Re: [PATCH] drivers/char/vt possible race

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Tue, 2004-02-10 at 15:34, Andrew Morton wrote:
> Benjamin Herrenschmidt <benh@kernel.crashing.org> wrote:
> >
> > Hi !
> > 
> > I falled again on the crash in con_do_write() with driver->data
> > beeing NULL. It happens during boot, when userland is playing
> > open/close games with tty's, I was intentionally typing keys like
> > mad during boot trying to trigger another problem when this one
> > poped up.
> 
> OK.  Was this patch confirmed to prevent any reoccurrences?

Well, I didn't see it again, and if for some reason, we still
enter the function with tty->driver_data == NULL (which may still
happen if the tty layer itself isn't serializing, which I suspect),
we will print a warning and bail out.

In the end, I suppose the warning can be removed, but I want to
make sure that if the race still happens, we behave properly now.

The patch makes sure the vt internal state stays consistent.

> > Andrew: I suggest putting that in -mm for a while, and if it
> > doesn't trigger any new problem, upstream, maybe without my
> > 2 printk's "argh" :)
> 
> Yup.  I'll also bring back the sysfs patch which somehow triggers
> this race.

Yup, let me know.

Ben.