From: Stefan Richter <stefanr@s5r6.in-berlin.de>
To: linux1394-user@lists.sourceforge.net, linux-kernel@vger.kernel.org
Cc: Thorsten Holtkaemper <openpub@web.de>
Subject: Oops in remove_wait_queue (Re: sbp2: sbp2util_node_write_no_wait failed)
Date: Sun, 06 Nov 2005 23:03:31 +0100 [thread overview]
Message-ID: <436E7DB3.1030103@s5r6.in-berlin.de> (raw)
In-Reply-To: <200511062007.13047.openpub@web.de>
Thorsten Holtkaemper wrote:
>>>I am running kernel 2.6.13.4
...
> After powering off and on the FireWire haddisk
> the login failed with the same messages. Trying to reload the FireWire
> related modules resulted in an oops. As far as I can remember such kind of
> oops also happened at least once during module reload without your patch.
>
> [...]
> ieee1394: Initialized config rom entry `ip1394'
> ohci1394: $Rev: 1299 $ Ben Collins <bcollins@debian.org>
> ACPI: PCI Interrupt 0000:00:0c.0[A] -> GSI 19 (level, low) -> IRQ 209
> ohci1394: fw-host0: Unexpected PCI resource length of 1000!
> ohci1394: fw-host0: OHCI-1394 1.1 (PCI): IRQ=[209] MMIO=[dfff7000-dfff77ff] Max Packet=[2048]
> eth1394: $Rev: 1264 $ Ben Collins <bcollins@debian.org>
> eth1394: eth2: IEEE-1394 IPv4 over 1394 Ethernet (fw-host0)
> sbp2: $Rev: 1306 $ Ben Collins <bcollins@debian.org>
> ieee1394: unsolicited response packet received - no tlabel match
> ieee1394: Node added: ID:BUS[0-00:1023] GUID[0050770e501e1452]
> ieee1394: Host added: ID:BUS[0-01:1023] GUID[000010dc00592003]
> scsi8 : SCSI emulation for IEEE-1394 SBP-2 Devices
> Unable to handle kernel NULL pointer dereference at virtual address 00000003
> printing eip:
> c012ec26
> *pde = 00000000
> Oops: 0002 [#1]
> Modules linked in: sbp2 ohci1394 eth1394 ieee1394 vfat fat udf isofs nfsd exportfs lockd nfs_acl sunrpc lp thermal fan button processor ac battery capability commoncap af_packet ipv6 ipt_MASQUERADE ipt_mark ip_nat_irc ip_nat_ftp iptable_mangle iptable_nat ipt_state ipt_REJECT ipt_limit ipt_LOG ip_conntrack_ftp ip_conntrack_irc ip_conntrack iptable_filter ip_tables parport_pc parport pcspkr rtc 8139cp ehci_hcd snd_intel8x0 snd_ac97_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc i2c_sis96x i2c_core shpchp pci_hotplug sis_agp agpgart sd_mod tsdev mousedev joydev evdev usbhid usb_storage scsi_mod hisax_fcpcipnp hisax_isac hisax crc_ccitt isdn slhc ohci_hcd usbcore 8139too mii subfs ide_cd cdrom ext3 jbd mbcache ide_disk ide_generic via82cxxx trm290 triflex slc90e66 sis5513 siimage serverworks sc1200 rz1000 piix pdc202xx_old pdc202xx_new opti621 ns87415 it821x hpt366 hpt34x generic cy82c693 cs5530 cs5520 cmd64x atiixp amd74xx alim15x3 aec62xx ide_
core unix
> CPU: 0
> EIP: 0060:[<c012ec26>] Not tainted VLI
> EFLAGS: 00210096 (2.6.13.4)
> EIP is at remove_wait_queue+0x16/0x40
> eax: ffffffff ebx: ffffc0f3 ecx: ffffffff edx: ffffc0e7
> esi: 00200296 edi: c73e3000 ebp: 00000000 esp: cade5ecc
> ds: 007b es: 007b ss: 0068
> Process klipper (pid: 8204, threadinfo=cade4000 task=ca905020)
> Stack: ffffc0e3 c73e3008 c016aa9e 00000000 00000000 00000009 c016ae1f cade5f38
> 00000000 00000000 00000000 00000178 00000000 00000000 00000000 00000000
> 00000178 cade4000 c14e45cc c14e45c8 c14e45c4 c14e45d8 c14e45d4 c14e45d0
> Call Trace:
> [<c016aa9e>] poll_freewait+0x2e/0x50
> [<c016ae1f>] do_select+0x18f/0x2c0
> [<c016aac0>] __pollwait+0x0/0xd0
> [<c016b238>] sys_select+0x2a8/0x440
> [<c0103009>] syscall_call+0x7/0xb
> Code: 04 89 0b 89 59 04 56 9d 8b 1c 24 8b 74 24 04 83 c4 08 c3 89 f6 83 ec 08 89 1c 24 89 74 24 04 9c 5e fa 8d 5a 0c 8b 42 0c 8b 4b 04 <89> 4804 89 01 c7 43 04 00 02 20 00 c7 42 0c 00 01 10 00 56 9d
> <3>ieee1394: sbp2: Error logging into SBP-2 device - login failed
> sbp2: probe of 0050770e501e1452-0 failed with error -16
> scsi9 : SCSI emulation for IEEE-1394 SBP-2 Devices
> ieee1394: sbp2: Error logging into SBP-2 device - login failed
> sbp2: probe of 0050770e501e1452-0 failed with error -16
> [...]
What is the klipper process, and what did it poll there?
One thing about sbp2 which isn't optimal yet is that it creates a
scsi_host before it logs in into the SBP-2 unit, instead of deferring
host creation until after a successful login. Maybe the short lifetime
of such a host reveals a vulnerability to race conditions somewhere.
(added linux-kernel to recipients)
--
Stefan Richter
-=====-=-=-= =-== --==-
http://arcgraph.de/sr/
next parent reply other threads:[~2005-11-06 22:06 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <200511061751.12787.openpub@web.de>
[not found] ` <436E4019.7000709@s5r6.in-berlin.de>
[not found] ` <200511062007.13047.openpub@web.de>
2005-11-06 22:03 ` Stefan Richter [this message]
2005-11-06 23:03 ` Thorsten Holtkaemper
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=436E7DB3.1030103@s5r6.in-berlin.de \
--to=stefanr@s5r6.in-berlin.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux1394-user@lists.sourceforge.net \
--cc=openpub@web.de \
/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
all inboxes | Powered by JetHome®