Jameson Nash bc7ea63e9c [MemCpyOpt] handle memcpy from memset for non-constant sizes (#143727) ...

Allows forwarding memset to memcpy for mismatching unknown sizes if
overread has undef contents. In that case we can refine the undef bytes
to the memset value.

Refs #140954 which laid some of the groundwork for this.

2025-06-11 20:04:27 -04:00

..

ADCE

[IR] Relax convergence requirements on call (#135794)

2025-05-02 08:17:58 +00:00

AddDiscriminators

[llvm] Remove undef from llvm/test/Transforms tests (#123889)

2025-01-22 14:47:15 +00:00

AggressiveInstCombine

[AggressiveInstCombine] Improve popcount matching if the input has known zero bits (#142501)

2025-06-03 12:56:53 -07:00

AlignmentFromAssumptions

[AMDGPU] Set AS8 address width to 48 bits

2025-05-19 17:26:05 -07:00

ArgumentPromotion

[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)

2025-03-14 15:50:49 +00:00

AtomicExpand

[AMDGPU][AtomicExpand] Use full flat emulation if a target supports f64 global atomic add instruction (#142859)

2025-06-05 00:45:42 -04:00

Attributor

[ValueTracking] Support scalable vector splats of ConstantInt/ConstantFP in isGuaranteedNotToBeUndefOrPoison. (#142894)

2025-06-05 22:08:03 -07:00

BDCE

[IR] Relax convergence requirements on call (#135794)

2025-05-02 08:17:58 +00:00

BlockExtractor

…

BranchFolding

…

CalledValuePropagation

…

CallSiteSplitting

Fix regression tests with bad FileCheck checks (#140373)

2025-05-22 07:59:57 +03:00

CanonicalizeAliases

…

CanonicalizeFreezeInLoops

…

CodeExtractor

[CodeExtractor] Improve debug info for input values. (#136016)

2025-04-26 10:12:44 +01:00

CodeGenPrepare

[CGP] Bail out if (Base|Scaled)Reg does not dominate insert point. (#142949)

2025-06-06 12:38:30 +01:00

ConstantHoisting

[RISCV] Correct RISCVTTIImpl::getIntImmCostInst for Zba (#128174)

2025-02-27 08:48:55 +01:00

ConstantMerge

…

ConstraintElimination

[ConstraintElim] Do not allow overflows in Decomposition (#140541)

2025-05-22 11:31:04 +08:00

Coroutines

[debuginfo][coro] Fix linkage name for clones of coro functions (#141889)

2025-06-11 13:50:32 +02:00

CorrelatedValuePropagation

[CVP] Keep ReachableCaseCount in sync with range of condition (#142302)

2025-06-02 17:42:02 +08:00

CrossDSOCFI

…

DCE

[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)

2025-03-14 15:50:49 +00:00

DeadArgElim

[DeadArgElim] fix verifier failure when changing musttail's function signature (#127366)

2025-04-10 07:08:09 -07:00

DeadStoreElimination

[IR][DSE] Support non-malloc functions in malloc+memset->calloc fold (#138299)

2025-06-04 09:35:20 +02:00

DFAJumpThreading

[DFAJumpThreading] Don't bail early after encountering unpredictable values (#119774)

2024-12-25 01:29:01 -08:00

DivRemPairs

…

DXILUpgrade

…

EarlyCSE

[EarlyCSE] Check attributes for commutative intrinsics (#142610)

2025-06-04 09:04:27 +02:00

EliminateAvailableExternally

[ctxprof] Only prune the profile in modules containing only context trees (#134340)

2025-04-07 19:52:03 -07:00

EmbedBitcode

[llvm][EmbedBitcodePass] Prevent modifying the module with ThinLTO (#139999)

2025-05-29 13:42:48 -07:00

EntryExitInstrumenter

EntryExitInstrumenter: skip available_externally linkage

2025-01-03 09:25:08 -08:00

ExpandLargeDivRem/X86

…

ExpandLargeFpConvert/X86

Rename ExpandLargeFpConvertPass to ExpandFpPass (#131128)

2025-03-14 13:11:45 +01:00

ExpandMemCmp

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

ExpandVariadics

…

FatLTOCleanup

[llvm][fatlto] Add FatLTOCleanup pass (#125911)

