Search results

[X86] bcmp with zero not vectorized

backend:X86

missed-optimization

Status: Open.

#144861 In llvm/llvm-project;

· nikic opened on Jun 19, 2025

[VectorCombine] foldShuffleOfShuffles - failure to fold shuffle(shuffle(x,y),undef) length changing masks

llvm:instcombine

missed-optimization

Status: Open.

#144656 In llvm/llvm-project;

· RKSimon opened on Jun 18, 2025

[VectorCombine] Recognise reduction patterns from extract+binop+shuffle chains

llvm:instcombine

missed-optimization

Status: Open.

Feature

#144654 In llvm/llvm-project;

· RKSimon opened on Jun 18, 2025

Copying of adjacent fields can be merged

llvm:optimizations

missed-optimization

Status: Open.

#144613 In llvm/llvm-project;

· hiraditya opened on Jun 17, 2025

Don't read from <code>vpcmpeqd xmm, xmm, xmm</code> and <code>vpternlogd zmm, zmm, zmm, 255</code> simultaneously

backend:X86

missed-optimization

Status: Open.

#144481 In llvm/llvm-project;

· Validark opened on Jun 17, 2025

NRVO not eliding copy on immediatly-invoked lambda

clang:frontendLanguage frontend issues, e.g. anything involving "Sema"Language frontend issues, e.g. anything involving "Sema"

missed-optimization

Status: Open.

#144323 In llvm/llvm-project;

· hcmh opened on Jun 16, 2025

[SLPVectorizer][X86] Addition no longer vectorized

llvm:SLPVectorizer

missed-optimization

Status: Open.

#144276 In llvm/llvm-project;

· nikic opened on Jun 15, 2025

[SelectionDAG][x86] Shuffle pyramid is not eliminated

backend:X86

missed-optimization

Status: Open.

#144227 In llvm/llvm-project;

· sutajo opened on Jun 14, 2025

[AArch64] Missed optimization: Make full use of smaddl instruction in AArch64

backend:AArch64

llvm:optimizations

missed-optimization

Status: Open.

#144208 In llvm/llvm-project;

· hstk30-hw opened on Jun 14, 2025

Missed optimization: eliminate redundant and+zext+or reconstruction in select arm

llvm:instcombine

missed-optimization

Status: Open.

#144020 In llvm/llvm-project;

· Cancelll opened on Jun 13, 2025

Missed optimization: fold (icmp eq a, c1) | (icmp ult f(a), c2) to icmp ult f(a), c2

llvm:instcombine

missed-optimization

Status: Open.

#143957 In llvm/llvm-project;

· Cancelll opened on Jun 12, 2025

Missed optimization: InstCombine is blocked by other uses

llvm:instcombine

missed-optimization

Status: Open.

#143658 In llvm/llvm-project;

· GINN-Imp opened on Jun 11, 2025