tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: SPL NOT ZERO ON SYSCALL ENTRY




On Mar 11, 2008, at 5:46 PM, Bill Stouder-Studenmund wrote:

On Wed, Mar 12, 2008 at 12:46:07AM +0100, Matthias Drochner wrote:

wrstuden%NetBSD.org@localhost said:
My guess is the problem was a missing splx() in an error-handling
case.

That's what I'd also had assumed if the reason was
"not lowered in syscall exit". But this was on entry...

Thinking about it, I think we went with mutexes that hold SPL state. So what might be happening is that we're releasing SPL mutexes (that's not the name, but I'm forgetting what exact term we used) in an order opposite
of the one in which we took them. ??

Except mutexes don't work that way.

spl is only dropped (to the old spl) once *all* mutex all released.
So order of release doesn't matter.


Home | Main Index | Thread Index | Old Index