diff options
| author | Andrew Trick <atrick@apple.com> | 2012-06-26 04:11:38 +0000 |
|---|---|---|
| committer | Andrew Trick <atrick@apple.com> | 2012-06-26 04:11:38 +0000 |
| commit | c9b1e25493b393013b28e5d457f2fb2845a4dd9f (patch) | |
| tree | c1193bd83841e829e703beeb00c1fe6415de5f05 /test/CodeGen/X86/multiple-loop-post-inc.ll | |
| parent | 5ac3f96c0e4a1f6b8253aabf74fe30b0439e9bdf (diff) | |
| download | external_llvm-c9b1e25493b393013b28e5d457f2fb2845a4dd9f.zip external_llvm-c9b1e25493b393013b28e5d457f2fb2845a4dd9f.tar.gz external_llvm-c9b1e25493b393013b28e5d457f2fb2845a4dd9f.tar.bz2 | |
Enable the new LoopInfo algorithm by default.
The primary advantage is that loop optimizations will be applied in a
stable order. This helps debugging and unit test creation. It is also
a better overall implementation without pathologically bad performance
on deep functions.
On large functions (llvm-stress --size=200000 | opt -loops)
Before: 0.1263s
After: 0.0225s
On deep functions (after tweaking llvm-stress, thanks Nadav):
Before: 0.2281s
After: 0.0227s
See r158790 for more comments.
The loop tree is now consistently generated in forward order, but loop
passes are applied in reverse order over the program. If we have a
loop optimization that prefers forward order, that can easily be
achieved by adding a different type of LoopPassManager.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@159183 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86/multiple-loop-post-inc.ll')
| -rw-r--r-- | test/CodeGen/X86/multiple-loop-post-inc.ll | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/test/CodeGen/X86/multiple-loop-post-inc.ll b/test/CodeGen/X86/multiple-loop-post-inc.ll index 7491190..9f7d036 100644 --- a/test/CodeGen/X86/multiple-loop-post-inc.ll +++ b/test/CodeGen/X86/multiple-loop-post-inc.ll @@ -1,9 +1,9 @@ ; RUN: llc -asm-verbose=false -disable-branch-fold -disable-code-place -disable-tail-duplicate -march=x86-64 -mcpu=nehalem < %s | FileCheck %s ; rdar://7236213 - -; Xfailed now that scheduler 2-address hack is disabled a lea is generated. -; The code isn't any worse though. -; XFAIL: * +; +; The scheduler's 2-address hack has been disabled, so there is +; currently no good guarantee that this test will pass until the +; machine scheduler develops an equivalent heuristic. ; CodeGen shouldn't require any lea instructions inside the marked loop. ; It should properly set up post-increment uses and do coalescing for |