| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Formatting | Eli Bendersky | 2013-04-04 | 1 | -2/+2 |
| * | Fix PR15632: No support for ppcf128 floating-point remainder on PowerPC. | Bill Schmidt | 2013-04-03 | 2 | -0/+12 |
| * | DAGCombiner: Merge store/loads when we have extload/truncstores | Arnold Schwaighofer | 2013-04-02 | 1 | -0/+19 |
| * | Merge load/store sequences with adresses: base + index + offset | Arnold Schwaighofer | 2013-04-01 | 1 | -25/+88 |
| * | DAGCombine: visitXOR can replace a node without returning it, bail out in tha... | Benjamin Kramer | 2013-03-30 | 1 | -3/+3 |
| * | [fast-isel] Add a preemptive fix for the case where we fail to materialize an | Chad Rosier | 2013-03-28 | 1 | -0/+2 |
| * | Enhance folding of (extract_subvec (insert_subvec V1, V2, IIdx), EIdx) | Michael Liao | 2013-03-25 | 1 | -21/+29 |
| * | Disable some unsafe-fp-math DAG-combine transformation after legalization. | Shuxin Yang | 2013-03-25 | 1 | -5/+17 |
| * | Remove the type legality check from the SelectionDAGBuilder when it lowers @l... | Owen Anderson | 2013-03-23 | 1 | -1/+0 |
| * | Make variable name more explicit and eliminate redundant lookup in SDNodeOrde... | Justin Holewinski | 2013-03-20 | 1 | -4/+4 |
| * | When computing the demanded bits of Load SDNodes, make sure that we are looki... | Nadav Rotem | 2013-03-20 | 1 | -12/+16 |
| * | Revert "pre-RA-sched: fix TargetOpcode usage" | Christian Konig | 2013-03-20 | 1 | -18/+12 |
| * | Move SDNode order propagation to SDNodeOrdering, which also fixes a missed | Justin Holewinski | 2013-03-20 | 4 | -23/+15 |
| * | pre-RA-sched: fix TargetOpcode usage | Christian Konig | 2013-03-20 | 1 | -12/+18 |
| * | Propagate DAG node ordering during type legalization and instruction selection | Justin Holewinski | 2013-03-20 | 3 | -2/+43 |
| * | Reset some of the target options which affect code generation. | Bill Wendling | 2013-03-13 | 1 | -0/+1 |
| * | Avoid generating ISD::SELECT for vector operands to SIGN_EXTEND | Richard Relph | 2013-03-12 | 1 | -2/+2 |
| * | Fix a crasher newly introduced in r176659/r176649, where fast-isel tries to | Nick Lewycky | 2013-03-11 | 1 | -0/+2 |
| * | Disable statistics on Release builds and move tests that depend on -stats. | Jan Wen Voung | 2013-03-08 | 2 | -18/+14 |
| * | Remove default from fully covered switch. | Benjamin Kramer | 2013-03-08 | 1 | -2/+1 |
| * | LegalizeDAG: Respect the result of TLI.getBooleanContents() when expanding SETCC | Tom Stellard | 2013-03-08 | 1 | -1/+13 |
| * | DAGCombiner: Use correct value type for checking legality of BR_CC v3 | Tom Stellard | 2013-03-08 | 1 | -1/+2 |
| * | Revert r176154 in favor of a better approach. | Bill Wendling | 2013-03-08 | 1 | -7/+0 |
| * | [fast-isel] Seriously, add support for the expect intrinsic. | Chad Rosier | 2013-03-07 | 1 | -3/+3 |
| * | [fast-isel] Add support for the expect intrinsic. | Chad Rosier | 2013-03-07 | 1 | -0/+5 |
| * | ArrayRefize some code. No functionality change. | Benjamin Kramer | 2013-03-07 | 2 | -5/+3 |
| * | pre-RA-sched debug-only fix | Andrew Trick | 2013-03-07 | 1 | -2/+4 |
| * | pre-RA-sched assertion fix. This bug was exposed by r176037. | Andrew Trick | 2013-03-07 | 1 | -3/+0 |
| * | No need to go through int64 and APInt when generating a new constant. | Nadav Rotem | 2013-03-07 | 1 | -3/+3 |
| * | SDAG: Handle scalarizing an extend of a <1 x iN> vector. | Jim Grosbach | 2013-03-07 | 2 | -0/+21 |
| * | Remove duplicate line and move another closer to its actual use | Eli Bendersky | 2013-03-01 | 1 | -3/+1 |
| * | Set properties for f128 type. | Akira Hatanaka | 2013-03-01 | 1 | -17/+62 |
| * | Generate an error message instead of asserting or segfaulting when we can't | Chad Rosier | 2013-03-01 | 1 | -0/+1 |
| * | Fix PR10475 | Michael Liao | 2013-03-01 | 3 | -12/+17 |
| * | A small refactoring + adding comments. | Eli Bendersky | 2013-02-28 | 2 | -10/+5 |
| * | Put some per-instruction statistics of fast isel under NDEBUG, together with | Eli Bendersky | 2013-02-28 | 1 | -3/+5 |
| * | Remove unnecessary cast to void. | Eric Christopher | 2013-02-27 | 1 | -1/+0 |
| * | Silence the unused variable warning. | Nadav Rotem | 2013-02-27 | 1 | -5/+7 |
| * | The FastISEL should be fast. But when we record statistics we use atomic oper... | Nadav Rotem | 2013-02-27 | 1 | -9/+10 |
| * | Reverted: r176136 - Have a way for a target to opt-out of target-independent ... | Michael Ilseman | 2013-02-27 | 1 | -1/+1 |
| * | SelectionDAG: If llvm.donothing has a landingpad, we should clear | Manman Ren | 2013-02-27 | 1 | -0/+7 |
| * | Have a way for a target to opt-out of target-independent fast isel | Michael Ilseman | 2013-02-26 | 1 | -1/+1 |
| * | Fix wording. | Chad Rosier | 2013-02-25 | 1 | -3/+4 |
| * | [fast-isel] Add X86FastIsel::FastLowerArguments to handle functions with 6 or | Chad Rosier | 2013-02-25 | 1 | -0/+10 |
| * | pre-RA-sched fix: only reevaluate physreg interferences when necessary. | Andrew Trick | 2013-02-25 | 1 | -33/+61 |
| * | 'Hexadecimal' has two 'a's and only one 'i'. | Matt Beaumont-Gay | 2013-02-25 | 1 | -2/+2 |
| * | Fix spelling noticed by Duncan. | Chandler Carruth | 2013-02-25 | 1 | -3/+3 |
| * | Fix the root cause of PR15348 by correctly handling alignment 0 on | Chandler Carruth | 2013-02-25 | 2 | -0/+9 |
| * | SelectionDAG compile time improvement. | Nadav Rotem | 2013-02-22 | 1 | -0/+19 |
| * | Fix isa<> check which could never be true. | Pete Cooper | 2013-02-22 | 1 | -1/+1 |