2025-02-13 11:39:02 -08:00

FixIrreducible

[AMDGPU] Fix for ControlFlowUtils reconnectPHIs. (#132954)

2025-04-09 13:58:33 +02:00

Float2Int

…

ForcedFunctionAttrs

ForceFunctionAttrs: Use reportFatalUsageError (#139473)

2025-05-15 15:00:05 +02:00

FunctionAttrs

Reland "[NVPTX] Unify and extend barrier{.cta} intrinsic support" (#141143)

2025-05-22 19:38:10 -07:00

FunctionImport

…

FunctionSpecialization

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

GCOVProfiling

[gcov,test] Update exit-block.ll now that exit block is always the second

2025-01-04 10:56:45 -08:00

GlobalDCE

…

GlobalMerge

[GlobalMerge] Fix handling of const options

2025-03-18 15:06:39 +01:00

GlobalOpt

[GlobalOpt] Bail out on non-ConstExprs in isSimpleEnoughtToCommit. (#143400)

2025-06-11 22:09:55 +01:00

GlobalSplit

…

GuardWidening

…

GVN

[llvm][GVN] Propagate trunc nuw to i1 equalities (#143273)

2025-06-08 18:58:16 +02:00

GVNHoist

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

GVNSink

[llvm][gvn-sink] Don't try to sink inline asm (#138414)

2025-05-05 18:16:33 -07:00

HardwareLoops

…

HelloNew

…

HipStdPar

[HIP][HIPSTDPAR] Re-work allocation interposition for hipstdpar (#138790)

2025-05-07 14:14:25 +01:00

HotColdSplit

[CodeExtractor] Improve debug info for input values. (#136016)

2025-04-26 10:12:44 +01:00

IndirectBrExpand

…

IndVarSimplify

[IndVars] Teach widenLoopCompare to use sext if narrow IV is positive and other operand is already sext. (#142703)

2025-06-10 12:52:39 -07:00

InferAddressSpaces

Reapply "Reapply "[AMDGPU] Make getAssumedAddrSpace return AS1 for pointer kernel arguments (#137488)""

2025-05-30 22:11:22 -04:00

InferAlignment

…

InferFunctionAttrs

[LibCall] Infer nocallback for libcalls (#135173)

2025-04-12 15:11:54 +08:00

Inline

[InlineCost]: Optimize inlining of recursive function. (#139982)

2025-06-04 11:41:55 +01:00

InstCombine

[InstCombine] Fold max/min when incrementing/decrementing by 1 (#142466)

2025-06-10 07:55:56 -07:00

InstNamer

…

InstSimplify

[ConstantFolding] Add folding for [de]interleave2, insert and extract (#141301)

2025-06-11 19:00:29 -04:00

InterleavedAccess

[IA] Add support for [de]interleave{4,6,8} (#141512)

2025-05-28 11:44:41 +01:00

Internalize

…

IRCE

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

IRNormalizer

IRNormalizer: Replace cl::opts with pass parameters (#133874)

2025-04-01 23:27:20 +07:00

IROutliner

[DebugInfo][RemoveDIs] Use autoupgrader to convert old debug-info (#143452)

2025-06-11 13:56:30 +01:00

JumpTableToSwitch

…

JumpThreading

Reland "[NVPTX] Unify and extend barrier{.cta} intrinsic support" (#141143)

2025-05-22 19:38:10 -07:00

KCFI

…

LCSSA

[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)

2025-03-14 15:50:49 +00:00

LICM

[LICM] Only check for provenance captures (#141731)

2025-05-28 11:57:30 +02:00

LoadStoreVectorizer

[NVPTX] Vectorize and lower 256-bit global loads/stores for sm_100+/ptx88+ (#139292)

2025-05-13 13:36:09 -07:00

LoopBoundSplit

…

LoopDataPrefetch

[RISCV] Enable prefetching writes (#130561)

2025-03-11 11:39:21 +08:00

LoopDeletion

[IndVarSimplify] Set samesign when converting signed comparison to unsigned comparison in eliminateIVComparison. (#138215)

2025-05-02 08:17:45 -07:00

LoopDistribute

[DebugInfo][LoopDistribute] Salvage debug values using dead instructions in the distributed loops (#137124)

2025-05-12 09:06:42 +08:00

LoopFlatten

…

LoopFusion

…

LoopIdiom

[SCEV] Add more tests with zext(add C, %var)<nsw>.

