projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: e2245ba) | patch
Allow x86 subtargets to use the GenericModel defined in X86Schedule.td.
authorAndrew Trick <atrick@apple.com>
Tue, 7 Aug 2012 00:25:30 +0000 (00:25 +0000)
committerAndrew Trick <atrick@apple.com>
Tue, 7 Aug 2012 00:25:30 +0000 (00:25 +0000)
commitc42a7017864fc62bfff36c1b8b7f4cd68e198861
treee7aba7708b1dacbac2d9f1884aac661d29b65667tree | snapshot
parente2245bab3ca29dc2142d8f254005f4ae7c40cde2commit | diff
Allow x86 subtargets to use the GenericModel defined in X86Schedule.td.

This allows codegen passes to query properties like
InstrItins->SchedModel->IssueWidth. It also ensure's that
computeOperandLatency returns the X86 defaults for loads and "high
latency ops". This should have no significant impact on existing
schedulers because X86 defaults happen to be the same as global
defaults.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@161370 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86Subtarget.cpp diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors