On Tue, Jan 3, 2012 at 4:37 PM, Diego Biurrun <[email protected]> wrote:
> On Tue, Jan 03, 2012 at 04:19:11PM -0800, Jason Garrett-Glaser wrote:
>> On Tue, Jan 3, 2012 at 3:58 PM, Diego Biurrun <[email protected]> wrote:
>> > On Tue, Jan 03, 2012 at 03:35:19PM -0800, Jason Garrett-Glaser wrote:
>> >> On Tue, Jan 3, 2012 at 3:32 PM, Diego Biurrun <[email protected]> wrote:
>> >> > On Tue, Jan 03, 2012 at 03:26:30PM -0800, Jason Garrett-Glaser wrote:
>> >> >> On Tue, Jan 3, 2012 at 2:29 PM, Diego Biurrun <[email protected]> wrote:
>> >> >> > This fixes compilation failures related to START_TIMER/STOP_TIMER 
>> >> >> > macros and
>> >> >> > -Werror=declaration-after-statement.  START_TIMER declares variables 
>> >> >> > and thus
>> >> >> > may not be placed after statements outside of a new block.
>> >> >>
>> >> >> START/STOP_TIMER are for development purposes only.  There is
>> >> >> absolutely no reason to make this sort of change for something that
>> >> >> should never be enabled except during development.
>> >> >
>> >> > Except that one cannot use it for development right now, because
>> >> > it will not compile?  How can development code be exempt from
>> >> > compilation requirements?
>> >>
>> >> -Werror=declaration-after-statement sounds like an extremely bad
>> >> option to enable during development, since placing strict rules on
>> >> variable declarations is just about the antithesis of hacking around
>> >> wildly with code.  START_TIMER in particular is dramatically more
>> >> obnoxious to use with that option.
>> >
>> > You may well have a point here, but -Werror=declaration-after-statement
>> > is now set by configure.  So without the braces I add or a change in
>> > the macros, compilation is currently failing.
>>
>> A better solution would be to revert the addition of such an insane
>> default compilation flag.
>
> Feel free to go and flame for such a thing, but please let me add two
> pairs of braces to fix compilation in the meantime.

Patch okay then.

Jason
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to