... > doing a TLB flush (I'm particularly thinking of those that > need to use IPI shootdown for the ... so that we don't > hold any locks while doing a TLB flush? ... two architectures which can sleep in their tlb flush handlers. -- Intel are signing ...

linux-kernel - Matthew Wilcox - Mar 26 2008 - 08:32

... holding an anon_vma_lock, the semaphore happened to sleep because some other processor was ... doing a TLB flush (I'm particularly thinking of those that need to ... implement some sort of spin_only_semaphore? -Tony -- To unsubscribe from this list: send the ...

linux-kernel - Luck, Tony - Mar 25 2008 - 16:49

... wrote: > 2) Is it feasible to rearrange the MM code so ... any locks while doing a TLB flush? Or should I implement ... holding any spinlocks. There would need to be additional measures for ... are called under pte lock. -- To unsubscribe from this list: send ...

linux-kernel - Christoph Lameter - Mar 26 2008 - 15:25

... wrote: > > > 2) Is it feasible to rearrange the MM code so ... any locks while doing a TLB flush? Or should I implement ... when you need a counting sempahore. Thanks, tglx -- To unsubscribe from ... in the body of a message to majordomo@vger.kernel.org ...

linux-kernel - Thomas Gleixner - Mar 26 2008 - 16:29

... large hole for real time to fall into. If you want ... do rt nicely you want to avoid semaphores and the corresponding ... any locks while doing a TLB flush? Or should I implement ... some sort of spin_only_semaphore? Better to keep ia64 perversions in the ...

linux-kernel - Alan Cox - Mar 25 2008 - 17:47

... between semaphores and rwsems > > rwsem is not a semaphore despite its name? What ... a sleeping rw lock might be a better name as opposed to the ... rw-mutex :-) But lets just call it a rwsem; we all know what that is. -- To unsubscribe from this list ...

linux-kernel - Peter Zijlstra - Mar 28 2008 - 05:59

... For those places that need semaphore semantics but cannot sleep Signed-off-by: ... (struct semaphore *sem); /* + * Attempt to acquire the semaphore. If another task is ... .c index bef977b..d3eb559 100644 --- a/kernel/semaphore.c +++ b/kernel/ ...

linux-kernel - Luck, Tony - Mar 26 2008 - 16:29

... those places that need semaphore semantics but cannot sleep > > Signed-off- ... semaphore *sem); > > /* > + * Attempt to acquire the semaphore. If another task is ... c > index bef977b..d3eb559 100644 > --- a/kernel/semaphore.c > +++ b/kernel ...

linux-kernel - Jens Axboe - Mar 27 2008 - 04:09

... > confusion between semaphores and rwsems rwsem is not a semaphore despite its name? What do you want to call it then? -- To unsubscribe from this list: ... -kernel" in the body of a message to majordomo@vger.kernel.org More majordomo ...

linux-kernel - Christoph Lameter - Mar 27 2008 - 14:44

... semaphores should only be used > when you need a counting sempahore. Seems that mutexes are ... concurrently. This means processors need to be able to concurrently walk potentially long lists of vmas. -- To unsubscribe from this list: send the line ...

linux-kernel - Christoph Lameter - Mar 26 2008 - 21:19

... should only be used > > when you need a counting sempahore. > > Seems that mutexes ... . This means > processors need to be able to concurrently walk potentially ... . confusion between semaphores and rwsems -- To unsubscribe from this list: send ...

linux-kernel - Peter Zijlstra - Mar 27 2008 - 09:20