ARM scheduling fix: compute predicated implicit use properly.
authorAndrew Trick <atrick@apple.com>
Fri, 22 Jun 2012 02:50:31 +0000 (02:50 +0000)
committerAndrew Trick <atrick@apple.com>
Fri, 22 Jun 2012 02:50:31 +0000 (02:50 +0000)
commitef2d9e59aba381c42e018df9c26f9025c1995a64
treefa88fc14ef313b27f36a9f50257d2ac321e30759
parentffccd923101ae177c1a2c407618d8f03a312d69b
ARM scheduling fix: compute predicated implicit use properly.

Minor drive by fix to cleanup latency computation. Calling
getOperandLatency with a deliberately incorrect operand index does not
give you the latency you want.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@158959 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/ARM/ARMBaseInstrInfo.cpp
test/CodeGen/ARM/neon_div.ll