On Tue, Jul 22, 2008 at 6:32 AM, Peter Teoh <htmldeveloper@xxxxxxxxx> wrote:
>> Also, how can I know by looking at the kernel code if a function is
>> preemptible or not ?
>>
>
> Generally, when u see the function: cond_resched(), or many other
> similar function (eg, schedule()), or any form of mutex usage (which
> will call schedule()) then it mean that CPU can be reschedule to
> execute other tasks - which means that the current tasks can be
> blocked.
>
> So in particular for you:
>
> This happened in memory.c:
>
> cond_resched();
> while (!(page = follow_page(vma, start, foll_flags))) {
> int ret;
> ret = handle_mm_fault(mm, vma, start,
> foll_flags & FOLL_WRITE);
> if (ret & VM_FAULT_ERROR) {
> if (ret & VM_FAULT_OOM)
> return i ? i : -ENOMEM;
> else if (ret
>
> where handle_mm_fault() will call handle_pte_fault(), so outside of
> handle_mm_fault(), it is potentially possible to be blocked.
>
Peter, thanks for your detailed response. I looked for the piece of
code you pasted above in memory.c (2.6.26) and I found it in
"get_user_pages()" which in turn is called by many other functions in
memory.c
So I assume that when a major fault happens, somehow get_user_pages()
gets called, is that correct ?
Basically what I need is, when a page gets swapped in, I want to find
out its corresponding process. i.e. which process needed to have that
page swapped in. There seems to be no direct way to do that as far as
I can see. Even the mm_struct doesn't have a mapping back to the
task_struct though there is something defined in 2.6.26 using
CONFIG_MM_OWNER, but I'm not sure if that is correct since it says it
points to the
"canonical user/owner of this mm". Here is the portion of mm_struct :
228 #ifdef CONFIG_MM_OWNER
229 /*
230 * "owner" points to a task that is regarded as the canonical
231 * user/owner of this mm. All of the following must be true in
232 * order for it to be changed:
233 *
234 * current == mm->owner
235 * current->mm != mm
236 * new_owner->mm == mm
237 * new_owner->alloc_lock is held
238 */
239 struct task_struct *owner;
240 #endif
Anyways, this is in 2.6.26 only.
In a nutshell, I want to be able to identify the process that needs
(caused) the swap in of a page from the swap space. How can I do it ?
Thanks
-arn
--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ
