This is the mail archive of the
mailing list for the GCC project.
Re: Patch for stricter implicit conversions between vectors
- From: Mark Mitchell <mark at codesourcery dot com>
- To: Andrew Pinski <pinskia at gmail dot com>
- Cc: Ian Lance Taylor <iant at google dot com>, Mark Shinwell <shinwell at codesourcery dot com>, gcc patches <gcc-patches at gcc dot gnu dot org>
- Date: Thu, 30 Nov 2006 23:46:10 -0800
- Subject: Re: Patch for stricter implicit conversions between vectors
- References: <4559B33A.email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com> <455A1A22.firstname.lastname@example.org> <email@example.com>
Andrew Pinski wrote:
>> >>> We do have such flags. For example, -fpermissive and
>> >>> -ffriend-injection.
>> > I disagree. I appreciate your message, but it doesn't add any new
>> > facts, and doesn't change my feelings on this matter. I won't bother
>> > to reply in detail, since it would add nothing to the conversation.
>> I agree with you that the flag is consistent with our precedent, and
>> that we should keep it.
> So again I don't think an option is useful for this case, it just
> allows for sloppy code.
When we remove support for even clearly non-conforming code, we get
complaints. Apple and AMD have publicly complained about this (citing
complaints from their users). Certainly, CodeSourcery has had to
explain various issues like this to our direct and indirect customers.
RMS has even complained to the SC about some cases.
It's clearly a tough call. From our perspective as developers, options
are bad; just remove the support and move on. But, for users, that's
painful. So, we have to decide these things case by case.
I think in this case we should have the option. Like Ian, I don't
really feel there's more to be usefully said. Unless a clear consensus
arises that the option is unnecessary, let's put it in.
(650) 331-3385 x713