| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | Make GlobalMerge honor the preferred alignment on globals without an ↵ | Eli Friedman | 2011-11-30 | 5 | -10/+32 | |
| | | | | | | | | | explicitly specified alignment. <rdar://problem/10497732>. llvm-svn: 145523 | |||||
| * | More cleanups. No content change. | Bill Wendling | 2011-11-30 | 1 | -9/+13 | |
| | | | | | llvm-svn: 145522 | |||||
| * | Minor cleanup. No content change. | Bill Wendling | 2011-11-30 | 1 | -7/+23 | |
| | | | | | llvm-svn: 145521 | |||||
| * | Remove the install-clang-c makefile target. | Bob Wilson | 2011-11-30 | 1 | -9/+0 | |
| | | | | | | | | When I did this before it broke a buildbot that was testing that target, but we've removed that buildbot now. llvm-svn: 145519 | |||||
| * | Remove an XXX which hasn't been fixed yet. It's too late now. | Bill Wendling | 2011-11-30 | 1 | -3/+2 | |
| | | | | | llvm-svn: 145518 | |||||
| * | Remove unused variable | Matt Beaumont-Gay | 2011-11-30 | 1 | -2/+0 | |
| | | | | | llvm-svn: 145517 | |||||
| * | Add some tests for all-lanes VLD1 parsing. | Jim Grosbach | 2011-11-30 | 1 | -0/+14 | |
| | | | | | llvm-svn: 145512 | |||||
| * | ARM parsing for VLD1 all lanes, with writeback. | Jim Grosbach | 2011-11-30 | 4 | -43/+82 | |
| | | | | | llvm-svn: 145510 | |||||
| * | Add a few functions to TargetLibraryInfo. | Chad Rosier | 2011-11-30 | 2 | -15/+148 | |
| | | | | | llvm-svn: 145508 | |||||
| * | ARM parsing for VLD1 two register all lanes, no writeback. | Jim Grosbach | 2011-11-30 | 6 | -5/+31 | |
| | | | | | llvm-svn: 145504 | |||||
| * | Add test arch to make it pass on non x86 targets | Nadav Rotem | 2011-11-30 | 1 | -1/+1 | |
| | | | | | llvm-svn: 145498 | |||||
| * | X86: Turns out bulldozer also supports sse42 and lzcnt. | Benjamin Kramer | 2011-11-30 | 1 | -11/+6 | |
| | | | | | | | | While at it remove the barcelona/instanbul/shanghai subtargets, they're unsupported by GCC and look pretty broken. llvm-svn: 145494 | |||||
| * | X86: Add subtargets for AMD's bulldozer. | Benjamin Kramer | 2011-11-30 | 1 | -0/+6 | |
| | | | | | llvm-svn: 145493 | |||||
| * | Add a tripple to the test | Nadav Rotem | 2011-11-30 | 1 | -0/+3 | |
| | | | | | llvm-svn: 145489 | |||||
| * | X86: PerformOrCombine introduced a vselect node with a wrong order of ↵ | Nadav Rotem | 2011-11-30 | 2 | -1/+23 | |
| | | | | | | | operands. This bug was introduced when a dedicated blend sdnode was replaced with the vselect node (in 139479). llvm-svn: 145488 | |||||
| * | Add instruction selection support for AVX2 horizontal add/sub instructions. | Craig Topper | 2011-11-30 | 1 | -24/+49 | |
| | | | | | llvm-svn: 145487 | |||||
| * | Mention that -O4 does result in more optimization when used with | Duncan Sands | 2011-11-30 | 1 | -1/+3 | |
| | | | | | | | | -fplugin-arg-dragonegg-enable-gcc-optzns, though it usually isn't a win. llvm-svn: 145486 | |||||
| * | Merge VPERM2F128/VPERM2I128 ISD node types. | Craig Topper | 2011-11-30 | 4 | -71/+50 | |
| | | | | | llvm-svn: 145485 | |||||
| * | Better test case found in duplicate PR10570. | Andrew Trick | 2011-11-30 | 1 | -43/+23 | |
| | | | | | llvm-svn: 145484 | |||||
| * | Merge decoding of VPERMILPD and VPERMILPS shuffle masks. Merge X86ISD node ↵ | Craig Topper | 2011-11-30 | 7 | -142/+135 | |
