2003-09-09 15:10:11

by Samuel Flory

Subject: sleeping function called from invalid context

I'm seeing this on arjanv's 2.6.0-0.test4.1.33 kernel.

Debug: sleeping function called from invalid context at
include/asm/uaccess.h:473
Call Trace:
[<c011b7dd>] __might_sleep+0x5d/0x70
[<c010d0ea>] save_v86_state+0x6a/0x200
[<c010c8b5>] do_IRQ+0xd5/0x110
[<c010acd2>] work_notifysig_v86+0x6/0x14
[<c010ac7f>] syscall_call+0x7/0xb

Debug: sleeping function called from invalid context at
include/asm/uaccess.h:473
Call Trace:
[<c011b7dd>] __might_sleep+0x5d/0x70
[<c010d0ea>] save_v86_state+0x6a/0x200
[<c010bba0>] do_general_protection+0x0/0xb0
[<c010acd2>] work_notifysig_v86+0x6/0x14
[<c010ac7f>] syscall_call+0x7/0xb


Any thoughts. The system seems to function despite the issue. It
seems to have also occurred under 2.6.0-0.test3.1.31:
Aug 25 16:28:49 goblin kernel: Call Trace:
Aug 25 16:28:49 goblin kernel: [<c011af2d>] __might_sleep+0x5d/0x70
Aug 25 16:28:49 goblin kernel: [<c010d06a>] save_v86_state+0x6a/0x200
Aug 25 16:28:49 goblin kernel: [<c010bb50>] do_general_protection+0x0/0xb0
Aug 25 16:28:49 goblin kernel: [<c010ac82>] work_notifysig_v86+0x6/0x14
Aug 25 16:28:49 goblin kernel: [<c010ac2f>] syscall_call+0x7/0xb
Aug 25 16:28:49 goblin kernel:
Aug 25 16:28:50 goblin kernel: Debug: sleeping function called from
invalid cont
ext at include/asm/uaccess.h:473
Aug 25 16:28:50 goblin kernel: Call Trace:
Aug 25 16:28:50 goblin kernel: [<c011af2d>] __might_sleep+0x5d/0x70
Aug 25 16:28:50 goblin kernel: [<c010d06a>] save_v86_state+0x6a/0x200
Aug 25 16:28:50 goblin kernel: [<c010bb50>] do_general_protection+0x0/0xb0
Aug 25 16:28:50 goblin kernel: [<c010ac82>] work_notifysig_v86+0x6/0x14
Aug 25 16:28:50 goblin kernel: [<c010ac2f>] syscall_call+0x7/0xb
Aug 25 16:28:50 goblin kernel:
Aug 25 16:28:51 goblin kernel: Debug: sleeping function called from
invalid cont
ext at include/asm/uaccess.h:473
Aug 25 16:28:51 goblin kernel: Call Trace:
Aug 25 16:28:51 goblin kernel: [<c011af2d>] __might_sleep+0x5d/0x70
Aug 25 16:28:51 goblin kernel: [<c010d06a>] save_v86_state+0x6a/0x200
Aug 25 16:28:51 goblin kernel: [<c010c865>] do_IRQ+0xd5/0x110
Aug 25 16:28:51 goblin kernel: [<c010ac82>] work_notifysig_v86+0x6/0x14
Aug 25 16:28:51 goblin kernel: [<c010ac2f>] syscall_call+0x7/0xb

--
Once you have their hardware. Never give it back.
(The First Rule of Hardware Acquisition)
Sam Flory <[email protected]>

2003-09-09 19:47:41

by Matt Mackall

Subject: Re: sleeping function called from invalid context

On Tue, Sep 09, 2003 at 08:03:01AM -0700, Samuel Flory wrote:
> I'm seeing this on arjanv's 2.6.0-0.test4.1.33 kernel.

> Debug: sleeping function called from invalid context at
> include/asm/uaccess.h:473
> Call Trace:
> [<c011b7dd>] __might_sleep+0x5d/0x70
> [<c010d0ea>] save_v86_state+0x6a/0x200

It's a warning about the possibility of hitting a very old but rarely
hit bug, system should work the same as it always has despite the
warning. I'm working on this, but it's ugly. Hope to post a patch in
the next week or so.

--
Matt Mackall : http://www.selenic.com : of or relating to the moon

2003-09-09 20:19:35

by Samuel Flory

Subject: Re: sleeping function called from invalid context

Matt Mackall wrote:

>On Tue, Sep 09, 2003 at 08:03:01AM -0700, Samuel Flory wrote:
>
>
>> I'm seeing this on arjanv's 2.6.0-0.test4.1.33 kernel.
>>
>>
>
>
>
>>Debug: sleeping function called from invalid context at
>>include/asm/uaccess.h:473
>>Call Trace:
>>[<c011b7dd>] __might_sleep+0x5d/0x70
>>[<c010d0ea>] save_v86_state+0x6a/0x200
>>
>>
>
>It's a warning about the possibility of hitting a very old but rarely
>hit bug, system should work the same as it always has despite the
>warning. I'm working on this, but it's ugly. Hope to post a patch in
>the next week or so.
>
>
>

That's good to know. I was actually running for at least a on an
older kernel before i noticed in in dmesg.

--
Once you have their hardware. Never give it back.
(The First Rule of Hardware Acquisition)
Sam Flory <[email protected]>