http://llvm.org/bugs/show_bug.cgi?id=16592
Bug ID: 16592
Summary: loop-vectorizer regression on x86-32
Product: new-bugs
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: new bugs
Assignee: [email protected]
Reporter: [email protected]
CC: [email protected]
Classification: Unclassified
Created attachment 10856
--> http://llvm.org/bugs/attachment.cgi?id=10856&action=edit
IR that used to loop-vectorize
This
void bar(float *A, float* B, float K) {
int i;
for (i = 0; i < 64; ++i)
A[i] *= B[i] + K;
}
used to vectorize with dragonegg on x86-32 (with -O3 -msse), but stopped
vectorizing recently. Note that it still vectorizes with clang, I think
because clang outputs inbounds GEPs for the array accesses while dragonegg
doesn't.
It still vectorizes on x86-64, but it's not clear to me why it is better on 64
bit.
IR attached.
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
LLVMbugs mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs
