oota-llvm.git
2005-03-25 Chris LattnerTreat free operations as volatile, since they cannot...
2005-03-25 Chris LattnerNew testcase that crashes licm.
2005-03-25 Chris LattnerFix a bug where LICM was not updating AA information...
2005-03-25 Chris Lattnerremove a debugging timer.
2005-03-25 Chris LattnerTwo changes here:
2005-03-25 Chris LattnerMake the spliceFrom case where one graph is completely...
2005-03-24 Chris Lattneradd a new DSGraph::spliceFrom method, which violently...
2005-03-24 Chris Lattneradd new spliceFrom methods.
2005-03-24 Nate BegemanImplement next round of Pattern ISel fixes
2005-03-24 Nate BegemanFix an incorrect argument being passed to BuildMI for...
2005-03-24 Chris LattnerThis replaces the correct but slow code with a more...
2005-03-24 Chris Lattneronly look at successors of globals. This gets us down...
2005-03-24 Chris LattnerUnfortunately, a previous patch was not safe. Revert...
2005-03-24 Nate BegemanCommit Gabor Greif's patch to use iterators in lowering...
2005-03-24 Chris Lattnerdon't bother |'ing in 0's
2005-03-24 Chris Lattnerbe more aggressive about incompleteness marking
2005-03-24 Chris Lattnernew testcase that DS-AA is causing to be miscompiled.
2005-03-24 Chris Lattnereliminate dead variables, patch contributed by Gabor...
2005-03-24 Nate BegemanImplement more of the PPC32 Pattern ISel:
2005-03-24 Chris LattnerFix silly "no newline at end of file" warning
2005-03-24 Chris LattnerFix compilation errors, patch contributed by the fabulo...
2005-03-24 Chris Lattnerfix a compilation error, patch contributed by Bill...
2005-03-24 Nate BegemanAddition of the PPC32 Pattern ISel. While it is far...
2005-03-24 Nate BegemanRemove comments that are now meaningless from the patte...
2005-03-24 Chris LattnerFix a crash while promoting a value out of a loop from...
2005-03-24 Chris Lattnerteach ds-aa about mod/ref for external function calls.
2005-03-24 Chris LattnerSimplify dead code into a fixme :)
2005-03-24 Chris Lattnerfix a compiler crash in runtime/libprofile
2005-03-23 Chris Lattnerwrap a long line
2005-03-23 Chris LattnerIf we are calling an external function, chain to anothe...
2005-03-23 Chris LattnerGlobalMR should be able to analyze this function.
2005-03-23 Chris Lattnerif a function doesn't access memory at all, it definite...
2005-03-23 Chris LattnerMake this more efficient by only making one virtual...
2005-03-23 Chris LattnerMake this a bit more aggressive
2005-03-23 Chris Lattnera hack to allow count-aa to work with ds-aa :(
2005-03-23 Chris LattnerAdd two options to allow -count-aa to print queries...
2005-03-23 Misha BrukmanFix grammar
2005-03-23 Chris Lattnerenable -debug-only=licm
2005-03-23 Chris Lattnerturn a dead conditional into an assert.
2005-03-23 Chris LattnerTotally gut mergeInGraph. There is absolutely no reaso...
2005-03-23 Chris Lattnerwrap a long line
2005-03-23 Andrew Lenharthdon't lie to the register allocator
2005-03-23 Chris LattnerMake -steens-aa more conservative (aka correct) by...
2005-03-23 Tanya LattnerAdded alias analysis.
2005-03-23 Chris Lattnerimplement Analysis/DSGraph/field-sensitive.ll
2005-03-23 Chris Lattnernew testcase to verify that we have field sensitive...
2005-03-23 Chris Lattnermake sure this test tests the intended target.
2005-03-23 Chris Lattnerdon't crash in some bad cases.
2005-03-22 Chris LattnerSeveral changes here:
2005-03-22 Chris LattnerWe might as well check the TD graphs as well, even...
2005-03-22 Chris LattnerMark external globals incomplete in the BU Globals...
2005-03-22 Chris LattnerNew testcase, the BU pass is marking the global complet...
2005-03-22 Tanya LattnerUpdating my entry.
2005-03-22 Chris LattnerFix a serious bug where we didn't insert globals into...
2005-03-22 Andrew Lenharthhum, it is good to use real instructions
2005-03-22 Chris LattnerDirectly count the number of memory instructions.
2005-03-22 Chris LattnerTimers SHOULD NOT record the time taken to count the...
2005-03-22 Chris LattnerRemove an iteration pass over the entire scalarmap...
2005-03-22 Chris LattnerWhen making a clone of a DSGraph from the BU pass,...
2005-03-22 Chris LattnerDon't use operator[], use the new method instead, which...
2005-03-22 Chris Lattneradd a method
2005-03-22 Chris LattnerNow that the dead ctor is gone, nothing uses the old...
2005-03-22 Chris Lattnerremove a dead ctor
2005-03-22 Chris Lattnernow that the second argument is always this->ReturnNode...
2005-03-22 Chris Lattnerinstead of using a local RetValMap, just use the graph...
2005-03-22 Andrew Lenharthtwo things: 1)evilness reduction patch, reduces the...
2005-03-22 Chris Lattnernow that the valuemapping is always the local scalar...
2005-03-22 Chris Lattnerremove the second argument to cloneInto
2005-03-22 Chris Lattnerdrop the second argument to cloneInto, which is always...
2005-03-22 Chris Lattneradd some timers, don't clone aux nodes
2005-03-22 Chris Lattnermove this out of line
2005-03-22 Chris Lattnermove this stuff out of line
2005-03-22 Chris Lattnerdon't generate temporary scalarmaps.
2005-03-21 Chris Lattnerallow passing clone flags down to cloneInto
2005-03-21 Chris Lattnerallow passing clone flags into the ctor
2005-03-21 Chris LattnerEnhance the TD pass to build composite graphs when...
2005-03-21 Chris Lattneradd a new map
2005-03-21 Chris Lattnermake this const correct
2005-03-21 Chris LattnerFix a major problem where we didn't add call graph...
2005-03-21 Chris LattnerNew testcase, which is important to handle correctly.
2005-03-21 Misha BrukmanWe may be adding functions to the Module during initial...
2005-03-21 Chris LattnerThis error message changed.
2005-03-21 Chris LattnerUgh, for some reason, I can't call this unless the...
2005-03-21 Chris LattnerThe reachability cloner should add arguments to merged...
2005-03-21 Chris LattnerMake the first operand of this method be modifiable.
2005-03-21 Chris Lattneradd a method
2005-03-21 Chris LattnerIf merging two calls like: foo(A) and bar(B, C), make...
2005-03-21 Chris Lattner'note to self'
2005-03-21 Chris LattnerRemove a bunch of cruft and dead code for handling...
2005-03-21 Chris LattnerNew testcase that the assembler incorrectly accepts.
2005-03-21 Tanya LattnerMake this method non-const. Making it const causes...
2005-03-21 Chris LattnerChange the way that the TD pass inlines graphs. Instea...
2005-03-21 Chris Lattnerrename a method add a data structure.
2005-03-21 Chris LattnerProvide a guaranteed definition of intptr_t. Thansk...
2005-03-20 Chris LattnerDon't strip modref bits when inlining down the call...
2005-03-20 Chris LattnerRemove the ability to keep track of inlined globals...
2005-03-20 Chris LattnerRemove the InlinedGlobals set which is always empty.
2005-03-20 Chris LattnerThis call is always a noop, remove it.
2005-03-20 Chris Lattnerremove some pointless asserts
2005-03-20 Chris Lattner-steens doesn't use the inlined globals facility.
next