| Age | Commit message (Expand) | Author |
|---|
| 2009-04-29 | Instead of passing in an unsigned value for the optimization level, use an enum, | Bill Wendling |
| 2009-04-29 | Second attempt: | Bill Wendling |
| 2009-04-28 | r70270 isn't ready yet. Back this out. Sorry for the noise. | Bill Wendling |
| 2009-04-28 | Massive check in. This changes the "-fast" flag to "-O#" in llc. If you want to | Bill Wendling |
| 2009-04-16 | In the list-burr's pseudo two-addr dependency heuristics, don't | Dan Gohman |
| 2009-04-16 | Handle SUBREG_TO_REG instructions with the same heuristics | Dan Gohman |
| 2009-03-24 | Minor compile-time optimization; don't bother checking | Dan Gohman |
| 2009-03-24 | Add a pre-pass to the burr-list scheduler which makes adjustments to | Dan Gohman |
| 2009-03-23 | When unfolding a load during scheduling, the new operator node has | Dan Gohman |
| 2009-03-23 | Fix canClobberPhysRegDefs to check all SDNodes grouped together | Dan Gohman |
| 2009-03-20 | For inline asm output operand that matches an input. Encode the input operand... | Evan Cheng |
| 2009-03-12 | Fix a typo in a comment. | Dan Gohman |
| 2009-03-06 | Fix ScheduleDAGRRList::CopyAndMoveSuccessors' handling of nodes | Dan Gohman |
| 2009-03-04 | Fix PR3701. 1. X86 target renamed eflags register to flags. This matches what... | Evan Cheng |
| 2009-02-12 | Oops. Last second clean up messed things up. | Evan Cheng |
| 2009-02-12 | Replace one of burr scheduling heuristic with something more sensible. Now ca... | Evan Cheng |
| 2009-02-11 | Fix a comment. | Dan Gohman |
| 2009-02-11 | Don't use special heuristics for nodes with no data predecessors | Dan Gohman |
| 2009-02-11 | Delete the heuristic for non-livein CopyFromReg nodes. Non-liveinness | Dan Gohman |
| 2009-02-11 | When scheduling a block in parts, keep track of the overall | Dan Gohman |
| 2009-02-10 | Factor out more code for computing register live-range informationfor | Dan Gohman |
| 2009-02-10 | Fix PR3457: Ignore control successors when looking for closest scheduled succ... | Evan Cheng |
| 2009-02-06 | Move ScheduleDAGSDNodes.h to be a private header. Front-ends | Dan Gohman |
| 2009-01-29 | Make a few things const, fix some comments, and simplify | Dan Gohman |
| 2009-01-15 | Move a few containers out of ScheduleDAGInstrs::BuildSchedGraph | Dan Gohman |
| 2009-01-13 | The list-td and list-tdrr schedulers don't yet support physreg | Dan Gohman |
| 2009-01-12 | Fix PR3241: Currently EmitCopyFromReg emits a copy from the physical register... | Evan Cheng |
| 2009-01-09 | Minor debug output tweak. | Evan Cheng |
| 2009-01-07 | Remove redundant 'else's. No functionality change. | Dan Gohman |
| 2009-01-06 | Update these argument lists for the isNormalMemory | Dan Gohman |
| 2009-01-06 | Use a latency value of 0 for the artificial edges inserted by | Dan Gohman |
| 2009-01-05 | Tidy up #includes, deleting a bunch of unnecessary #includes. | Dan Gohman |
| 2009-01-03 | CommuteNodesToReducePressure() is now removed. | Dan Gohman |
| 2009-01-03 | Remove the code from the scheduler that commuted two-address | Dan Gohman |
| 2008-12-23 | Rename BuildSchedUnits to BuildSchedGraph, and refactor the | Dan Gohman |
| 2008-12-16 | Fix some register-alias-related bugs in the post-RA scheduler liveness | Dan Gohman |
| 2008-12-16 | Make addPred and removePred return void, since the return value is not | Dan Gohman |
| 2008-12-09 | Rewrite the SDep class, and simplify some of the related code. | Dan Gohman |
| 2008-12-04 | Make debug output more informative. | Dan Gohman |
| 2008-11-25 | Initial support for anti-dependence breaking. Currently this code does not | Dan Gohman |
| 2008-11-24 | Check in the rest of this change. The isAntiDep flag needs to be passed | Dan Gohman |
| 2008-11-21 | Add a flag to SDep for tracking which edges are anti-dependence edges. | Dan Gohman |
| 2008-11-21 | Rename SDep's isSpecial to isArtificial, to make this field a little | Dan Gohman |
| 2008-11-21 | Remove the CycleBound computation code from the ScheduleDAGRRList | Dan Gohman |
| 2008-11-21 | Change these schedulers to not emit no-ops. It turns out that | Dan Gohman |
| 2008-11-20 | Remove a remnant of list-burr's fast mode. | Dan Gohman |
| 2008-11-20 | Factor out the SethiUllman numbering logic from the list-burr and | Dan Gohman |
| 2008-11-20 | Remove the "fast" form of the list-burr scheduler, and use the | Dan Gohman |
| 2008-11-20 | Facter AddPseudoTwoAddrDeps and associated infrasructure out of | Dan Gohman |
| 2008-11-20 | Factor out the code for verifying the work of the scheduler, | Dan Gohman |
