summaryrefslogtreecommitdiffstats
path: root/llvm/lib
Commit message (Collapse)AuthorAgeFilesLines
* Use explicit add_subdirectory's for LLVM target sublibraries insteadOscar Fuentes2011-02-2016-10/+41
| | | | | | | | | of testing for its presence at cmake time. This way the build automatically regenerates the makefiles when a svn update brings in a new sublibrary. llvm-svn: 126068
* PR9218: SimplifyDemandedVectorElts can return a non-null value that is notEli Friedman2011-02-191-2/+7
| | | | | | | the instruction passed in. Make sure to account for this correctly, instead of looping infinitely. llvm-svn: 126058
* Minor x86 README updates.Eli Friedman2011-02-191-20/+14
| | | | llvm-svn: 126054
* implement PR9264: disambiguating 'bt mem, imm' as a btl.Chris Lattner2011-02-191-0/+3
| | | | | | | This is reasonable to do since all bt-mem forms do the same thing. llvm-svn: 126047
* rewrite the memset_pattern pattern generation stuff to accept any 2/4/8/16-byteChris Lattner2011-02-191-32/+12
| | | | | | | | | | | | | | | | | | constant, including globals. This makes us generate much more "pretty" pattern globals as well because it doesn't break it down to an array of bytes all the time. This enables us to handle stores of relocatable globals. This kicks in about 48 times in 254.gap, giving us stuff like this: @.memset_pattern40 = internal constant [2 x %struct.TypHeader* (%struct.TypHeader*, %struct.TypHeader*)*] [%struct.TypHeader* (%struct.TypHeader*, %struct .TypHeader*)* @IsFalse, %struct.TypHeader* (%struct.TypHeader*, %struct.TypHeader*)* @IsFalse], align 16 ... call void @memset_pattern16(i8* %scevgep5859, i8* bitcast ([2 x %struct.TypHeader* (%struct.TypHeader*, %struct.TypHeader*)*]* @.memset_pattern40 to i8* ), i64 %tmp75) nounwind llvm-svn: 126044
* Stores of null pointers should turn into memset, we weren't recognizingChris Lattner2011-02-191-0/+5
| | | | | | them as splat values. llvm-svn: 126041
* Implement rdar://9009151, transforming strided loop stores ofChris Lattner2011-02-191-32/+125
| | | | | | | | | | | unsplatable values into memset_pattern16 when it is available (recent darwins). This transforms lots of strided loop stores of ints for example, like 5 in vpr: Formed memset: call void @memset_pattern16(i8* %4, i8* getelementptr inbounds ([16 x i8]* @.memset_pattern9, i32 0, i32 0), i64 %tmp25) from store to: {%3,+,4}<%11> at: store i32 3, i32* %scevgep, align 4, !tbaa !4 llvm-svn: 126040
* Fix typos.Eric Christopher2011-02-191-2/+2
| | | | llvm-svn: 126018
* Do not emit empty DW_TAG_lexical_block DIEs. In one test case, size of ↵Devang Patel2011-02-191-17/+20
| | | | | | debug info reduced by almost 7%. llvm-svn: 126009
* Give SplitAnalysis a VRM member to access VirtRegMap::getOriginal().Jakob Stoklund Olesen2011-02-193-5/+7
| | | | llvm-svn: 126005
* Avoid dangling else warnings.Joerg Sonnenberger2011-02-191-1/+2
| | | | llvm-svn: 126004
* Missed member rename for naming convention.Jakob Stoklund Olesen2011-02-192-9/+9
| | | | llvm-svn: 126003
* This method belonged in VirtRegMap.Jakob Stoklund Olesen2011-02-193-10/+10
| | | | llvm-svn: 126002
* Separate timers for local and global splitting.Jakob Stoklund Olesen2011-02-191-2/+5
| | | | llvm-svn: 126001
* These tags are now covered by dwarf::TagString().Devang Patel2011-02-181-9/+1
| | | | llvm-svn: 125987
* Provide tag strings for llvm specific tags.Devang Patel2011-02-181-0/+4
