Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | Nuke trailing whitespace. | Eric Christopher | 2009-08-26 | 1 | -14/+13 | |
| | | | | llvm-svn: 80145 | |||||
* | llvm-mc/Mach-O: Set .subsections_via_symbols flag properly. | Daniel Dunbar | 2009-08-26 | 2 | -10/+29 | |
| | | | | llvm-svn: 80144 | |||||
* | CMake: Removed outdated TODO. | Oscar Fuentes | 2009-08-26 | 1 | -2/+0 | |
| | | | | llvm-svn: 80124 | |||||
* | Generate section for bss and enable weak symbols | Venkatraman Govindaraju | 2009-08-26 | 1 | -0/+6 | |
| | | | | llvm-svn: 80121 | |||||
* | Alter 79292 to produce output that actually assembles. | Dale Johannesen | 2009-08-26 | 1 | -0/+2 | |
| | | | | llvm-svn: 80119 | |||||
* | Remove unneeded ARM-specific DAG nodes for VLD* and VST* Neon operations. | Bob Wilson | 2009-08-26 | 3 | -150/+133 | |
| | | | | | | | | The instructions can be selected directly from the intrinsics. We will need to add some ARM-specific nodes for VLD/VST of 3 and 4 128-bit vectors, but those are not yet implemented. llvm-svn: 80117 | |||||
* | Add extload expansion for f128 | Anton Korobeynikov | 2009-08-26 | 1 | -3/+6 | |
| | | | | llvm-svn: 80116 | |||||
* | Unbreak FP128 stuff in cbe | Anton Korobeynikov | 2009-08-26 | 1 | -1/+2 | |
| | | | | llvm-svn: 80115 | |||||
* | Unbreak CMake build | Douglas Gregor | 2009-08-26 | 1 | -0/+1 | |
| | | | | llvm-svn: 80109 | |||||
* | Expand scalar_to_vector - we don't have any isel logic for it now | Anton Korobeynikov | 2009-08-26 | 1 | -1/+1 | |
| | | | | llvm-svn: 80107 | |||||
* | Add comments detailing a known bug, so that people writing other | Dan Gohman | 2009-08-26 | 1 | -0/+8 | |
| | | | | | | backends don't use it as an example. llvm-svn: 80105 | |||||
* | -fast is now -O0. -fast-isel is no longer experimental. | Dan Gohman | 2009-08-26 | 1 | -2/+2 | |
| | | | | llvm-svn: 80104 | |||||
* | Move ProfileInfo::Edge's operator<< out of line. Among other benefits, | Dan Gohman | 2009-08-26 | 1 | -0/+8 | |
| | | | | | | | this eliminates the ATTRIBUTE_USED, which wasn't being used in a manner acceptable to some GCC versions, according to the buildbots. llvm-svn: 80103 | |||||
* | Implemented comments from Daniel Dunbar. | Andreas Neustifter | 2009-08-26 | 1 | -37/+45 | |
| | | | | | | (See http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20090817/084958.html) llvm-svn: 80100 | |||||
* | Don't use INSERT_SUBREG to model anyext operations on x86-64, as it | Dan Gohman | 2009-08-26 | 2 | -34/+31 | |
| | | | | | | | | leads to partial-register definitions. To help avoid redundant zero-extensions, also teach the h-register matching patterns that use movzbl to match anyext as well as zext. llvm-svn: 80099 | |||||
* | Create a ScalarEvolution-based AliasAnalysis implementation. | Dan Gohman | 2009-08-26 | 1 | -0/+128 | |
| | | | | | | | | | | | | | | | | This is a simple AliasAnalysis implementation which works by making ScalarEvolution queries. ScalarEvolution has a more complete understanding of arithmetic than BasicAA's collection of ad-hoc checks, so it handles some cases that BasicAA misses, for example p[i] and p[i+1] within the same iteration of a loop. This is currently experimental. It may be that the main use for this pass will be to help find cases where BasicAA can be profitably extended, or to help in the development of the overall AliasAnalysis infrastructure, however it's also possible that it could grow up to become a directly useful pass. llvm-svn: 80098 | |||||
