On Monday 03 January 2005 1:06 pm, Alan Cox wrote: > > As the maintainer, I decided to go > > in favor of simplicity and ease-of-maintenance. > > Thats unfortunate because we know from years of history its always been > the wrong decision, that older gcc gets volatile wrong and that > processor speculation which is not covered by volatile keywords still > eats you alive on PPC platforms.
I'm certainly more comfortable with explicit barriers myself, and only use "volatile" with hardware register access (where the standard readl/writel macros use it for me, anyway). Alan, could you explain that bit about speculative execution on PPC? (That's what I assume you meant.) Maybe a simple example would help make it clearer why "volatile" isn't the preferred solution. - Dave ------------------------------------------------------- The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
