Badness at mm/backing-dev.c:179

Badness at mm/backing-dev.c:179

[Mark Jackson]

The following commit causes a "Badness at mm/backing-dev.c:179":-

f1d0b063d993527754f062c589b73f125024d216 is first bad commit
commit f1d0b063d993527754f062c589b73f125024d216
Author: Kay Sievers <kay.sievers@vrfy.org>
Date:   Tue Dec 2 10:31:50 2008 -0800

     bdi: register sysfs bdi device only once per queue

     <snip>

     Tested-by: Peter Korsgaard <jacmet@sunsite.dk>
     Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
     Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
     Cc: David Woodhouse <dwmw2@infradead.org>
     Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
     Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Below is the extract from dmesg:-

physmap platform flash device: 00800000 at 00000000
physmap-flash.0: Found 1 x16 devices at 0x0 in 16-bit bank
  Amd/Fujitsu Extended Query Table at 0x0040
number of CFI chips: 1
cfi_cmdset_0002: Disabling erase-suspend-program due to code brokenness.
RedBoot partition parsing not available
Using physmap partition information
Creating 4 MTD partitions on "physmap-flash.0":
0x00000000-0x00020000 : "u-boot"
0x00020000-0x007e0000 : "root"
------------[ cut here ]------------
Badness at mm/backing-dev.c:179
PC is at bdi_register+0xe/0x44
LR is at bdi_register_dev+0x16/0x20
pc : [<9003f07a>]    lr : [<9003f0c6>]    Not tainted
sp : 91c19ccc  r12: 91c5409c  r11: 00000000
r10: 901c56c8  r9 : 0000001f  r8 : 00000001
r7 : 91c19cd0  r6 : 91c5409c  r5 : 91d0bc00  r4 : 00000000
r3 : 91e9b200  r2 : 901fd3fc  r1 : 00000001  r0 : 00000004
Flags: qvNzc
Mode bits: hjmde....g
CPU Mode: Supervisor
Process: swapper [1] (task: 91c16000 thread: 91c18000)
Call trace:
  [<9003f0c6>] bdi_register_dev+0x16/0x20
  [<900d7cc4>] add_disk+0x88/0xb0
  [<90100148>] add_mtd_blktrans_dev+0x178/0x188
  [<901004ea>] mtdblock_add_mtd+0x36/0x3c
  [<900ffec4>] blktrans_notify_add+0x1a/0x3a
  [<900fe16c>] add_mtd_device+0x60/0xa0
  [<900fef2e>] add_mtd_partitions+0x37e/0x3a4
  [<90105614>] physmap_flash_probe+0x1ec/0x21c
  [<900f9eb4>] platform_drv_probe+0x10/0x12
  [<900f9680>] driver_probe_device+0x84/0xf0
  [<900f9726>] __driver_attach+0x3a/0x50
  [<900f904a>] bus_for_each_dev+0x2e/0x4c
  [<900f9566>] driver_attach+0x12/0x14
  [<900f9324>] bus_add_driver+0x6c/0x170
  [<900f9860>] driver_register+0x58/0xb0
  [<900fa0b6>] platform_driver_register+0x56/0x5c
  [<9000adca>] physmap_init+0xa/0x10
  [<900146da>] do_one_initcall+0x36/0x10c
  [<90000380>] kernel_init+0x48/0x94
  [<9001fca0>] do_exit+0x0/0x49c

0x007e0000-0x007f0000 : "splash"
------------[ cut here ]------------
Badness at mm/backing-dev.c:179
PC is at bdi_register+0xe/0x44
LR is at bdi_register_dev+0x16/0x20
pc : [<9003f07a>]    lr : [<9003f0c6>]    Tainted: G        W
sp : 91c19ccc  r12: 91c5409c  r11: 00000000
r10: 901c56c8  r9 : 0000001f  r8 : 00000002
r7 : 91c19cd0  r6 : 91c5409c  r5 : 91d0bc00  r4 : 00000000
r3 : 91e9b800  r2 : 901fd414  r1 : 00000002  r0 : 00000004
Flags: qvNzc
Mode bits: hjmde....g
CPU Mode: Supervisor
Process: swapper [1] (task: 91c16000 thread: 91c18000)
Call trace:
  [<9003f0c6>] bdi_register_dev+0x16/0x20
  [<900d7cc4>] add_disk+0x88/0xb0
  [<90100148>] add_mtd_blktrans_dev+0x178/0x188
  [<901004ea>] mtdblock_add_mtd+0x36/0x3c
  [<900ffec4>] blktrans_notify_add+0x1a/0x3a
  [<900fe16c>] add_mtd_device+0x60/0xa0
  [<900fef2e>] add_mtd_partitions+0x37e/0x3a4
  [<90105614>] physmap_flash_probe+0x1ec/0x21c
  [<900f9eb4>] platform_drv_probe+0x10/0x12
  [<900f9680>] driver_probe_device+0x84/0xf0
  [<900f9726>] __driver_attach+0x3a/0x50
  [<900f904a>] bus_for_each_dev+0x2e/0x4c
  [<900f9566>] driver_attach+0x12/0x14
  [<900f9324>] bus_add_driver+0x6c/0x170
  [<900f9860>] driver_register+0x58/0xb0
  [<900fa0b6>] platform_driver_register+0x56/0x5c
  [<9000adca>] physmap_init+0xa/0x10
  [<900146da>] do_one_initcall+0x36/0x10c
  [<90000380>] kernel_init+0x48/0x94
  [<9001fca0>] do_exit+0x0/0x49c

