http://llvm.org/bugs/show_bug.cgi?id=21296
Bug ID: 21296
Summary: Performance regression on ARM within (219545, 219569]
Product: new-bugs
Version: unspecified
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: new bugs
Assignee: [email protected]
Reporter: [email protected]
CC: [email protected]
Classification: Unclassified
Our benchmark buildbot detected a regression on ARM:
http://llvm.org/perf/db_default/v4/nts/graph?plot.0=49.128.2&highlight_run=31861
According to Hal:
Based on the commit logs, it does seem like the best candidates are:
r219566, r219567, r219568 - InstCombine bug fixes re: no-wrap flags
r219550, r219562 - Fixes to trip-count computations in the
vectorizer/unroller
and given the magnitude of the change, I think that the trip-count changes are
more likely.
Since those were bug fixes, we should proceed with caution and make sure there
isn't a fix on top of those patches before considering anything more radical.
It may very well be that the ARM back-end is not dealing with some new patterns
correctly, or that the transformation has generated unoptimal sequences.
--
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
