diff options
author | Nadav Rotem <nrotem@apple.com> | 2013-04-12 21:15:03 +0000 |
---|---|---|
committer | Nadav Rotem <nrotem@apple.com> | 2013-04-12 21:15:03 +0000 |
commit | 9eb366acba65b5779d2129db3a6fb6a0414572d4 (patch) | |
tree | 993baaba80d0870be37cf9e98091f3a2066b339c /test/Analysis | |
parent | 196ee11f85ce0148d2c2e33fbe1f1171ac5a8828 (diff) |
CostModel: increase the default cost of supported floating point operations from 1 to two. Fixed a few tests that changes because now the cost of one insert + a vector operation on two doubles is lower than two scalar operations on doubles.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179413 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/Analysis')
-rw-r--r-- | test/Analysis/CostModel/X86/arith.ll | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/test/Analysis/CostModel/X86/arith.ll b/test/Analysis/CostModel/X86/arith.ll index 85b442533f..92f5a1ec3a 100644 --- a/test/Analysis/CostModel/X86/arith.ll +++ b/test/Analysis/CostModel/X86/arith.ll @@ -66,9 +66,9 @@ define void @avx2mull() { ; CHECK: fmul define i32 @fmul(i32 %arg) { - ;CHECK: cost of 1 {{.*}} fmul + ;CHECK: cost of 2 {{.*}} fmul %A = fmul <4 x float> undef, undef - ;CHECK: cost of 1 {{.*}} fmul + ;CHECK: cost of 2 {{.*}} fmul %B = fmul <8 x float> undef, undef ret i32 undef } |