[LoopVectorizer] Refine loop vectorizer's register usage calculator by ignoring speci...
[oota-llvm.git] / test / Transforms / StraightLineStrengthReduce /
2015-08-13 Jingyue Wu[SeparateConstOffsetFromGEP] strengthen the inbounds...
2015-07-16 Jingyue Wu[NVPTX] enable SpeculativeExecution in NVPTX
2015-06-28 Jingyue Wu[SLSR] S's basis must have the same type as S
2015-06-19 Matt ArsenaultAMDGPU: Fix some places missed in rename
2015-06-11 Matt ArsenaultSLSR: Pass address space to isLegalAddressingMode
2015-05-15 Jingyue Wu[SLSR] handle (B | i) * S
2015-05-13 Jingyue Wu[SLSR] handles non-canonicalized Mul candidates
2015-04-23 Jingyue Wu[NVPTX] run SeparateConstOffsetFromGEP before SLSR
2015-04-21 Jingyue Wu[SLSR] garbage-collect unused instructions
2015-04-15 Jingyue Wu[NFC] [SLSR] clean up some tests
2015-04-15 Jingyue Wu[SLSR] handle candidate form (B + i * S)
2015-04-06 Jingyue Wu[SLSR] consider &B[S << i] as &B[(1 << i) * S]
2015-04-02 Jingyue Wu[SLSR] handles off bounds GEPs
2015-03-26 Jingyue Wu[SLSR] handle candidate form &B[i * S]
2015-02-03 Jingyue WuAdd straight-line strength reduction to LLVM