This is the mail archive of the
mailing list for the GCC project.
Re: improve unaligned loads with no special insns
- From: Richard Henderson <rth at redhat dot com>
- To: Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Sat, 6 Nov 2004 12:22:41 -0800
- Subject: Re: improve unaligned loads with no special insns
- References: <10411062012.AA21639@vlsi1.ultra.nyu.edu>
On Sat, Nov 06, 2004 at 03:12:59PM -0500, Richard Kenner wrote:
> Can you give an example where it matters?
Not off the top of my head, but given that you're the one advocating
lying to the compiler, which basically always comes back to bite you,
I think the onus should be on you to prove that it doesn't.
> I'm concerned about pessimizing the code by having the alias set zero:
> it could well be faster to have better-scheduled multiple loads...
I doubt it. The difference in the amount of code is really startling.
Plus we'll have done the bulk dce and cse at the tree level before this