summaryrefslogtreecommitdiffstats
path: root/llvm
Commit message (Collapse)AuthorAgeFilesLines
* Push twines deeper into SourceMgr's error handling methods.Benjamin Kramer2010-09-2713-52/+54
| | | | llvm-svn: 114847
* MMX parameters aren't handled here yet.Dale Johannesen2010-09-271-0/+3
| | | | llvm-svn: 114844
* Don't try to make a vector of x86mmx; this won't work,Dale Johannesen2010-09-271-2/+4
| | | | | | and asserts. llvm-svn: 114843
* writeGraph doesn't need its ShortNames argument.Dan Gohman2010-09-271-3/+2
| | | | llvm-svn: 114842
* Delete an unused function.Dan Gohman2010-09-271-2/+0
| | | | llvm-svn: 114841
* Delete an unused declaration.Dan Gohman2010-09-271-2/+0
| | | | llvm-svn: 114839
* Don't construct a redundant GraphWriter object.Dan Gohman2010-09-271-7/+4
| | | | llvm-svn: 114838
* Factor out code from the standalone WriteGraph function into a helperDan Gohman2010-09-271-10/+20
| | | | | | function on GraphWriter. llvm-svn: 114837
* Remove reference to nonexistent test in CMake makefile for unit testsDouglas Gregor2010-09-271-2/+1
| | | | llvm-svn: 114835
* Constify properlyDominates in the same manner as dominates.Dan Gohman2010-09-271-3/+22
| | | | | | Add constified overloads for findNearestCommonDominator. llvm-svn: 114834
* Add support for viewing graphviz graphs with xdot.py.Dan Gohman2010-09-274-5/+105
| | | | llvm-svn: 114832
* Move ValueMapTest from ADT to VMCore so that ADT doesn't needDan Gohman2010-09-272-1/+1
| | | | | | to link in "core". llvm-svn: 114831
* Add an all() method to BitVector, for testing whether all bits are set.Dan Gohman2010-09-274-0/+27
| | | | llvm-svn: 114830
* Remove WriteGraph's Name argument, which it didn't use, andDan Gohman2010-09-272-8/+7
| | | | | | | rename writeHeader's Name argument to Title, to be consistent with WriteGraph. llvm-svn: 114829
* Make this code 65-bit clean.Dan Gohman2010-09-271-1/+1
| | | | llvm-svn: 114828
* Fix indentation.Dan Gohman2010-09-271-2/+2
| | | | llvm-svn: 114827
* Removed remnant target library dependencies from LLVMLibDeps.cmakeOscar Fuentes2010-09-271-2/+0
| | | | llvm-svn: 114826
* Updated LLVMLibDeps.cmakeOscar Fuentes2010-09-271-9/+11
| | | | llvm-svn: 114825
* MC-COFF: Drop empty sections, and label symbols. Convert relocationsMichael J. Spencer2010-09-273-143/+196
| | | | | | | | targeted at symbols into relocations relative to the containing section. Patch by Nathan Jeffords! llvm-svn: 114823
* yet more aliases.Chris Lattner2010-09-272-0/+7
| | | | llvm-svn: 114822
* add a couple more aliases, rdar://8456378Chris Lattner2010-09-272-0/+7
| | | | llvm-svn: 114821
* fix rdar://8470918 - llvm-mc can't assemble smovlChris Lattner2010-09-272-0/+11
| | | | llvm-svn: 114819
* Fix rdar://8468087 - llvm-mc commutes fmul (and friend) operands.Chris Lattner2010-09-272-5/+20
| | | | | | | | My previous fix for rdar://8456371 should only apply to fmulp/faddp, not to fmul/fadd. Instruction set orthogonality is overrated or something. llvm-svn: 114818
* the latest assembler that runs on powerpc 10.4 machines doesn'tChris Lattner2010-09-276-2/+40
| | | | | | | support aligned comm. Detect when compiling for 10.4 and don't emit an alignment for comm. THis will hopefully fix PR8198. llvm-svn: 114817
* improve indentationChris Lattner2010-09-271-3/+3
| | | | llvm-svn: 114815
* Insert missing coherency in comment. Add a quick check for hardwareEric Christopher2010-09-271-3/+7
| | | | | | divide support also. llvm-svn: 114813
* Mass rename for Jim.Eric Christopher2010-09-271-33/+33
| | | | llvm-svn: 114812
* This code should never fire on non-darwin subtargets.Eric Christopher2010-09-271-3/+3
| | | | llvm-svn: 114811
* implement support for 'clr' alias. This is part of rdar://8416805,Chris Lattner2010-09-272-0/+25
| | | | | | but balrog was wanting it on irc. llvm-svn: 114809
* Fixed some tests to avoid LiveIntervals::getInstructionFromIndex(..) ↵Lang Hames2010-09-263-9/+8
