summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Transforms/Utils/LoopUnroll.cpp
Commit message (Collapse)AuthorAgeFilesLines
* LoopUnroll: always check for NULL LoopPassManagerAndrew Trick2012-06-051-3/+5
| | | | llvm-svn: 158007
* Allow NULL LoopPassManager argument in UnrollLoop. PR12734.Andrew Trick2012-05-081-20/+24
| | | | llvm-svn: 156358
* Fix 12513: Loop unrolling breaks with indirect branches.Andrew Trick2012-04-101-0/+6
| | | | | | | | Take this opportunity to generalize the indirectbr bailout logic for loop transformations. CFG transformations will never get indirectbr right, and there's no point trying. llvm-svn: 154386
* Avoid a confusing assert for silly options: -unroll-runtime -unroll-count=1.Andrew Trick2011-12-161-0/+5
| | | | | | No need for an explicit test case for an unsupported combination of options. llvm-svn: 146721
* Add -unroll-runtime for unrolling loops with run-time trip counts.Andrew Trick2011-12-091-7/+22
| | | | | | | | | | | | | Patch by Brendon Cahoon! This extends the existing LoopUnroll and LoopUnrollPass. Brendon measured no regressions in the llvm test suite with -unroll-runtime enabled. This implementation works by using the existing loop unrolling code to unroll the loop by a power-of-two (default 8). It generates an if-then-else sequence of code prior to the loop to execute the extra iterations before entering the unrolled loop. llvm-svn: 146245
* Comments. Thanks for the spell check Nick!Andrew Trick2011-08-101-2/+2
| | | | | | Also, my apologies for spoiling the autocomplete on SimplifyInstructions.cpp. I couldn't think of a better filename. llvm-svn: 137229
* Invoke SimplifyIndVar when we partially unroll a loop. Fixes PR10534.Andrew Trick2011-08-101-1/+24
| | | | llvm-svn: 137203
* Cleanup. Added LoopBlocksDFS::perform for simple clients.Andrew Trick2011-08-101-7/+2
| | | | llvm-svn: 137195
* Fix the LoopUnroller to handle nontrivial loops and partial unrolling.Andrew Trick2011-08-101-52/+60
| | | | | | | | | | These are not individual bug fixes. I had to rewrite a good chunk of the unroller to make it sane. I think it was getting lucky on trivial completely unrolled loops with no early exits. I included some fairly simple unit tests for partial unrolling. I didn't do much stress testing, so it may not be perfect, but should be usable now. llvm-svn: 137190
* LoopUnroll looks like it has some stale code. Remove it to prove my sanity ↵Andrew Trick2011-08-091-6/+1
| | | | | | and avoid further confusion. llvm-svn: 137106
* SCEV: Use AssertingVH to catch dangling BasicBlock* when passes forgetAndrew Trick2011-08-031-2/+9
| | | | | | to notify SCEV of a change. Add forgetLoop in a couple of those places. llvm-svn: 136797
* Add clarifying comments for the new arguments to UnrollLoop.Andrew Trick2011-07-251-0/+12
| | | | llvm-svn: 135988
* Move trip count discovery outside of the generic LoopUnroll helper. ThisAndrew Trick2011-07-231-12/+2
| | | | | | removes its dependence on canonical induction variables. llvm-svn: 135829
* whitespaceAndrew Trick2011-07-231-7/+7
| | | | llvm-svn: 135828
* Reinstate r133513 (reverted in r133700) with an additional fix for aJay Foad2011-06-231-39/+46
| | | | | | -Wshorten-64-to-32 warning in Instructions.h. llvm-svn: 133708
* Revert r133513:Eric Christopher2011-06-231-46/+39
| | | | | | | | | "Reinstate r133435 and r133449 (reverted in r133499) now that the clang self-hosted build failure has been fixed (r133512)." Due to some additional warnings. llvm-svn: 133700
* Remove unused variables.Benjamin Kramer2011-06-211-2/+2
| | | | llvm-svn: 133514
* Reinstate r133435 and r133449 (reverted in r133499) now that the clangJay Foad2011-06-211-32/+39
| | | | | | self-hosted build failure has been fixed (r133512). llvm-svn: 133513
* Revert r133435 and r133449 to appease buildbots.Chad Rosier2011-06-211-39/+32
| | | | llvm-svn: 133499
* Change how PHINodes store their operands.Jay Foad2011-06-201-32/+39
