| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Make the GraphWriter be more consistent about the string | Dan Gohman | 2008-07-21 | 1 | -19/+8 | 
| | | | | | | | | | used for the graph "title" and the graph "label", as there are differences in interpretation of these strings between viewers. llvm-svn: 53871 | ||||
| * | Fix uses of underscore-capital names. | Dan Gohman | 2008-07-21 | 1 | -2/+2 | 
| | | | | | llvm-svn: 53870 | ||||
| * | Added initial support for small sections on Mips. | Bruno Cardoso Lopes | 2008-07-21 | 7 | -29/+87 | 
| | | | | | | | | Added gp_rel relocations to support addressing small section contents. Added command line to specify small section threshold in bytes. llvm-svn: 53869 | ||||
| * | Now that the MachineInstr leaks are fixed, enable leak checking | Dan Gohman | 2008-07-21 | 1 | -0/+2 | 
| | | | | | | | in the MachineInstr clone code. llvm-svn: 53868 | ||||
| * | Temporary hack to build with GCC 4.0 instead of 4.2. | Bill Wendling | 2008-07-21 | 1 | -1/+2 | 
| | | | | | llvm-svn: 53860 | ||||
| * | Use better variable names | Anton Korobeynikov | 2008-07-21 | 1 | -3/+3 | 
| | | | | | llvm-svn: 53859 | ||||
| * | Don't use larger alignment. | Anton Korobeynikov | 2008-07-21 | 1 | -1/+1 | 
| | | | | | llvm-svn: 53857 | ||||
| * | Add VerifyNode, a place to put sanity checks on | Duncan Sands | 2008-07-21 | 6 | -12/+54 | 
| | | | | | | | | | | | | | generic SDNode's (nodes with their own constructors should do sanity checking in the constructor). Add sanity checks for BUILD_VECTOR and fix all the places that were producing bogus BUILD_VECTORs, as found by "make check". My favorite is the BUILD_VECTOR with only two operands that was being used to build a vector with four elements! llvm-svn: 53850 | ||||
| * | Use movaps instead of movups to spill 16-byte vector values when default ↵ | Evan Cheng | 2008-07-21 | 1 | -9/+19 | 
| | | | | | | | alignment is >= 16. This fixes some massive performance regressions. llvm-svn: 53844 | ||||
| * | Revert r53812 -- premature. LegalizeTypes isn't actually on yet! | Nick Lewycky | 2008-07-21 | 1 | -8/+22 | 
| | | | | | llvm-svn: 53816 | ||||
| * | Switch on the use of arbitrary precision integers in scalar evolution. This will | Nick Lewycky | 2008-07-21 | 1 | -22/+8 | 
| | | | | | | | | | | | bail after 256-bits to avoid producing code that the backends can't handle. Previously, we capped it at 64-bits, preferring to miscompile in those cases. This change also reverts much of r52248 because the invariants the code was expecting are now being met. llvm-svn: 53812 | ||||
| * | This header isn't necessary now. | Wojciech Matyjewicz | 2008-07-20 | 1 | -2/+0 | 
| | | | | | llvm-svn: 53811 | ||||
| * | Fix PR2088. Use modulo linear equation solver to compute loop iteration | Wojciech Matyjewicz | 2008-07-20 | 5 | -24/+102 | 
| | | | | | | | count. llvm-svn: 53810 | ||||
| * | Updated VC++ projects | Cedric Venet | 2008-07-20 | 1 | -0/+8 | 
| | | | | | llvm-svn: 53809 | ||||
| * | Fix for first part of PR2562. Generate the "pinsrw" instruction for inserts | Bill Wendling | 2008-07-20 | 2 | -0/+17 | 
| | | | | | | | into v4i16 vectors. llvm-svn: 53807 | ||||
| * | Pull r53795 from Gaz into mainline: | Bill Wendling | 2008-07-20 | 1 | -14/+29 | 
