From: Andrew Morton <akpm@zip.com.au>
To: "David S. Miller" <davem@redhat.com>
Cc: bcrl@redhat.com, hugh@veritas.com, marcelo@conectiva.com.br,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] __free_pages_ok oops
Date: Wed, 06 Feb 2002 23:07:35 -0800 [thread overview]
Message-ID: <3C6227B7.37BFA2EC@zip.com.au> (raw)
In-Reply-To: <3C621C74.8A005EB6@zip.com.au>, <3C6214F0.A66C89CF@zip.com.au> <20020206.215539.33252283.davem@redhat.com> <3C621C74.8A005EB6@zip.com.au> <20020206.224940.122062252.davem@redhat.com>
"David S. Miller" wrote:
>
> From: Andrew Morton <akpm@zip.com.au>
> Date: Wed, 06 Feb 2002 22:19:32 -0800
>
> It's only a problem if this is the final put_page(). In the
> case of sendfile(), process-context code can be taught to take
> a temporary reference on the page, and only release it after the network
> stack is known to have finished with the page. sendfile is synchronous, yes?
>
> Userspace can return long before the SKBs are free'd up. That SKB
> free doesn't happen until the ACKs come back from the receiver.
I feel that presence on the lru list should contribute to
page->count. It seems a bit weird and kludgy that this
is not so.
If we were to do this then would this not fix networking's
problem? The skb free wouldn't release the page - it would
be left on the LRU with ->count == 1 and kswapd would reap it.
(Says me, hoping that Hugh will code it :))
-
next prev parent reply other threads:[~2002-02-07 7:08 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-02-06 19:06 [PATCH] __free_pages_ok oops Hugh Dickins
2002-02-06 19:47 ` Andrew Morton
2002-02-06 20:15 ` Hugh Dickins
2002-02-06 21:11 ` Andrew Morton
2002-02-07 20:31 ` Manfred Spraul
2002-02-07 5:09 ` Benjamin LaHaise
2002-02-07 5:47 ` Andrew Morton
2002-02-07 5:55 ` David S. Miller
2002-02-07 6:19 ` Andrew Morton
2002-02-07 6:49 ` David S. Miller
2002-02-07 7:07 ` Andrew Morton [this message]
2002-02-07 11:52 ` Hugh Dickins
2002-02-07 12:34 ` Rik van Riel
2002-02-07 12:37 ` David S. Miller
2002-02-07 12:44 ` Rik van Riel
2002-02-07 13:19 ` Hugh Dickins
2002-02-07 13:27 ` Rik van Riel
2002-02-07 13:55 ` Daniel Phillips
2002-02-07 14:28 ` Hugh Dickins
2002-02-07 14:56 ` Rik van Riel
2002-02-07 20:21 ` Hugh Dickins
2002-02-07 20:58 ` Andrea Arcangeli
2002-02-07 21:09 ` Andrew Morton
2002-02-07 22:18 ` Andrea Arcangeli
2002-02-07 22:31 ` Andrew Morton
2002-02-07 23:09 ` Andrea Arcangeli
2002-02-07 23:27 ` Andrew Morton
2002-02-08 17:46 ` Hugh Dickins
2002-02-09 14:14 ` Gerd Knorr
2002-02-09 15:47 ` arjan
2002-02-09 14:33 ` Benjamin LaHaise
2002-02-12 20:19 ` Hugh Dickins
2002-02-13 18:52 ` Marcelo Tosatti
2002-02-14 10:47 ` Hugh Dickins
2002-02-14 11:10 ` Gerd Knorr
2002-02-14 13:10 ` Andrea Arcangeli
2002-02-14 14:01 ` Hugh Dickins
2002-02-14 15:17 ` Andrea Arcangeli
2002-02-14 16:27 ` Linus Torvalds
2002-02-25 18:32 ` Benjamin LaHaise
2002-02-25 19:35 ` Linus Torvalds
2002-02-07 9:48 ` Benjamin LaHaise
-- strict thread matches above, loose matches on Subject: below --
2002-02-09 8:52 alad
2002-02-09 10:46 ` Hugh Dickins
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=3C6227B7.37BFA2EC@zip.com.au \
--to=akpm@zip.com.au \
--cc=bcrl@redhat.com \
--cc=davem@redhat.com \
--cc=hugh@veritas.com \
--cc=linux-kernel@vger.kernel.org \
--cc=marcelo@conectiva.com.br \
/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.