linux-kernel - Re: hibernate event order question

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20080517210325.GB3675@yamamaya.is-a-geek.org>
Date:	Sat, 17 May 2008 23:03:26 +0200
From:	Tobias Diedrich <ranma+kernel@...edrich.de>
To:	"Rafael J. Wysocki" <rjw@...k.pl>
Cc:	linux-netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
	Ayaz Abdulla <aabdulla@...dia.com>
Subject: Re: hibernate event order question

Rafael J. Wysocki wrote:
> > Now for my question:
> > AFAICS the ordering of events is as follows:
> > 1) User reqests hibernate
> > 2) Tasks are frozen
> > 3) Device suspend callbacks get called
> > 4) Device resume callbacks get called
> > 5) Memory image is written to disk
> 
> 6) Device suspend callbacks get called again to prepare for the system sleep
>    state.

Hmm, I guess I don't see that because by then the console is disabled or
something like that?
(It also doesn't help that the system immediately powers down after that)
Can I get debugging output during that phase out onto the serial
console and/or plain text console & simply turn the poweroff into a
halt? (halt in addition to shutdown and reboot in /sys/power/disk
might be nice)

The problem is I'm really seeing/have seen multiple issues/situations here:
1) When the device is part of a bridge, connectivity is down after
   Hibernate
2) Boot->Hibernate->WOL works->Hibernate->WOL only works with
   byte-reversed MAC-address
3) I have to reconfirm this, but I think after fixing bug 1) with
   the patch of the previous mail bug 2) turns into 4):
4) WOL doesn't work at all after Hibernate
5) nforce ethernet and netgear switch don't like each other and
   sometimes negotiate only 100Mbit instead of 1Gbit, while my
   notebook on a much longer cable never has negotiation problems
6) In the past sometimes negotiation would take very long, that one
   I haven't seen in some time so I assume it's fixed in recent
   kernels.  This one only happened if both ports were enabled IIRC

"Device suspend callbacks get called again" means I have to retest 3)
Maybe I'm imagining things or seeing a heisenbug here?

1) Is fixed by the patch in the previous mail
2) I assume to be related to DEV_HAS_CORRECT_MACADDR,
NVREG_TRANSMITPOLL_MAC_ADDR_REV and some magic byteswapping code in
the driver.
1) and 2)/4) can be worked around (for the bridge case) by removing
the forcedeth module prior to hibernate and reinserting the module
and readding the devices to the bridge after resume.

-- 
Tobias						PGP: http://9ac7e0bc.uguu.de
このメールは十割再利用されたビットで作られています。
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/