2015-02-10 |
Jonas Paulsson | Two comment typo fixes in lib/CodeGen/SelectionDAG... |
blob | commitdiff | raw |
2015-02-10 |
Chandler Carruth | [x86] Fix PR22524: the DAG combiner was incorrectly... |
blob | commitdiff | raw | diff to current |
2015-02-05 |
Ahmed Bougacha | [CodeGen] Add hook/combine to form vector extloads... |
blob | commitdiff | raw | diff to current |
2015-01-27 |
Quentin Colombet | Revert r227242 - Merge vector stores into wider vector... |
blob | commitdiff | raw | diff to current |
2015-01-27 |
Sanjay Patel | Merge vector stores into wider vector stores (PR21711) |
blob | commitdiff | raw | diff to current |
2015-01-22 |
Sanjay Patel | merge consecutive stores of extracted vector elements... |
blob | commitdiff | raw | diff to current |
2015-01-22 |
Michael Kuperstein | [DAGCombine] Produce better code for constant splats |
blob | commitdiff | raw | diff to current |
2015-01-22 |
Michael Kuperstein | Revert r226811, MSVC accepts code sane compilers don't. |
blob | commitdiff | raw | diff to current |
2015-01-22 |
Michael Kuperstein | [DAGCombine] Produce better code for constant splats |
blob | commitdiff | raw | diff to current |
2015-01-22 |
Elena Demikhovsky | Fixed a bug in type legalizer for masked load/store... |
blob | commitdiff | raw | diff to current |
2015-01-22 |
Elena Demikhovsky | Fixed a comment |
blob | commitdiff | raw | diff to current |
2015-01-22 |
Elena Demikhovsky | Fixed a bug in narrowing store operation. |
blob | commitdiff | raw | diff to current |
2015-01-21 |
Tim Northover | DAGCombine: fold (or (and X, M), (and X, N)) -> (and... |
blob | commitdiff | raw | diff to current |
2015-01-21 |
Tim Northover | Revert "DAGCombine: fold (or (and X, M), (and X, N... |
blob | commitdiff | raw | diff to current |
2015-01-21 |
Tim Northover | DAGCombine: fold (or (and X, M), (and X, N)) -> (and... |
blob | commitdiff | raw | diff to current |
2015-01-17 |
Mehdi Amini | Improve DAG combine pass on certain IR vector patterns |
blob | commitdiff | raw | diff to current |
2015-01-14 |
Chandler Carruth | [cleanup] Re-sort all the #include lines in LLVM using |
blob | commitdiff | raw | diff to current |
2015-01-14 |
Mehdi Amini | DAG Combiner: Fold SelectCC When Cond is UNDEF |
blob | commitdiff | raw | diff to current |
2015-01-13 |
Matthias Braun | DAGCombiner: simplify by using condition variables... |
blob | commitdiff | raw | diff to current |
2015-01-13 |
Matt Arsenault | R600: Implement getRecipEstimate |
blob | commitdiff | raw | diff to current |
2015-01-13 |
Olivier Sallenave | Added TLI hook for isFPExtFree. Some of the FMA combine... |
blob | commitdiff | raw | diff to current |
2015-01-13 |
Matt Arsenault | Combine fcmp + select to fminnum / fmaxnum if no nans... |
blob | commitdiff | raw | diff to current |
2015-01-09 |
Hal Finkel | [DAGCombine] Remainder of fix to r225380 (More FMA... |
blob | commitdiff | raw | diff to current |
2015-01-09 |
Hal Finkel | Partial fix to r225380 (More FMA folding opportunities) |
blob | commitdiff | raw | diff to current |
2015-01-08 |
Ahmed Bougacha | [SelectionDAG] Allow targets to specify legality of... |
blob | commitdiff | raw | diff to current |
2015-01-07 |
Olivier Sallenave | More FMA folding opportunities. |
blob | commitdiff | raw | diff to current |
2015-01-07 |
Olivier Sallenave | Test commit |
blob | commitdiff | raw | diff to current |
2015-01-05 |
Craig Topper | Replace several 'assert(false' with 'llvm_unreachable... |
blob | commitdiff | raw | diff to current |
2014-12-31 |
Alexey Samsonov | Revert "merge consecutive stores of extracted vector... |
blob | commitdiff | raw | diff to current |
2014-12-23 |
Mehdi Amini | Always assert in DAGCombine and not only when -debug... |
blob | commitdiff | raw | diff to current |
2014-12-23 |
Michael Kuperstein | [DagCombine] Improve DAGCombiner BUILD_VECTOR when... |
blob | commitdiff | raw | diff to current |
2014-12-19 |
Sanjay Patel | merge consecutive stores of extracted vector elements |
blob | commitdiff | raw | diff to current |
2014-12-17 |
Michael Kuperstein | [DAGCombine] Slightly improve lowering of BUILD_VECTOR... |
blob | commitdiff | raw | diff to current |
2014-12-12 |
