lkml.org 
[lkml]   [2000]   [Mar]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: new IRQ scalability changes in 2.3.48
On Fri, 3 Mar 2000, Ingo Molnar wrote:

> yes, the local APIC queues pending interrupts, and these pending requests
> serve as a 'focus' for subsequent interrupts, just in case they happen. I
> have not actually seen any big effect from this, but it does make sense
> conceptually - if the IRQ rate is very high. Having said that, for
> level-triggered IRQs this probably has no effect (because no new IRQs are
> sent for a not yet acked in-service or pending IRQ), but if a

There is one issue here, although it does not happen within Linux,
seemingly. If two different sources (e.g. two redirection entries of an
I/O APIC) issue a level triggered interrupt of the same vector in an
overlapping manner, due to the focus feature they will both be delivered
to the same CPU. However the first remote IRQ source deactivation will
remove the IRQ request from the local APIC (i.e. the relevant IRR bit gets
cleared) regardless of the state of the other IRQ source.

--
+ Maciej W. Rozycki, Technical University of Gdansk, Poland +
+--------------------------------------------------------------+
+ e-mail: macro@ds2.pg.gda.pl, PGP key available +


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:56    [W:0.099 / U:1.484 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site