This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Fix -m64 on powerpc-darwin for pic code
- From: Zack Weinberg <zack at codesourcery dot com>
- To: Andrew Pinski <pinskia at physics dot uc dot edu>
- Cc: "gcc-patches at gcc dot gnu dot org Patches" <gcc-patches at gcc dot gnu dot org>, shebs at apple dot com
- Date: Mon, 01 Nov 2004 20:42:23 -0800
- Subject: Re: [PATCH] Fix -m64 on powerpc-darwin for pic code
- References: <F2AB3A59-2C87-11D9-8250-000A95D692F4@physics.uc.edu>
Andrew Pinski <firstname.lastname@example.org> writes:
> Even for the following simple code we ICE:
> int i = 0;
> void f(void)
> i = 0;
> This patch fixes the problem plus adds some needed cleanup
> to the darwin code with respect with -m64 and makes it harder
> to make the mistake to generate only the 32bit code.
> Even though this is a semi big patch, I am either making a new
> expand which makes sure that we can just call the non _si/_di
> functions versions of the generation of the instructions or
> the removal of the code which just to call those functions.
The changes look vaguely sensible, but I am not approving the patch,
because you didn't explain it in adequate detail. You need to tell us
what the ICE was, why it happened, and why this is the correct fix.