On Wed, Feb 25, 2009 at 11:51 AM, Matthew Garrett <mjg59@xxxxxxxxxxxxx> wrote: > On Wed, Feb 25, 2009 at 04:18:25PM +0000, Jonathan Buzzard wrote: > >> Because it makes far more sense to have a user mode program to drive the >> features, and for the kernel to provide the necessary thin layer to >> access the features. > > It makes sense for the kernel to provide a consistent abstraction of > hardware functionality where possible. Almost every kernel driver could > be rewritten in userspace - that doesn't make it a good idea. That argument would be most applicable if we were talking about adding some new /dev interface-- but we're not. Toshiba's HCI has been around for a couple of decades; /dev/toshiba, which exposes it, has been around for over 10 years; toshset, a useful and popular tool which depends on /dev/toshiba, nearly as long as /dev/toshiba. The proposed patch makes /dev/toshiba, and hence toshset, work on more laptops-- it's a no-brainer to anyone with a practical mindset. You've got Jonathan and Charles-- the two most-knowledgeable people on Toshiba's firmware and /dev/toshiba in the free software community-- saying that it's not practical to jam all that functionality into the kernel. Please listen to them. Regards, --John -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
