|
About the Loop Optimizations category
|
|
0
|
493
|
January 8, 2022
|
|
RFC: Statement-Seeded Partitioning for Loop Distribution
|
|
1
|
63
|
October 21, 2026
|
|
Ordered floating-point reductions on X86: enablement and recurrence-aware costing
|
|
2
|
97
|
October 13, 2026
|
|
[LoopFusion] Initiating a Cost Model for LoopFusion
|
|
8
|
473
|
October 9, 2026
|
|
RFC: Extending GVN's loop-load PRE
|
|
10
|
363
|
October 1, 2026
|
|
[RFC] GPU wave counts for spill-placement costs
|
|
0
|
71
|
September 24, 2026
|
|
LoopVectorizer: loop tail vectorization & RISCV calling conventions
|
|
6
|
298
|
August 22, 2026
|
|
[RFC] loop vectorization of "compress store"/"expand load" patterns
|
|
12
|
1129
|
August 6, 2026
|
|
[RFC] Using Cross-Part CSE to guide Loop Interleaving
|
|
0
|
187
|
July 31, 2026
|
|
[RFC] Universal Profile-Guided Automatic Software Loop Data Prefetcher for LLVM
|
|
7
|
1038
|
July 27, 2026
|
|
[RFC] [LoopVectorizer] Check-First Early-Exit Loop Vectorization
|
|
1
|
340
|
July 21, 2026
|
|
[RFC] Vectorizing loops which have load based tripcount
|
|
0
|
159
|
July 15, 2026
|
|
A general-purpose loop iteration-space splitting utility (LoopSplitUtils)
|
|
3
|
286
|
July 13, 2026
|
|
[RFC] Use (de)interleaving intrinsics for fixed-length vectors
|
|
4
|
339
|
June 11, 2026
|
|
[RFC] Enable MaximizeBandwidth by default for X86 in the Loop Vectorizer
|
|
2
|
273
|
June 9, 2026
|
|
Allowing LoopVectorizer to vectorize loops with uniform OpenMP runtime calls
|
|
7
|
291
|
June 6, 2026
|
|
[RFC] Put "LLVM Loop Optimization Discussion" on community calendar and avoid conflict
|
|
4
|
307
|
June 2, 2026
|
|
[RFC] Public Roadmap board for the Vectorizers
|
|
11
|
725
|
May 1, 2026
|
|
[Loop Vectorizer] Live-out IVs stopping epilogue vectorization
|
|
1
|
168
|
April 7, 2026
|
|
[LoopInterchange] Some questions and a potential missing case?
|
|
2
|
263
|
March 23, 2026
|
|
[RFC] Reducing Runtime Pointer Comparisons for Stencil Codes via Stencil Group Merging in LoopAccessAnalysis
|
|
2
|
191
|
March 18, 2026
|
|
[RFC] `@llvm.parallel_alloca` intrinsic for vectorizable stack allocations
|
|
4
|
241
|
March 12, 2026
|
|
[LICM] Is it correct to move such a load instruction to the loop preheader?
|
|
2
|
168
|
March 9, 2026
|
|
Expand vector_reduce_fminimum/fmaximum
|
|
1
|
119
|
March 2, 2026
|
|
Semantics of llvm.loop.parallel_accesses and interaction with alloca
|
|
5
|
275
|
February 17, 2026
|
|
Why are ordered reductions enabled by default on ARM and RISCV, but not on other targets?
|
|
6
|
410
|
December 22, 2025
|
|
[RFC] Plan to Improve LoopInterchange by Undoing Simple Reductions
|
|
12
|
685
|
December 22, 2025
|
|
LoopVectorizer: epilogue vectorization "with length" (SystemZ)
|
|
4
|
284
|
December 19, 2025
|
|
Vectorizing matrix transpose with runtime stride on AArch64 (VPlan / VPRecipe questions)
|
|
2
|
252
|
December 2, 2025
|
|
[RFC] A New Pass as an Alternative to DependenceAnalysis
|
|
28
|
1912
|
October 7, 2025
|