| | | | | | | | type for VPERMILPD/PS. Add instruction selection support for VINSERTI128/VEXTRACTI128. llvm-svn: 145483 | |||||
| * | LSR: handle the expansion of phi operands that use postinc forms of the IV. | Andrew Trick | 2011-11-30 | 3 | -36/+126 | |
| | | | | | | | Fixes PR11431: SCEVExpander::expandAddRecExprLiterally(const llvm::SCEVAddRecExpr*): Assertion `(!isa<Instruction>(Result) || SE.DT->dominates(cast<Instruction>(Result), Builder.GetInsertPoint())) && "postinc expansion does not dominate use"' failed. llvm-svn: 145482 | |||||
| * | Whitespace. | Chad Rosier | 2011-11-30 | 1 | -3/+1 | |
| | | | | | llvm-svn: 145470 | |||||
| * | Alphabetize TargetLibraryInfo enum and fix doxygen comments. No functional | Chad Rosier | 2011-11-30 | 2 | -27/+27 | |
| | | | | | | | change intended. llvm-svn: 145468 | |||||
| * | llvm_unreachable() is not for user diagnostics.... | Jim Grosbach | 2011-11-30 | 1 | -1/+1 | |
| | | | | | llvm-svn: 145465 | |||||
| * | ARM parsing aliases for VLD1 single register all lanes. | Jim Grosbach | 2011-11-30 | 6 | -6/+142 | |
| | | | | | llvm-svn: 145464 | |||||
| * | Add support for sqrt, sqrtl, and sqrtf in TargetLibraryInfo. Disable | Chad Rosier | 2011-11-29 | 5 | -3/+35 | |
| | | | | | | | | | (fptrunc (sqrt (fpext x))) -> (sqrtf x) transformation if -fno-builtin is specified. rdar://10466410 llvm-svn: 145460 | |||||
| * | Tidy up a bit. | Jim Grosbach | 2011-11-29 | 2 | -33/+10 | |
| | | | | | llvm-svn: 145458 | |||||
| * | Add comment. | Jim Grosbach | 2011-11-29 | 1 | -0/+1 | |
| | | | | | llvm-svn: 145456 | |||||
| * | ARM parsing aliases for data-size suffices on VST1. | Jim Grosbach | 2011-11-29 | 1 | -19/+56 | |
| | | | | | llvm-svn: 145454 | |||||
| * | FileCheckize. | Jakob Stoklund Olesen | 2011-11-29 | 2 | -6/+23 | |
| | | | | | llvm-svn: 145452 | |||||
| * | Change names for MIPS "generic" processors defined in Mips.td to match what GNU | Akira Hatanaka | 2011-11-29 | 8 | -12/+12 | |
| | | | | | | | | | | | tools use. Patch by Simon Atanasyan. "mips32r1" => "mips32" "4ke" => mips32r2" "mips64r1" => "mips64" llvm-svn: 145451 | |||||
| * | ARM assembly parsing and encoding for four-register VST1. | Jim Grosbach | 2011-11-29 | 5 | -52/+53 | |
| | | | | | llvm-svn: 145450 | |||||
| * | llvm-config: Replace with C++ version (was llvm-config-2). | Daniel Dunbar | 2011-11-29 | 15 | -942/+65 | |
| | | | | | | | - Reapply of r144300, with lots of fixes/migration easement in between. llvm-svn: 145449 | |||||
| * | Add another missing pattern. llvm-gcc likes f64 but clang likes i64 so it ↵ | Evan Cheng | 2011-11-29 | 2 | -1/+26 | |
| | | | | | | | was generating poor code for some SSE builtins. llvm-svn: 145448 | |||||
| * | Enable some VST1 tests and add a few more. | Jim Grosbach | 2011-11-29 | 1 | -14/+20 | |
| | | | | | llvm-svn: 145443 | |||||
| * | ARM assembly parsing and encoding for three-register VST1. | Jim Grosbach | 2011-11-29 | 4 | -44/+47 | |
| | | | | | llvm-svn: 145442 | |||||
| * | Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions. | Jakob Stoklund Olesen | 2011-11-29 | 12 | -81/+71 | |
| | | | | | | | | | | Like V_SET0, these instructions are expanded by ExpandPostRA to xorps / vxorps so they can participate in execution domain swizzling. This also makes the AVX variants redundant. llvm-svn: 145440 | |||||
