| Commit message (Expand) | Author | Age | Files | Lines |
* | Rename TargetSubtarget to TargetSubtargetInfo for consistency. | Evan Cheng | 2011-07-01 | 1 | -2/+2 |
* | Sink SubtargetFeature and TargetInstrItineraries (renamed MCInstrItineraries)... | Evan Cheng | 2011-06-29 | 1 | -0/+1 |
* | - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and | Evan Cheng | 2011-06-28 | 1 | -7/+7 |
* | pre-RA-sched: Cleanup register pressure tracking. | Andrew Trick | 2011-06-27 | 1 | -9/+1 |
* | The scheduler needs to be aware on the existence of untyped nodes when it per... | Owen Anderson | 2011-06-24 | 1 | -1/+2 |
* | Don't allocate empty read-only SmallVectors during SelectionDAG deallocation. | Benjamin Kramer | 2011-06-18 | 1 | -1/+1 |
* | Added -stress-sched flag in the Asserts build. | Andrew Trick | 2011-06-15 | 1 | -1/+1 |
* | Be careful about scheduling nodes above previous calls. It increase usages of | Evan Cheng | 2011-04-26 | 1 | -0/+19 |
* | Fix a ton of comment typos found by codespell. Patch by | Chris Lattner | 2011-04-15 | 1 | -1/+1 |
* | In the pre-RA scheduler, maintain cmp+br proximity. | Andrew Trick | 2011-04-14 | 1 | -0/+8 |
* | Recommit r129383. PreRA scheduler heuristic fixes: VRegCycle, TokenFactor lat... | Andrew Trick | 2011-04-13 | 1 | -46/+14 |
* | Revert 129383. It causes some targets to hit a scheduler assert. | Andrew Trick | 2011-04-12 | 1 | -7/+46 |
* | PreRA scheduler heuristic fixes: VRegCycle, TokenFactor latency. | Andrew Trick | 2011-04-12 | 1 | -46/+7 |
* | Added a check in the preRA scheduler for potential interference on a | Andrew Trick | 2011-04-07 | 1 | -0/+46 |
* | Improve pre-RA-sched register pressure tracking for duplicate operands. | Andrew Trick | 2011-03-09 | 1 | -1/+5 |
* | Fix some latent bugs if the nodes are unschedulable. We'd gotten away | Eric Christopher | 2011-03-08 | 1 | -0/+4 |
* | Fix for -sched-high-latency-cycles in sched=list-ilp mode. | Andrew Trick | 2011-03-05 | 1 | -1/+3 |
* | Increased the register pressure limit on x86_64 from 8 to 12 | Andrew Trick | 2011-03-05 | 1 | -1/+13 |
* | Introducing a new method of tracking register pressure. We can't | Andrew Trick | 2011-02-04 | 1 | -1/+73 |
* | whitespace | Andrew Trick | 2011-02-03 | 1 | -18/+18 |
* | Reapply 124301 | Devang Patel | 2011-01-27 | 1 | -1/+5 |
* | Revert 124301. | Devang Patel | 2011-01-26 | 1 | -5/+1 |
* | Process valid SDDbgValues even if the node does not have any order assigned. | Devang Patel | 2011-01-26 | 1 | -1/+5 |
* | Refactor. | Devang Patel | 2011-01-26 | 1 | -19/+30 |
* | This assertion is too restrictive, it does not apply for dangling dbg value n... | Devang Patel | 2011-01-25 | 1 | -8/+0 |
* | flags -> glue for selectiondag | Chris Lattner | 2010-12-23 | 1 | -49/+48 |
* | rename MVT::Flag to MVT::Glue. "Flag" is a terrible name for | Chris Lattner | 2010-12-21 | 1 | -7/+7 |
* | Two sets of changes. Sorry they are intermingled. | Evan Cheng | 2010-11-03 | 1 | -4/+7 |
* | Avoiding overly aggressive latency scheduling. If the two nodes share an | Evan Cheng | 2010-10-29 | 1 | -0/+9 |
* | Re-commit 117518 and 117519 now that ARM MC test failures are out of the way. | Evan Cheng | 2010-10-28 | 1 | -0/+3 |
* | Revert 117518 and 117519 for now. They changed scheduling and cause MC tests ... | Evan Cheng | 2010-10-28 | 1 | -3/+0 |
* | Fix a major bug in operand latency computation. The use index must be adjusted | Evan Cheng | 2010-10-28 | 1 | -0/+3 |
* | - Add TargetInstrInfo::getOperandLatency() to compute operand latencies. This | Evan Cheng | 2010-10-06 | 1 | -19/+1 |
* | Model Cortex-a9 load to SUB, RSB, ADD, ADC, SBC, RSC, CMN, MVN, or CMP | Evan Cheng | 2010-09-29 | 1 | -17/+17 |
* | Teach if-converter to be more careful with predicating instructions that would | Evan Cheng | 2010-09-10 | 1 | -9/+7 |
* | Add missing null check reported by Amaury Pouly. | Evan Cheng | 2010-08-10 | 1 | -2/+3 |
* | Fix a bug in the code which re-inserts DBG_VALUE nodes after scheduling; | Dan Gohman | 2010-07-10 | 1 | -1/+3 |
* | Reapply bottom-up fast-isel, with several fixes for x86-32: | Dan Gohman | 2010-07-10 | 1 | -8/+5 |
* | --- Reverse-merging r107947 into '.': | Bob Wilson | 2010-07-09 | 1 | -5/+8 |
* | Re-apply bottom-up fast-isel, with fixes. Be very careful to avoid emitting | Dan Gohman | 2010-07-09 | 1 | -8/+5 |
* | grammar tweak in comment. | Jim Grosbach | 2010-06-30 | 1 | -1/+1 |
* | Add a VT argument to getMinimalPhysRegClass and replace the copy related uses | Rafael Espindola | 2010-06-29 | 1 | -1/+1 |
* | Remove variables which are assigned to but for which the value | Duncan Sands | 2010-06-25 | 1 | -8/+1 |
* | It's possible that a flag is added to the SDNode that points back to the | Bill Wendling | 2010-06-24 | 1 | -11/+19 |
* | MorphNodeTo doesn't preserve the memory operands. Because we're morphing a node | Bill Wendling | 2010-06-23 | 1 | -0/+21 |
* | Use A.append(...) instead of A.insert(A.end(), ...) when A is a | Dan Gohman | 2010-06-21 | 1 | -1/+1 |
* | Code refactoring, no functionality changes. | Evan Cheng | 2010-06-10 | 1 | -82/+83 |
* | Fix some latency computation bugs: if the use is not a machine opcode do not ... | Evan Cheng | 2010-05-28 | 1 | -6/+15 |
* | Allow targets more controls on what nodes are scheduled by reg pressure, what... | Evan Cheng | 2010-05-20 | 1 | -0/+20 |
* | Add a hybrid bottom up scheduler that reduce register usage while avoiding | Evan Cheng | 2010-05-20 | 1 | -1/+36 |