| Commit message (Expand) | Author | Age | Files | Lines |
... | |
* | Fixed up my changes to add support for cloning Machine Instructions. | Tanya Lattner | 2004-05-23 | 1 | -2/+3 |
* | Adding support to clone MachineInstr | Tanya Lattner | 2004-05-23 | 1 | -0/+19 |
* | Inline both direct and indirect callees in the CBU phase because | Vikram S. Adve | 2004-05-23 | 1 | -34/+34 |
* | Complete rewrite of the code that merges DS graphs for equivalence classes | Vikram S. Adve | 2004-05-23 | 1 | -0/+428 |
* | Fix a typo in a comment. | Brian Gaeke | 2004-05-23 | 1 | -1/+1 |
* | Fix for PR340: Verifier misses malformed switch instruction | Chris Lattner | 2004-05-21 | 1 | -0/+12 |
* | Fix to make the CBE always emit comparisons inline. Hopefully this will | Chris Lattner | 2004-05-20 | 2 | -0/+8 |
* | Fix a bug in outputting branches to constant PCs. Since the PC is supplied as | Brian Gaeke | 2004-05-20 | 1 | -2/+3 |
* | Minor simplification: eliminate a dyn_cast. | Brian Gaeke | 2004-05-19 | 1 | -9/+4 |
* | Add CloneTraceInto(), which is based on (and has mostly the same | Brian Gaeke | 2004-05-19 | 1 | -0/+33 |
* | Move RemapInstruction() to ValueMapper, so that it can be shared with | Brian Gaeke | 2004-05-19 | 3 | -22/+26 |
* | Fold two assertions with backwards error messages into one with a | Brian Gaeke | 2004-05-17 | 1 | -6/+3 |
* | Clean up this pass somewhat: | Brian Gaeke | 2004-05-14 | 1 | -42/+38 |
* | Don't keep track of references to LLVM BasicBlocks while emitting; use | Brian Gaeke | 2004-05-14 | 1 | -11/+9 |
* | Support MachineBasicBlock operands on RawFrm instructions. | Brian Gaeke | 2004-05-14 | 2 | -54/+24 |
* | Generate branch machine instructions with MachineBasicBlock operands instead of | Brian Gaeke | 2004-05-14 | 2 | -14/+18 |
* | This was not meant to be committed | Chris Lattner | 2004-05-13 | 1 | -7/+0 |
* | Fix a nasty bug that caused us to unroll EXTREMELY large loops due to overflow | Chris Lattner | 2004-05-13 | 2 | -3/+10 |
* | Two more improvements for null pointer handling: storing a null pointer | Chris Lattner | 2004-05-13 | 2 | -6/+24 |
* | Second half of my fixed-sized-alloca patch. This folds the LEA to compute | Chris Lattner | 2004-05-13 | 2 | -186/+358 |
* | Substantially improve code generation for address exposed locals (aka fixed | Chris Lattner | 2004-05-13 | 2 | -128/+202 |
* | Fix a really nasty bug from my changes on Monday to PHIElim. These changes | Chris Lattner | 2004-05-12 | 1 | -11/+19 |
* | Start NextMBBNumber out at zero. | Brian Gaeke | 2004-05-12 | 1 | -1/+1 |
* | Add non-const MachineBasicBlock::getParent() accessor method. | Brian Gaeke | 2004-05-12 | 1 | -0/+19 |
* | Pass boolean constants into function calls more efficiently, generating: | Chris Lattner | 2004-05-12 | 2 | -2/+14 |
* | Do not pass in the same argument to the extracted function more than once, and | Chris Lattner | 2004-05-12 | 1 | -1/+8 |
* | Implement support for code extracting basic blocks that have a return | Chris Lattner | 2004-05-12 | 1 | -2/+12 |
* | Implement splitting of PHI nodes, allowing block extraction of BB's that have | Chris Lattner | 2004-05-12 | 1 | -7/+96 |
* | * Pull some code out into the definedInRegion/definedInCaller methods | Chris Lattner | 2004-05-12 | 1 | -78/+78 |
* | Generate substantially better code when there are a limited number of exits | Chris Lattner | 2004-05-12 | 1 | -22/+61 |
* | Two minor improvements: | Chris Lattner | 2004-05-12 | 1 | -18/+14 |
* | Switch this from using an std::map to using a DenseMap. This speeds up | Chris Lattner | 2004-05-10 | 1 | -6/+5 |
* | Use a new VRegPHIUseCount to compute uses of PHI values by other phi values | Chris Lattner | 2004-05-10 | 1 | -14/+21 |
* | Now that we use an ilist of machine instructions, iterators are more robust | Chris Lattner | 2004-05-10 | 1 | -5/+8 |
* | Fix a fairly serious pessimizaion that was preventing us from efficiently | Chris Lattner | 2004-05-10 | 2 | -4/+6 |
* | Patch to fix PR337. Make sure to mark all aliased physical registers as used | Chris Lattner | 2004-05-10 | 1 | -3/+8 |
* | Fix some comments, avoid sign extending booleans when zero extend works fine | Chris Lattner | 2004-05-09 | 2 | -10/+10 |
* | Generate more efficient code for casting booleans to integers (no sign extens... | Chris Lattner | 2004-05-09 | 2 | -2/+2 |
* | syntactically loopify natural loops so that the GCC loop optimizer can find t... | Chris Lattner | 2004-05-09 | 2 | -104/+170 |
* | Do not emit prototypes for setjmp/longjmp, as they are handled specially | Chris Lattner | 2004-05-09 | 2 | -2/+4 |
* | Fine grainify namespacification | Chris Lattner | 2004-05-09 | 1 | -4/+1 |
* | Make the floating point constant pools local to each function, split the | Chris Lattner | 2004-05-09 | 2 | -198/+224 |
* | Get this looking more like a function pass. | Chris Lattner | 2004-05-09 | 2 | -64/+66 |
* | Implement the AddPrototypes method | Chris Lattner | 2004-05-09 | 2 | -2/+86 |
* | Print all PHI copies for successor blocks before the terminator, whether it b... | Chris Lattner | 2004-05-09 | 2 | -20/+46 |
* | Fix stupid bug in my checkin yesterday | Chris Lattner | 2004-05-08 | 1 | -2/+1 |
* | Changed CPUResource to allow access to maxnum users. | Tanya Lattner | 2004-05-08 | 1 | -6/+16 |
* | Updating my versions of ModuloScheduling in cvs. Still not complete. | Tanya Lattner | 2004-05-08 | 8 | -606/+1720 |
* | Add support for widening integral casts. | Brian Gaeke | 2004-05-08 | 4 | -136/+304 |
* | Add a bunch more branches | Brian Gaeke | 2004-05-08 | 2 | -8/+24 |