oota-llvm.git
2005-10-11 Chris LattnerFix another lsr-is-nondeterministic case
2005-10-11 Chris LattnerFix a powerpc crash on CodeGen/Generic/llvm-ct-intrinsi...
2005-10-11 Chris LattnerAdd a canonicalization that got lost, fixing PowerPC...
2005-10-10 Chris Lattnerclean up some corner cases
2005-10-10 Chris LattnerImplement trivial DSE. If two stores are neighbors...
2005-10-10 Chris LattnerAdd support for CombineTo, allowing the dag combiner...
2005-10-10 Chris Lattnernew testcase
2005-10-10 Nate BegemanTeach the DAGCombiner several new tricks, teaching...
2005-10-10 Chris LattnerA testcase sitting in my tree
2005-10-10 Chris LattnerFix comment
2005-10-10 Chris LattnerAdd ISD::ADD to MaskedValueIsZero
2005-10-10 Chris LattnerThis function is now dead
2005-10-10 Chris LattnerEnable Nate's excellent DAG combiner work by default...
2005-10-10 Chris LattnerThese definitions have been moved to common code.
2005-10-10 Chris LattnerPull DAG ISel generation nodes out of the PowerPC backe...
2005-10-09 Chris Lattneradd a todo for something I noticed
2005-10-09 Chris Lattner(X & Y) & C == 0 if either X&C or Y&C are zero
2005-10-09 Chris LattnerMake MaskedValueIsZero a bit more aggressive
2005-10-09 Andrew LenharthThis seems useful from the original patch that added...
2005-10-09 Chris LattnerFix funky xcode indentation
2005-10-09 Chris LattnerHrm, you didn't see this.
2005-10-09 Chris LattnerFix a source of non-determinism in the backend: the...
2005-10-09 Chris LattnerWhen emiting a CopyFromReg and the source is already...
2005-10-09 Chris LattnerDisable formation of rlwinm instructions from SRA bases...
2005-10-09 Chris LattnerTestcase that is miscompiled by the ppc backend
2005-10-08 Nate BegemanRemove another unused file. Preparing for the great...
2005-10-08 Chris Lattnerremove a comma to compile with pedantic gcc
2005-10-08 Nate BegemanRemove a file that is no longer used
2005-10-08 Nate BegemanLo and behold, the last bits of SelectionDAG.cpp have...
2005-10-07 Chris LattnerWhen preselecting, favor things that have low depth...
2005-10-07 Chris Lattnerremove debugging code
2005-10-07 Chris Lattnerimplement CodeGen/PowerPC/div-2.ll:test2-4 by propagati...
2005-10-07 Chris LattnerThese don't need to be sdivs.
2005-10-07 Chris Lattnerfix indentation
2005-10-07 Chris LattnerTurn sdivs into udivs when we can prove the sign bits...
2005-10-07 Chris LattnerNew testcase, should turn into a ushr
2005-10-07 Jeff CohenRemove prolific source of VC++ truncation warnings.
2005-10-07 Jeff CohenRemove useless variable.
2005-10-07 Chris Lattneradd a hack to work around broken VC++ scoping rules...
2005-10-07 Chris LattnerFix a CQ regression from my patch to split F32/F64...
2005-10-07 Chris Lattnerdouble to float casts should nto result in an fmr
2005-10-07 Chris LattnerFix CodeGen/Generic/bool-to-double.ll
2005-10-07 Chris Lattnertest broken in the IA64 backend.
2005-10-06 Chris LattnerPull out Call, reducing stack frame size from 6032...
2005-10-06 Chris LattnerPull out setcc, this reduces stack frame size from...
2005-10-06 Chris LattnerPull two more methods out, reducing stack frame size...
2005-10-06 Chris LattnerAdd a recursive-iterative hybrid stage to attempt to...
2005-10-06 Chris Lattnersilence a bogus GCC warning
2005-10-06 Chris LattnerFix the LLC regressions on X86 last night. In particul...