| | | | llvm-svn: 125986
* Do not lose debug info of an inlined function argument even if the argument ↵Devang Patel2011-02-184-14/+15
| | | | | | | | is only used through GEPs. This time with a fix that avoids using invalidated DenseMap iterator. llvm-svn: 125984
* Use VirtRegMap's Virt2SplitMap to keep track of the original live range ↵Jakob Stoklund Olesen2011-02-183-2/+12
| | | | | | | | before splitting. All new virtual registers created for spilling or splitting point back to their original. llvm-svn: 125980
* add a way to disable all builtins, wire it up to opt's ↵Chris Lattner2011-02-181-0/+6
| | | | | | -disable-simplifylibcalls flag. llvm-svn: 125978
* Make loop-idiom use TargetLibraryInfo to determine whether it is allowedChris Lattner2011-02-181-1/+18
| | | | | | to hack on memset, memcpy etc. llvm-svn: 125974
* Move library stuff out of the toplevel CMakeLists.txt file.Oscar Fuentes2011-02-188-0/+84
| | | | llvm-svn: 125968
* Add VirtRegMap::rewrite() and use it in the new register allocators.Jakob Stoklund Olesen2011-02-184-6/+81
| | | | | | | | | | The rewriter works almost identically to -rewriter=trivial, except it also eliminates any identity copies. This makes the new register allocators independent of VirtRegRewriter.cpp which will be going away at the same time as RegAllocLinearScan. llvm-svn: 125967
* introduce a new TargetLibraryInfo pass, which transformations can use toChris Lattner2011-02-183-0/+51
| | | | | | | | query about available library functions. For now this just has memset_pattern16, which exists on darwin, but it can be extended for a bunch of other things in the future. llvm-svn: 125965
* Reapply r114997 now that the buildbots have been updated.Bill Wendling2011-02-181-8/+4
| | | | llvm-svn: 125960
* Fix style and a typoBruno Cardoso Lopes2011-02-181-4/+4
| | | | llvm-svn: 125949
* Add assembly parsing support for "msr" and also fix its encoding. Also addBruno Cardoso Lopes2011-02-187-47/+169
| | | | | | testcases for the disassembler to make sure it still works for "msr". llvm-svn: 125948
* improve support for OpenBSD, patch by Amit Kulkarni!Chris Lattner2011-02-181-2/+4
| | | | llvm-svn: 125943
* Expose getTypeName to the C API. Patch by Patrick Walton.Rafael Espindola2011-02-181-0/+4
| | | | llvm-svn: 125845
* Add some transforms of the kind X-Y>X -> 0>Y which are valid when there is noDuncan Sands2011-02-181-18/+17
| | | | | | overflow. These subsume some existing equality transforms, so zap those. llvm-svn: 125843
* add a poor division by constant case.Chris Lattner2011-02-181-0/+37
| | | | llvm-svn: 125832
* Roll out r125794 to help diagnose the llvm-gcc-i386-linux-selfhost failure.Cameron Zwarich2011-02-184-10/+13
| | | | llvm-svn: 125830
* prevent jump threading from merging blocks when their address isChris Lattner2011-02-181-1/+11
| | | | | | | | | | | | | | | | | taken (and used!). This prevents merging the blocks (invalidating the block addresses) in a case like this: #define _THIS_IP_ ({ __label__ __here; __here: (unsigned long)&&__here; }) void foo() { printf("%p\n", _THIS_IP_); printf("%p\n", _THIS_IP_); printf("%p\n", _THIS_IP_); } which fixes PR4151. llvm-svn: 125829
* hoist GlobalValue::removeDeadConstantUsers up to being a method on Constant.Chris Lattner2011-02-182-49/+53
| | | | llvm-svn: 125828
* Don't unroll loops whose header block's address is taken.Chris Lattner2011-02-181-2/+10
