llvm-project/llvm/lib/Transforms
Michael Zolotukhin 73957179d3 [LoopUnrolling] Try harder to avoid rebuilding LCSSA when possible.
In r255133 (reapplied r253126) we started to avoid redundant
recomputation of LCSSA after loop-unrolling. This patch moves one step
further in this direction - now we can avoid it for much wider range of
loops, as we start to look at IR and try to figure out if the
transformation actually breaks LCSSA phis or makes it necessary to
insert new ones.

Differential Revision: http://reviews.llvm.org/D16838

llvm-svn: 259869
2016-02-05 02:17:36 +00:00
..
Hello Remove autoconf support 2016-01-26 21:29:08 +00:00
IPO LowerBitSets: Don't bother to do any work if the llvm.bitset.test intrinsic is unused. 2016-02-03 03:48:46 +00:00
InstCombine [InstCombine] Revert r238452: Fold IntToPtr and PtrToInt into preceding loads. 2016-02-03 18:04:13 +00:00
Instrumentation [InstrProfiling] Fix a comment (NFC) 2016-02-03 23:22:43 +00:00
ObjCARC Add support for objc_unsafeClaimAutoreleasedReturnValue to the 2016-01-27 19:05:08 +00:00
Scalar [RS4GC] Pass DenseMap by reference, NFC 2016-02-05 01:42:52 +00:00
Utils [LoopUnrolling] Try harder to avoid rebuilding LCSSA when possible. 2016-02-05 02:17:36 +00:00
Vectorize [SCEV] Try to reuse existing value during SCEV expansion 2016-02-04 01:27:38 +00:00
CMakeLists.txt
LLVMBuild.txt