diff options
| author | Evan Cheng <evan.cheng@apple.com> | 2006-02-28 23:38:49 +0000 |
|---|---|---|
| committer | Evan Cheng <evan.cheng@apple.com> | 2006-02-28 23:38:49 +0000 |
| commit | 0e69f45b07346f876ef8600a8ee4170c5b54a3dd (patch) | |
| tree | 4aa45a6a6caaf0db943cffe6df31aa5a4ecdeb25 /llvm/lib | |
| parent | 990c3602bd35ac52fcda1c2b9d34c19b793bd57c (diff) | |
| download | bcm5719-llvm-0e69f45b07346f876ef8600a8ee4170c5b54a3dd.tar.gz bcm5719-llvm-0e69f45b07346f876ef8600a8ee4170c5b54a3dd.zip | |
Another entry.
llvm-svn: 26430
Diffstat (limited to 'llvm/lib')
| -rw-r--r-- | llvm/lib/Target/X86/README.txt | 16 |
1 files changed, 16 insertions, 0 deletions
diff --git a/llvm/lib/Target/X86/README.txt b/llvm/lib/Target/X86/README.txt index 6c9a235ebbb..c85477c1a68 100644 --- a/llvm/lib/Target/X86/README.txt +++ b/llvm/lib/Target/X86/README.txt @@ -517,3 +517,19 @@ It would be better to emit "cmp %al, 1" than a xor and test. //===---------------------------------------------------------------------===// Enable X86InstrInfo::convertToThreeAddress(). + +//===---------------------------------------------------------------------===// + +Investigate whether it is better to codegen the following + + %tmp.1 = mul int %x, 9 +to + + movl 4(%esp), %eax + leal (%eax,%eax,8), %eax + +as opposed to what llc is currently generating: + + imull $9, 4(%esp), %eax + +Currently the load folding imull has a higher complexity than the LEA32 pattern. |
