From: Kalle Valo <kalle.valo@iki.fi>
To: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: linux-kernel@vger.kernel.org, linux-wireless@vger.kernel.org,
linux-omap@vger.kernel.org, Helge Deller <deller@gmx.de>,
rusty@rustcorp.com.au, akpm@linux-foundation.org,
roland@redhat.com, dave@hiauly1.hia.nrc.ca,
Parisc List <linux-parisc@vger.kernel.org>
Subject: Re: regression: crash from 'ls /sys/modules/wl1251_spi/notes'
Date: Wed, 30 Dec 2009 19:20:26 +0200 [thread overview]
Message-ID: <87d41wwf8l.fsf@purkki.valot.fi> (raw)
In-Reply-To: <1262188157.2749.21.camel@mulgrave.site> (James Bottomley's message of "Wed\, 30 Dec 2009 09\:49\:17 -0600")
James Bottomley <James.Bottomley@HansenPartnership.com> writes:
> On Wed, 2009-12-30 at 13:41 +0200, Kalle Valo wrote:
>
>> Also removing wl1251_spi causes a crash. The reason for this is that a
>> sysfs file with a null string as name is trying to be removed from the
>> notes directory.
>
> Yes, this is because the notes sections describe the text sections.
> When we ignore an empty text section, we'd need to ignore its
> corresponding notes section.
>
> The reason we didn't see this on parisc is because our compiler doesn't
> actually produce any notes sections.
Also on omap not all modules can reproduce the problem. For example
cfg80211, mac80211 and wl1251 modules work just fine, but with
wl1251_spi I see the problem everytime.
> A better, and more comprehensive patch would be to try not to count the
> empty text sections when we're building the notes section (and actually
> anywhere else in the file). This patch actually relies on the fact that
> if sh_size is zero for the text section it should be for the
> corresponding notes section. If that doesn't work, we'd actually have
> to do the matching in the construction piece.
>
> Can you try it to see if it works for you? If it doesn't, I'll try
> matching notes to text. It works fine on parisc, but as we don't have a
> notes section, that's not saying much ...
Thanks. I tested the patch on my setup, this time on top of Linus'
tree (commit 6b7b284). And the patch fixes the issue, I can't
reproduce the problem at all, both 'rmmod wl1251_spi' and ls work
without any problems. So here's my tested-by:
Tested-by: Kalle Valo <kalle.valo@iki.fi>
Also please consider sending the fix to stable.
Thank you for fixing this so fast.
--
Kalle Valo
next prev parent reply other threads:[~2009-12-30 17:20 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-12-30 11:41 regression: crash from 'ls /sys/modules/wl1251_spi/notes' Kalle Valo
2009-12-30 15:49 ` James Bottomley
2009-12-30 17:20 ` Kalle Valo [this message]
2009-12-31 21:15 ` Helge Deller
2010-01-06 1:15 ` Andrew Morton
2010-01-01 3:08 ` Américo Wang
2010-01-02 16:34 ` James Bottomley
2010-01-04 18:23 ` Roland McGrath
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=87d41wwf8l.fsf@purkki.valot.fi \
--to=kalle.valo@iki.fi \
--cc=James.Bottomley@HansenPartnership.com \
--cc=akpm@linux-foundation.org \
--cc=dave@hiauly1.hia.nrc.ca \
--cc=deller@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-omap@vger.kernel.org \
--cc=linux-parisc@vger.kernel.org \
--cc=linux-wireless@vger.kernel.org \
--cc=roland@redhat.com \
--cc=rusty@rustcorp.com.au \
/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).