| Commit message (Expand) | Author | Age | Files | Lines |
* | Source order scheduler should not preschedule nodes with multiple uses. rdar:... | Evan Cheng | 2012-03-22 | 1 | -7/+11 |
* | Assign node orders to target intrinsics which do not produce results. rdar://... | Evan Cheng | 2012-03-22 | 1 | -0/+6 |
* | [fast-isel] Fold "urem x, pow2" -> "and x, pow2-1". This should fix the 271% | Chad Rosier | 2012-03-22 | 1 | -0/+7 |
* | Checking a build_vector for an all-ones value. | Jim Grosbach | 2012-03-21 | 1 | -5/+14 |
* | When combining (vextract shuffle (load ), <1,u,u,u>), 0) -> (load ), add user... | Craig Topper | 2012-03-20 | 1 | -0/+1 |
* | Do everything up to generating code to try to get a register for | Eric Christopher | 2012-03-20 | 1 | -1/+6 |
* | Untabify. | Eric Christopher | 2012-03-20 | 1 | -2/+2 |
* | Add another debugging statement here. | Eric Christopher | 2012-03-20 | 1 | -0/+4 |
* | Use lookUpRegForValue here instead of duplicating the code. | Eric Christopher | 2012-03-20 | 1 | -9/+2 |
* | f16 FDIV can now be legalized by promoting to f32 | Pete Cooper | 2012-03-19 | 1 | -1/+2 |
* | Fix DAG combine which creates illegal vector shuffles. Patch by Heikki Kultala. | Duncan Sands | 2012-03-19 | 1 | -0/+6 |
* | Revert r152613 (and r152614), "Inline the d'tor and add an anchor instead." f... | NAKAMURA Takumi | 2012-03-16 | 1 | -1/+1 |
* | We actually handle AllocaInst via getRegForValue below just fine. | Eric Christopher | 2012-03-15 | 1 | -1/+1 |
* | Add some debugging output into fast isel as well. | Eric Christopher | 2012-03-15 | 1 | -2/+6 |
* | Add another debug statement. | Eric Christopher | 2012-03-15 | 1 | -1/+3 |
* | When optimizing certain BUILD_VECTOR nodes into other BUILD_VECTOR nodes, add... | Nadav Rotem | 2012-03-15 | 1 | -0/+4 |
* | Add a xform to the DAG combiner. | Bill Wendling | 2012-03-15 | 1 | -0/+17 |
* | Insert the debugging instructions in one fell-swoop so that it doesn't call the | Bill Wendling | 2012-03-14 | 1 | -7/+8 |
* | Fortify r152675 a bit. Although I'm not able to come up with a test case that... | Evan Cheng | 2012-03-13 | 1 | -3/+11 |
* | DAG combine incorrectly optimize (i32 vextract (v4i16 load $addr), c) to | Evan Cheng | 2012-03-13 | 1 | -4/+19 |
* | Add a return type. | Bill Wendling | 2012-03-13 | 1 | -1/+1 |
* | Inline the d'tor and add an anchor instead. | Bill Wendling | 2012-03-13 | 1 | -1/+1 |
* | Refactor the SelectionDAG's 'dump' methods into their own .cpp file. | Bill Wendling | 2012-03-13 | 3 | -633/+632 |
* | llvm::SwitchInst | Stepan Dyatkovskiy | 2012-03-11 | 1 | -1/+1 |
* | Give dagcombiner's worklist some inline capacity. | Benjamin Kramer | 2012-03-10 | 1 | -3/+2 |
* | Use uint16_t to store instruction implicit uses and defs. Reduces static data. | Craig Topper | 2012-03-08 | 2 | -8/+8 |
* | Taken into account Duncan's comments for r149481 dated by 2nd Feb 2012: | Stepan Dyatkovskiy | 2012-03-08 | 1 | -4/+4 |
* | misched preparation: rename core scheduler methods for consistency. | Andrew Trick | 2012-03-07 | 6 | -37/+37 |
* | misched preparation: clarify ScheduleDAG and ScheduleDAGInstrs roles. | Andrew Trick | 2012-03-07 | 3 | -14/+22 |
* | misched preparation: modularize schedule emission. | Andrew Trick | 2012-03-07 | 2 | -3/+51 |
* | misched preparation: modularize schedule printing. | Andrew Trick | 2012-03-07 | 3 | -0/+17 |
* | misched preparation: modularize schedule verification. | Andrew Trick | 2012-03-07 | 5 | -3/+22 |
* | whitespace | Andrew Trick | 2012-03-07 | 1 | -5/+5 |
* | Cleanup in preparation for misched: Move DAG visualization logic. | Andrew Trick | 2012-03-07 | 2 | -0/+7 |
* | whitespace | Andrew Trick | 2012-03-07 | 1 | -3/+3 |
* | Cleanup: DAG building is specific to either SD or MI scheduling. Not part of ... | Andrew Trick | 2012-03-07 | 1 | -1/+1 |
* | Extend r148086 to check for [r +/- reg] address mode. This fixes queens perfo... | Evan Cheng | 2012-03-06 | 1 | -4/+7 |
* | Make it possible for a target to mark FSUB as Expand. This requires providin... | Owen Anderson | 2012-03-06 | 2 | -16/+39 |
* | Fix warnings about adding a bool to a string. | Bill Wendling | 2012-03-05 | 1 | -2/+2 |
* | Use uint16_t to store register overlaps to reduce static data. | Craig Topper | 2012-03-04 | 2 | -2/+2 |
* | Fix a codegen fault in which log2 or exp2 could be dead-code eliminated even ... | James Molloy | 2012-03-01 | 1 | -2/+4 |
* | LegalizeIntegerTypes: Reorder operations in the "big shift by small amount" o... | Benjamin Kramer | 2012-02-29 | 1 | -4/+4 |
* | Re-commit r151623 with fix. Only issue special no-return calls if it's a dire... | Evan Cheng | 2012-02-28 | 5 | -20/+29 |
* | Fix off-by one in comment. | Benjamin Kramer | 2012-02-28 | 1 | -1/+1 |
* | LegalizeIntegerTypes: Reenable the large shift with small amount optimization. | Benjamin Kramer | 2012-02-28 | 1 | -11/+21 |
* | Revert r151623 "Some ARM implementaions, e.g. A-series, does return stack pre... | Daniel Dunbar | 2012-02-28 | 5 | -29/+20 |
* | Code cleanup following CR by Duncan. | Nadav Rotem | 2012-02-28 | 1 | -5/+3 |
* | Fix a bug in the code that builds SDNodes from vector GEPs. | Nadav Rotem | 2012-02-28 | 1 | -0/+4 |
* | Some ARM implementaions, e.g. A-series, does return stack prediction. That is, | Evan Cheng | 2012-02-28 | 5 | -20/+29 |
* | Don't crash when a glue node contains an internal CopyToReg | Hal Finkel | 2012-02-24 | 1 | -0/+3 |