llvm-svn: 26021

llvm-svn: 26019

llvm-svn: 26018

llvm-svn: 26017

llvm-svn: 26016

llvm-svn: 26015

- Manually select SDOperand's returned by TryFoldLoad which make up the
  load address.

llvm-svn: 26012

llvm-svn: 26010

llvm-svn: 26009

llvm-svn: 26007

llvm-svn: 26006

llvm-svn: 26005

llvm-svn: 26004

Turn A / (C1 << N), where C1 is "1<<C2" into A >> (N+C2) [udiv only].

Tested with: rem.ll:test5, div.ll:test10

llvm-svn: 26003

llvm-svn: 26001

llvm-svn: 26000

llvm-svn: 25998

llvm-svn: 25997

llvm-svn: 25992

llvm-svn: 25991

* Fix hasNUsesOfValue(), it should be const.

llvm-svn: 25990

llvm-svn: 25985

llvm-svn: 25984

llvm-svn: 25983

llvm-svn: 25981

llvm-svn: 25980

This patch speeds up 172.mgrid from 31.81s to 11.39s on darwin/ppc.
Many many thanks to Nate for tracking down the root cause of the issue.

llvm-svn: 25979

llvm-svn: 25978

llvm-svn: 25976

llvm-svn: 25975

#LLVM LOC, and auto-cse's cast instructions.

llvm-svn: 25974

llvm-svn: 25973

tblgen patch reverted.

llvm-svn: 25971

llvm-svn: 25969

Select them afterwards if it returns true.

llvm-svn: 25968

case, the double being loaded may not be 8-byte aligned, so we have to use
our standard bit_convert game.

llvm-svn: 25967

llvm-svn: 25966

llvm-svn: 25965

1. When rewriting code in outer loops, sometimes we would insert code into
   inner loops that is invariant in that loop.
2. Notice that 4*(2+x) is 8+4*x and use that to simplify expressions.

This is a performance neutral change.

llvm-svn: 25964

llvm-svn: 25963