Florian Hahn1-0/+2065
5 days[SLP]Cast incoming value to a propr type for int nodes, bitcasted to fpAlexey Bataev1-0/+52
5 daysRevert "[InstCombine] Always fold alignment assumptions into operand bundles ...Nico Weber3-19/+60
5 daysInstCombine: Handle multiple use copysign (#176917)Matt Arsenault1-7/+7
5 days[VPlan] Mark VPActiveLaneMaskPHIRecipe as readnone. (#177886)Florian Hahn3-31/+85
5 days[VectorCombine] Trim low end of loads used in shufflevector rebroadcasts. (#1...Leon Clark1-23/+32
5 daysNFC: Cleanup AArch64/partial-reduce-chained.llSander de Smalen1-15/+16
5 daysInstCombine: Handle nsz in copysign SimplifyDemandedFPClass (#176916)Matt Arsenault1-4/+2
5 daysInstCombine: Add baseline tests for SimplifyDemandedFPClass copysign improvem...Matt Arsenault1-0/+651
5 daysInstCombine: Improve single-use fneg(fabs(x)) SimplifyDemandedFPClass handlin...Matt Arsenault1-11/+265
5 days[VectorCombine] Fix typo in foldPermuteOfBinops cost calculation (#178072)calebwat1-0/+18
generated by cgit v1.1 at 2026-02-04 05:10:02 +0000