llvm-6502/lib/Analysis
Chris Lattner df14a04b5c Fix a problem that Nate noticed with LSR:
When inserting code for an addrec expression with a non-unit stride, be
more careful where we insert the multiply.  In particular, insert the multiply
in the outermost loop we can, instead of the requested insertion point.

This allows LSR to notice the mul in the right loop, reducing it when it gets
to it.  This allows it to reduce the multiply, where before it missed it.

This happens quite a bit in the test suite, for example, eliminating 2
multiplies in art, 3 in ammp, 4 in apsi, reducing from 1050 multiplies to
910 muls in galgel (!), from 877 to 859 in applu, and 36 to 30 in bzip2.

This speeds up galgel from 16.45s to 16.01s, applu from 14.21 to 13.94s and
fourinarow from 66.67s to 63.48s.

This implements Transforms/LoopStrengthReduce/nested-reduce.ll


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24102 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-30 06:24:33 +00:00
..
DataStructure don't bother building the archive version of this library 2005-10-24 01:08:20 +00:00
IPA Remove this pass, it is not useful 2005-10-24 02:35:43 +00:00
AliasAnalysis.cpp
AliasAnalysisCounter.cpp
AliasAnalysisEvaluator.cpp
AliasSetTracker.cpp
BasicAliasAnalysis.cpp floor/ceil don't read/write memory. This allows gcse to eliminate 6 calls 2005-08-24 16:58:56 +00:00
CFGPrinter.cpp * Unbreak optimized build (noticed by Eric van Riet Paap) 2005-08-04 14:16:48 +00:00
ConstantFolding.cpp Move some constant folding functions into LLVMAnalysis since they are used 2005-10-27 16:00:10 +00:00
ConstantRange.cpp
InstCount.cpp expose a ctor 2005-10-24 01:00:45 +00:00
Interval.cpp
IntervalPartition.cpp
LoadValueNumbering.cpp
LoopInfo.cpp Add a new getLoopLatch() method. 2005-09-12 17:03:55 +00:00
Makefile DONT_BUILD_RELINKED is gone and implied by BUILD_ARCHIVE now 2005-10-24 02:26:13 +00:00
PostDominators.cpp
ProfileInfo.cpp
ProfileInfoLoader.cpp
ProfileInfoLoaderPass.cpp
ScalarEvolution.cpp Move some constant folding code shared by Analysis and Transform passes 2005-10-27 15:54:34 +00:00
ScalarEvolutionExpander.cpp Fix a problem that Nate noticed with LSR: 2005-10-30 06:24:33 +00:00
Trace.cpp
ValueNumbering.cpp