diff options
| author | Evan Cheng <evan.cheng@apple.com> | 2006-03-29 03:03:46 +0000 |
|---|---|---|
| committer | Evan Cheng <evan.cheng@apple.com> | 2006-03-29 03:03:46 +0000 |
| commit | c999c745c0e94b15c14e5dca44d23c572b89f9ef (patch) | |
| tree | f764fdfbd613812ac2d3007c25b92119f7f03fef | |
| parent | 4f5633883b2e9ba187885e09e81564a53e549d60 (diff) | |
Another entry about shuffles.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27245 91177308-0d34-0410-b5e6-96231b3b80d8
| -rw-r--r-- | lib/Target/X86/README.txt | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/lib/Target/X86/README.txt b/lib/Target/X86/README.txt index 09d9bb6ef8..060552b93d 100644 --- a/lib/Target/X86/README.txt +++ b/lib/Target/X86/README.txt @@ -667,3 +667,9 @@ directly %esp[0] if there are no other uses. Use movhps to update upper 64-bits of a v4sf value. Also movlps on lower half of a v4sf value. + +//===---------------------------------------------------------------------===// + +Better codegen for vector_shuffles like this { x, 0, 0, 0 } or { x, 0, x, 0}. +Perhaps use pxor / xorp* to clear a XMM register first? + |