Skip to contentSkip to conversation

r8169: Kernel Panic when a lot of data is transferred through network interface

Bug 16142 [email protected] 1 participant 15 messages Jun 14, 2010 – Jul 20, 2012 1 message not in the archive libata

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]/

[email protected]

author
#1 lore raw

https://bugzilla.kernel.org/show_bug.cgi?id=16142

Andrew Morton <[email protected]> changed:

What |Removed |Added
----------------------------------------------------------------------------

          Component|Network                     |IDE
         AssignedTo|drivers_network@kernel-bugs |[email protected]
                   |.osdl.org                   |
            Product|Drivers                     |IO/Storage

--- Comment #8 from Andrew Morton <[email protected]> 2010-06-14 22:17:35 --- Seems to be an oops in ide_complete_rq(), unrelated to the network driver.

I'll recategorise the bug to IDE, thanks.

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#2 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

Borislav Petkov <[email protected]> changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #9 from Borislav Petkov <[email protected]> 2010-06-15 06:38:58 --- Hi,

ok, here's what you can do:

1. Switch to libata and done, no bug. IDE is deprecated anyways.

If you'd still like to help us fix this and keep using ide, here are some more things to do :)

1. Can you connect your machine to another over a serial console so that you can catch a clean OOPS. If you don't have a serial cable, you can try doing this with a netconsole, see <Documentation/networking/netconsole.txt> in the kernel sources. And if even that is not an option, try making a picture which contains the first oops from the beginning and doesn't cut off the register dump. Something like your https://bugzilla.kernel.org/attachment.cgi?id=26730
but without the shaky effects :)

2. Go to your kernel source tree and do

objdump -d drivers/ide/ide-io.o | less

then look for the line similar to

"000000000000070b <ide_complete_rq>:"

and copy/paste here the whole block all the way to the end of that function, i.e. line that looks like

" 755: c3 retq"

3. Just to make sure, I think your "shaky" picture showed that there are no proprietary kernel modules your using, e.g. nvidia GPU drivers, but let me make sure I see it correctly: you're not installing any closed-source, proprietary crap, right?

I think this is all for now, thanks for your help.

Regards,
Boris.

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#3 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

--- Comment #10 from [email protected] 2010-06-15 11:17:33 --- First of all: Thank you all for your quick responses.

Switching to libdata solved the problem for me.
(at least until now ;) )

----
But:
If you are still interested in the bug, I would try to help you as far as I can.

Answers to your questions:
1. Proprietary drivers: I am using the nvidia driver, but it was not loaded when the bug occurred. (for the 2.6.34 kernel I didn't even installed them)

2. Serial console: sorry this computer does not have an serial port (there is one on the motherboard, but I don't have the matching adapter.)

3. Netconsole: I tried to use it; normal dmesg logging worked, but I did not receive the oops messages.

Right now my questions are:
1. Do you still want to take care of this bug?
2. If so, is there a possibility to use firewire to capture the oops messages?

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#4 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

--- Comment #11 from Borislav Petkov <[email protected]> 2010-06-15 11:49:59 --- (In reply to comment #10)

Quoted 2 lines

If you are still interested in the bug, I would try to help you as far as I can.

Cool.

Quoted 3 lines

Answers to your questions:
1. Proprietary drivers: I am using the nvidia driver, but it was not loaded when the bug occurred. (for the 2.6.34 kernel I didn't even installed them)

Ok, please remove the proprietary drivers when reproducing the oops.

Quoted 5 lines

2. Serial console: sorry this computer does not have an serial port (there is one on the motherboard, but I don't have the matching adapter.)

3. Netconsole: I tried to use it; normal dmesg logging worked, but I did not receive the oops messages.

was the netconsole built-in or a module? Try building it into the kernel if not.

Quoted 2 lines

Right now my questions are:
1. Do you still want to take care of this bug?

Sure, I'll give it a try.

Quoted 1 line

2. If so, is there a possibility to use firewire to capture the oops messages?

There is something related in <Documentation/debugging-via-ohci1394.txt> but I haven't tried it yet.

Ok, so what we need now is a cleaner OOPS. If nothing else works, try making a picture but make sure it is readable and complete. Also, send me the objdump part I asked you for in the earlier message.

Thanks,
Boris.

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#5 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

--- Comment #12 from [email protected] 2010-06-15 15:18:58 --- Created an attachment (id=26776)
--> (https://bugzilla.kernel.org/attachment.cgi?id=26776)
kernel oops v3

this one is done via firewire

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#6 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

--- Comment #13 from [email protected] 2010-06-15 15:20:47 --- Created an attachment (id=26777)
--> (https://bugzilla.kernel.org/attachment.cgi?id=26777)
part of 'objdump -d drivers/ide/ide-io.'

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#7 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

[email protected] changed:

What |Removed |Added
----------------------------------------------------------------------------

  Attachment #26777|application/octet-stream    |text/plain
          mime type|                            |
Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#11 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

--- Comment #14 from [email protected] 2010-06-15 15:26:20 ---

Quoted 1 line

Ok, please remove the proprietary drivers when reproducing the oops.

I hope I did so, but I am not sure; Therefore please tell me if you think for any reason I did not remove the drivers.

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#12 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

--- Comment #15 from Andrew Morton <[email protected]> 2010-06-15 19:12:28 --- (In reply to comment #12)

Quoted 1 line

Created an attachment (id=26776)

--> (https://bugzilla.kernel.org/attachment.cgi?id=26776) [details]

Quoted 3 lines

kernel oops v3

this one is done via firewire

This seems impossible. I think the oops happened in

int ide_complete_rq(ide_drive_t *drive, int error, unsigned int nr_bytes)

{
        ide_hwif_t *hwif = drive->hwif;

accessing 0x00000040, so `drive' is NULL. But it's clearly non-NULL up in cdrom_newpc_intr(). Possibly it's a miscompile. But you've tried several different kernel version, which I assume were compiled by someone at gentoo.

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#13 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

--- Comment #16 from [email protected] 2010-06-15 19:39:35 --- Well as gentoo does not provide precompiled packages, I compiled everything by myself.
I could try to use another version of gcc for compiling the kernel.

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#14 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

--- Comment #17 from Borislav Petkov <[email protected]> 2010-07-03 08:26:01 --- See thread at http://marc.info/?l=linux-ide&m=127676206925930 for how it all turns out. In short:

Quoted 4 lines

But as I wrote in my original bugreport, I tested with a PCI NIC. (But the onboard NIC was not completely disabled, it was only disabled via ifconfig ... down)
When using the PCI NIC, the whole problem (kernel panik) did not occure.

Ok, this confirms my suspicion that it is shared-irq related. Also, we already verified that switching to libata does fix the issue for you so you are good to go. Considering the DEPRECATED status of ide, I have a very little incentive in hunting this thing further down, so let's leave it at that.

Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
author
#15 lore raw
In reply to a message not in the archive

https://bugzilla.kernel.org/show_bug.cgi?id=16142

Alan <[email protected]> changed:

What |Removed |Added
----------------------------------------------------------------------------

             Status|NEW                         |RESOLVED
                 CC|                            |[email protected]
         Resolution|                            |OBSOLETE
Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.