All of lore.kernel.org
 help / color / mirror / Atom feed
* paging in realtime context ...
@ 2013-06-28 14:14 Lars Segerlund
  2013-06-28 15:37 ` Stanislav Meduna
  0 siblings, 1 reply; 2+ messages in thread
From: Lars Segerlund @ 2013-06-28 14:14 UTC (permalink / raw)
  To: linux-rt-users

 Quick question, if I mlock my app and prefault the stacks, what
happens if I haven't linked static and use a library ?

 Surely a bunch of pagefaults ?

 If not my thinking is broken, which might a possibility ... I just
wanted to make sure since I really don't trust logical conclusions in
this case.

 / regards, Lars Segerlund.

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: paging in realtime context ...
  2013-06-28 14:14 paging in realtime context Lars Segerlund
@ 2013-06-28 15:37 ` Stanislav Meduna
  0 siblings, 0 replies; 2+ messages in thread
From: Stanislav Meduna @ 2013-06-28 15:37 UTC (permalink / raw)
  To: Lars Segerlund, linux-rt-users

On 28.06.2013 16:14, Lars Segerlund wrote:

>  Quick question, if I mlock my app and prefault the stacks, what
> happens if I haven't linked static and use a library ?
> 
>  Surely a bunch of pagefaults ?

I am no kernel expert but as I understand it should not matter
whether the application is linked statically or dynamically.
What is mapped gets locked.

After mlockall returns you should only get pagefaults if the
mapping changes (heap grows, you call mmap, dlopen()
a library, start a new thread, ...).

-- 
                                            Stano


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2013-06-28 15:37 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-06-28 14:14 paging in realtime context Lars Segerlund
2013-06-28 15:37 ` Stanislav Meduna

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.