summaryrefslogtreecommitdiffstats
path: root/llvm
Commit message (Collapse)AuthorAgeFilesLines
...
* Revert my last patch until I consult with Evan about it.Owen Anderson2008-11-031-10/+0
| | | | llvm-svn: 58591
* Changes from Duncan's review:Nick Lewycky2008-11-022-9/+39
| | | | | | | | * merge two weak functions by making them both alias a third non-weak fn * don't reimplement CallSite::hasArgument * whitelist the safe linkage types llvm-svn: 58568
* Testcase for PR2691Anton Korobeynikov2008-11-021-0/+5
| | | | llvm-svn: 58567
* Silence a warningAnton Korobeynikov2008-11-021-4/+6
| | | | llvm-svn: 58563
* Get this building on 64 bit machines (error:Duncan Sands2008-11-021-5/+5
| | | | | | | cast from ‘const llvm::PointerType*’ to ‘unsigned int’ loses precision). llvm-svn: 58561
* Don't do pre-splitting if doing so would create a value join that did notOwen Anderson2008-11-021-0/+10
| | | | | | | | | exist before. Updating the live intervals in that care is tricky in the general case. Evan, if you see a tighter guard condition for this, let me know. llvm-svn: 58560
* CMake: added a source file.Oscar Fuentes2008-11-021-0/+1
| | | | llvm-svn: 58559
* Add a new MergeFunctions pass. It finds identical functions and merges them.Nick Lewycky2008-11-026-3/+429
| | | | | | | | This triggers only 60 times in llvm-test (look at .llvm.bc, not .linked.rbc) and so it probably wont be turned on by default. Also, may of those are likely to go away when PR2973 is fixed. llvm-svn: 58557
* Fix demanded bits analysis with srem by negative number. Based on a patchNick Lewycky2008-11-022-3/+11
| | | | | | by Richard Osborne. llvm-svn: 58555
* Fix this recently moved code to use the correct type. CI is now aDan Gohman2008-11-022-2/+14
| | | | | | | ConstantInt, and SI is the original cast instruction. This fixes PR2996. llvm-svn: 58549
* Added interface to allow clients to create a MemIntrinsicNode forMon P Wang2008-11-012-5/+39
| | | | | | target intrinsics that touches memory llvm-svn: 58548
* Reverting back 58505. Will commit it once I have the bc reader/writer/docsSanjiv Gupta2008-11-011-2/+1
| | | | | | ready. llvm-svn: 58547
* Whitespace fixes. No functionality change.Bill Wendling2008-10-311-27/+27
| | | | llvm-svn: 58539
* Testcase for PR2613Anton Korobeynikov2008-10-311-0/+61
| | | | llvm-svn: 58537
* Invalidate debug/eh/gc labels when unreachable MBB is deleted.Anton Korobeynikov2008-10-311-13/+27
| | | | | | Based on patch by Martin Nowack! llvm-svn: 58536
* Add comment.Evan Cheng2008-10-311-0/+2
| | | | llvm-svn: 58533
* Use better data structure for ConstPoolId2AddrMap.Evan Cheng2008-10-312-15/+20
| | | | llvm-svn: 58532
* Actually make debug output understandable.Evan Cheng2008-10-311-4/+4
| | | | llvm-svn: 58529
* x86_64 rip-relative and magic mode addressMon P Wang2008-10-311-4/+8
| | | | llvm-svn: 58528
* Forgot this in last commit.Evan Cheng2008-10-311-1/+1
| | | | llvm-svn: 58527
* Encode PICADD; some code clean up.Evan Cheng2008-10-311-76/+101
| | | | llvm-svn: 58526
* Remove some unused virtual function bodies.Dan Gohman2008-10-311-7/+5
| | | | llvm-svn: 58524
* Revert r58489. It isn't correct for all cases.Bill Wendling2008-10-312-3/+4
| | | | llvm-svn: 58523
* Symbol visibility is unsupported on cygwin too.Anton Korobeynikov2008-10-311-1/+1
| | | | | | Patch by Jay Foad! llvm-svn: 58520
* Add an assert to catch user errors like:Torok Edwin2008-10-311-2/+6
| | | | | | | | | MyFunctionPass() : FunctionPass(ID) {} when the user actually meant to write: MyFunctionPass() : FunctionPass(&ID) {} llvm-svn: 58518
* Change x86 register allocation ordering to match that of gcc. Otherwise some ↵Evan Cheng2008-10-311-19/+19
| | | | | | tools get confused by prologue generated by llvm. llvm-svn: 58517
* Add a fixme.Evan Cheng2008-10-311-0/+3
| | | | llvm-svn: 58514