2025-06-03 22:03:20 +01:00

LoopInstSimplify

…

LoopInterchange

[LoopInterchange] Handle confused dependence correctly (#140709)

2025-06-05 16:44:02 +09:00

LoopLoadElim

[AMDGPU] Set AS8 address width to 48 bits

2025-05-19 17:26:05 -07:00

LoopPredication

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

LoopRotate

[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)

2025-03-14 15:50:49 +00:00

LoopSimplify

…

LoopSimplifyCFG

[LoopSimplifyCFG] Fix SCEV invalidation after removing dead exit (#127536)

2025-02-21 12:26:39 +01:00

LoopStrengthReduce

MachineScheduler: Improve instruction clustering (#137784)

2025-06-05 15:28:04 +08:00

LoopTransformWarning

…

LoopUnroll

[LoopPeel] Add tests for last iteration peeling of min/max intrinsics

2025-06-10 13:08:36 -07:00

LoopUnrollAndJam

[UnrollAndJam] Do not preserve loop nests if a loop was fully unrolled. (#133510)

2025-03-29 19:21:34 +00:00

LoopVectorize

[RISCV][TTI] Allow partial reduce with mismatched extends (#143608)

2025-06-10 15:26:54 -07:00

LoopVersioning

LoopVersioning: improve a test, regen with UTC (#122876)

2025-01-14 12:06:22 +00:00

LoopVersioningLICM

[LAA] Keep pointer checks on partial analysis (#139719)

2025-06-04 16:47:20 +01:00

LowerAtomic

Reland [llvm] Add support for llvm IR atomicrmw fminimum/fmaximum instructions (#137701)

2025-04-30 22:06:37 +01:00

LowerConstantIntrinsics

[LLVM][ConstantFold] Undefined values are not constant (#130713)

2025-03-12 09:50:50 -07:00

LowerExpectIntrinsic

…

LowerGlobalDestructors

…

LowerGuardIntrinsic

…

LowerIFunc

…

LowerInvoke

…

LowerMatrixIntrinsics

[Matrix] Propagate shape information through cast insts (#141869)

2025-06-10 12:15:41 -07:00

LowerSwitch

…

LowerTypeTests

Require x86 target for test.

2025-06-09 17:06:33 -07:00

LowerWidenableCondition

…

MakeGuardsExplicit

…

Mem2Reg

IR: Use poison in dropDroppableUse (#134576)

2025-04-07 14:59:34 +07:00

MemCpyOpt

[MemCpyOpt] handle memcpy from memset for non-constant sizes (#143727)

2025-06-11 20:04:27 -04:00

MemProfContextDisambiguation

[MemProf] Add dot graph dumping immediately after stack node update (#143025)

2025-06-05 13:49:11 -07:00

MergedLoadStoreMotion

[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)

2025-03-14 15:50:49 +00:00

MergeFunc

LowerTypeTests: Shrink check size by 1 instruction on x86.

2025-06-06 12:43:24 -07:00

MergeICmps

…

MetaRenamer

…

MoveAutoInit

…

NameAnonGlobals

…

NaryReassociate

[NVPTX] use untyped loads and stores where ever possible (#137698)

2025-05-10 08:26:26 -07:00

NewGVN

[NewGVN] Check intrinsic ID before accessing operands (#141571)

2025-05-30 07:57:01 +02:00

ObjCARC

[DebugInfo][RemoveDIs] Use autoupgrader to convert old debug-info (#143452)

2025-06-11 13:56:30 +01:00

OpenMP

[FIX] Update check lines of llvm/test/Transforms/OpenMP/remove_globalization.ll

2025-06-09 10:59:53 -04:00

PartialInlining

…

PartiallyInlineLibCalls

…

PGOProfile

[MemProf] Optionally save context size info on largest cold allocations (#142837)

2025-06-04 13:08:56 -07:00

PhaseOrdering

[PhaseOrdering] Add test for #139050 (NFC)