| * | Potential bug in RewriteLoopBodyWithConditionConstant: use iterator should ↵ | Stepan Dyatkovskiy | 2011-11-29 | 1 | -1/+5 | |
| | | | | | | | not be changed inside the uses enumeration loop. llvm-svn: 145432 | |||||
| * | Update the docs for some of the test-suite configure changes and | Eric Christopher | 2011-11-29 | 1 | -16/+11 | |
| | | | | | | | be more clear about what to do and how to do it. llvm-svn: 145426 | |||||
| * | If fast-isel fails, remove dead instructions generated during the failed | Chad Rosier | 2011-11-29 | 3 | -0/+54 | |
| | | | | | | | attempt. llvm-svn: 145425 | |||||
| * | grammar. | Rafael Espindola | 2011-11-29 | 1 | -1/+1 | |
| | | | | | llvm-svn: 145423 | |||||
| * | comment. | Andrew Trick | 2011-11-29 | 1 | -0/+4 | |
| | | | | | llvm-svn: 145422 | |||||
| * | build/CMake: Finish removal of add_llvm_library_dependencies. | Daniel Dunbar | 2011-11-29 | 78 | -574/+0 | |
| | | | | | llvm-svn: 145420 | |||||
| * | Add a link to Bill's blog post. | Benjamin Kramer | 2011-11-29 | 1 | -2/+3 | |
| | | | | | llvm-svn: 145419 | |||||
| * | Release notes for segmented stacks. | Rafael Espindola | 2011-11-29 | 2 | -8/+5 | |
| | | | | | | | Patch by Sanjoy Das. llvm-svn: 145416 | |||||
| * | Fix a theoretical problem (not seen in the wild): if different instances of a | Duncan Sands | 2011-11-29 | 3 | -4/+15 | |
| | | | | | | | | | | | | | | | | weak variable are compiled by different compilers, such as GCC and LLVM, while LLVM may increase the alignment to the preferred alignment there is no reason to think that GCC will use anything more than the ABI alignment. Since it is the GCC version that might end up in the final program (as the linkage is weak), it is wrong to increase the alignment of loads from the global up to the preferred alignment as the alignment might only be the ABI alignment. Increasing alignment up to the ABI alignment might be OK, but I'm not totally convinced that it is. It seems better to just leave the alignment of weak globals alone. llvm-svn: 145413 | |||||
| * | MC/X86/COFF: Allow quotes in names when targeting MS/Windows, | Michael J. Spencer | 2011-11-29 | 6 | -5/+58 | |
| | | | | | | | | | | | as MC is the only assembler we support. This splits MS/Windows and GNU/Windows ASM infos into two seperate classes. While there is currently only one difference, full MS C++ ABI support will require many more. llvm-svn: 145409 | |||||
| * | Fixed ObjectFile functions: | Danil Malyshev | 2011-11-29 | 13 | -48/+192 | |
| | | | | | | | | | | | | - getSymbolOffset() renamed as getSymbolFileOffset() - getSymbolFileOffset(), getSymbolAddress(), getRelocationAddress() returns same result for ELFObjectFile, MachOObjectFile and COFFObjectFile. - added getRelocationOffset() - fixed MachOObjectFile::getSymbolSize() - fixed MachOObjectFile::getSymbolSection() - fixed MachOObjectFile::getSymbolOffset() for symbols without section data. llvm-svn: 145408 | |||||
| * | Fixed vsqrt.ss intrinsic usage - order of input operands was wrong. | Elena Demikhovsky | 2011-11-29 | 2 | -1/+12 | |
| | | | | | | | | Added a test. Thanks Bruno for reviewing the patch. llvm-svn: 145403 | |||||
| * | Fix shuffle decoding for memory forms for (V)SHUFPS/D. | Craig Topper | 2011-11-29 | 2 | -6/+30 | |
| | | | | | llvm-svn: 145392 | |||||