| | | | | | overhead where possible. Thanks to Jakob for the suggestions. llvm-svn: 114798
* New cmake options LLVM_INCLUDE_X (X = {TOOLS, EXAMPLES, TESTS}) forOscar Fuentes2010-09-252-9/+30
| | | | | | not generating build targets for those parts of the build. llvm-svn: 114797
* Avoid warnings about implicit conversions to `bool' in MSVC. This timeOscar Fuentes2010-09-253-4/+4
| | | | | | | | for real. Patch by Nathan Jeffords! llvm-svn: 114796
* Reverting "CMake: Don't include tools, unittets, or examples asOscar Fuentes2010-09-254-28/+14
| | | | | | | | | | | | available targets unless LLVM_INCLUDE_X is ON. LLVM_BUILD_X implies LLVM_INCLUDE_X" It breaks the configuration phase when cmake is invoked without parameters, it is too complex for the purpose and introduces an incovenience for the user (as both LLVM_BUILD_X and LLVM_INCLUDE_X must set to OFF for not including X on the build) llvm-svn: 114795
* Avoid using VNInfo::getCopy as much as possible. I want to get rid of it.Jakob Stoklund Olesen2010-09-252-10/+10
| | | | llvm-svn: 114794
* temporarily revert developer policy change as a courtesy to vikram.Chris Lattner2010-09-251-18/+14
| | | | llvm-svn: 114792
* Removed VNInfo::isDefAccurate(). Def "accuracy" can be checked by testing ↵Lang Hames2010-09-2511-104/+78
| | | | | | whether LiveIntervals::getInstructionFromIndex(def) returns NULL. llvm-svn: 114791
* Add test case for PTX ret instructionChe-Liang Chiou2010-09-251-0/+6
| | | | llvm-svn: 114789
* Add ret instruction to PTX backendChe-Liang Chiou2010-09-2511-4/+53
| | | | llvm-svn: 114788
* Remove trailing spaces of CallingConv.hChe-Liang Chiou2010-09-251-5/+5
| | | | llvm-svn: 114787
* Move ELF to HasReliableSymbolDifference=true. Also take the opportunity to putRafael Espindola2010-09-255-12/+105
| | | | | | symbols defined in merge sections in independent atoms. llvm-svn: 114786
* LoadPRE was not properly checking that the load it was PRE'ing ↵Owen Anderson2010-09-252-2/+20
| | | | | | | | | | | | | | post-dominated the block it was being hoisted to. Splitting critical edges at the merge point only addressed part of the issue; it is also possible for non-post-domination to occur when the path from the load to the merge has branches in it. Unfortunately, full anticipation analysis is time-consuming, so for now approximate it. This is strictly more conservative than real anticipation, so we will miss some cases that real PRE would allow, but we also no longer insert loads into paths where they didn't exist before. :-) This is a very slight net positive on SPEC for me (0.5% on average). Most of the benchmarks are largely unaffected, but when it pays off it pays off decently: 181.mcf improves by 4.5% on my machine. llvm-svn: 114785
* Fix IIC_iEXTAr itinerary class of Cortex-A9.Evan Cheng2010-09-251-1/+1
| | | | llvm-svn: 114784
* Remove a unused instruction itinerary class.Evan Cheng2010-09-254-4/+0
| | | | llvm-svn: 114782
* If we're changing the source of a memcpy we need to use the alignmentEric Christopher2010-09-251-3/+5
| | | | | | | | | of the source, not the original alignment since it may no longer be valid. Fixes rdar://8400094 llvm-svn: 114781
* Fix zero and sign extension instructions scheduling itineraries.Evan Cheng2010-09-255-31/+45
| | | | llvm-svn: 114780
* Remove SlotIndex::PHI_BIT. It is no longer used by anything.Jakob Stoklund Olesen2010-09-254-39/+13
| | | | llvm-svn: 114779
* Remove the only use of SlotIndex::isPHI. This bit is not being set consistentlyJakob Stoklund Olesen2010-09-251-1/+1
| | | | | | and it will be removed shortly. llvm-svn: 114778
* Terminator gaps were unused. Might as well delete them.Jakob Stoklund Olesen2010-09-243-37/+1
| | | | llvm-svn: 114776
* More pseudo instruction scheduling itinerary fixes.Evan Cheng2010-09-245-3/+25
| | | | llvm-svn: 114768
* Fix for test/CodeGen/PowerPC/2008-10-17-AsmMatchingOperands.ll crash.John Thompson2010-09-241-0/+20
| | | | llvm-svn: 114767
OpenPOWER on IntegriCloud