Re: [GIT PATCH] USB autosuspend fixes for 2.6.23-rc6

From: Adrian Bunk
Date: Thu Sep 13 2007 - 18:05:46 EST


On Thu, Sep 13, 2007 at 11:28:42PM +0200, Adrian Bunk wrote:
> On Thu, Sep 13, 2007 at 01:44:23PM -0700, Linus Torvalds wrote:
> >
> >
> > On Thu, 13 Sep 2007, Adrian Bunk wrote:
> > >
> > > No, what I'm concerned about is that this would require userspace for
> > > something that is completely in-kernel.
> >
> > If done right (and autosuspend now is), there is no "required" userspace.
> >
> > If you want autosuspend, you just say so. The kernel doesn't do it by
> > default. This is not about "user space required" - it's about "user space
> > can ask for it if it wants to".
> >
> > Notice? There doesn't even have to be any blacklists/whitelists at all. It
> > really can be just an application that allows the user to check or uncheck
> > the capability (with a warning saying something like: "Some USB devices
> > may disconnect when suspended - if this affects you, uncheck this").
> >
> > That's why the kernel shouldn't set policy. It's a *good* thing to just
> > expose the capabilities, but not necessarily use them!
>
> What is not policy is the blacklist or whitelist information.
>
> And I'm also a bit concerned why "is policy" is that much a reason
> against setting *reasonable default policies* without requiring the user
> to do various things in userspace.
>
> Especially since this creates some nasty interdependencies between the
> kernel and userspace.
>
> And as an example, couldn't you equally say it's wrong that the kernel
> enables DMA on disks instead of leaving it to userspace?
>
> We've already seen the udev disaster where upgrading from Debian 3.1 to
> Debian 4.0 means upgrading from kernel 2.6.8 to 2.6.18 with the udev
> version in Debian 3.1 not supporting kernel 2.6.18 and the udev version
> in Debian 4.0 not supporting kernel 2.6.8, and I don't have a good
> feeling about outsourcing more and more things to userspace tools not
> distributed with the kernel.

Let me paraphrase the latter:

Given a distribution shipping with kernel 2.6.23 released in 2007.

What is the maximum amount of userspace I might have to upgrade
if I'll want to use kernel 2.6.43 released in 2011 (sic) with this
distribution?

E.g. when looking at the reverse dependencies of libhal, it would not be
funny if kernel 2.6.43 required a more recent version of HAL.

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/