| | | | | | | | | | | | | | | | | This is part of a futile attempt to not "break" bizzaro code like this: l1: printf("l1: %p\n", &&l1); ++x; if( x < 3 ) goto l1; Previously we'd fold &&l1 to 1, which is fine per our semantics but not helpful to the user. llvm-svn: 125827
* Recognize monitor/mwait with explicit register argumentsJoerg Sonnenberger2011-02-181-0/+8
| | | | llvm-svn: 125805
* Check the errorcode.Argyrios Kyrtzidis2011-02-181-0/+1
| | | | llvm-svn: 125804
* Trim debugging output.Jakob Stoklund Olesen2011-02-182-29/+6
| | | | llvm-svn: 125802
* Recognize leavel and leaveq aliases for leave.Joerg Sonnenberger2011-02-171-0/+3
| | | | | | Validate encoding of leave in 64bit mode. llvm-svn: 125795
* Do not lose debug info of an inlined function argument even if the argument ↵Devang Patel2011-02-174-13/+10
| | | | | | is only used through GEPs. llvm-svn: 125794
* have instcombine preserve nsw/nuw/exact when sinkingChris Lattner2011-02-171-11/+53
| | | | | | common operations through a phi. llvm-svn: 125790
* Add basic register allocator statistics.Jakob Stoklund Olesen2011-02-172-0/+18
| | | | llvm-svn: 125789
* fix typoChris Lattner2011-02-171-1/+1
| | | | llvm-svn: 125787
* fix instcombine merging GEPs through a PHI to only make theChris Lattner2011-02-171-3/+7
| | | | | | result inbounds if all of the inputs are inbounds. llvm-svn: 125785
* Enhance constant folding of bitcast operations on vectors of floats.Nadav Rotem2011-02-173-3/+14
| | | | | | | Add getAllOnesValue of FP numbers to Constants and APFloat. Add more tests. llvm-svn: 125776
* add is always integer, thanks to Frits for noticing this.Chris Lattner2011-02-171-1/+1
| | | | llvm-svn: 125774
* [AVX] Recorganize X86ShuffleDecode into its own libraryDavid Greene2011-02-177-33/+116
| | | | | | | | | | | | (LLVMX86Utils.a) to break cyclic library dependencies between LLVMX86CodeGen.a and LLVMX86AsmParser.a. Previously this code was in a header file and marked static but AVX requires some additional functionality here that won't be used by all clients. Since including unused static functions causes a gcc compiler warning, keeping it as a header would break builds that use -Werror. Putting this in its own library solves both problems at once. llvm-svn: 125765
* Split local live ranges.Jakob Stoklund Olesen2011-02-172-2/+280
| | | | | | | | A local live range is live in a single basic block. If such a range fails to allocate, try to find a sub-range that would get a larger spill weight than its interference. llvm-svn: 125764
* The labyrinthine X86 backend no longer appears to requireDan Gohman2011-02-171-37/+0
| | | | | | these patterns. llvm-svn: 125759
* Fix wrong logic in promotion of signed mul-with-overflow (I pointed this out atDuncan Sands2011-02-171-18/+12
| | | | | | | | | the time but presumably my email got lost). Examples where the previous logic got it wrong: (1) a signed i8 multiply of 64 by 2 overflows, but the high part is zero; (2) a signed i8 multiple of -128 by 2 overflows, but the high part is all ones. llvm-svn: 125748
* Triple::MinGW64 is deprecated and removed. We can use Triple::MinGW32 generally.NAKAMURA Takumi2011-02-176-11/+2
| | | | | | No one uses *-mingw64. mingw-w64 is represented as {i686|x86_64}-w64-mingw32. In llvm side, i686 and x64 can be treated as similar way. llvm-svn: 125747
OpenPOWER on IntegriCloud