On Friday 08 November 2002 13:58, Jens Axboe wrote:

> Here's a patch that includes that feature, puts the tunables in sysfs
> (so you obviously need that mounted). In
>
> /sys/block/<disk>/iosched

Stupid question time:

A great deal of text has been expended over the years by people desperately 
trying to make sure you didn't need /proc mounted to have a usable system, 
for some definition of usable.  Now with rootfs, initramfs, sysfs, and the 
libfs inspired "make a filesystem rather than an ioctl" policy, the main 
argument against requiring the use of /proc is that it has a lot more gunk in 
it (left over from the days when it was the only ramfs type system to export 
values in) than anyone is comfortable with.  (The argument against /dev/pty 
largely seems to be inertia, now that the "number of ptys" issue as a config 
tunable seems to have been cleared up).

There seems to be some sort of nebulous plan for eventually stripping down 
/proc, perhaps making a "crapfs" that's a union mount on top of /proc 
providing deprecated legacy support for a release or two.  But I haven't 
heard it explicitly stated.

So my questions are:

1) will some subset of /proc, /sys, /dev/pty, etc become required at some 
point in the future on everything but the most customized embedded systems?  
Or is keeping the system usable without them still a goal?

2) Is there a plan to rehabilitate /proc?

(I ask because I don't know.  Maybe I missed some important posts...)

Rob

-- 
http://penguicon.sf.net - Terry Pratchett, Eric Raymond, Pete Abrams, Illiad, 
CmdrTaco, liquid nitrogen ice cream, and caffienated jello.  Well why not?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/