index
:
emscripten-fastcomp
master
LLVM with the emscripten fastcomp javascript backend
git repository hosting
about
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
lib
/
CodeGen
/
MachineLICM.cpp
Age
Commit message (
Expand
)
Author
2011-10-20
As Evan suggested, loads from constant pool are safe to speculate.
Devang Patel
2011-10-20
Add a comment.
Devang Patel
2011-10-17
Constraint register class with constrainRegClass() to CSE a virtual into anot...
Evan Cheng
2011-10-17
It is safe to speculate load from GOT. This fixes performance regression caus...
Devang Patel
2011-10-13
Tabs to spaces.
Nick Lewycky
2011-10-12
Disable machine LICM speculation check (for profitability) until I have time ...
Evan Cheng
2011-10-12
Expand the check for a landing pad so that it looks at the basic block's
Bill Wendling
2011-10-12
Fix r141744.
Evan Cheng
2011-10-11
Refine r141689 with a tri-state variable.
Evan Cheng
2011-10-11
N.B. This is with the new EH scheme:
Bill Wendling
2011-10-11
Add dominance check for the instruction being hoisted.
Devang Patel
2011-10-10
Revert r141569 and r141576.
Devang Patel
2011-10-10
If loop header is also loop exiting block then it may not be safe to hoist in...
Devang Patel
2011-10-10
Add dominance check for the instruction being hoisted.
Devang Patel
2011-09-01
Teach MachineLICM reg pressure tracking code to deal with MVT::untyped. Sorry...
Evan Cheng
2011-06-29
Sink SubtargetFeature and TargetInstrItineraries (renamed MCInstrItineraries)...
Evan Cheng
2011-06-28
- Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and
Evan Cheng
2011-06-27
More refactoring. Move getRegClass from TargetOperandInfo to TargetInstrInfo.
Evan Cheng
2011-04-11
Look pass copies when determining whether hoisting would end up inserting mor...
Evan Cheng
2011-03-07
Move getRegPressureLimit() from TargetLoweringInfo to TargetRegisterInfo.
Cameron Zwarich
2011-01-20
Sorry, several patches in one.
Evan Cheng
2011-01-10
Simplify a bunch of isVirtualRegister() and isPhysicalRegister() logic.
Jakob Stoklund Olesen
2010-11-11
Add a FIXME comment.
Dan Gohman
2010-11-01
MachineLICM should not claim to be preserving the CFG when it can split critical
Jakob Stoklund Olesen
2010-10-26
Use instruction itinerary to determine what instructions are 'cheap'.
Evan Cheng
2010-10-20
More accurate estimate / tracking of register pressure.
Evan Cheng
2010-10-20
Make CodeGen TBAA-aware.
Dan Gohman
2010-10-19
Re-enable register pressure aware machine licm with fixes. Hoist() may have
Evan Cheng
2010-10-19
Get rid of static constructors for pass registration. Instead, every pass ex...
Owen Anderson
2010-10-19
Revert r116781 "- Add a hook for target to determine whether an instruction def
Daniel Dunbar
2010-10-19
Fix for machine licm assert: RCCost <= RegPressure[RCId]
Andrew Trick
2010-10-19
- Add a hook for target to determine whether an instruction def is
Evan Cheng
2010-10-16
More machine LICM work. It now tracks register pressure for path from prehead...
Evan Cheng
2010-10-14
Register pressure and instruction latency aware machine LICM. Work in progress.
Evan Cheng
2010-10-12
Begin adding static dependence information to passes, which will allow us to
Owen Anderson
2010-10-08
Don't waste time unfolding simple loads. The unfolded copy won't be hoisted.
Evan Cheng
2010-10-07
Now with fewer extraneous semicolons!
Owen Anderson
2010-08-06
Reapply r110396, with fixes to appease the Linux buildbot gods.
Owen Anderson
2010-08-06
Revert r110396 to fix buildbots.
Owen Anderson
2010-08-05
Don't use PassInfo* as a type identifier for passes. Instead, use the addres...
Owen Anderson
2010-07-29
Comment typo.
Dale Johannesen
2010-07-21
Fix batch of converting RegisterPass<> to INTIALIZE_PASS().
Owen Anderson
2010-07-20
Fix test for switch statements and increase
Dale Johannesen
2010-07-20
Don't hoist things out of a large switch inside a
Dale Johannesen
2010-07-14
Teach ProcessImplicitDefs to transform more COPY instructions into IMPLICIT_D...
Evan Cheng
2010-07-12
change machinelicm to use MachineInstr::isSafeToMove. No
Chris Lattner
2010-07-09
Fix MachineLICM to actually visit inner loops.
Dan Gohman
2010-06-22
Move PHIElimination's SplitCriticalEdge for MachineBasicBlocks out
Dan Gohman
2010-05-29
Doh. Machine LICM is re-initializing the CSE map over and over. Patch by Anna...
Evan Cheng
2010-05-13
Teach MachineLICM and MachineSink how to clear kill flags conservatively
Dan Gohman
[next]