Messages in this thread | | | Date | Wed, 4 Oct 2000 20:59:27 -0300 (BRST) | From | Rik van Riel <> | Subject | Re: Weightless process class |
| |
On Thu, 5 Oct 2000, Matti Aarnio wrote: > On Wed, Oct 04, 2000 at 06:16:57PM -0300, Rik van Riel wrote:
[priority inversion] > > We don't need that. > > > > We just need one boolean per thread ... is it holding a kernel > > lock or not? > > The BKL or *any* (kernel) lock ? > > For my knowledge there is no limitation on how many > locks a thread can hold. Having a single bool might > not be enough. A counter is better ?
On return_to_userspace, you /know/ the thread has released the kernel locks.
Though this might not work for IPC semaphores...
Then again, if the Montavista people (maybe together with SGI or IBM folks?) can get a light-weight priority inversion scheme working, then we'll have everything right ;)
> > If it is, make sure its scheduling latency isn't too high. > > e.g. all processes having *any* locks are raised to the highest > possible class to make sure they are not starved out ?
SCHED_OTHER is enough for most purposes. OTOH, you could have one SCHED_FIFO process block execution of a higher priority SCHED_FIFO process that wants a lock ... priority inversion anyway?
> > If it isn't holding any lock, we can do with it what we want, > > including completely starving the task for several seconds > > (or even minutes) if scheduling latency or VM pressure warrants > > it. > > Yes, that is obvious.
*grin*
regards,
Rik -- "What you're running that piece of shit Gnome?!?!" -- Miguel de Icaza, UKUUG 2000
http://www.conectiva.com/ http://www.surriel.com/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
| |