On Fri, Oct 17, 2008 at 08:36:56PM +0200, Olivier Guilyardi wrote:
> I don't see why it's fragile. There's no reason for the compiler to modify
> rb->read_ptr if not asked to. Buf if that's the case, because of some sort of
> dark voodoo optimization, then we could use a temporary variable to hold the
> intermediary result. Better, all computation could be done on this temporary
> variable and transfered into rb->read_ptr once finished, so that modifying the
> later only involves memory copy.
There is no reason indeed, and only a completely braindead
compiler would do it unless forced to, but it is not formally
excluded unless you make read_ptr volatile. Using a temporary
variable doesn't change the picture at all.
Ciao,
-- FA Laboratorio di Acustica ed Elettroacustica Parma, Italia Lascia la spina, cogli la rosa. _______________________________________________ Linux-audio-user mailing list Linux-audio-user@email-addr-hidden http://lists.linuxaudio.org/mailman/listinfo/linux-audio-userReceived on Sat Oct 18 00:15:05 2008
This archive was generated by hypermail 2.1.8 : Sat Oct 18 2008 - 00:15:06 EEST