Matt Arsenault | Add target hook for whether it is profitable to reduce... |
blob | commitdiff | raw | diff to current |
2014-12-09 |
Owen Anderson | Fix a few instances found in SelectionDAG where we... |
blob | commitdiff | raw | diff to current |
2014-12-04 |
Simon Pilgrim | [InstCombine] Minor optimization for bswap with binary ops |
blob | commitdiff | raw | diff to current |
2014-12-04 |
Elena Demikhovsky | Masked Load / Store Intrinsics - the CodeGen part. |
blob | commitdiff | raw | diff to current |
2014-11-28 |
Duncan P. N. Exon... | Revert "Masked Vector Load and Store Intrinsics." |
blob | commitdiff | raw | diff to current |
2014-11-23 |
Elena Demikhovsky | Masked Vector Load and Store Intrinsics. |
blob | commitdiff | raw | diff to current |
2014-11-21 |
Andrea Di Biagio | [DAG] Teach how to turn a build_vector into a shuffle... |
blob | commitdiff | raw | diff to current |
2014-11-21 |
Andrea Di Biagio | [DAG] Refactor the shuffle combining logic in DAGCombin... |
blob | commitdiff | raw | diff to current |
2014-11-21 |
Hao Liu | DAGCombiner: Allow the DAGCombiner to combine multiple... |
blob | commitdiff | raw | diff to current |
2014-11-19 |
David Blaikie | Update SetVector to rely on the underlying set's insert... |
blob | commitdiff | raw | diff to current |
2014-11-17 |
Oliver Stannard | Fix optimisations of SELECT_CC which assumed result... |
blob | commitdiff | raw | diff to current |
2014-11-15 |
Andrea Di Biagio | [DAG] Improved target independent vector shuffle foldin... |
blob | commitdiff | raw | diff to current |
2014-11-11 |
Oliver Stannard | LLVM incorrectly folds xor into select |
blob | commitdiff | raw | diff to current |
2014-11-05 |
Andrea Di Biagio | [X86] Teach method 'isVectorClearMaskLegal' how to... |
blob | commitdiff | raw | diff to current |
2014-11-03 |
Paul Robinson | Normally an 'optnone' function goes through fast-isel... |
blob | commitdiff | raw | diff to current |
2014-10-30 |
Louis Gerbarg | Fix incorrect invariant check in DAG Combine |
blob | commitdiff | raw | diff to current |
2014-10-29 |
NAKAMURA Takumi | Whitespace. |
blob | commitdiff | raw | diff to current |
2014-10-24 |
Sanjay Patel | Use rsqrt (X86) to speed up reciprocal square root... |
blob | commitdiff | raw | diff to current |
2014-10-22 |
Benjamin Kramer | Strength reduce constant-sized vectors into arrays... |
blob | commitdiff | raw | diff to current |
2014-10-21 |
Matt Arsenault | Add minnum / maxnum codegen |
blob | commitdiff | raw | diff to current |
2014-10-17 |
Jan Vesely | SelectionDAG: Add sext_inreg optimizations |
blob | commitdiff | raw | diff to current |
2014-10-09 |
Sanjay Patel | Improve sqrt estimate algorithm (fast-math) |
blob | commitdiff | raw | diff to current |
2014-10-08 |
Eric Christopher | Remove unnecessary include. |
blob | commitdiff | raw | diff to current |
2014-10-08 |
Eric Christopher | Use both the cached TLI and the subtarget off of the... |
blob | commitdiff | raw | diff to current |
2014-10-06 |
Hal Finkel | [DAGCombine] Remove SIGN_EXTEND-related inf-loop |
blob | commitdiff | raw | diff to current |
2014-10-06 |
Sanjay Patel | Fast-math fold: x / (y * sqrt(z)) -> x * (rsqrt(z... |
blob | commitdiff | raw | diff to current |
2014-10-05 |
Chandler Carruth | [x86, dag] Teach the DAG combiner to prune inputs toa... |
blob | commitdiff | raw | diff to current |
2014-09-30 |
Sanjay Patel | Use the target-specified iteration count to opt out... |
blob | commitdiff | raw | diff to current |
2014-09-30 |
Sanjay Patel | Split the estimate() interface into separate functions... |
blob | commitdiff | raw | diff to current |
2014-09-30 |
Andrea Di Biagio | [DAG] Check in advance if a build_vector has a legal... |
blob | commitdiff | raw | diff to current |
2014-09-27 |
James Molloy | [AArch64] Redundant store instructions should be remove... |
blob | commitdiff | raw | diff to current |
2014-09-26 |
Sanjay Patel | Refactor reciprocal and reciprocal square root estimate... |
blob | commitdiff | raw | diff to current |
2014-09-26 |
David Xu | Revert patch ofr218493 |
blob | commitdiff | raw | diff to current |
2014-09-26 |
David Xu | Redundant store instructions should be removed as dead... |
blob | commitdiff | raw | diff to current |
