Re: manipulating sigmask from filesystems and drivers

[Olivier Galibert]

On Thu, Aug 01, 2002 at 04:30:40PM -0700, Linus Torvalds wrote:
> And yes, these logging programs are mission-critical, and they do have
> signals going on, and they rely on well-defined and documented interfaces
> that say that doing a write() to a filesystem is _not_ going to return in
> the middle just because a signal came in.

How hard and/or insane would it be to somehow special-case SIGKILL?
It is a tad annoying not to be able to get rid of D state processes,
especially ones blocking unmounts because the filesystem is busy.

Of course, an alternative is a real, brutal, forced unmount that
leaves a clean filesystem and dead/dying processes.

  OG.