From: Steffen Klassert <steffen.klassert@secunet.com>
To: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Alexey Dobriyan <adobriyan@gmail.com>,
linux-crypto@vger.kernel.org, netdev@vger.kernel.org,
ken@codelabs.ch
Subject: Re: sha512: make it work, undo percpu message schedule
Date: Fri, 13 Jan 2012 07:22:56 +0100 [thread overview]
Message-ID: <20120113062256.GC12501@secunet.com> (raw)
In-Reply-To: <20120111003611.GA12257@gondor.apana.org.au>
On Wed, Jan 11, 2012 at 11:36:11AM +1100, Herbert Xu wrote:
> On Wed, Jan 11, 2012 at 03:00:40AM +0300, Alexey Dobriyan wrote:
> > commit f9e2bca6c22d75a289a349f869701214d63b5060
> > aka "crypto: sha512 - Move message schedule W[80] to static percpu area"
> > created global message schedule area.
> >
> > If sha512_update will ever be entered twice, hilarity ensures.
>
> Hmm, do you know why this happens? On the face of it this shouldn't
> be possible as preemption is disabled.
>
I did not try to reproduce, but this looks like a race of the 'local out'
and the receive packet path. On 'lokal out' bottom halves are enabled,
so could be interrupted by the NET_RX_SOFTIRQ while doing a sha512_update.
The NET_RX_SOFTIRQ could invoke sha512_update too, that would corrupt the
hash value. My guess could be checked easily by disabling the bottom halves
before the percpu value is fetched.
next prev parent reply other threads:[~2012-01-13 6:22 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-11 0:00 sha512: make it work, undo percpu message schedule Alexey Dobriyan
2012-01-11 0:12 ` Alexey Dobriyan
2012-01-11 0:36 ` Herbert Xu
2012-01-12 23:55 ` Alexey Dobriyan
2012-01-13 0:19 ` Herbert Xu
2012-01-13 7:08 ` Herbert Xu
2012-01-13 10:35 ` Eric Dumazet
2012-01-13 10:41 ` Eric Dumazet
2012-01-13 10:57 ` Eric Dumazet
2012-01-13 11:33 ` Alexey Dobriyan
2012-01-13 12:34 ` Eric Dumazet
2012-01-14 18:20 ` Alexey Dobriyan
2012-01-14 18:27 ` [PATCH 1/3] " Alexey Dobriyan
2012-01-14 18:40 ` [PATCH 2/3] sha512: reduce stack usage to safe number Alexey Dobriyan
2012-01-14 19:08 ` Linus Torvalds
2012-01-14 20:41 ` Alexey Dobriyan
2012-01-14 21:14 ` Linus Torvalds
2012-01-16 9:56 ` David Laight
2012-01-16 10:20 ` Alexey Dobriyan
2012-01-16 10:23 ` Eric Dumazet
2012-01-16 11:37 ` David Laight
2012-01-17 12:03 ` Alexey Dobriyan
2012-01-18 18:02 ` [PATCH 4/3] sha512: reduce stack usage even on i386 Alexey Dobriyan
2012-01-26 2:35 ` Herbert Xu
2012-01-27 17:51 ` Alexey Dobriyan
2012-01-27 22:32 ` Herbert Xu
2012-01-30 11:10 ` Alexey Dobriyan
2012-02-03 3:34 ` Herbert Xu
2012-01-14 21:46 ` [PATCH 1/3] sha512: make it work, undo percpu message schedule Eric Dumazet
2012-01-14 21:52 ` Linus Torvalds
2012-01-14 22:00 ` Eric Dumazet
2012-01-15 1:43 ` Herbert Xu
2012-01-13 11:02 ` Steffen Klassert
2012-01-15 1:43 ` Herbert Xu
2012-01-13 11:45 ` David Laight
2012-01-13 12:35 ` Eric Dumazet
2012-01-13 6:22 ` Steffen Klassert [this message]
2012-01-13 6:46 ` Herbert Xu
2012-01-13 6:48 ` Eric Dumazet
2012-01-13 6:50 ` Herbert Xu
2012-01-13 9:45 ` David Laight
2012-01-11 1:12 ` Adrian-Ken Rueegsegger
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=20120113062256.GC12501@secunet.com \
--to=steffen.klassert@secunet.com \
--cc=adobriyan@gmail.com \
--cc=herbert@gondor.apana.org.au \
--cc=ken@codelabs.ch \
--cc=linux-crypto@vger.kernel.org \
--cc=netdev@vger.kernel.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).