* | Fix a missing newline, now that Value's operator<< doesn't add one of its own. | Dan Gohman | 2009-08-26 | 1 | -1/+1 | |
| | | | | llvm-svn: 80096 | |||||
* | Use SetVector instead of std::set so that alias relations are tested and | Dan Gohman | 2009-08-26 | 1 | -8/+7 | |
| | | | | | | printed in a deterministic order. llvm-svn: 80095 | |||||
* | llvm-mc/Mach-O: Add support for relocations. | Daniel Dunbar | 2009-08-26 | 1 | -13/+225 | |
| | | | | | | | | | | - I haven't really tried to find the "right" way to store the fixups or apply them, yet. This works, but isn't particularly elegant or fast. - Still no evaluation support, so we don't actually ever not turn a fixup into a relocation entry. llvm-svn: 80089 | |||||
* | llvm-mc/Mach-O: Move symbol indices into the MCSymbolData structure. | Daniel Dunbar | 2009-08-26 | 1 | -21/+13 | |
| | | | | llvm-svn: 80088 | |||||
* | llvm-mc: Add symbol entries for undefined symbols used in .fill and .org. | Daniel Dunbar | 2009-08-26 | 1 | -2/+10 | |
| | | | | llvm-svn: 80086 | |||||
* | Add dummy inline asm handling for 'r' constraint. This fixes PR4778 | Anton Korobeynikov | 2009-08-26 | 3 | -0/+59 | |
| | | | | llvm-svn: 80085 | |||||
* | Moved isDeclaration() check further down to allow for function counts for | Andreas Neustifter | 2009-08-26 | 1 | -1/+4 | |
| | | | | | | declarations if necessary. llvm-svn: 80084 | |||||
* | llvm-mc: Change MCContext value table to take const MCSymbol*s. | Daniel Dunbar | 2009-08-26 | 1 | -4/+4 | |
| | | | | llvm-svn: 80079 | |||||
* | Revert 79977. It causes llvm-gcc bootstrap failures on some platforms. | Devang Patel | 2009-08-26 | 21 | -419/+607 | |
| | | | | llvm-svn: 80073 | |||||
* | test commit | Venkatraman Govindaraju | 2009-08-26 | 1 | -1/+1 | |
| | | | | llvm-svn: 80070 | |||||
* | llvm-mc/Mach-O: Add section padding where needed (to align the next section). | Daniel Dunbar | 2009-08-26 | 1 | -39/+45 | |
| | | | | | | Also, simplify some of Mach-O writer code which can now use section addresses. llvm-svn: 80067 | |||||
* | llvm-mc/Mach-O: Set addresses for symbols. | Daniel Dunbar | 2009-08-26 | 2 | -10/+32 | |
| | | | | llvm-svn: 80065 | |||||
* | Add an 'inline hint' attribute to represent source | Dale Johannesen | 2009-08-26 | 5 | -0/+6 | |
| | | | | | | | | code hints that it would be a good idea to inline a function ("inline" keyword). No functional change yet; FEs do not emit this and inliner does not use it. llvm-svn: 80063 | |||||
* | Add isClosure() predicate. This is used to add DW_AT_APPLE_block attribute. | Devang Patel | 2009-08-26 | 1 | -4/+1 | |
| | | | | | | Patch by Caroline Tice. llvm-svn: 80061 | |||||
* | llvm-mc: Improve indirect symbol support (add the indirect index table). | Daniel Dunbar | 2009-08-26 | 2 | -11/+59 | |
| | | | | llvm-svn: 80059 | |||||
* | Remove unused variables. | Dan Gohman | 2009-08-26 | 2 | -5/+0 | |
| | | | | llvm-svn: 80058 | |||||
* | Fix the InsertAtEnd form of ShuffleVectorInst constructor to use | Dan Gohman | 2009-08-25 | 1 | -4/+6 | |
| | | | | | | the correct type. llvm-svn: 80050 | |||||
* | Eliminate the unused Context argument on one of the ICmpInst and FCmpInst | Dan Gohman | 2009-08-25 | 7 | -150/+146 | |
| | | | | | | constructors. llvm-svn: 80049 | |||||
* | Updated i128 sext support for CellSPU backend, contributed by Ken Werner (IBM) | Scott Michel | 2009-08-25 | 1 | -16/+32 | |
| | | | | llvm-svn: 80042 | |||||
* | Use covariant return types for Instruction::clone, and eliminate | Dan Gohman | 2009-08-25 | 1 | -12/+12 | |
