This is the mail archive of the
mailing list for the GCC project.
Re: Deprecating basic asm in a function - What now?
- From: Jakub Jelinek <jakub at redhat dot com>
- To: Bernd Edlinger <bernd dot edlinger at hotmail dot de>
- Cc: Andrew Haley <aph at redhat dot com>, Jeff Law <law at redhat dot com>, Michael Matz <matz at suse dot de>, "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>, David Wohlferd <dw at LimeGreenSocks dot com>
- Date: Tue, 21 Jun 2016 21:31:00 +0200
- Subject: Re: Deprecating basic asm in a function - What now?
- Authentication-results: sourceware.org; auth=none
- References: <AM4PR0701MB216259CD17DFCAA81D39ED79E42B0 at AM4PR0701MB2162 dot eurprd07 dot prod dot outlook dot com>
- Reply-to: Jakub Jelinek <jakub at redhat dot com>
On Tue, Jun 21, 2016 at 06:29:45PM +0000, Bernd Edlinger wrote:
> I do not like the idea to deprecate the basic asm at all, I must admit,
> but I think if we added a warning, that just contains a positive information, like
> "warning: basic asm semantic has been changed to implicitly clobber memory,
> if you have a problem with that, please convert this asm statement to extended
> asm syntax."
I don't see any point of such a warning.
And, I agree that deprecation of basic asm is a bad idea.