* Totally disable the setSubgraphColor calls temporarily, as they'reDan Gohman2008-10-311-2/+2
| | | | | | currently troublesome even for #ifndef NDEBUG builds. llvm-svn: 58512
* Add a bunch of libcalls for ppcf128 that were somehowDuncan Sands2008-10-314-38/+286
| | | | | | completely forgotten about when writing LegalizeTypes. llvm-svn: 58508
* For some targets pointer and int are 16-bits. Allow 16-bits as a valid indexSanjiv Gupta2008-10-311-1/+2
| | | | | | in such cases. llvm-svn: 58505
* Don't skip over all "terminator" instructions when determining where to put theBill Wendling2008-10-312-4/+3
| | | | | | | callee-saved restore code. It could skip over conditional jumps accidentally. Instead, just skip the "return" instructions. llvm-svn: 58489
* Fix warning.Daniel Dunbar2008-10-311-1/+1
| | | | llvm-svn: 58486
* CMake: Fix some unwanted commentary line wraps on the last change.Oscar Fuentes2008-10-311-19/+14
| | | | llvm-svn: 58485
* CMake: Establish dependencies among executables and libraries, usingOscar Fuentes2008-10-311-3/+123
| | | | | | data manually extracted from llvm-config. llvm-svn: 58484
* Use MOVSSmr instead of EXTRACTPSmr in the case of extractingDan Gohman2008-10-312-4/+16
| | | | | | vector element 0 for a store, as it's smaller and faster. llvm-svn: 58483
* Revert errant deletion. The target needs to be able to specify that it ↵Jim Grosbach2008-10-302-0/+8
| | | | | | doesn't want the generic constant pool to be emitted. llvm-svn: 58475
* I think we got non-machine specific constpool entries covered.Evan Cheng2008-10-301-1/+2
| | | | llvm-svn: 58474
* Change the name of the generated solution file for a CMake build. (from ↵Cedric Venet2008-10-301-0/+1
| | | | | | | | | 'project' to 'LLVM'). !!!!!!!!!! Warning !!!!!!!!!!!!!!! If you already have created a solution with cmake, you will need to manually open to the new one. llvm-svn: 58461
* Testcase for PR2987.Duncan Sands2008-10-301-0/+9
| | | | llvm-svn: 58459
* Canonicalize sext(i1) to i1?-1:0, and update various instcombineDan Gohman2008-10-303-41/+162
| | | | | | optimizations accordingly. llvm-svn: 58457
* Testcase for PR2986.Duncan Sands2008-10-301-0/+14
| | | | llvm-svn: 58456
* Fix PR2986: do not use a potentially illegalDuncan Sands2008-10-301-1/+6
| | | | | | | | | | type for the shift amount type. Add a check that shifts and rotates use the type returned by getShiftAmountTy for the amount. This exposed some problems in CellSPU and PPC, which have already been fixed. llvm-svn: 58455
* Shift amounts should have type getShiftAmountTyDuncan Sands2008-10-301-9/+9
| | | | | | | (i32 for PPC, not i8). Correct this, and some formatting while there. llvm-svn: 58451
* Add InlineCost class for represent the estimated cost of inlining aDaniel Dunbar2008-10-308-18/+97
| | | | | | | | | function. - This explicitly models the costs for functions which should "always" or "never" be inlined. This fixes bugs where such costs were not previously respected. llvm-svn: 58450
* Shift amounts should have the type given byDuncan Sands2008-10-301-5/+5
| | | | | | getShiftAmountTy (i32 in the case of CellSPU). llvm-svn: 58449
* Add missing vsetcc expansion for wideningMon P Wang2008-10-301-1/+12
| | | | llvm-svn: 58443
* CMake: Reverted some unintentional changes on the previous commit.Oscar Fuentes2008-10-301-4/+1
| | | | llvm-svn: 58435
* CMake: Cygwin pretends to be Unix.Oscar Fuentes2008-10-301-3/+11
| | | | | | Patch by Jay Foad. llvm-svn: 58434
* ARM JIT should observe -relocation-model command line option.Evan Cheng2008-10-302-2/+6
| | | | llvm-svn: 58433
* Add initial support for vector widening. Logic is set to widen for X86.Mon P Wang2008-10-306-33/+938
| | | | | | | One will only see an effect if legalizetype is not active. Will move support to LegalizeType soon. llvm-svn: 58426
OpenPOWER on IntegriCloud