This is the mail archive of the
mailing list for the GCC project.
Re: [patch] Resubmission in an attempt to get someones attention
- From: Janis Johnson <janis187 at us dot ibm dot com>
- To: Steve Ellcey <sje at cup dot hp dot com>
- Cc: janis187 at us dot ibm dot com, gcc-patches at gcc dot gnu dot org
- Date: Thu, 6 Dec 2001 14:15:38 -0800
- Subject: Re: [patch] Resubmission in an attempt to get someones attention
- References: <200112062116.NAA20686@hpsje.cup.hp.com>
I had forgotten that it only affected interaction with another compiler;
I should have looked at your original message again.
On Thu, Dec 06, 2001 at 01:16:20PM -0800, Steve Ellcey wrote:
> I am not quite sure how to write a test for this. Tests that involve
> just GCC will work before and after the changes because while the
> parameter passing/returns was done "wrong" with respect to the ABI they
> were done consistently within GCC. It is in the interaction between GCC
> and HP C that the problem arises but I don't believe that GCC supports
> that type of test involving GCC and the system compiler.
> I suppose I could write a simple structure passing / returning test that
> at least makes sure GCC still works in general and that nothing broke,
> but it wouldn't be testing that this change had the intended affect. Do
> you think that would be useful?
> Steve Ellcey
> > If someone (Jim?) approves this I'll volunteer to commit the changes
> > after running a bootstrap and test run on ia64-linux.
> > This probably deserves at least one new test, even if it only affects
> > ia64-hpux.
> > Janis