| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | Implement more aggressive folding of constant GEP instructions | Chris Lattner | 2003-06-26 | 1 | -6/+25 | |
| | | | | | llvm-svn: 6913 | |||||
| * | Add support for elimination of load instruction from global constants | Chris Lattner | 2003-06-26 | 1 | -0/+48 | |
| | | | | | llvm-svn: 6912 | |||||
| * | Integrate the C writer's name mangler. | Brian Gaeke | 2003-06-25 | 1 | -15/+82 | |
| | | | | | llvm-svn: 6908 | |||||
| * | First draft of X86 LLC backend. This should be OK for small programs like | Brian Gaeke | 2003-06-25 | 1 | -30/+354 | |
| | | | | | | | Shootout, but it has some issues with bigger programs. Work in progress. llvm-svn: 6907 | |||||
| * | Instcombine: X * -1 -> -X | Chris Lattner | 2003-06-25 | 1 | -3/+4 | |
| | | | | | llvm-svn: 6904 | |||||
| * | Fix bug: LevelRaise/2003-06-25-ExprAnalysis.ll | Chris Lattner | 2003-06-25 | 1 | -2/+0 | |
| | | | | | llvm-svn: 6903 | |||||
| * | Fix bug: Mem2Reg/2003-06-26-IterativePromote.ll | Chris Lattner | 2003-06-25 | 1 | -10/+17 | |
| | | | | | llvm-svn: 6901 | |||||
| * | - Fixed name mangling conditions to handle 'linkonce' linkage type. In | Joel Stanley | 2003-06-25 | 1 | -30/+44 | |
| | | | | | | | | | | | | | particular, name mangling for GlobalValues only occurs when the linkage type is internal or when the name must be mangled to avoid a collision. See comments in CWriter::getValueName for more information. - 'inline' keyword is now emitted for functions with 'linkonce' linkage type. - Fixed typos. llvm-svn: 6898 | |||||
| * | Add argument to DAE to allow operation on non-internal functions | Chris Lattner | 2003-06-25 | 1 | -9/+21 | |
| | | | | | llvm-svn: 6895 | |||||
| * | It seems likely that floats would need a cast too, because they are | Brian Gaeke | 2003-06-25 | 1 | -1/+2 | |
| | | | | | | | ordinarily promoted to doubles. llvm-svn: 6894 | |||||
| * | Fix bug: ADCE/2003-06-24-BadSuccessor.ll | Chris Lattner | 2003-06-24 | 1 | -12/+35 | |
| | | | | | llvm-svn: 6891 | |||||
| * | Make assertion more descriptive | Chris Lattner | 2003-06-24 | 1 | -1/+1 | |
| | | | | | llvm-svn: 6889 | |||||
| * | Do not mark ALL terminators live if any instruciton in the block is live. ↵ | Chris Lattner | 2003-06-24 | 1 | -2/+23 | |
| | | | | | | | | | | We only want to mark it live if it is an unconditional branch. This fixes bug: ADCE/2002-05-28-Crash.ll and makes this pass _much_ more useful. llvm-svn: 6887 | |||||
| * | Fix bug: SCCP/2003-06-24-OverdefinedPHIValue.ll | Chris Lattner | 2003-06-24 | 1 | -4/+5 | |
| | | | | | llvm-svn: 6883 | |||||
| * | Fix bug: TailDup/2003-06-24-Simpleloop.ll | Chris Lattner | 2003-06-24 | 1 | -1/+2 | |
| | | | | | llvm-svn: 6881 | |||||
| * | Implement new transforms: | Chris Lattner | 2003-06-23 | 1 | -4/+33 | |
| | | | | | | | | Replace (cast (sub A, B) to bool) -> (setne A, B) Replace (cast (add A, B) to bool) -> (setne A, -B) llvm-svn: 6873 | |||||
| * | Fix 2003-06-23-PromotedExprs.llx -- if we are adding two bytes we better | Brian Gaeke | 2003-06-23 | 1 | -0/+14 | |
| | | | | | | | | explicitly cast the result to be a byte, or C will gleefully promote it to int. llvm-svn: 6869 | |||||
