RPCSX/llvm
1
0
Fork 0
mirror of https://github.com/RPCSX/llvm.git synced 2024-12-04 18:06:49 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

A couple new README entries.

Browse Source 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127786 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Eli Friedman 2011-03-17 01:22:09 +00:00
parent f21b1058a1
commit 3d831381a9
1 changed files with 36 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
lib/Target/X86/README.txt
View File

@ -1947,3 +1947,39 @@ which is "perfect".
//===---------------------------------------------------------------------===//
For the branch in the following code:
int a();
int b(int x, int y) {
if (x & (1<<(y&7)))
return a();
return y;
}
We currently generate:
movb %sil, %al
andb $7, %al
movzbl %al, %eax
btl %eax, %edi
jae .LBB0_2
movl+andl would be shorter than the movb+andb+movzbl sequence.
//===---------------------------------------------------------------------===//
For the following:
struct u1 {
float x, y;
};
float foo(struct u1 u) {
return u.x + u.y;
}
We currently generate:
movdqa %xmm0, %xmm1
pshufd $1, %xmm0, %xmm0 # xmm0 = xmm0[1,0,0,0]
addss %xmm1, %xmm0
ret
We could save an instruction here by commuting the addss.
//===---------------------------------------------------------------------===//