This is the mail archive of the
mailing list for the GCC project.
Re: FAIL: gcc.c-torture/execute/950628-1.c execution, -O1
- To: Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>
- Subject: Re: FAIL: gcc.c-torture/execute/950628-1.c execution, -O1
- From: Richard Henderson <rth at redhat dot com>
- Date: Fri, 26 Jan 2001 13:59:56 -0800
- Cc: gcc at gcc dot gnu dot org
- References: <10101262148.AA14555@vlsi1.ultra.nyu.edu>
On Fri, Jan 26, 2001 at 04:47:55PM -0500, Richard Kenner wrote:
> I think the former needs more discussion, but I'd say that for 3.0
> the *safest* approach is to "allow" multiple stores by changing the
> remaining dependence function. That's both the smallest and safest change.
I don't agree, mostly because of the useage for function invariance
I mentioned before. IMO the *safest* change is not to combine /u
slots. I have a patch for objects_must_conflict_p to do that, but
have been waiting to see how the discussion resolves before doing
anything with it.
I'd also advocate removing the non-source-language /u bits from your
Dec 30 change from the branch, but leave them in mainline.