2005-10-06 Andrew LenharthThis is suppose to work now
2005-10-06 Andrew Lenharthremove VAX compatibility instruction, we will never...
2005-10-06 Chris LattnerMake the legalizer completely non-recursive
2005-10-05 Nate BegemanLet the combiner handle more cases
2005-10-05 Nate BegemanRemove some bad code from Legalize
2005-10-05 Nate BegemanCheck in some more DAGCombiner pieces
2005-10-05 Chris LattnerFix a bug in the local spiller, where we could take...
2005-10-05 Chris Lattnersilence some warnings
2005-10-05 Chris Lattnerimplement visitBR_CC so that PowerPC/inverted-bool...
2005-10-05 Chris Lattnerfix some pastos
2005-10-05 Chris LattnerAdd a new HandleNode class, which is used to handle...
2005-10-05 Chris Lattneradd a helper class
2005-10-05 Chris LattnerImplement the code for PowerPC/inverted-bool-compares...
2005-10-05 Chris Lattnermake sure that -view-isel-dags is the input to the...
2005-10-05 Chris LattnerFix a crash compiling Olden/tsp
2005-10-05 Chris LattnerAdd some rules for building preprocessed files
2005-10-04 Chris Lattnerrefactor a bit of code.
2005-10-04 Chris LattnerMinor speedup to avoid array searches given a Use*...
2005-10-04 Chris LattnerChange the signature of replaceUsesOfWithOnConstant...
2005-10-04 Chris LattnerChange the signature of replaceUsesOfWithOnConstant...
2005-10-04 Chris LattnerFor large constants (e.g. arrays and structs with many...
2005-10-04 Chris Lattnerminor cleanup/fastpath for the bcreader. This speeds...
2005-10-04 Jim LaskeyReverting to version - until problem isolated.
2005-10-04 Chris LattnerAdd a forward def
2005-10-04 Nate BegemanFix some faulty logic in the libcall inserter.
2005-10-04 Chris Lattnerimplement the struct version of the array speedup,...
2005-10-04 Chris LattnerFix DemoteRegToStack on an invoke. This fixes PR634.
2005-10-04 Nate BegemanAdd back a workaround that fixes some breakages from...
2005-10-03 Chris LattnerClean up the code a bit. Use isInstructionTriviallyDea...
2005-10-03 Chris Lattnernew testcase for PR635
2005-10-03 Chris LattnerChange ConstantArray::replaceUsesOfWithOnConstant to...
2005-10-03 Chris Lattnermove some methods, no other changes
2005-10-03 Chris Lattnerminor microoptimizations
2005-10-03 Chris LattnerUse a map to cache the ModuleType information, so we...
2005-10-03 Jim LaskeyRefactor gathering node info and emission.
2005-10-03 Chris Lattnerclean up this code a bit, no functionality change
2005-10-03 Chris LattnerSpeed up the asm printer a lot by not printing formatte...
2005-10-03 Chris LattnerBreak the body of the loop out into a new method
2005-10-03 Chris LattnerFix case of path
2005-10-03 Chris LattnerMake IVUseShouldUsePostIncValue more aggressive when...
2005-10-03 Chris LattnerRefactor some code into a function
2005-10-03 Chris LattnerThis break is bogus and I have no idea why it was there...
2005-10-03 Chris Lattnerwhen checking if we should move a split edge block...
2005-10-03 Chris LattnerThis member can be const too
2005-10-02 Chris Lattnerput the right labels on the data
2005-10-02 Chris LattnerFix a problem where the legalizer would run out of...
2005-10-02 Chris Lattnersilence a bogus warning
2005-10-02 Chris Lattnersilence some warnings
2005-10-02 Chris Lattnersilence a warning
2005-10-02 Chris Lattneradd patterns for float binops and fma ops
2005-10-02 Chris LattnerSort the cpu and features table, so that the alpha...
next