From: kanoj@google.engr.sgi.com (Kanoj Sarcar)
To: "Stephen C. Tweedie" <sct@redhat.com>
Cc: andrea@suse.de, torvalds@transmeta.com, linux-mm@kvack.org
Subject: Re: filecache/swapcache questions [RFC] [RFT] [PATCH] kanoj-mm12-2.3.8 Fix swapoff races
Date: Mon, 28 Jun 1999 10:25:45 -0700 (PDT) [thread overview]
Message-ID: <199906281725.KAA72836@google.engr.sgi.com> (raw)
In-Reply-To: <14199.41900.732658.354175@dukat.scot.redhat.com> from "Stephen C. Tweedie" at Jun 28, 99 05:32:44 pm
>
> Hi,
>
> On Sun, 27 Jun 1999 18:48:47 -0700 (PDT), kanoj@google.engr.sgi.com
> (Kanoj Sarcar) said:
>
> > Linus/Andrea/Stephen,
> > This is the patch that tries to cure the swapoff races with processes
> > forking, exiting, and (readahead) swapping by faulting.
>
> > Basically, all these operations are synchronized by the process
> > mmap_sem. Unfortunately, swapoff has to visit all processes, during
> > which it must hold tasklist_lock, a spinlock. Hence, it can not take
> > the mmap_sem, a sleeping mutex.
>
> But it can atomic_inc(&mm->count) to pin the mm, drop the task lock and
> take the mm semaphore, and mmput() once it has finished.
>
Hmm, hadn't thought about that one. Of course, as soon as you drop
the task_lock, in theory, you have to resume your search from the
beginning of the task list, since the list might have changed while
you dropped the task_lock (assume for a moment that the vm code does
not know how the task list is managed). That prevents any forward
progress by swapoff.
I did think of other ways to maintain a hold on the process,
preventing it from forking or exitting, but my judgement was they
were going to be more heavyweight than my current solution.
> > So, the patch links up all active mm's in a list that swapoff can
> > visit
>
> There shouldn't be need for a new data structure. A bit of extra work
> in swapoff should be all that is needed, and that avoids adding any
> extra code at all on the hot paths.
>
> Adding extra locks is the sort of thing other unixes do to solve
> problems like this: we don't want to fall into that trap on Linux. :)
>
Agreed ... if you can come up with a reasonably simple and lightweight
solution without using locks.
Thanks.
Kanoj
kanoj@engr.sgi.com
> --Stephen
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://humbolt.geo.uu.nl/Linux-MM/
next prev parent reply other threads:[~1999-06-28 17:25 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
1999-06-21 5:29 filecache/swapcache questions Kanoj Sarcar
1999-06-21 11:25 ` Stephen C. Tweedie
1999-06-21 16:46 ` Kanoj Sarcar
1999-06-21 16:57 ` Stephen C. Tweedie
1999-06-21 17:36 ` Kanoj Sarcar
1999-06-21 17:49 ` Stephen C. Tweedie
1999-06-21 18:46 ` Kanoj Sarcar
1999-06-21 23:44 ` Kanoj Sarcar
1999-06-24 22:23 ` Andrea Arcangeli
1999-06-24 23:55 ` Kanoj Sarcar
1999-06-25 0:26 ` Andrea Arcangeli
1999-06-28 1:48 ` filecache/swapcache questions [RFC] [RFT] [PATCH] kanoj-mm12-2.3.8 Fix swapoff races Kanoj Sarcar
1999-06-28 10:35 ` Andrea Arcangeli
1999-06-28 17:11 ` filecache/swapcache questions [RFC] [RFT] [PATCH] kanoj-mm12-2.3.8 Kanoj Sarcar
1999-06-28 16:32 ` filecache/swapcache questions [RFC] [RFT] [PATCH] kanoj-mm12-2.3.8 Fix swapoff races Stephen C. Tweedie
1999-06-28 17:25 ` Kanoj Sarcar [this message]
1999-06-28 20:40 ` Stephen C. Tweedie
1999-06-28 21:11 ` Kanoj Sarcar
1999-06-28 22:12 ` Stephen C. Tweedie
1999-06-28 23:43 ` Kanoj Sarcar
1999-06-29 11:44 ` Stephen C. Tweedie
1999-06-29 22:01 ` Kanoj Sarcar
1999-06-30 17:28 ` Stephen C. Tweedie
1999-06-30 18:05 ` Kanoj Sarcar
1999-06-28 19:39 ` Chuck Lever
1999-06-28 19:55 ` filecache/swapcache questions [RFC] [RFT] [PATCH] kanoj-mm12-2.3.8 Kanoj Sarcar
1999-06-28 20:33 ` Chuck Lever
1999-06-28 20:51 ` Kanoj Sarcar
1999-06-28 21:32 ` Chuck Lever
1999-06-28 21:38 ` Kanoj Sarcar
1999-06-28 21:50 ` Chuck Lever
1999-06-28 22:15 ` Kanoj Sarcar
1999-06-29 11:23 ` Stephen C. Tweedie
1999-06-29 17:36 ` Kanoj Sarcar
1999-06-28 22:22 ` Stephen C. Tweedie
1999-06-28 22:21 ` Stephen C. Tweedie
1999-06-28 22:57 ` Andrea Arcangeli
1999-06-29 2:13 ` Chuck Lever
1999-06-29 12:01 ` Stephen C. Tweedie
1999-06-29 12:32 ` Andrea Arcangeli
1999-06-30 15:59 ` Stephen C. Tweedie
1999-06-29 1:00 ` Chuck Lever
1999-06-28 22:08 ` Stephen C. Tweedie
1999-06-28 22:59 ` Andrea Arcangeli
1999-06-29 0:53 ` Chuck Lever
1999-06-29 11:14 ` Stephen C. Tweedie
1999-06-28 22:09 ` Stephen C. Tweedie
1999-06-28 20:45 ` filecache/swapcache questions [RFC] [RFT] [PATCH] kanoj-mm12-2.3.8 Fix swapoff races Stephen C. Tweedie
1999-06-28 21:14 ` Chuck Lever
1999-06-28 21:25 ` filecache/swapcache questions [RFC] [RFT] [PATCH] kanoj-mm12-2.3.8 Kanoj Sarcar
1999-06-28 22:15 ` filecache/swapcache questions [RFC] [RFT] [PATCH] kanoj-mm12-2.3.8 Fix swapoff races Stephen C. Tweedie
1999-06-28 22:48 ` Andrea Arcangeli
1999-06-29 1:29 ` Chuck Lever
1999-06-29 11:58 ` Stephen C. Tweedie
1999-06-29 12:09 ` Andrea Arcangeli
1999-06-29 15:27 ` Chuck Lever
1999-06-29 11:55 ` Stephen C. Tweedie
1999-06-29 20:08 ` Andrea Arcangeli
1999-06-28 22:36 ` filecache/swapcache questions Stephen C. Tweedie
1999-06-28 23:24 ` Kanoj Sarcar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=199906281725.KAA72836@google.engr.sgi.com \
--to=kanoj@google.engr.sgi.com \
--cc=andrea@suse.de \
--cc=linux-mm@kvack.org \
--cc=sct@redhat.com \
--cc=torvalds@transmeta.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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.