From: Arnd Bergmann <arnd@arndb•de>
To: Rob Herring <robherring2@gmail•com>
Cc: "Eric Dumazet" <eric.dumazet@gmail•com>,
linux-arm-kernel@lists•infradead.org,
"Russell King - ARM Linux" <linux@arm•linux.org.uk>,
"Jon Masters" <jonathan@jonmasters•org>,
netdev@vger•kernel.org, "Måns Rullgård" <mans@mansr•com>,
"David Laight" <David.Laight@aculab•com>
Subject: Re: alignment faults in 3.6
Date: Thu, 11 Oct 2012 13:35:42 +0000 [thread overview]
Message-ID: <201210111335.43177.arnd@arndb.de> (raw)
In-Reply-To: <5076C78E.1020408@gmail.com>
On Thursday 11 October 2012, Rob Herring wrote:
>
> On 10/11/2012 07:40 AM, Eric Dumazet wrote:
> > On Thu, 2012-10-11 at 12:28 +0000, Arnd Bergmann wrote:
> >
> >>
> >> Rob Herring as the original reporter has dropped off the Cc list, adding
> >> him back.
> >>
> >> I assume that the calxeda xgmac driver is the culprit then. It uses
> >> netdev_alloc_skb() rather than netdev_alloc_skb_ip_align() in
> >> xgmac_rx_refill but it is not clear whether it does so intentionally
> >> or by accident.
>
> This in fact does work and eliminates the unaligned traps. However, not
> all h/w can do IP aligned DMA (i.MX FEC for example), so I still think
> this is a questionable optimization by the compiler. We're saving 1 load
> instruction here for data that is likely already in the cache. It may be
> legal per the ABI, but the downside of this optimization is much greater
> than the upside.
>
What about the other approach that Eric suggested for such hardware
in http://www.spinics.net/lists/arm-kernel/msg200206.html ?
Arnd
next prev parent reply other threads:[~2012-10-11 13:35 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <506E1762.3010601@gmail.com>
[not found] ` <CANLjY-no-L9=LRCvCetC5XCHicOoSWOfRZmUd7itAiUHWBhAyw@mail.gmail.com>
[not found] ` <CAG5Tg6Wuybruc31x1eh6mFafsg=7cOgYLgpZUBuJUo75ap3osQ@mail.gmail.com>
[not found] ` <506E3E58.80703@gmail.com>
[not found] ` <CAG5Tg6V0Vj9HTy3=AU-q+Qa3DuJs6R1Kw+3uOWU_jOGfkgyXzw@mail.gmail.com>
[not found] ` <20121005071216.GD4625@n2100.arm.linux.org.uk>
[not found] ` <CAG5Tg6VciuxJ=k+fatzLugcehF54mGNOLHpUwdMX99oCnAVg8Q@mail.gmail.com>
[not found] ` <20121005082439.GF4625@n2100.arm.linux.org.uk>
[not found] ` <506ED18C.3010009@gmail.com>
[not found] ` <20121005140556.GQ4625@n2100.arm.linux.org.uk>
[not found] ` <506EEFBB.3060705@gmail.com>
[not found] ` <507619FA.6080001@jonmasters.org>
2012-10-11 2:27 ` alignment faults in 3.6 Måns Rullgård
2012-10-11 2:34 ` Jon Masters
2012-10-11 8:21 ` David Laight
2012-10-11 8:53 ` Russell King - ARM Linux
2012-10-11 9:45 ` Måns Rullgård
2012-10-11 10:00 ` Eric Dumazet
2012-10-11 10:20 ` Måns Rullgård
2012-10-11 10:22 ` Eric Dumazet
2012-10-11 10:32 ` Russell King - ARM Linux
2012-10-11 10:49 ` Eric Dumazet
2012-10-11 10:56 ` Maxime Bizon
2012-10-11 11:28 ` Eric Dumazet
2012-10-11 11:47 ` Maxime Bizon
2012-10-11 11:54 ` Eric Dumazet
2012-10-11 12:00 ` Eric Dumazet
2012-10-11 12:51 ` Maxime Bizon
2012-10-11 12:59 ` Eric Dumazet
2012-10-11 12:28 ` Arnd Bergmann
2012-10-11 12:40 ` Eric Dumazet
2012-10-11 13:20 ` Rob Herring
2012-10-11 13:32 ` Måns Rullgård
2012-10-11 13:35 ` Arnd Bergmann [this message]
2012-10-11 13:47 ` Eric Dumazet
2012-10-11 15:23 ` Rob Herring
2012-10-11 15:39 ` David Laight
2012-10-11 16:18 ` Måns Rullgård
2012-10-12 8:11 ` Arnd Bergmann
2012-10-12 9:03 ` Russell King - ARM Linux
2012-10-12 10:04 ` Eric Dumazet
2012-10-12 12:24 ` Russell King - ARM Linux
2012-10-12 11:00 ` Måns Rullgård
2012-10-12 11:07 ` Russell King - ARM Linux
2012-10-12 11:18 ` Måns Rullgård
2012-10-12 11:44 ` Russell King - ARM Linux
2012-10-12 12:08 ` Eric Dumazet
2012-10-12 14:22 ` Benjamin LaHaise
2012-10-12 14:36 ` David Laight
2012-10-12 14:48 ` Eric Dumazet
2012-10-12 15:00 ` Benjamin LaHaise
2012-10-12 15:04 ` Ben Hutchings
2012-10-12 15:47 ` David Laight
2012-10-12 16:13 ` Ben Hutchings
2012-10-12 12:16 ` Måns Rullgård
2012-10-12 11:19 ` Russell King - ARM Linux
2012-10-11 16:15 ` Eric Dumazet
2012-10-11 16:59 ` Catalin Marinas
2012-10-11 10:16 ` David Laight
2012-10-11 10:46 ` Måns Rullgård
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=201210111335.43177.arnd@arndb.de \
--to=arnd@arndb$(echo .)de \
--cc=David.Laight@aculab$(echo .)com \
--cc=eric.dumazet@gmail$(echo .)com \
--cc=jonathan@jonmasters$(echo .)org \
--cc=linux-arm-kernel@lists$(echo .)infradead.org \
--cc=linux@arm$(echo .)linux.org.uk \
--cc=mans@mansr$(echo .)com \
--cc=netdev@vger$(echo .)kernel.org \
--cc=robherring2@gmail$(echo .)com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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