| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | convert LoopInfo.h and GraphWriter.h to use raw_ostream | Chris Lattner | 2009-08-23 | 1 | -2/+2 |
| * | Fix counting of Post-RA scheduling stalls. Improve debug output. | David Goodwin | 2009-08-12 | 1 | -14/+25 |
| * | This logic was accidentally inverted in r78767. | Dan Gohman | 2009-08-12 | 1 | -3/+3 |
| * | Factor out the code for finding an available register for use | Dan Gohman | 2009-08-12 | 1 | -54/+70 |
| * | Use DEBUG macro for debug output. | David Goodwin | 2009-08-11 | 1 | -11/+7 |
| * | Add some debug output. | David Goodwin | 2009-08-11 | 1 | -2/+14 |
| * | Replace DOUT. | David Goodwin | 2009-08-11 | 1 | -11/+12 |
| * | Post RA scheduler changes. Introduce a hazard recognizer that uses the target... | David Goodwin | 2009-08-10 | 1 | -71/+21 |
| * | Use setPreservesAll and setPreservesCFG in CodeGen passes. | Dan Gohman | 2009-07-31 | 1 | -0/+1 |
| * | inline the global 'getInstrOperandRegClass' function into its callers | Chris Lattner | 2009-07-29 | 1 | -4/+7 |
| * | llvm_unreachable->llvm_unreachable(0), LLVM_UNREACHABLE->llvm_unreachable. | Edwin Török | 2009-07-14 | 1 | -1/+1 |
| * | assert(0) -> LLVM_UNREACHABLE. | Edwin Török | 2009-07-11 | 1 | -1/+2 |
| * | Move getInstrOperandRegClass from the scheduler to TargetInstrInfo. | Evan Cheng | 2009-05-05 | 1 | -12/+0 |
| * | Fix pr3954. The register scavenger asserts for inline assembly with | Bob Wilson | 2009-04-09 | 1 | -1/+1 |
| * | Add parentheses to pacify gcc-4.3. | Duncan Sands | 2009-03-11 | 1 | -1/+1 |
| * | Fix a post-RA scheduling liveness bug. When a basic block is being | Dan Gohman | 2009-03-10 | 1 | -9/+22 |
| * | When scheduling a block in parts, keep track of the overall | Dan Gohman | 2009-02-11 | 1 | -10/+28 |
| * | Consider any instruction that modifies the stack pointer to be | Dan Gohman | 2009-02-10 | 1 | -0/+10 |
| * | Factor out more code for computing register live-range informationfor | Dan Gohman | 2009-02-10 | 1 | -191/+292 |
| * | Move getPointerRegClass from TargetInstrInfo to TargetRegisterInfo. | Evan Cheng | 2009-02-06 | 1 | -5/+4 |
| * | Move ScheduleDAGInstrs.h to be a private header. Front-ends | Dan Gohman | 2009-02-06 | 1 | -1/+1 |
| * | Change the post-RA scheduler to iterate through the | Dan Gohman | 2009-02-03 | 1 | -9/+11 |
| * | Instead of adding dependence edges between terminator instructions | Dan Gohman | 2009-01-16 | 1 | -2/+10 |
| * | If an anti-dependence uses a non-allocatable register, set AntiDepReg | Dan Gohman | 2009-01-16 | 1 | -1/+3 |
| * | Fix the check for an empty basic block to check for an empty SUnits | Dan Gohman | 2009-01-16 | 1 | -2/+2 |
| * | Fix a "comparison between signed and unsigned integer expressions" | Dan Gohman | 2009-01-16 | 1 | -1/+1 |
| * | Initial hazard recognizer support in post-pass scheduling. This includes | Dan Gohman | 2009-01-16 | 1 | -8/+119 |
| * | Move a few containers out of ScheduleDAGInstrs::BuildSchedGraph | Dan Gohman | 2009-01-15 | 1 | -11/+14 |
| * | Tidy up #includes, deleting a bunch of unnecessary #includes. | Dan Gohman | 2009-01-05 | 1 | -3/+0 |
| * | Rename BuildSchedUnits to BuildSchedGraph, and refactor the | Dan Gohman | 2008-12-23 | 1 | -3/+3 |
| * | Use ~0u instead of -1u as the special value, to hopefully avoid | Dan Gohman | 2008-12-19 | 1 | -20/+20 |
| * | Eliminate the loop that walks the critical path. Instead, just track the | Dan Gohman | 2008-12-16 | 1 | -54/+64 |
| * | Enable anti-dependence breaking by default when post-RA scheduling is enabled. | Dan Gohman | 2008-12-16 | 1 | -1/+1 |
| * | When breaking an anti-dependency, don't use a register which has seen | Dan Gohman | 2008-12-16 | 1 | -0/+1 |
| * | Fix some register-alias-related bugs in the post-RA scheduler liveness | Dan Gohman | 2008-12-16 | 1 | -37/+61 |
| * | Rewrite the SDep class, and simplify some of the related code. | Dan Gohman | 2008-12-09 | 1 | -22/+17 |
| * | Drop the reg argument to isRegReDefinedByTwoAddr, which was redundant. | Dan Gohman | 2008-12-05 | 1 | -1/+1 |
| * | Ignore IMPLICIT_DEF instructions when computing physreg liveness. | Dan Gohman | 2008-12-05 | 1 | -0/+7 |
| * | Use register names instead of numbers in debug output. | Dan Gohman | 2008-12-04 | 1 | -2/+3 |
| * | Rewrite the liveness bookkeeping code to fix a bunch of | Dan Gohman | 2008-12-03 | 1 | -34/+42 |
| * | Fix an inconsistency in a comment. | Dan Gohman | 2008-12-03 | 1 | -4/+3 |
| * | Don't charge the full latency for anti and output dependencies. This is | Dan Gohman | 2008-12-03 | 1 | -3/+6 |
| * | When looking for anti-dependences on the critical path, don't bother | Dan Gohman | 2008-12-03 | 1 | -0/+4 |
| * | Add a comment about callee-saved registers. | Dan Gohman | 2008-12-03 | 1 | -0/+2 |
| * | Silence a warning. | Nick Lewycky | 2008-11-27 | 1 | -1/+1 |
| * | Suppress warnings. | Dan Gohman | 2008-11-25 | 1 | -7/+7 |
| * | Initial support for anti-dependence breaking. Currently this code does not | Dan Gohman | 2008-11-25 | 1 | -23/+416 |
| * | Factor out the code for verifying the work of the scheduler, | Dan Gohman | 2008-11-20 | 1 | -12/+1 |
| * | Experimental post-pass scheduling support. Post-pass scheduling | Dan Gohman | 2008-11-19 | 1 | -8/+195 |
| * | Tidy up several unbeseeming casts from pointer to intptr_t. | Dan Gohman | 2008-09-04 | 1 | -1/+1 |