| | | | | | | | | | If .loc and .file aren't used, always emit the "debug_line" section. This requires at least one entry in the line matrix. So if there's nothing to emit into the matrix, emit an end of matrix value anyway. llvm-svn: 53803 | ||||
| * | Unbreak build: 'DarwinTargetAsmInfo' was already taken as PPC TAI flavour. | Anton Korobeynikov | 2008-07-19 | 3 | -12/+14 | 
| | | | | | llvm-svn: 53801 | ||||
| * | Make sure custom lowering for LegalizeTypes | Duncan Sands | 2008-07-19 | 1 | -4/+10 | 
| | | | | | | | | | | | returns a node with the right number of return values. This fixes codegen of Generic/cast-fp.ll, Generic/fp_to_int.ll and PowerPC/multiple-return-values.ll when using -march=ppc32 -mattr=+64bit. llvm-svn: 53794 | ||||
| * | XFAIL this test. | Nick Lewycky | 2008-07-19 | 1 | -0/+1 | 
| | | | | | llvm-svn: 53793 | ||||
| * | While testing particular algorithms to compute loop iteration count the brute | Wojciech Matyjewicz | 2008-07-19 | 5 | -7/+10 | 
| | | | | | | | | | | | | | | force evaluation (ComputeIterationCountExhaustively) should be turned off. It doesn't apply to trip-count2.ll because this file tests the brute force evaluation. The test for PR2364 (2008-05-25-NegativeStepToZero.ll) currently fails showing that the patch for this bug doesn't work. I'll fix it in a few hours with a patch for PR2088. llvm-svn: 53792 | ||||
| * | Use chars, where possible | Anton Korobeynikov | 2008-07-19 | 1 | -25/+25 | 
| | | | | | llvm-svn: 53791 | ||||
| * | Switch MIPS to new ELFTargetAsmInfo. Add few FIXMEs. | Anton Korobeynikov | 2008-07-19 | 4 | -138/+114 | 
| | | | | | llvm-svn: 53790 | ||||
| * | Fix a FIXME :) | Anton Korobeynikov | 2008-07-19 | 2 | -6/+7 | 
| | | | | | llvm-svn: 53789 | ||||
| * | Use generic ELFTargetAsmInfo and DarwinTargetAsmInfo for X86 code | Anton Korobeynikov | 2008-07-19 | 5 | -274/+32 | 
| | | | | | llvm-svn: 53788 | ||||
| * | Add TargetAsmInfo stuff for all darwin-based targets | Anton Korobeynikov | 2008-07-19 | 2 | -0/+172 | 
| | | | | | llvm-svn: 53787 | ||||
| * | Add TargetAsmInfo for all ELF-based targets | Anton Korobeynikov | 2008-07-19 | 2 | -0/+202 | 
| | | | | | llvm-svn: 53786 | ||||
| * | Testcase for PR2549 | Anton Korobeynikov | 2008-07-19 | 1 | -0/+636 | 
| | | | | | llvm-svn: 53785 | ||||
| * | Use aligned stack spills, where possible. This fixes PR2549. | Anton Korobeynikov | 2008-07-19 | 1 | -19/+23 | 
| | | | | | llvm-svn: 53784 | ||||
| * | Remove bogus assertion. This unbreaks mingw, where ConstantSDNode | Anton Korobeynikov | 2008-07-19 | 1 | -4/+0 | 
| | | | | | | | has alignment 8 and LoadSDNode (used as LargestT template) - 4. llvm-svn: 53782 | ||||
| * | tentative definitions are default visibility | Nick Kledzik | 2008-07-19 | 1 | -1/+1 | 
| | | | | | llvm-svn: 53780 | ||||
| * | Fix a memory leak in LiveIntervalAnalysis. | Evan Cheng | 2008-07-19 | 2 | -11/+13 | 
| | | | | | llvm-svn: 53779 | ||||
| * | typo. | Devang Patel | 2008-07-18 | 1 | -2/+2 | 
| | | | | | llvm-svn: 53778 | ||||
| * | If all linker input files are native object files then lto-bugpoint is not ↵ | Devang Patel | 2008-07-18 | 4 | -8/+48 | 
| | | | | | | | useful. llvm-svn: 53777 | ||||
| * | Start writing lto-bugpoint tool. | Devang Patel | 2008-07-18 | 4 | -0/+140 | 
| | | | | | llvm-svn: 53774 | ||||
