| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | fix two strange things in the default passmgr: | Chris Lattner | 2009-11-01 | 1 | -2/+8 |
| * | IPSCCP apparently is not a superset of IPCP, this is bad, | Chris Lattner | 2009-11-01 | 1 | -0/+1 |
| * | Improve the other instance of the comment. | Duncan Sands | 2009-11-01 | 1 | -1/+1 |
| * | only run GlobalDCE at -O3 and run it late instead of early. | Chris Lattner | 2009-11-01 | 1 | -4/+9 |
| * | We currently only run ipsccp at LTO time, which is silly. It subsumes | Chris Lattner | 2009-11-01 | 1 | -2/+1 |
| * | remove a bunch of locking from LLVMContextImpl. Since only one thread | Chris Lattner | 2009-11-01 | 1 | -2/+7 |
| * | the verifier shouldn't modify the IR. | Chris Lattner | 2009-11-01 | 2 | -3/+4 |
| * | Reverting 85714, 85715, 85716, which are breaking the build | Douglas Gregor | 2009-11-01 | 4 | -10/+3 |
| * | Add a function to Passes.h to allow clients to create instances | Dan Gohman | 2009-11-01 | 1 | -0/+7 |
| * | Don't #include Pass.h from CallGraph.h. | Dan Gohman | 2009-11-01 | 1 | -1/+1 |
| * | Remove the #include of Pass.h from PassManager.h. This breaks a significant | Dan Gohman | 2009-11-01 | 2 | -2/+2 |
| * | fix an issue where the verifier would reject a function whose entry | Chris Lattner | 2009-11-01 | 1 | -0/+4 |
| * | Revert 85678/85680. The decision is to stay with the current form of | Chris Lattner | 2009-11-01 | 1 | -4/+0 |
| * | update the comment above llvm::SplitCriticalEdge, and make | Chris Lattner | 2009-10-31 | 1 | -3/+9 |
| * | Make blockaddress(@func, null) be valid, and make 'deleting a basic | Chris Lattner | 2009-10-31 | 1 | -0/+4 |
| * | Revert r85667. LoopUnroll currently can't call utility functions which | Dan Gohman | 2009-10-31 | 1 | -5/+3 |
| * | Merge the enhancements from LoopUnroll's FoldBlockIntoPredecessor into | Dan Gohman | 2009-10-31 | 1 | -3/+5 |
| * | Rename forgetLoopBackedgeTakenCount to forgetLoop, because it | Dan Gohman | 2009-10-31 | 1 | -5/+4 |
| * | Remove CodeGenLICM. It's largely obsoleted by MachineLICM's new ability | Dan Gohman | 2009-10-31 | 2 | -7/+0 |
| * | make hasAddressTaken() constant time by storing a refcount in BB's subclass d... | Chris Lattner | 2009-10-30 | 1 | -2/+11 |
| * | Fix MachineLICM to use the correct virtual register class when | Dan Gohman | 2009-10-30 | 1 | -2/+5 |
| * | Add option to createGVNPass to disable PRE. | Evan Cheng | 2009-10-30 | 1 | -1/+1 |
| * | Rather than having llvm-gcc changing the meaning of OptimizeSize, just make s... | Evan Cheng | 2009-10-30 | 1 | -1/+1 |
| * | Initial target-independent CodeGen support for BlockAddresses. | Dan Gohman | 2009-10-30 | 5 | -2/+44 |
| * | Remove dead code. | Devang Patel | 2009-10-30 | 2 | -27/+2 |
| * | Add a BlockAddress MachineOperand kind. | Dan Gohman | 2009-10-30 | 1 | -10/+27 |
| * | Remove a redundant copy constructor. | Dan Gohman | 2009-10-30 | 1 | -1/+0 |
| * | Move some code from being emitted as boilerplate duplicated in every | Dan Gohman | 2009-10-29 | 1 | -0/+8 |
| * | Rename usesCustomDAGSchedInserter to usesCustomInserter, and update a | Dan Gohman | 2009-10-29 | 3 | -8/+9 |
| * | add interpreter support for indirect goto / blockaddress. The interpreter | Chris Lattner | 2009-10-29 | 1 | -0/+6 |
| * | add newline to make cl.exe happy. | Zhongxing Xu | 2009-10-29 | 1 | -1/+1 |
| * | Don't put in these EH changes. | Bill Wendling | 2009-10-29 | 2 | -14/+1 |
| * | A switch-on-string-literal construct that is a nice alternative to | Douglas Gregor | 2009-10-29 | 1 | -0/+83 |
| * | Reverting r85338 for now. It's causing a bootstrap failure on PPC darwin9. | Bill Wendling | 2009-10-29 | 2 | -1/+14 |
| * | Add a hasAddressTaken for BasicBlock. | Dan Gohman | 2009-10-29 | 1 | -0/+4 |
| * | add IRBuilder support for IndirectBr | Chris Lattner | 2009-10-28 | 1 | -0/+7 |
| * | Revert r85346 change to control tail merging by CodeGenOpt::Level. | Bob Wilson | 2009-10-28 | 1 | -4/+3 |
| * | Extend getMallocArraySize() to determine the array size if the malloc argumen... | Victor Hernandez | 2009-10-28 | 1 | -19/+6 |
| * | Make AntiDepReg.h internal. | David Goodwin | 2009-10-28 | 1 | -62/+0 |
| * | add a new 'SetCurrentDebugType' API (requested by Andrew Haley for JIT | Chris Lattner | 2009-10-28 | 1 | -32/+35 |
| * | Add ABCD, a generalized implementation of the Elimination of Array Bounds | Nick Lewycky | 2009-10-28 | 2 | -0/+7 |
| * | bitcode writer support for blockaddress. | Chris Lattner | 2009-10-28 | 1 | -1/+2 |
| * | Previously, all operands to Constant were themselves constant. | Chris Lattner | 2009-10-28 | 2 | -11/+12 |
| * | basic blocks can now have non-instruction users. | Chris Lattner | 2009-10-28 | 1 | -4/+4 |
| * | Add a second ValueType argument to isFPImmLegal. | Evan Cheng | 2009-10-28 | 1 | -1/+1 |
| * | Remove getIEEEFloatParts and getIEEEDoubleParts. They are not needed. | Evan Cheng | 2009-10-28 | 1 | -7/+0 |
| * | rename indbr -> indirectbr to appease the residents of #llvm. | Chris Lattner | 2009-10-28 | 4 | -31/+31 |
| * | IR support for the new BlockAddress constant kind. This is | Chris Lattner | 2009-10-28 | 2 | -0/+41 |
| * | Record CodeGen optimization level in the BranchFolding pass so that we can | Bob Wilson | 2009-10-27 | 1 | -3/+4 |
| * | Revert the API changes from r85295 to make it easier for people to build | Jeffrey Yasskin | 2009-10-27 | 1 | -3/+8 |
