| Commit message (Expand) | Author | Age | Files | Lines |
... | |
* | [InstCombine] reduce more checks for power-of-2-or-zero using ctpop | Sanjay Patel | 2019-07-01 | 1 | -1/+7 |
* | [InstCombine] Shift amount reassociation in bittest (PR42399) | Roman Lebedev | 2019-07-01 | 1 | -0/+60 |
* | [InstCombine] Omit 'urem' where possible | Roman Lebedev | 2019-07-01 | 1 | -4/+20 |
* | [InstCombine] Simplify icmp ult/uge (shl %x, C2), C1 iff C1 is power of two -... | Huihui Zhang | 2019-06-25 | 1 | -0/+21 |
* | [InstCombine] reduce checks for power-of-2-or-zero using ctpop | Sanjay Patel | 2019-06-25 | 1 | -12/+10 |
* | [InstCombine] Fold icmp eq/ne (and %x, C), 0 iff (-C) is power of two -> %x ... | Huihui Zhang | 2019-06-25 | 1 | -11/+9 |
* | [InstCombine] fix typo in comment; NFC | Sanjay Patel | 2019-06-20 | 1 | -1/+1 |
* | [InstCombine] canonicalize check for power-of-2 | Sanjay Patel | 2019-06-20 | 1 | -0/+20 |
* | [InstCombine] Fold icmp eq/ne (and %x, signbit), 0 -> %x s>=/s< 0 earlier | Huihui Zhang | 2019-06-19 | 1 | -10/+17 |
* | [InstCombine] foldICmpWithLowBitMaskedVal(): 'icmp sgt/sle': avoid miscompiles | Roman Lebedev | 2019-06-09 | 1 | -0/+8 |
* | [InstCombine] Avoid use after free in DenseMap, when built with GCC | Martin Storsjo | 2019-05-30 | 1 | -1/+4 |
* | [ValueTracking][ConstantRange] Distinguish low/high always overflow | Nikita Popov | 2019-05-28 | 1 | -1/+2 |
* | [InstCombine] Refactor OptimizeOverflowCheck; NFCI | Nikita Popov | 2019-05-26 | 1 | -78/+60 |
* | [InstCombine] Remove OverflowCheckFlavor; NFC | Nikita Popov | 2019-05-26 | 1 | -17/+14 |
* | [InstCombine] allow sinking fneg operands through an FP min/max | Sanjay Patel | 2019-05-07 | 1 | -5/+5 |
* | Move Value *RHSCIOp def into the scope where its actually used. NFCI. | Simon Pilgrim | 2019-05-05 | 1 | -2/+1 |
* | [InstCombine] reduce code duplication; NFC | Sanjay Patel | 2019-04-29 | 1 | -5/+7 |
* | [InstCombine] visitFCmpInst - appease copy+paste pattern warning. NFCI. | Simon Pilgrim | 2019-04-29 | 1 | -1/+1 |
* | [InstCombine] Remove redundant/bogus mul_with_overflow combines | Nikita Popov | 2019-04-13 | 1 | -8/+0 |
* | [InstCombine] Handle ssubo always overflow | Nikita Popov | 2019-04-10 | 1 | -3/+3 |
* | [InstCombine] Handle saddo always overflow | Nikita Popov | 2019-04-10 | 1 | -3/+3 |
* | [InstCombine] Handle usubo always overflow | Nikita Popov | 2019-04-10 | 1 | -0/+3 |
* | [InstCombine] Directly call computeOverflow methods in OptimizeOverflowCheck;... | Nikita Popov | 2019-04-10 | 1 | -6/+13 |
* | [InstCombine] Restructure OptimizeOverflowCheck; NFC | Nikita Popov | 2019-04-09 | 1 | -31/+28 |
* | [InstCombine] Combine no-wrap sub and icmp w/ constant. | Luqman Aden | 2019-04-04 | 1 | -1/+10 |
* | [InstCombine] Fix crashing from (icmp (bitcast ([su]itofp X)), Y) | Sanjay Patel | 2019-02-07 | 1 | -29/+33 |
* | [InstCombine] refactor folds for (icmp (bitcast X), Y); NFCI | Sanjay Patel | 2019-02-07 | 1 | -75/+69 |
* | [InstCombine] X | C == C --> (X & ~C) == 0 | Sanjay Patel | 2019-02-06 | 1 | -9/+18 |
* | [opaque pointer types] Pass function types to CallInst creation. | James Y Knight | 2019-02-01 | 1 | -4/+4 |
* | [InstCombine] Simplify cttz/ctlz + icmp ugt/ult | Nikita Popov | 2019-01-19 | 1 | -10/+66 |
* | Update the file headers across all of the LLVM projects in the monorepo | Chandler Carruth | 2019-01-19 | 1 | -4/+3 |
* | [InstCombine] Simplify cttz/ctlz + icmp eq/ne into mask check | Nikita Popov | 2018-12-18 | 1 | -3/+22 |
* | [InstCombine] Fix negative GEP offset evaluation for 32-bit pointers | Nikita Popov | 2018-12-12 | 1 | -5/+3 |
* | [InstCombine] foldICmpWithLowBitMaskedVal(): don't miscompile -1 vector elts | Roman Lebedev | 2018-12-06 | 1 | -0/+4 |
* | [InstCombine] simplify icmps with same operands based on dominating cmp | Sanjay Patel | 2018-12-05 | 1 | -0/+5 |
* | [InstCombine] rearrange foldICmpWithDominatingICmp; NFC | Sanjay Patel | 2018-12-04 | 1 | -33/+45 |
* | [InstCombine] add helper for icmp with dominator; NFC | Sanjay Patel | 2018-12-04 | 1 | -20/+23 |
* | [InstCombine] foldICmpWithLowBitMaskedVal(): disable 2 faulty folds. | Roman Lebedev | 2018-12-03 | 1 | -0/+4 |
* | [InstCombine] propagate FMF for fcmp+fabs folds | Sanjay Patel | 2018-11-07 | 1 | -7/+13 |
* | [InstCombine] peek through fabs() when checking isnan() | Sanjay Patel | 2018-11-07 | 1 | -1/+6 |
* | [InstCombine] add folds for fcmp Pred fabs(X), 0.0 | Sanjay Patel | 2018-11-07 | 1 | -0/+8 |
* | [InstCombine] add fold for fabs(X) u< 0.0 | Sanjay Patel | 2018-11-07 | 1 | -0/+5 |
* | [IR] add optional parameter for copying IR flags to compare instructions | Sanjay Patel | 2018-11-07 | 1 | -29/+14 |
* | [InstCombine] allow vector types for fcmp+fpext fold | Sanjay Patel | 2018-11-06 | 1 | -9/+9 |
* | [InstCombine] propagate fast-math-flags when folding fcmp+fpext, part 2 | Sanjay Patel | 2018-11-06 | 1 | -3/+6 |
* | [InstCombine] rearrange code for fcmp+fpext; NFCI | Sanjay Patel | 2018-11-06 | 1 | -29/+27 |
* | [InstCombine] propagate fast-math-flags when folding fcmp+fpext | Sanjay Patel | 2018-11-06 | 1 | -5/+7 |
* | [InstCombine] propagate fast-math-flags when folding fcmp+fneg, part 2 | Sanjay Patel | 2018-11-06 | 1 | -2/+3 |
* | [InstCombine] reduce code; NFC | Sanjay Patel | 2018-11-06 | 1 | -1/+1 |
* | [InstCombine] propagate fast-math-flags when folding fcmp+fneg | Sanjay Patel | 2018-11-06 | 1 | -11/+16 |