This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] misalignment and loop peeling (vectorizer)
- From: Devang Patel <dpatel at apple dot com>
- To: Dorit Nuzman <DORIT at il dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Ira Rosen <IRAR at il dot ibm dot com>
- Date: Wed, 1 Feb 2006 10:53:37 -0800
- Subject: Re: [PATCH] misalignment and loop peeling (vectorizer)
- References: <OFC2D9E65E.AA277A9A-ONC2257108.002DEAC9-C2257108.002EC172@il.ibm.com>
On Feb 1, 2006, at 12:30 AM, Dorit Nuzman wrote:
DR_MISALIGNMENT is probably what you want to use - that's where the
misalignment of the first access in the loop is stored, and what
peeling
tries to align (it's -1 if it's unknown at compile time).
IIUC, DR_OFFSET_MISALIGNMENT also represents the same thing. One is
an int value, where as DR_OFFSET_MISALIGNMENT is a tree node. Is
there any other difference between these two? Either way it is fine
for me.
Also, do you have a testcase to commit with this patch?
Yes. The one I included for mainline (vect-align-1.c) is also meant
for autovect-branch.
-
Devang