From: Andrew Morton <akpm@linux-foundation.org>
To: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Chris Snook <csnook@redhat.com>,
Dave Jones <davej@codemonkey.org.uk>,
Linux Kernel <linux-kernel@vger.kernel.org>,
netdev@vger.kernel.org, Peter Zijlstra <peterz@infradead.org>
Subject: Re: GFP_ATOMIC page allocation failures.
Date: Wed, 2 Apr 2008 22:32:54 -0700 [thread overview]
Message-ID: <20080402223254.5ac50337.akpm@linux-foundation.org> (raw)
In-Reply-To: <200804031622.26973.nickpiggin@yahoo.com.au>
On Thu, 3 Apr 2008 16:22:26 +1100 Nick Piggin <nickpiggin@yahoo.com.au> wrote:
> > It also tells us when we mucked up the net driver - I doubt if we (or at
> > least, I) would have discovered that e1000 does a 32k allocation for a
> > 5k(?) frame if this warning wasn't coming out.
>
> Is that right? If it is allocating for 9K MTU, then the slab allocator
> (slub in this case) will bump that up to the 16K kmalloc slab. If it is
> a 5K frame, then it would get the 8K kmalloc slab I think.
>
> Oh, but SLUB's default MIN_OBJECTS is 4, so 4*8 is 32 indeed. So slub
> is probably deciding to round the kmalloc-8192 allocations up to order-3.
> I think. How did you know it was a 5k frame? :)
urgh, it was a while ago, and I don't know if e1000e retains the behaviour.
iirc the issue was with some errant versions of the hardware needing
exorbitant alignment and additional padding at the end because of
occasional DMA overruns. Something like that.
next prev parent reply other threads:[~2008-04-03 5:33 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20080401235609.GA6947@codemonkey.org.uk>
[not found] ` <200804021228.16875.nickpiggin@yahoo.com.au>
[not found] ` <47F32789.2070703@redhat.com>
[not found] ` <20080402005646.f8df1c1b.akpm@linux-foundation.org>
2008-04-02 8:17 ` GFP_ATOMIC page allocation failures KOSAKI Motohiro
2008-04-02 8:24 ` David Miller
2008-04-02 8:43 ` Andrew Morton
2008-04-02 10:00 ` KOSAKI Motohiro
2008-04-02 10:56 ` KOSAKI Motohiro
2008-04-02 18:44 ` David Miller
2008-04-02 20:12 ` Michael Chan
2008-04-02 20:53 ` David Miller
2008-04-02 11:04 ` Peter Zijlstra
2008-04-02 18:45 ` David Miller
2008-04-02 19:06 ` Jeff Garzik
[not found] ` <200804022012.58760.nickpiggin@yahoo.com.au>
2008-04-02 15:54 ` Andrew Morton
2008-04-03 5:22 ` Nick Piggin
2008-04-03 5:32 ` Andrew Morton [this message]
2008-04-03 8:59 ` Evgeniy Polyakov
2008-06-26 21:06 ` Dave Jones
2008-06-26 22:26 ` Chris Snook
2008-06-27 10:01 ` Evgeniy Polyakov
2008-04-02 17:21 ` Jeff Garzik
2008-04-02 17:33 ` Andrew Morton
2008-04-02 18:18 ` Jeff Garzik
2008-04-02 18:37 ` Kok, Auke
2008-04-03 5:57 ` Nick Piggin
2008-04-03 18:20 ` Jeff Garzik
[not found] <adYyJ-20N-11@gated-at.bofh.it>
[not found] ` <aef6w-6rx-45@gated-at.bofh.it>
[not found] ` <aefJ9-7KR-15@gated-at.bofh.it>
[not found] ` <aeqF6-45P-29@gated-at.bofh.it>
2008-04-04 9:52 ` Bodo Eggert
2008-04-04 10:59 ` Nick Piggin
2008-04-04 11:35 ` Bodo Eggert
2008-04-05 1:06 ` Nick Piggin
2008-04-06 12:12 ` Bodo Eggert
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=20080402223254.5ac50337.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=csnook@redhat.com \
--cc=davej@codemonkey.org.uk \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=nickpiggin@yahoo.com.au \
--cc=peterz@infradead.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).