| | | | | | | | the forms of ExtractElementInst and InsertElementInst that are equivalent to clone. llvm-svn: 80041 | |||||
* | Get rid of this horrible "benign race" by exploiting ManagedStatic to initialize | Owen Anderson | 2009-08-25 | 1 | -7/+14 | |
| | | | | | | the array on its first access. llvm-svn: 80040 | |||||
* | This should use isIndenticalToWhenDefined. | Dan Gohman | 2009-08-25 | 1 | -2/+2 | |
| | | | | llvm-svn: 80039 | |||||
* | Rename Instruction::isIdenticalTo to Instruction::isIdenticalToWhenDefined, | Dan Gohman | 2009-08-25 | 6 | -71/+157 | |
| | | | | | | | | | | | | | | | | and introduce a new Instruction::isIdenticalTo which tests for full identity, including the SubclassOptionalData flags. Also, fix the Instruction::clone implementations to preserve the SubclassOptionalData flags. Finally, teach several optimizations how to handle SubclassOptionalData correctly, given these changes. This fixes the counterintuitive behavior of isIdenticalTo not comparing the full value, and clone not returning an identical clone, as well as some subtle bugs that could be caused by these. Thanks to Nick Lewycky for reporting this, and for an initial patch! llvm-svn: 80038 | |||||
* | Revert last patch. We need to put this into TargetLowering. There will be a lot | Bill Wendling | 2009-08-25 | 2 | -2/+0 | |
| | | | | | | of EH stuff going into there, so we can wait to add them all then. llvm-svn: 80036 | |||||
* | llvm-mc: Add statistic for number of fragments emitted by the assembler. | Daniel Dunbar | 2009-08-25 | 1 | -0/+6 | |
| | | | | llvm-svn: 80033 | |||||
* | Add the #include here. | Bill Wendling | 2009-08-25 | 1 | -0/+1 | |
| | | | | llvm-svn: 80032 | |||||
* | Add a target asm info hook to specify that particular bits of data in the FDE | Bill Wendling | 2009-08-25 | 2 | -0/+2 | |
| | | | | | | | | | | should be forced to 32-bits (.long) even on 64-bit architectures. Darwin wants these bits to be 64-bits (.quad). However, other platforms may disagree. This is just the info right now and is part of a work-in-progress which needs this. We'll add the actual *use* of this soon. llvm-svn: 80024 | |||||
* | Start refactoring PIC16 TargetObjectFile code. Eventually, all the stuff from | Sanjiv Gupta | 2009-08-25 | 3 | -12/+58 | |
| | | | | | | PIC16Section will move to MCSectionPIC16. llvm-svn: 80021 | |||||
* | Don't assume that two identical instructions that read from memory | Dan Gohman | 2009-08-25 | 1 | -1/+1 | |
| | | | | | | | | will always return the same value. This isn't currently necessary, since this code doesn't currently ever get called under circumstances where it would matter, but it may some day. llvm-svn: 80017 | |||||
* | Remove some unused SDNode definitions. | Bob Wilson | 2009-08-25 | 1 | -27/+0 | |
| | | | | llvm-svn: 80015 | |||||
* | Teach ScalarEvolution about GlobalAliases. | Dan Gohman | 2009-08-25 | 1 | -0/+3 | |
| | | | | llvm-svn: 80014 | |||||
* | Use X86II::MO_NO_FLAG. | Dan Gohman | 2009-08-25 | 1 | -2/+3 | |
| | | | | llvm-svn: 80012 | |||||
* | Expose the instruction contraint string as an argument to the NLdSt class. | Bob Wilson | 2009-08-25 | 2 | -24/+20 | |
| | | | | llvm-svn: 80011 | |||||
* | Special-case static allocas in IndVarSimplify's loop invariant | Dan Gohman | 2009-08-25 | 1 | -0/+5 | |
| | | | | | | | | sinking code, since they are special. If the loop preheader happens to be the entry block of a function, don't sink static allocas out of it. This fixes PR4775. llvm-svn: 80010 |