http://llvm.org/bugs/show_bug.cgi?id=17678
Bug ID: 17678
Summary: [Vectorizer] Implement non-interleaved stride
vectorization
Product: libraries
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: Loop Optimizer
Assignee: [email protected]
Reporter: [email protected]
CC: [email protected]
Classification: Unclassified
After non-unit stride is implemented, make sure more complex cases, such as:
for (i..N/3) {
a[3*i] = b[3*i] + K;
a[3*i+1] = b[3*i+1] + K;
a[3*i+2] = b[3*i+2] + K;
}
with the exact same operations on all "lanes" can be vectorized, thus not
requiring interleaving of loads/stores.
--
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