| | | | | | | | | | | | | | | | | | | Change PHINodes to store simple pointers to their incoming basic blocks, instead of full-blown Uses. Note that this loses an optimization in SplitCriticalEdge(), because we can no longer walk the use list of a BasicBlock to find phi nodes. See the comment I removed starting "However, the foreach loop is slow for blocks with lots of predecessors". Extend replaceAllUsesWith() on a BasicBlock to also update any phi nodes in the block's successors. This mimics what would have happened when PHINodes were proper Users of their incoming blocks. (Note that this only works if OldBB->replaceAllUsesWith(NewBB) is called when OldBB still has a terminator instruction, so it still has some successors.) llvm-svn: 133435
* Don't unroll loops whose header block's address is taken.Chris Lattner2011-02-181-2/+10
| | | | | | | | | | | | | | | | | This is part of a futile attempt to not "break" bizzaro code like this: l1: printf("l1: %p\n", &&l1); ++x; if( x < 3 ) goto l1; Previously we'd fold &&l1 to 1, which is fine per our semantics but not helpful to the user. llvm-svn: 125827
* random cleanupsChris Lattner2011-01-111-2/+2
| | | | llvm-svn: 123221
* Remove all uses of the "ugly" method BranchInst::setUnconditionalDest().Jay Foad2011-01-071-1/+3
| | | | llvm-svn: 123025
* Replace calls to ConstantFoldInstruction with calls to SimplifyInstructionDuncan Sands2010-11-231-5/+6
| | | | | | | in two places that are really interested in simplified instructions, not constants. llvm-svn: 120044
* Be more consistent in using ValueToValueMapTy.Rafael Espindola2010-10-131-4/+3
| | | | llvm-svn: 116387
* Preserve ScalarEvolution in the loop unroller.Dan Gohman2010-07-261-0/+6
| | | | llvm-svn: 109412
* Use ValueMap instead of DenseMap.Devang Patel2010-06-241-3/+3
| | | | | | The ValueMapper used by various cloning utility maps MDNodes also. llvm-svn: 106706
* Cosmetic change.Devang Patel2010-06-231-9/+9
| | | | | | Do not use "ValueMap" as a name for a local variable or an argument. llvm-svn: 106698
* Rename ValueMapTy as ValueToValueMapTy to clearly indicate that this has no ↵Devang Patel2010-04-201-4/+4
| | | | | | replationship with ADT/ValueMap. llvm-svn: 101950
* Add a DominatorTree argument to isLCSSA so that it doesn't have toDan Gohman2010-03-101-6/+0
| | | | | | | compute a set of reachable blocks for itself each time it is called, which is fairly frequently. llvm-svn: 98179
* Don't bother with sprintf, just pass the Twine through.Benjamin Kramer2010-01-271-5/+1
| | | | llvm-svn: 94684
* Change errs() to dbgs().David Greene2010-01-051-11/+11
| | | | llvm-svn: 92600
* Add Loop contains utility methods for testing whether a loopDan Gohman2009-12-181-4/+4
| | | | | | | | contains another loop, or an instruction. The loop form is substantially more efficient on large loops than the typical code it replaces. llvm-svn: 91654
* remove a bunch of extraneous LLVMContext argumentsChris Lattner2009-11-061-2/+1
| | | | | | from various APIs, addressing PR5325. llvm-svn: 86231
* Teach LoopUnroll how to bail if LoopSimplify can't give it what it needs.Dan Gohman2009-11-051-1/+12
| | | | llvm-svn: 86164
* Revert r85667. LoopUnroll currently can't call utility functions whichDan Gohman2009-10-311-1/+47
| | | | | | | auto-update the DominatorTree because it doesn't keep the DominatorTree current while it works. llvm-svn: 85670
* Merge the enhancements from LoopUnroll's FoldBlockIntoPredecessor intoDan Gohman2009-10-311-47/+1
| | | | | | | MergeBlockIntoPredecessor. This makes SimplifyCFG slightly more aggressive, and makes it unnecessary for LoopUnroll to have its own copy of this code. llvm-svn: 85667
* Simplify this code.Dan Gohman2009-10-311-2/+2
| | | | llvm-svn: 85662
* Rename UnrollLoop.cpp to LoopUnroll.cpp, and LoopUnroll.cpp toDan Gohman2009-10-311-0/+372
LoopUnrollPass.cpp, for consistency with other passes which are similarly split. llvm-svn: 85659
OpenPOWER on IntegriCloud