0x007f0000-0x00800000 : "env"
------------[ cut here ]------------
Badness at mm/backing-dev.c:179
PC is at bdi_register+0xe/0x44
LR is at bdi_register_dev+0x16/0x20
pc : [<9003f07a>]    lr : [<9003f0c6>]    Tainted: G        W
sp : 91c19ccc  r12: 91c5409c  r11: 00000000
r10: 901c56c8  r9 : 0000001f  r8 : 00000003
r7 : 91c19cd0  r6 : 91c5409c  r5 : 91d0bc00  r4 : 00000000
r3 : 91e9be00  r2 : 901fd42c  r1 : 00000003  r0 : 00000004
Flags: qvNzc
Mode bits: hjmde....g
CPU Mode: Supervisor
Process: swapper [1] (task: 91c16000 thread: 91c18000)
Call trace:
  [<9003f0c6>] bdi_register_dev+0x16/0x20
  [<900d7cc4>] add_disk+0x88/0xb0
  [<90100148>] add_mtd_blktrans_dev+0x178/0x188
  [<901004ea>] mtdblock_add_mtd+0x36/0x3c
  [<900ffec4>] blktrans_notify_add+0x1a/0x3a
  [<900fe16c>] add_mtd_device+0x60/0xa0
  [<900fef2e>] add_mtd_partitions+0x37e/0x3a4
  [<90105614>] physmap_flash_probe+0x1ec/0x21c
  [<900f9eb4>] platform_drv_probe+0x10/0x12
  [<900f9680>] driver_probe_device+0x84/0xf0
  [<900f9726>] __driver_attach+0x3a/0x50
  [<900f904a>] bus_for_each_dev+0x2e/0x4c
  [<900f9566>] driver_attach+0x12/0x14
  [<900f9324>] bus_add_driver+0x6c/0x170
  [<900f9860>] driver_register+0x58/0xb0
  [<900fa0b6>] platform_driver_register+0x56/0x5c
  [<9000adca>] physmap_init+0xa/0x10
  [<900146da>] do_one_initcall+0x36/0x10c
  [<90000380>] kernel_init+0x48/0x94
  [<9001fca0>] do_exit+0x0/0x49c

physmap platform flash device: 00800000 at 08000000
physmap-flash.1: Found 1 x16 devices at 0x0 in 16-bit bank
  Amd/Fujitsu Extended Query Table at 0x0040
number of CFI chips: 1
cfi_cmdset_0002: Disabling erase-suspend-program due to code brokenness.
RedBoot partition parsing not available
Using physmap partition information
Creating 1 MTD partitions on "physmap-flash.1":
0x00000000-0x00800000 : "data"
------------[ cut here ]------------
Badness at mm/backing-dev.c:179
PC is at bdi_register+0xe/0x44
LR is at bdi_register_dev+0x16/0x20
pc : [<9003f07a>]    lr : [<9003f0c6>]    Tainted: G        W
sp : 91c19ccc  r12: 91c5409c  r11: 00000000
r10: 901c56c8  r9 : 0000001f  r8 : 00000004
r7 : 91c19cd0  r6 : 91c5409c  r5 : 91d0bc00  r4 : 00000000
r3 : 91cb9400  r2 : 901fd444  r1 : 00000000  r0 : 00000001
Flags: qvNzc
Mode bits: hjmde....g
CPU Mode: Supervisor
Process: swapper [1] (task: 91c16000 thread: 91c18000)
Call trace:
  [<9003f0c6>] bdi_register_dev+0x16/0x20
  [<900d7cc4>] add_disk+0x88/0xb0
  [<90100148>] add_mtd_blktrans_dev+0x178/0x188
  [<901004ea>] mtdblock_add_mtd+0x36/0x3c
  [<900ffec4>] blktrans_notify_add+0x1a/0x3a
  [<900fe16c>] add_mtd_device+0x60/0xa0
  [<900fef2e>] add_mtd_partitions+0x37e/0x3a4
  [<90105614>] physmap_flash_probe+0x1ec/0x21c
  [<900f9eb4>] platform_drv_probe+0x10/0x12
  [<900f9680>] driver_probe_device+0x84/0xf0
  [<900f9726>] __driver_attach+0x3a/0x50
  [<900f904a>] bus_for_each_dev+0x2e/0x4c
  [<900f9566>] driver_attach+0x12/0x14
  [<900f9324>] bus_add_driver+0x6c/0x170
  [<900f9860>] driver_register+0x58/0xb0
  [<900fa0b6>] platform_driver_register+0x56/0x5c
  [<9000adca>] physmap_init+0xa/0x10
  [<900146da>] do_one_initcall+0x36/0x10c
  [<90000380>] kernel_init+0x48/0x94
  [<9001fca0>] do_exit+0x0/0x49c

