On Mon, 3 Jan 2005 17:05:25 -0500 (EST), Alan Stern <[EMAIL PROTECTED]> wrote:
> But first you should read my reply to Pete, and you should see how the > driver is affected. You will be hard-pressed to find any spots where the > "volatile" causes abominable code generation. I saw your reply, and did not want to argue, because in the narrow case you're presenting you're right. I just do not trust anyone else to use that volatile correctly once you turn away for a second. Its effects are non-local. That is, SMP is not an issue now, but then someone rearranges locking and it becomes an issue, when this causes breakage in unrelated part of UHCI. We're balancing my fears of the future against your fears of the future. I am not afraid of anyone forgetting to use an accessor, especially if you rename the field now, for example. And also, someone will see that volatile in UHCI code and will think it's a good idea in general without going into the analysis you did for the narrow case. That's one of my concerns. -- Pete ------------------------------------------------------- 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
