diff options
| author | Evan Cheng <evan.cheng@apple.com> | 2006-04-18 01:22:57 +0000 |
|---|---|---|
| committer | Evan Cheng <evan.cheng@apple.com> | 2006-04-18 01:22:57 +0000 |
| commit | 74e955d9312f4d0dcc9c2022e58c71341afce1f4 (patch) | |
| tree | f1ea2a57cda2a97f6f31f9bca3bf2e78564e9089 | |
| parent | 309db81c107d72a4762e781d44b07f7fc830c055 (diff) | |
Another entry
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27786 91177308-0d34-0410-b5e6-96231b3b80d8
| -rw-r--r-- | lib/Target/X86/README.txt | 35 |
1 files changed, 35 insertions, 0 deletions
diff --git a/lib/Target/X86/README.txt b/lib/Target/X86/README.txt index 33be39ee91..39a4407f8b 100644 --- a/lib/Target/X86/README.txt +++ b/lib/Target/X86/README.txt @@ -961,3 +961,38 @@ selector should select pshufd or The register allocator can made the two-address to three-address transformation. It also exposes some other problems. See MOV32ri -3 and the spills. + +//===---------------------------------------------------------------------===// + +http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25500 + +LLVM is producing bad code. + +LBB_main_4: # cond_true44 + addps %xmm1, %xmm2 + subps %xmm3, %xmm2 + movaps (%ecx), %xmm4 + movaps %xmm2, %xmm1 + addps %xmm4, %xmm1 + addl $16, %ecx + incl %edx + cmpl $262144, %edx + movaps %xmm3, %xmm2 + movaps %xmm4, %xmm3 + jne LBB_main_4 # cond_true44 + +There are two problems. 1) No need to two loop induction variables. We can +compare against 262144 * 16. 2) Poor register allocation decisions. We should +be able eliminate one of the movaps: + + addps %xmm1, %xmm2 + subps %xmm3, %xmm2 + movaps (%ecx), %xmm4 + movaps %xmm2, %xmm2 <=== Eliminate! + addps %xmm4, %xmm2 + addl $16, %ecx + incl %edx + cmpl $262144, %edx + movaps %xmm3, %xmm1 + movaps %xmm4, %xmm3 + jne LBB_main_4 # cond_true44 |