| * | Softfloat support for FDIV. Patch by | Duncan Sands | 2008-07-18 | 3 | -0/+20 | 
| | | | | | | | Richard Pennington. llvm-svn: 53773 | ||||
| * | Eliminate unused variable. | Duncan Sands | 2008-07-18 | 1 | -1/+1 | 
| | | | | | llvm-svn: 53772 | ||||
| * | Supress a gcc-4.3 warning. | Duncan Sands | 2008-07-18 | 1 | -1/+1 | 
| | | | | | llvm-svn: 53771 | ||||
| * | Revert 53729, after waking up in the middle of | Duncan Sands | 2008-07-18 | 1 | -4/+3 | 
| | | | | | | | | | | | | | | | | | the night realising that it was wrong :) I think the reason the same type was being used for the shufflevec of indices as for the actual indices is so that if one of them needs splitting then so does the other. After my patch it might be that the indices need splitting but not the rest, yet there is no good way of handling that. I think the right solution is to not have the shufflevec be an operand at all: just have it be the list of numbers it actually is, stored as extra info in the node. llvm-svn: 53768 | ||||
| * | In the CBackend, use casts to force integer add, subtract, and | Dan Gohman | 2008-07-18 | 3 | -1/+42 | 
| | | | | | | | | multiply to be done as unsigned, so that they have well defined behavior on overflow. This fixes PR2408. llvm-svn: 53767 | ||||
| * | Fix a LocalSpiller leak. This fixes tramp3d-v4. | Dan Gohman | 2008-07-18 | 1 | -0/+1 | 
| | | | | | llvm-svn: 53766 | ||||
| * | Make PRE actually handle critical edges (by splitting them). Confirmed that ↵ | Owen Anderson | 2008-07-18 | 1 | -3/+6 | 
| | | | | | | | bootstrap passes with this change. llvm-svn: 53762 | ||||
| * | Reapply r53735. My last patch fixed the failures Dan observed. | Owen Anderson | 2008-07-18 | 1 | -52/+9 | 
| | | | | | llvm-svn: 53761 | ||||
| * | Add some checks that got lost in the shuffle. This fixes 464.h264ref. | Owen Anderson | 2008-07-18 | 1 | -0/+4 | 
| | | | | | llvm-svn: 53760 | ||||
| * | Revert r53735. It broke SPEC 464.h264ref. | Dan Gohman | 2008-07-18 | 1 | -9/+52 | 
| | | | | | llvm-svn: 53757 | ||||
| * | verify limits-fndefn.c from the GCC testsuite before the | Chris Lattner | 2008-07-18 | 1 | -1/+1 | 
| | | | | | | | | heat death of the universe, fixing an O(N^2) problem in the size of a basic block. llvm-svn: 53749 | ||||
| * | Re-introduce LeakDetector support for MachineInstrs and MachineBasicBlocks. | Dan Gohman | 2008-07-17 | 5 | -7/+27 | 
| | | | | | | | | Fix a leak that this turned up in LowerSubregs.cpp. And, comment a leak in LiveIntervalAnalysis.cpp. llvm-svn: 53746 | ||||
| * | When printing MemOperand nodes, only use print() for | Dan Gohman | 2008-07-17 | 1 | -3/+10 | 
| | | | | | | | | | PseudoSourceValue values, which never have names. Use getName() for all other values, because we want to print just a short summary of the value, not the entire instruction. llvm-svn: 53738 | ||||
| * | Use MergeBlockIntoPredecessor to simplify some code. | Owen Anderson | 2008-07-17 | 1 | -52/+9 | 
| | | | | | llvm-svn: 53735 | ||||
| * | Subreg live interval valno may not have a corresponding def machineinstr ↵ | Evan Cheng | 2008-07-17 | 2 | -1/+35 | 
| | | | | | | | since it's less precise. llvm-svn: 53734 | ||||
| * | Add nounwind. | Evan Cheng | 2008-07-17 | 1 | -1/+1 | 
| | | | | | llvm-svn: 53733 | ||||

