aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/MachineLICM.cpp
AgeCommit message (Expand)Author
2010-04-13Fast path implicit_def check.Evan Cheng
2010-04-13Avoid variable shadowing.Evan Cheng
2010-04-13Expand postra machine licm's capability a little more. If an instruction's re...Evan Cheng
2010-04-13Teach postra machine licm to hoist more obvious invariants, e.g. instructions...Evan Cheng
2010-04-13Eliminate MachineBasicBlock::const_livein_iterator and makeDan Gohman
2010-04-12Plug trivial leak.Benjamin Kramer
2010-04-09Delete this obsolete comment.Dan Gohman
2010-04-08Make post regalloc machine licm functional. It now passes all of MultiSource.Evan Cheng
2010-04-07Add comments for missed opportunities.Evan Cheng
2010-04-07Fix typo.Evan Cheng
2010-04-07Post regalloc LICM. Work in progress.Evan Cheng
2010-03-03- Change MachineInstr::isIdenticalTo to take a new option that determines whe...Evan Cheng
2010-02-28Don't try to replace physical registers when doing CSE.Dan Gohman
2010-02-28Don't unconditionally suppress hoisting of instructions with implicitDan Gohman
2010-02-09move target-independent opcodes out of TargetInstrInfoChris Lattner
2010-01-05Change errs() to dbgs().David Greene
2009-12-18Add Loop contains utility methods for testing whether a loopDan Gohman
2009-11-20Enable hoisting load from constant memories.Evan Cheng
2009-11-20Add option -licm-const-load to hoist all loads from constant memory.Evan Cheng
2009-11-20Add MachineBasicBlock::getName, and use it in place of getBasicBlock()->getName.Jakob Stoklund Olesen
2009-11-17Fix comment.Evan Cheng
2009-11-07- Add TargetInstrInfo::isIdentical(). It's similar to MachineInstr::isIdenticalEvan Cheng
2009-11-05Code refactoring.Evan Cheng
2009-11-03Re-apply 85799. It turns out my code isn't buggy.Evan Cheng
2009-11-02Revert 85799 for now. It might be breaking llvm-gcc driver.Evan Cheng
2009-11-02Initilize the machine LICM CSE map upon the first time an instruction is hois...Evan Cheng
2009-10-30Fix MachineLICM to use the correct virtual register class whenDan Gohman
2009-10-29Refactor the code for unfolding a load into a separate function.Dan Gohman
2009-10-28Simplify this code: if the unfolded load can't be hoisted, just deleteDan Gohman
2009-10-28Teach MachineLICM to unfold loads from constant memory fromDan Gohman
2009-10-25Remove includes of Support/Compiler.h that are no longer needed after theNick Lewycky
2009-10-25Remove VISIBILITY_HIDDEN from class/struct found inside anonymous namespaces.Nick Lewycky
2009-10-09Factor out LiveIntervalAnalysis' code to determine whether an instructionDan Gohman
2009-10-09isTriviallyReMaterializable checks theDan Gohman
2009-10-09Give Dan and my recent changes, machine LICM is now code size neutral.Evan Cheng
2009-10-09Fix a logic error that caused non-rematable loop invariants loads to be licm'...Evan Cheng
2009-10-07Replace TargetInstrInfo::isInvariantLoad and its target-specificDan Gohman
2009-09-26Don't hoist or sink instructions with physreg uses if the physreg isDan Gohman
2009-09-25Unbreak MachineLICM for instructions that reference RIP on x86-64 too.Dan Gohman
2009-08-22Convert DOUT to DEBUG(errs()...).Bill Wendling
2009-08-11Tidy #includes.Dan Gohman
2009-07-25More migration to raw_ostream, the water has dried up around the iostream hole.Daniel Dunbar
2009-02-27MachineLICM CSE should match destination register classes; avoid hoisting imp...Evan Cheng
2009-02-05Machine LICM increases register pressure and it almost always increase code s...Evan Cheng
2009-02-05Teach machine licm to CSE hoisted instructions.Evan Cheng
2009-02-04Don't call isInvariantLoad twice.Evan Cheng
2009-02-04Don't bother hoisting out a "cheap" instruction if all of its uses are PHIs. ...Evan Cheng
2009-02-04For now, only hoist re-materilizable instructions. LICM will increase registe...Evan Cheng
2009-01-15Simplify the MachineLICM pass by having it only traverse outerDan Gohman
2009-01-05Tidy up #includes, deleting a bunch of unnecessary #includes.Dan Gohman