2014-09-23 |
Sanjay Patel | Use SDValue bool operator to reduce code. No functional... |
blob | commitdiff | raw | diff to current |
2014-09-21 |
Sanjay Patel | Refactor reciprocal square root estimate into target... |
blob | commitdiff | raw | diff to current |
2014-09-19 |
Hal Finkel | Optionally enable more-aggressive FMA formation in... |
blob | commitdiff | raw | diff to current |
2014-09-15 |
Sanjay Patel | Replace dead links to "Hacker's Delight" with general... |
blob | commitdiff | raw | diff to current |
2014-09-11 |
Matt Arsenault | Add DAG combine for shl + add of constants. |
blob | commitdiff | raw | diff to current |
2014-09-11 |
Sanjay Patel | Combine fmul vector FP constants when unsafe math is... |
blob | commitdiff | raw | diff to current |
2014-09-11 |
David Xu | Build correct vector filled with undef nodes |
blob | commitdiff | raw | diff to current |
2014-09-08 |
Sanjay Patel | Group unsafe fmul math folds together for easier readin... |
blob | commitdiff | raw | diff to current |
2014-09-08 |
Sanjay Patel | Fix the FIXME that was just added in r217390 - remove... |
blob | commitdiff | raw | diff to current |
2014-09-08 |
Sanjay Patel | group unsafe math folds together for easier reading |
blob | commitdiff | raw | diff to current |
2014-09-05 |
Sanjay Patel | Allow vector fsub ops with constants to get the same... |
blob | commitdiff | raw | diff to current |
2014-09-05 |
Sanjay Patel | clean up; NFC |
blob | commitdiff | raw | diff to current |
2014-09-02 |
Matt Arsenault | Fix interference caused by fmul 2, x -> fadd x, x |
blob | commitdiff | raw | diff to current |
2014-09-02 |
Matt Arsenault | Fix comment and unnecessary check for FP build_vectors. |
blob | commitdiff | raw | diff to current |
2014-09-02 |
Hal Finkel | Enable splitting indexing from loads with TargetConstants |
blob | commitdiff | raw | diff to current |
2014-09-02 |
Hal Finkel | Revert "Revert '[DAGCombiner] Split up an indexed load... |
blob | commitdiff | raw | diff to current |
2014-08-28 |
Sanjay Patel | Move FNEG next to FABS and make them more similar,... |
blob | commitdiff | raw | diff to current |
2014-08-28 |
Owen Anderson | Do not introduce new shuffle patterns after operation... |
blob | commitdiff | raw | diff to current |
2014-08-28 |
Sanjay Patel | Janitorial services: "Don’t duplicate function or class... |
blob | commitdiff | raw | diff to current |
2014-08-28 |
Sanjay Patel | Remove local TLI vars that are just duplicates of the... |
blob | commitdiff | raw | diff to current |
2014-08-28 |
Sanjay Patel | Use local vars to improve readability. No functional... |
blob | commitdiff | raw | diff to current |
2014-08-27 |
Sanjay Patel | Use local variable in visitFADD. No functional change. |
blob | commitdiff | raw | diff to current |
2014-08-27 |
Sanjay Patel | Group unsafe-math optimizations for fsub into one block... |
blob | commitdiff | raw | diff to current |
2014-08-27 |
Sanjay Patel | Use local variable to improve readability. |
blob | commitdiff | raw | diff to current |
2014-08-27 |
Craig Topper | Simplify creation of a bunch of ArrayRefs by using... |
blob | commitdiff | raw | diff to current |
2014-08-24 |
Craig Topper | Use range based for loops to avoid needing to re-mentio... |
blob | commitdiff | raw | diff to current |
2014-08-21 |
Sanjay Patel | name change: isPow2DivCheap -> isPow2SDivCheap |
blob | commitdiff | raw | diff to current |
2014-08-21 |
Benjamin Kramer | DAGCombiner: Make concat_vector combine safe for EVTs... |
blob | commitdiff | raw | diff to current |
2014-08-16 |
Matt Arsenault | Fix fmul combines with constant splat vectors |
blob | commitdiff | raw | diff to current |
2014-08-16 |
Andrea Di Biagio | [DAGCombiner] Improve the folding of target independet... |
blob | commitdiff | raw | diff to current |
2014-08-14 |
Sanjay Patel | optimize vector fneg of bitcasted integer value |
blob | commitdiff | raw | diff to current |
2014-08-14 |
Chandler Carruth | [SDAG] Fix a bug in the DAG combiner where we would... |
blob | commitdiff | raw | diff to current |
2014-08-13 |
Andrea Di Biagio | [DAGCombiner] Improved target independent vector shuffl... |
blob | commitdiff | raw | diff to current |
next |