| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Update CMake dependencies. | Daniel Dunbar | 2009-10-24 | 1 | -1/+0 |
| * | Teach macho-dump to dump UUIDs. | Daniel Dunbar | 2009-10-24 | 1 | -0/+3 |
| * | Make DominanceFrontierBase::print's output prettier. | Dan Gohman | 2009-10-24 | 1 | -3/+5 |
| * | Make DominanceFrontier::addBasicBlock return the iterator for the newly | Dan Gohman | 2009-10-24 | 1 | -2/+2 |
| * | Add an explicit keyword. | Dan Gohman | 2009-10-24 | 1 | -1/+1 |
| * | Revert back 85006 for now as it breaks PIC16 tests. | Sanjiv Gupta | 2009-10-24 | 6 | -51/+3 |
| * | Adding support for placing global objects in shared data memory. | Sanjiv Gupta | 2009-10-24 | 6 | -3/+51 |
| * | various cleanups suggested by Duncan | Chris Lattner | 2009-10-24 | 1 | -3/+5 |
| * | fix PR5287, a serious regression from my previous patches. Thanks to | Chris Lattner | 2009-10-24 | 2 | -24/+39 |
| * | Auto-upgrade free instructions to calls to the builtin free function. | Victor Hernandez | 2009-10-24 | 30 | -269/+169 |
| * | 80 col violation. | Evan Cheng | 2009-10-24 | 1 | -1/+2 |
| * | Add some asserts to catch copyRegToReg() fails early | Anton Korobeynikov | 2009-10-24 | 1 | -4/+9 |
| * | Restrict Thumb1 register allocation to low registers, even for instructions that | Jim Grosbach | 2009-10-24 | 1 | -0/+16 |
| * | Identity copies should not contribute to spill weight. | Evan Cheng | 2009-10-23 | 2 | -4/+28 |
| * | FIXME no longer applies. R12 and R3 are available for allocation | Jim Grosbach | 2009-10-23 | 1 | -3/+0 |
| * | Fix http://llvm.org/PR4822: allow module deletion after a function has been | Jeffrey Yasskin | 2009-10-23 | 5 | -56/+106 |
| * | Remove AllocationInst. Since MallocInst went away, AllocaInst is the only su... | Victor Hernandez | 2009-10-23 | 21 | -146/+138 |
| * | Fix stylistic and documentation problems in ValueMap found by Nick Lewycky and | Jeffrey Yasskin | 2009-10-23 | 3 | -26/+36 |
| * | APInt-ify the gep scaling code, so that it correctly handles the case where | Dan Gohman | 2009-10-23 | 2 | -4/+17 |
| * | Make LoopDeletion check the maximum backedge taken count, rather than the | Dan Gohman | 2009-10-23 | 2 | -1/+28 |
| * | some stuff is done, we still have constantexpr simplification to do. | Chris Lattner | 2009-10-23 | 1 | -31/+15 |
| * | teach libanalysis to simplify vector loads with bitcast sources. This | Chris Lattner | 2009-10-23 | 2 | -7/+23 |
| * | enhance FoldReinterpretLoadFromConstPtr to handle loads of up to 32 | Chris Lattner | 2009-10-23 | 2 | -6/+19 |
| * | teach libanalysis to fold int and fp loads from almost arbitrary | Chris Lattner | 2009-10-23 | 2 | -8/+253 |
| * | Add 2.6 release note. | Tanya Lattner | 2009-10-23 | 1 | -0/+1 |
| * | Update tests for 84931. | Evan Cheng | 2009-10-23 | 7 | -14/+14 |
| * | X86 needs critical path anti-dependency breaking. | Evan Cheng | 2009-10-23 | 1 | -1/+1 |
| * | Commit fixes for half precision I noted in review, so | Dale Johannesen | 2009-10-23 | 1 | -3/+3 |
| * | This is passing on Darwin PPC. | Bill Wendling | 2009-10-23 | 1 | -1/+1 |
| * | Minor code cleanup. | Victor Hernandez | 2009-10-23 | 1 | -3/+2 |
| * | Neuter stack protectors by only checking character arrays. This is what GCC | Bill Wendling | 2009-10-23 | 1 | -1/+6 |
| * | Allow the target to select the level of anti-dependence breaking that should ... | David Goodwin | 2009-10-22 | 11 | -23/+44 |
| * | Use 'waitpid' instead of 'wait'. Basing Program::Wait() on 'wait()' prevents... | Ted Kremenek | 2009-10-22 | 1 | -1/+1 |
| * | Try r84890 again (adding ValueMap<>), now that I've tested the compile on | Jeffrey Yasskin | 2009-10-22 | 4 | -0/+683 |
| * | size_t, not unsigned here to silence a warning. | Eric Christopher | 2009-10-22 | 1 | -26/+26 |
| * | Random include cleanup. | Benjamin Kramer | 2009-10-22 | 3 | -2/+1 |
| * | Fix OProfileJITEventListener after r84054 renamed CompileUnit to Scope. | Jeffrey Yasskin | 2009-10-22 | 1 | -6/+6 |
| * | Tidying up some code and comments. No functionality change. | Bill Wendling | 2009-10-22 | 1 | -24/+6 |
| * | Revert r84890, which broke the linux build. | Jeffrey Yasskin | 2009-10-22 | 4 | -683/+0 |
| * | Add a ValueMap<ValueOrSubclass*, T> type. ValueMap<Value*, T> is safe to use | Jeffrey Yasskin | 2009-10-22 | 4 | -0/+683 |
| * | Hide MetadataContext implementation details. | Devang Patel | 2009-10-22 | 5 | -53/+192 |
| * | Fix getMDs() interface such that it does not expose implementation details. | Devang Patel | 2009-10-22 | 5 | -26/+35 |
| * | Using TrackingVH instead of WeakVH or WeakMetadataVH. | Devang Patel | 2009-10-22 | 6 | -44/+22 |
| * | Sort handler names to ensure deterministic behavior. | Devang Patel | 2009-10-22 | 2 | -5/+6 |
| * | Trying again to tweak the top-level Makefile to facilitate an Apple-style build. | Stuart Hastings | 2009-10-22 | 3 | -1/+17 |
| * | Revert 84843. Evan, this was breaking some of the if-conversion tests. | Bob Wilson | 2009-10-22 | 2 | -5/+6 |
| * | Include config.h in order to have HAVE_STDINT_H be defined. | Duncan Sands | 2009-10-22 | 1 | -0/+1 |
| * | Verify that the function and exception table have been allocated | Nicolas Geoffray | 2009-10-22 | 1 | -2/+2 |
| * | Check that accessing a struct field that occurs before the start | Duncan Sands | 2009-10-22 | 1 | -0/+6 |
| * | Test handling of record fields with negative offsets. | Duncan Sands | 2009-10-22 | 1 | -0/+10 |
