From: Daniel Walker on
On Mon, 2010-05-24 at 13:31 -0700, Michel Lespinasse wrote:
> This helps in the following situation:
> - Thread A takes a page fault while reading or writing memory.
> do_page_fault() acquires the mmap_sem for read and blocks on disk
> (either reading the page from file, or hitting swap) for a long time.
> - Thread B does an mmap call and blocks trying to acquire the mmap_sem
> for write
> - Thread C is a monitoring process trying to read every /proc/pid/maps
> in the system. This requires acquiring the mmap_sem for read. Thread C
> blocks behind B, waiting for A to release the rwsem. If thread C
> could be allowed to run in parallel with A, it would probably get done
> long before thread A's disk access completes, thus not actually slowing
> down thread B.
>
> Test results with down_read_critical_test (10 seconds):

Did you try using queuing on scheduling priorities in some way to do
this? It seems like your setting it up so Thread C is more important
than Thread B, but your using code to dictate that instead of scheduling
priorities. It would make more sense to me if the threads had priorities
to dictate what's "critical" and what's not.

Daniel

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Arjan van de Ven on
On Mon, 24 May 2010 13:31:21 -0700
Michel Lespinasse <walken(a)google.com> wrote:

> This helps in the following situation:
> - Thread A takes a page fault while reading or writing memory.
> do_page_fault() acquires the mmap_sem for read and blocks on disk
> (either reading the page from file, or hitting swap) for a long
> time.
> - Thread B does an mmap call and blocks trying to acquire the mmap_sem
> for write
> - Thread C is a monitoring process trying to read every /proc/pid/maps
> in the system. This requires acquiring the mmap_sem for read.
> Thread C blocks behind B, waiting for A to release the rwsem. If
> thread C could be allowed to run in parallel with A, it would
> probably get done long before thread A's disk access completes, thus
> not actually slowing down thread B.
>
> Test results with down_read_critical_test (10 seconds):


this is a really bad idea btw

we've had many issues in the past, when this was an unfair lock, that
"top" or other similar things, caused basically a DoS......
now any process that can get to /proc/<pid>/exe or maps, can do this in
a tight enough loop so that the actual process will never get the lock
for write. BAD IDEA ;-)


--
Arjan van de Ven Intel Open Source Technology Centre
For development, discussion and tips for power savings,
visit http://www.lesswatts.org
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/