| * | Include <cmath> instead of <math.h> | Brian Gaeke | 2003-06-23 | 1 | -10/+1 | |
| | | | | | | | Remove isnan; it's too unportable to handle cleanly at this point. llvm-svn: 6866 | |||||
| * | Add support for GCC 3.3 | Chris Lattner | 2003-06-23 | 1 | -2/+2 | |
| | | | | | llvm-svn: 6865 | |||||
| * | avoid dividing by zero when dealing with zero sized types (like [0 x double]) | Chris Lattner | 2003-06-23 | 2 | -0/+2 | |
| | | | | | llvm-svn: 6862 | |||||
| * | Add the padding needed for variable-size alloca's, which should work now. | Vikram S. Adve | 2003-06-23 | 1 | -6/+14 | |
| | | | | | llvm-svn: 6859 | |||||
| * | Fix bug: InstCombine/2003-06-22-ConstantExprCrash.ll | Chris Lattner | 2003-06-22 | 1 | -1/+7 | |
| | | | | | llvm-svn: 6857 | |||||
| * | Add paranoia checking | Chris Lattner | 2003-06-22 | 1 | -1/+1 | |
| | | | | | llvm-svn: 6856 | |||||
| * | Test change | Chris Lattner | 2003-06-22 | 1 | -0/+1 | |
| | | | | | llvm-svn: 6852 | |||||
| * | Initial checkin of Tail duplication pass. | Chris Lattner | 2003-06-22 | 1 | -0/+324 | |
| | | | | | llvm-svn: 6846 | |||||
| * | Do not use the inefficient, fixed function, CBW, CDQ, ... instructions | Chris Lattner | 2003-06-22 | 1 | -2/+4 | |
| | | | | | llvm-svn: 6844 | |||||
| * | Remove a ton of extraneous #includes | Chris Lattner | 2003-06-22 | 2 | -5/+1 | |
| | | | | | llvm-svn: 6842 | |||||
| * | Remove support for the MultiObject flag, which was fundamentally broken | Chris Lattner | 2003-06-22 | 4 | -14/+3 | |
| | | | | | llvm-svn: 6840 | |||||
| * | Instcombine cast (getelementptr Ptr, 0, 0, 0) to ... into: cast Ptr to ... | Chris Lattner | 2003-06-21 | 1 | -0/+17 | |
| | | | | | | | | This fixes type safety problems in a variety of benchmarks that were confusing DSA. llvm-svn: 6837 | |||||
| * | Minor simplification to 64-bit instsel | Chris Lattner | 2003-06-21 | 1 | -4/+4 | |
| | | | | | llvm-svn: 6831 | |||||
| * | For 16 and 32-bit multiplies, use the IMUL instruction instead of the MUL ↵ | Chris Lattner | 2003-06-21 | 1 | -21/+11 | |
| | | | | | | | | | instruction. This allows us to not force the use of the EAX/AX registers! llvm-svn: 6830 | |||||
| * | Add IMULr16 & IMULr32 instructions | Chris Lattner | 2003-06-21 | 1 | -7/+9 | |
| | | | | | llvm-svn: 6829 | |||||
| * | Generate code for LONG indexes to getelementptr instructions more efficiently | Chris Lattner | 2003-06-21 | 1 | -1/+25 | |
| | | | | | llvm-svn: 6828 | |||||
| * | Some preprocessors doen't support // comments and get confused | Chris Lattner | 2003-06-20 | 1 | -1/+1 | |
| | | | | | llvm-svn: 6821 | |||||
| * | Fix the build. :( | Chris Lattner | 2003-06-20 | 1 | -3/+3 | |
| | | | | | llvm-svn: 6797 | |||||
| * | RDCCR defines arg. #1, not arg. #2. | Vikram S. Adve | 2003-06-20 | 1 | -1/+1 | |
| | | | | | llvm-svn: 6796 | |||||
| * | Changes to privatize NodeType | Chris Lattner | 2003-06-19 | 1 | -6/+6 | |
| | | | | | llvm-svn: 6795 | |||||
| * | * Changes to make NodeType be private to DSNode. | Chris Lattner | 2003-06-19 | 5 | -76/+107 | |
| | | | | | | | | | * Add new MultiObject flag to DSNode which keeps track of whether or not multiple objects have been merged into the node, allowing must-alias info to be tracked. llvm-svn: 6794 | |||||
| * | Fix output bug: Intel asm comment character is #, not ;. | Brian Gaeke | 2003-06-19 | 1 | -1/+1 | |
| | | | | | llvm-svn: 6792 | |||||
| * | RECAPITALIZE all the names of pseudo-instructions, and add a comment explaining | Brian Gaeke | 2003-06-19 | 1 | -14/+17 | |
| | | | | | | | | | the convention. :-) llvm-svn: 6791 | |||||
| * | Rename many instructions to their Intel asm-compatible names. This | Brian Gaeke | 2003-06-19 | 1 | -49/+49 | |
| | | | | | | | | | | involves removing the [bwl] suffixes from instruction names, as well as some other distinguishing marks (32/64/80 on fp insns, _i suffixes, etc.) Lowercase all instr. names as well for consistency's sake. llvm-svn: 6790 | |||||
| * | Add Initialization and Finalization methods for the Printer pass, | Brian Gaeke | 2003-06-19 | 1 | -3/+26 | |
| | | | | | | | | | | | | | to print various things on a module-by-module basis (currently, only the former is used). Don't print < > around names. The assembler can't take it. Print pseudoinstructions only as comments. The poor little assembler can't take that, either. llvm-svn: 6789 | |||||
| * | Remove usage of sys/unistd.h | Chris Lattner | 2003-06-19 | 2 | -2/+1 | |
| | | | | | llvm-svn: 6788 | |||||
| * | Implement the functionality of InstCombine/call.ll | Chris Lattner | 2003-06-19 | 1 | -1/+145 | |
| | | | | | llvm-svn: 6783 | |||||
| * | Remove a bunch of complicated code. The functionality is implemented in ↵ | Chris Lattner | 2003-06-19 | 1 | -118/+6 | |
| | | | | | | | instcombine instead llvm-svn: 6782 | |||||
| * | lib/Target/X86/X86TargetMachine.{cpp,h}: Add initial version | Brian Gaeke | 2003-06-18 | 2 | -0/+15 | |
| | | | | | | | (non-working) of llc guts for X86, and add a prototype for it. llvm-svn: 6779 | |||||
| * | lib/Target/Sparc/Sparc.cpp: | Brian Gaeke | 2003-06-18 | 1 | -1/+22 | |
| | | | | | | | | | | | | | | | | | | | | | | Move LowerAllocations, PrintFunction, and SymbolStripping passes, and the corresponding -disable-strip and -d options, over here to the SPARC target-specific bits of llc. Rename -d to -dump-asm. tools/llc/Makefile: Reindent. Add x86 library so that llc compiles again. tools/llc/llc.cpp: Remove support for running arbitrary optimization passes. Use opt instead. Remove LowerAllocations, PrintFunction, and SymbolStripping passes, as noted above. Allow user to select a backend (x86 or SPARC); default to guessing from the endianness/pointer size of the input bytecode file. Fix typos. Delete empty .s file and exit with error status if target does not support static compilation. llvm-svn: 6776 | |||||
| * | Detemplatize the PATypeHandle class, which was only really instantiated on ↵ | Chris Lattner | 2003-06-18 | 4 | -20/+20 | |
| | | | | | | | 'Type'. llvm-svn: 6774 | |||||
| * | Handle arguments passed in through the va_arg area | Chris Lattner | 2003-06-18 | 1 | -1/+7 | |
| | | | | | llvm-svn: 6769 | |||||
| * | These instructions really take three operands. This fixes some assertions | Chris Lattner | 2003-06-18 | 1 | -2/+2 | |
| | | | | | llvm-svn: 6765 | |||||