2025-06-06 17:50:59 +02:00

PlaceSafepoints

…

PreISelIntrinsicLowering

[LibCall] Infer nocallback for libcalls (#135173)

2025-04-12 15:11:54 +08:00

Reassociate

[Reassociate] Move Disjoint flag handling to OverflowTracking. (#140406)

2025-05-23 14:59:18 +01:00

Reg2Mem

[IR] Don't allow values of opaque type (#137625)

2025-04-30 15:01:00 +02:00

RelLookupTableConverter

[RelLookupTableConverter] Fix test cases

2025-06-01 22:36:20 +08:00

RewriteStatepointsForGC

GCStrategy: Use Twine properly for error message (#132760)

2025-04-08 21:57:29 +07:00

SafeStack

[clang][IR] Overload @llvm.thread.pointer to support non-AS0 targets (#132489)

2025-05-14 21:51:56 +01:00

SampleProfile

[SampleFDO] Match functions with the same base function name (#126688)

2025-03-24 10:11:55 -07:00

SandboxVectorizer

Reapply "[SandboxVec] Add a simple pack reuse pass (#141848)"

2025-06-05 09:14:17 -07:00

ScalarizeMaskedMemIntrin

…

Scalarizer

[DebugInfo][RemoveDIs] Remove debug-intrinsic printing cmdline options (#131855)

2025-04-01 14:27:11 +01:00

SCCP

[SCCP] Remove masking operations (#142736)

2025-06-04 22:31:08 +08:00

SeparateConstOffsetFromGEP

Revert "[SeparateConstOffsetFromGEP] Decompose constant xor operand if possible (#135788)"

2025-06-11 08:49:13 +09:00

SimpleLoopUnswitch

[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)

2025-03-14 15:50:49 +00:00

SimplifyCFG

[SimplifyCFG] Bail out on vector GEPs in passingValueIsAlwaysUndefined (#142526)

2025-06-04 12:37:30 +08:00

SimplifyTypeTests

LowerTypeTests: Shrink check size by 1 instruction on x86.

2025-06-06 12:43:24 -07:00

Sink

…

SLPVectorizer

[SLP]Check if the user node has instructions, used only outside

2025-05-29 10:09:49 -07:00

SpeculativeExecution

[RemoveDIs] Remove "try-debuginfo-iterators..." test flags (#130298)

2025-03-14 15:50:49 +00:00

SROA

[LLVM][SROA] Teach SROA how to "bitcast" between fixed and scalable vectors. (#130973)

2025-06-11 11:02:32 +01:00

StraightLineStrengthReduce

[NVPTX] Remove Float register classes (#140487)

2025-05-21 11:33:57 -07:00

StripDeadPrototypes

…

StripSymbols

[llvm-extract] Delete dead CG Profile edges (#134940)

2025-04-09 10:36:12 -07:00

StructurizeCFG

[StructurizeCFG] Stop setting DebugLocs in flow blocks (#139088)

2025-05-09 14:22:14 -04:00

TailCallElim

[DebugInfo][TailCallElim] Use ret DILocation for return value selects (#134825)

2025-04-09 12:48:59 +01:00

ThinLTOBitcodeWriter

Add -funique-source-file-identifier option.

2025-06-05 10:52:01 -07:00

TypePromotion

[TypePromotion] Do not zero-extend getelementptr indexes since signed

2025-04-05 16:49:05 +02:00

UnifyFunctionExitNodes

…

UnifyLoopExits

…

Util

[LLVM][VecLib] Refactor LIBMVEC integration to be target neutral. (#138262)

2025-05-07 11:05:25 +01:00

VectorCombine

[VectorCombine] Add test cases for scalarizing extracts of extends.

2025-06-05 10:08:07 +01:00

WholeProgramDevirt

[llvm] Use ABI instead of preferred alignment for const prop checks (#142500)

2025-06-04 10:57:59 -07:00

lower-builtin-allow-check-remarks.ll

…

lower-builtin-allow-check.ll

[ubsan] Parse and use <cutoffs[0,1,2]=70000;cutoffs[5,6,8]=90000> in LowerAllowCheckPass (#124211)

2025-01-27 20:08:53 -08:00