This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug target/77468] [7 Regression] C-ray regression on Aarch64
- From: "pinskia at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Mon, 05 Sep 2016 08:18:15 +0000
- Subject: [Bug target/77468] [7 Regression] C-ray regression on Aarch64
- Auto-submitted: auto-generated
- References: <bug-77468-4@http.gcc.gnu.org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77468
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Component|middle-end |target
--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
This sounds like the generic tuning is wrong.
Can you try -mcpu=cortex-a53 and/or -mcpu=cortex-a57. One thing I noticed
about the generic tuning is the cost model for the vectorizer is just bogus and
needs a lot of tuning. I noticed this when I was comparing GCC 5 and GCC 6.
Also can you try out GCC 6 to see if an issue there too (I bet it is).