2010-12-01 |
Chris Lattner | fix a bozo bug I introduced in r119930, causing a miscompile of
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | Enhance DSE to handle the variable index case in PR8657.
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | remove fixme comment too.
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | check in *all* files. This is now handled by my previous...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | teach DSE to use GetPointerBaseWithConstantOffset to...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | move GetPointerBaseWithConstantOffset out of GVN into...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | remove a fixed fixme
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | Make DeleteDeadInstruction be a static function, move...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | switch RemoveAccessedObjects to use AliasAnalysis:...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | two improvements to RemoveAccessedObjects:
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | remove the "undead" terminology, which is nonstandard...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | move call handling in handleEndBlock up a bit, and...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | various cleanups and code simplification
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | make getPointerSize a static function. Add ivars to...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | reduce indentation, clean up TD use a bit.
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | forbid rtti and exceptions
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | enhance isRemovable to refuse to delete volatile mem...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | add TLI support indicating that jumps are more expensive...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | Rewrite the main DSE loop to be written in terms of...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | death to extraneous \n's.
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | rename a function and reduce some indentation, no functional...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | remove a use of llvm-dis
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | remove the pointless check of MemoryUseIntrinsic from
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | strength reduce this.
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | getLocationForDest should work for memset as well.
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | rename doesClobberMemory -> hasMemoryWrite to be more...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | clean up handling of 'free', detangling it from everything...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | merge one more away
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | I already merged partial-overwrite.ll -> PartialStore.ll
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | clean up DSE tests, removing some poorly reduced and...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | enhance basicaa to return "Mod" for a memcpy call when the
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | Teach basicaa that memset's modref set is at worst...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | my previous patch would cause us to start deleting...
|
commit | commitdiff | tree |
2010-11-30 |
Chris Lattner | two changes to DSE that shouldn't affect anything:
|
commit | commitdiff | tree |
2010-11-29 |
Chris Lattner | prune an llvmcontext include and simplify some code.
|
commit | commitdiff | tree |
2010-11-29 |
Chris Lattner | Generalize the darwin wrapper hack to work with generic...
|
commit | commitdiff | tree |
2010-11-29 |
Chris Lattner | convert llvm-ar and llvm-ranlib to raw_ostream from...
|
commit | commitdiff | tree |
2010-11-29 |
Chris Lattner | fix PR8677, patch by Jakub Staszak!
|
commit | commitdiff | tree |
2010-11-28 |
Chris Lattner | fix PR8686, accepting a 'b' suffix at the end of all...
|
commit | commitdiff | tree |
2010-11-28 |
Chris Lattner | add a function to the C api to get the context out...
|
commit | commitdiff | tree |
2010-11-23 |
Chris Lattner | add a MemoryBuffer::getOpenFile method, which turns...
|
commit | commitdiff | tree |
2010-11-23 |
Chris Lattner | Revert functionality doug added in r98575 that was...
|
commit | commitdiff | tree |
2010-11-23 |
Chris Lattner | fix this harder.
|
commit | commitdiff | tree |
2010-11-23 |
Chris Lattner | fix a bug I introduced on the other side of the #ifdef
|
commit | commitdiff | tree |
2010-11-23 |
Chris Lattner | reimplement SwapByteOrder.h in terms of overloading...
|
commit | commitdiff | tree |
2010-11-23 |
Chris Lattner | add a 'LLVMConstIntOfArbitraryPrecision' api to the...
|
commit | commitdiff | tree |
2010-11-23 |
Chris Lattner | duncan's spider sense was right, I completely reversed...
|
commit | commitdiff | tree |
2010-11-23 |
Chris Lattner | filecheckize
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | fix comment
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | apparently tailcalls are better on darwin/x86-64 than...
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | apply Dan's fix for PR8268 which allows constant folding...
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | add some helper methods for asmprinter flags, from...
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | implement PR8524, apparently mainline gas accepts movq...
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | rework some DSE paths to use the newly-public "getPointerDep...
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | the getLocationForSource/Dest methods can be static.
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | add "getLocation" method to AliasAnalysis for getting...
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | implement PR8576, deleting dead stores with intervening...
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | file checkize
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | add some random notes.
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | optimize:
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | tail calls on x86 are implemented.
|
commit | commitdiff | tree |
2010-11-21 |
Chris Lattner | Implement PR8644: forwarding a memcpy value to a byval,
|
commit | commitdiff | tree |
2010-11-20 |
Chris Lattner | a byval argument without an align can have an arbitrary...
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | slightly simplify code and substantially improve comment...
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | remove a pointless restriction from memcpyopt. It was
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | filecheckize, this is still not optimal, see PR8643
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | remove another pointless noalias check: M is a memcpy...
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | use AA::isNoAlias instead of open coding it. Remove...
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | finish a thought.
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | rearrange some code, splitting memcpy/memcpy optimization
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | allow eliminating an alloca that is just copied from...
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | enhance the "alloca is just a memcpy from constant...
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | fix a small oversight in the "eliminate memcpy from...
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | filecheckize
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | trivial QoI improvement. On this invalid input:
|
commit | commitdiff | tree |
2010-11-18 |
Chris Lattner | patch from Frits van Bommel:
|
commit | commitdiff | tree |
2010-11-17 |
Chris Lattner | add some justification for "using namespace llvm;"
|
commit | commitdiff | tree |
2010-11-17 |
Chris Lattner | various cleanups and other improvements, patch by Zhanyong...
|
commit | commitdiff | tree |
2010-11-17 |
Chris Lattner | With the newly simplified SourceMgr interfaces and...
|
commit | commitdiff | tree |
2010-11-17 |
Chris Lattner | now that AsmPrinter::EmitInlineAsm is factored right...
|
commit | commitdiff | tree |
2010-11-17 |
Chris Lattner | rearrange how the handler in SourceMgr is installed...
|
commit | commitdiff | tree |
2010-11-17 |
Chris Lattner | refactor the interface to EmitInlineAsm a bit, no functional...
|
commit | commitdiff | tree |
2010-11-17 |
Chris Lattner | fix PR8613 - Copy constructor of SwitchInst does not...
|
commit | commitdiff | tree |
2010-11-17 |
Chris Lattner | tidy up
|
commit | commitdiff | tree |
2010-11-16 |
Chris Lattner | update coding standards. Partial specialization is...
|
commit | commitdiff | tree |
2010-11-16 |
Chris Lattner | libc++ and compiler_rt are now dual licensed under...
|
commit | commitdiff | tree |
2010-11-16 |
Chris Lattner | Fix a bug I introduced in the ppc refactoring, which...
|
commit | commitdiff | tree |
2010-11-16 |
Chris Lattner | add copy of comment to the code that will survive the...
|
commit | commitdiff | tree |
2010-11-16 |
Chris Lattner | relax an assertion a bit, allowing the GPR argument of
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | fix a pasto that massively broke the ppc jit while...
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | Wire up primitive support in the assembler backend...
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | split out an encoder for memri operands, allowing a...
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | add support for encoding the lo14 forms used for a...
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | fix a regression with the new instprinter: we lost...
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | pull the code to get the operand value out of the loop.
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | split the giant encoder loop into two new helper functions.
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | reduce nesting and minor cleanups, no functionality...
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | implement the start of support for lo16 and ha16, allowing...
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | add a fixup for conditional branches, giving us output...
|
commit | commitdiff | tree |
2010-11-15 |
Chris Lattner | change direct branches to encode with the same encoding...
|
commit | commitdiff | tree |
next |