| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | DAGCombiner: Make the post-legalize vector op optimization more aggressive. | Benjamin Kramer | 2013-02-21 | 1 | -10/+0 |
| * | DAGCombiner: Fold pointless truncate, bitcast, buildvector series | Arnold Schwaighofer | 2013-02-20 | 1 | -0/+32 |
| * | Dont merge consecutive loads/stores into vectors when noimplicitfloat is used. | Nadav Rotem | 2013-02-14 | 1 | -6/+6 |
| * | Add some legality checks for SETCC before introducing it in the DAG combiner ... | Owen Anderson | 2013-02-14 | 1 | -2/+6 |
| * | PR14562 - Truncation of left shift became undef | Paul Redmond | 2013-02-12 | 1 | -2/+9 |
| * | Check type for legality before forming a select from loads. | Pete Cooper | 2013-02-12 | 1 | -1/+3 |
| * | DAGCombiner: Constant folding around pre-increment loads/stores | Hal Finkel | 2013-02-08 | 1 | -0/+89 |
| * | Reapply r174343, with a fix for a scary DAG combine bug where it failed to di... | Owen Anderson | 2013-02-05 | 1 | -2/+5 |
| * | Revert r174343, "When the target-independent DAGCombiner inferred a higher al... | NAKAMURA Takumi | 2013-02-05 | 1 | -5/+2 |
| * | When the target-independent DAGCombiner inferred a higher alignment for a load, | Owen Anderson | 2013-02-05 | 1 | -2/+5 |
| * | rdar://13126763 | Shuxin Yang | 2013-02-02 | 1 | -13/+20 |
| * | Revert 172708. | Nadav Rotem | 2013-01-20 | 1 | -13/+5 |
| * | Optimization for the following SIGN_EXTEND pairs: | Elena Demikhovsky | 2013-01-17 | 1 | -5/+13 |
| * | This patch addresses an incorrect transformation in the DAG combiner. | Bill Schmidt | 2013-01-14 | 1 | -5/+15 |
| * | Fix a DAG combine bug visitBRCOND() is transforming br(xor(x, y)) to br(x != y). | Evan Cheng | 2013-01-09 | 1 | -12/+18 |
| * | Sink AddrMode back into TargetLowering, removing one of the most | Chandler Carruth | 2013-01-07 | 1 | -1/+1 |
| * | DAGCombiner: Avoid generating illegal vector INT_TO_FP nodes | Tom Stellard | 2013-01-02 | 1 | -4/+5 |
| * | Move all of the header files which are involved in modelling the LLVM IR | Chandler Carruth | 2013-01-02 | 1 | -4/+4 |
| * | Remove the Function::getFnAttributes method in favor of using the AttributeSet | Bill Wendling | 2012-12-30 | 1 | -2/+3 |
| * | Refactor DAGCombinerInfo. Change the different booleans that indicate if we a... | Nadav Rotem | 2012-12-27 | 1 | -2/+2 |
| * | Do not introduce vector operations in functions marked with noimplicitfloat. | Bob Wilson | 2012-12-20 | 1 | -2/+5 |
| * | Change TargetLowering::isCondCodeLegal to take an MVT, instead of EVT. | Patrik Hagglund | 2012-12-19 | 1 | -3/+6 |
| * | Optimized load + SIGN_EXTEND patterns in the X86 backend. | Elena Demikhovsky | 2012-12-19 | 1 | -0/+1 |
| * | Fix a bug in DAGCombiner::MatchBSwapHWord. Make sure the node has operands be... | Evan Cheng | 2012-12-13 | 1 | -1/+2 |
| * | DAGCombine: clamp hi bit in APInt::getBitsSet to avoid assertion | Manman Ren | 2012-12-12 | 1 | -1/+2 |
| * | Revert EVT->MVT changes, r169836-169851, due to buildbot failures. | Patrik Hagglund | 2012-12-11 | 1 | -6/+3 |
| * | Change TargetLowering::isCondCodeLegal to take an MVT, instead of EVT. | Patrik Hagglund | 2012-12-11 | 1 | -3/+6 |
| * | Fix a miscompile in the DAG combiner. Previously, we would incorrectly | Chandler Carruth | 2012-12-11 | 1 | -2/+6 |
| * | Teach DAG combine to handle vector add/sub with vectors of all 0s. | Craig Topper | 2012-12-10 | 1 | -0/+10 |
| * | Remove extra blank line. | Craig Topper | 2012-12-09 | 1 | -1/+0 |
| * | Teach DAG combine to handle vector logical operations with vectors of all 1s ... | Craig Topper | 2012-12-08 | 1 | -0/+30 |
| * | Fix a bug in the code that merges consecutive stores. Previously we did not | Nadav Rotem | 2012-12-06 | 1 | -10/+14 |
| * | Use the new script to sort the includes of every file under lib. | Chandler Carruth | 2012-12-03 | 1 | -9/+9 |
| * | Allow merging multiple store sequences on the same chain. | Nadav Rotem | 2012-12-02 | 1 | -2/+15 |
| * | When combining consecutive stores allow loads in between the stores, if the l... | Nadav Rotem | 2012-11-29 | 1 | -3/+61 |
| * | Handle DAG CSE adding new uses during ReplaceAllUsesWith. Fixes PR14333. | Rafael Espindola | 2012-11-14 | 1 | -1/+12 |
| * | Be careful not to optimize a SELECT_CC into a SETCC post-legalization if the ... | Owen Anderson | 2012-11-03 | 1 | -23/+27 |
| * | Add a few more simple fast-math constant propagations and cancellations. | Owen Anderson | 2012-11-01 | 1 | -0/+18 |
| * | In various places throughout the code generator, there were special | Ulrich Weigand | 2012-10-29 | 1 | -20/+16 |
| * | Teach DAG combine to fold (buildvec (Xint2fp x)) to (Xint2fp (buildvec x)) | Michael Liao | 2012-10-24 | 1 | -0/+71 |
| * | Keep coding standard. Don't evaluate getNumOperands() every time. | Jakub Staszak | 2012-10-24 | 1 | -1/+1 |
| * | Clean up code and put transformation on (build_vec (ext x)) into a helper func | Michael Liao | 2012-10-23 | 1 | -55/+69 |
| * | Simplify condition checking as CONCAT assume all inputs of the same type. | Michael Liao | 2012-10-19 | 1 | -3/+3 |
| * | In SimplifySelectOps we pulled two loads through a select node despite the fa... | Nadav Rotem | 2012-10-18 | 1 | -0/+4 |
| * | Revert part of r166049 back and enable test case in r166125. | Michael Liao | 2012-10-17 | 1 | -0/+42 |
| * | Revert r166049 | Michael Liao | 2012-10-17 | 1 | -46/+0 |
| * | Teach DAG combine to fold (extract_subvec (concat v1, ..) i) to v_i | Michael Liao | 2012-10-17 | 1 | -2/+17 |
| * | Teach DAG combine to fold (trunc (fptoXi x)) to (fptoXi x) | Michael Liao | 2012-10-16 | 1 | -0/+46 |
| * | Refactor the AddrMode class out of TLI to its own header file. | Nadav Rotem | 2012-10-08 | 1 | -1/+1 |
| * | Move TargetData to DataLayout. | Micah Villmow | 2012-10-08 | 1 | -8/+8 |
