From: Thomas Gleixner <tglx@linutronix.de>
To: "Brandeburg, Jesse" <jesse.brandeburg@intel.com>
Cc: Olaf Kirch <okir@suse.de>, Jiri Kosina <jkosina@suse.cz>,
linux-kernel@vger.kernel.org, linux-netdev@vger.kernel.org,
kkeil@suse.de, agospoda@redhat.com, arjan@linux.intel.com,
"Graham, David" <david.graham@intel.com>,
"Allan, Bruce W" <bruce.w.allan@intel.com>,
"Ronciak, John" <john.ronciak@intel.com>,
chris.jones@canonical.com, tim.gardner@intel.com,
airlied@gmail.com
Subject: RE: [RFC PATCH 07/12] e1000e: debug contention on NVM SWFLAG
Date: Thu, 2 Oct 2008 20:02:30 +0200 (CEST) [thread overview]
Message-ID: <alpine.LFD.1.10.0810021959030.5549@apollo.tec.linutronix.de> (raw)
In-Reply-To: <36D9DB17C6DE9E40B059440DB8D95F52064FB25C@orsmsx418.amr.corp.intel.com>
On Thu, 2 Oct 2008, Brandeburg, Jesse wrote:
> Olaf Kirch wrote:
> > Looks like the e1000 watchdog racing with some dhclient activity
> > (upping the interface).
>
> > I just noticed that the driver actually uses register pages. So it
> > looks like it's possible to have something like this without the
> > mutex:
> >
> > process A selects page A
> > process B selects page B
> > process A writes to register at offset A'
>
> I think that is possible, which is why the mutex patch would be good for
> the future. However we have not shown that to be happening as a root
> cause, but I don't rule it out.
Nevertheless I vote strongly for putting that check in _NOW_. It has
proven that there is concurrent access and that's definitely a bug by
all means.
> so, why now? Drivers since before the e1000/e1000e split had this same
> code, with no reports of problems. This code has been heavily tested,
> and one of the platforms easily reproducing this has been available for
> 3 years now (ich8), with code that is basically unchanged in the driver.
Well, timing of events changes slightly over time and we definitely
had some major changes in the last three years which influence timing
(high res timers, dynticks, NAPI ....)
Thanks,
tglx
next prev parent reply other threads:[~2008-10-02 18:02 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-09-30 3:19 [RFC PATCH 00/12] e1000e debug and protection patches Jesse Brandeburg
2008-09-30 3:19 ` [RFC PATCH 01/12] x86: export set_memory_ro and set_memory_rw Jesse Brandeburg
2008-09-30 7:07 ` Ingo Molnar
2008-09-30 3:19 ` [RFC PATCH 02/12] On Tue, 23 Sep 2008, David Miller wrote: Jesse Brandeburg
2008-10-02 22:23 ` Jesse Barnes
2008-10-03 20:46 ` David Miller
2008-10-03 21:29 ` Jesse Barnes
2008-10-03 21:45 ` Jiri Kosina
2008-10-03 23:28 ` Jesse Brandeburg
2008-10-03 23:30 ` Jesse Brandeburg
2008-10-04 10:21 ` Jiri Kosina
2008-10-04 11:02 ` Thomas Gleixner
2008-10-05 1:24 ` Jesse Brandeburg
2008-10-05 8:51 ` Thomas Gleixner
2008-10-05 15:05 ` Arjan van de Ven
2008-10-05 15:55 ` Thomas Gleixner
2008-10-05 16:02 ` Arjan van de Ven
2008-10-05 16:16 ` Thomas Gleixner
2008-10-05 17:01 ` Arjan van de Ven
2008-10-07 23:19 ` David Miller
2008-09-30 3:19 ` [RFC PATCH 03/12] e1000e: reset swflag after resetting hardware Jesse Brandeburg
2008-09-30 3:19 ` [RFC PATCH 04/12] e1000e: do not ever sleep in interrupt context Jesse Brandeburg
2008-09-30 3:19 ` [RFC PATCH 05/12] e1000e: fix lockdep issues Jesse Brandeburg
2008-09-30 3:19 ` [RFC PATCH 06/12] e1000e: drop stats lock Jesse Brandeburg
2008-09-30 3:19 ` [RFC PATCH 07/12] e1000e: debug contention on NVM SWFLAG Jesse Brandeburg
2008-10-02 14:28 ` Jiri Kosina
2008-10-02 15:03 ` Olaf Kirch
2008-10-02 16:27 ` Brandeburg, Jesse
2008-10-02 17:33 ` Olaf Kirch
2008-10-02 18:58 ` Thomas Gleixner
2008-10-02 19:07 ` Olaf Kirch
2008-10-02 19:08 ` Olaf Kirch
2008-10-02 18:02 ` Thomas Gleixner [this message]
2008-10-02 23:42 ` [PATCH] e1000e: prevent concurrent access to NVRAM Thomas Gleixner
2008-10-03 0:19 ` Jesse Brandeburg
2008-10-03 0:28 ` Thomas Gleixner
2008-09-30 3:19 ` [RFC PATCH 08/12] e1000e: allow bad checksum Jesse Brandeburg
2008-09-30 8:38 ` Jiri Kosina
2008-09-30 3:20 ` [RFC PATCH 09/12] e1000e: dump eeprom to dmesg for ich8/9 Jesse Brandeburg
2008-09-30 3:20 ` [RFC PATCH 10/12] e1000e: Use set_memory_ro()/set_memory_rw() to protect flash memory Jesse Brandeburg
2008-09-30 3:20 ` [RFC PATCH 11/12] e1000e: write protect ICHx NVM to prevent malicious write/erase Jesse Brandeburg
2008-09-30 12:40 ` Jiri Kosina
2008-09-30 15:47 ` Allan, Bruce W
2008-10-01 13:29 ` Jiri Kosina
2008-10-01 19:13 ` Allan, Bruce W
2008-09-30 3:20 ` [RFC PATCH 12/12] update version Jesse Brandeburg
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=alpine.LFD.1.10.0810021959030.5549@apollo.tec.linutronix.de \
--to=tglx@linutronix.de \
--cc=agospoda@redhat.com \
--cc=airlied@gmail.com \
--cc=arjan@linux.intel.com \
--cc=bruce.w.allan@intel.com \
--cc=chris.jones@canonical.com \
--cc=david.graham@intel.com \
--cc=jesse.brandeburg@intel.com \
--cc=jkosina@suse.cz \
--cc=john.ronciak@intel.com \
--cc=kkeil@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-netdev@vger.kernel.org \
--cc=okir@suse.de \
--cc=tim.gardner@intel.com \
/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).