Regards
Mark

Re: Badness at mm/backing-dev.c:179

[Kay Sievers]

On Wed, Dec 3, 2008 at 13:07, Mark Jackson <mpfj@mimc.co.uk> wrote:
> The following commit causes a "Badness at mm/backing-dev.c:179":-
>
> f1d0b063d993527754f062c589b73f125024d216 is first bad commit
> commit f1d0b063d993527754f062c589b73f125024d216
> Author: Kay Sievers <kay.sievers@vrfy.org>
> Date:   Tue Dec 2 10:31:50 2008 -0800
>
>    bdi: register sysfs bdi device only once per queue

> Below is the extract from dmesg:-
>
> ------------[ cut here ]------------
> Badness at mm/backing-dev.c:179

That's a known issue, and will not cause any harm besides printing
this warning. Andrew added the warning to catch drivers who use one
single request queue for multiple devices. Currently known are mtd and
floppy (triggers only if you have more than one device per driver).
The drivers need to be changed to use a separate queue per device, to
get rid of this warning.

Thanks,
Kay

Re: Badness at mm/backing-dev.c:179

[Mark Jackson]

Kay Sievers wrote:
> On Wed, Dec 3, 2008 at 13:07, Mark Jackson <mpfj@mimc.co.uk> wrote:
>> The following commit causes a "Badness at mm/backing-dev.c:179":-
>>
>> f1d0b063d993527754f062c589b73f125024d216 is first bad commit
>> commit f1d0b063d993527754f062c589b73f125024d216
>> Author: Kay Sievers <kay.sievers@vrfy.org>
>> Date:   Tue Dec 2 10:31:50 2008 -0800
>>
>>    bdi: register sysfs bdi device only once per queue
> 
>> Below is the extract from dmesg:-
>>
>> ------------[ cut here ]------------
>> Badness at mm/backing-dev.c:179
> 
> That's a known issue, and will not cause any harm besides printing
> this warning. Andrew added the warning to catch drivers who use one
> single request queue for multiple devices. Currently known are mtd and
> floppy (triggers only if you have more than one device per driver).
> The drivers need to be changed to use a separate queue per device, to
> get rid of this warning.

Okay ... thanks for the info.

Mark

Re: Badness at mm/backing-dev.c:179

[Andrew Morton]

On Wed, 3 Dec 2008 13:22:47 +0100 "Kay Sievers" <kay.sievers@vrfy.org> wrote:

> On Wed, Dec 3, 2008 at 13:07, Mark Jackson <mpfj@mimc.co.uk> wrote:
> > The following commit causes a "Badness at mm/backing-dev.c:179":-
> >
> > f1d0b063d993527754f062c589b73f125024d216 is first bad commit
> > commit f1d0b063d993527754f062c589b73f125024d216
> > Author: Kay Sievers <kay.sievers@vrfy.org>
> > Date:   Tue Dec 2 10:31:50 2008 -0800
> >
> >    bdi: register sysfs bdi device only once per queue
> 
> > Below is the extract from dmesg:-
> >
> > ------------[ cut here ]------------
> > Badness at mm/backing-dev.c:179
> 
> That's a known issue, and will not cause any harm besides printing
> this warning. Andrew added the warning to catch drivers who use one
> single request queue for multiple devices. Currently known are mtd and
> floppy (triggers only if you have more than one device per driver).
> The drivers need to be changed to use a separate queue per device, to
> get rid of this warning.

Actually, adding that warning wasn't such a great idea - we'll just get
heaps of email about things we already know about.  I think I'll take
it out again.

Which means this stuff won't get fixed for ages :(

Re: Badness at mm/backing-dev.c:179

[Geert Uytterhoeven]

On Wed, 3 Dec 2008, Kay Sievers wrote:
> On Wed, Dec 3, 2008 at 13:07, Mark Jackson <mpfj@mimc.co.uk> wrote:
> > The following commit causes a "Badness at mm/backing-dev.c:179":-
> >
> > f1d0b063d993527754f062c589b73f125024d216 is first bad commit
> > commit f1d0b063d993527754f062c589b73f125024d216
> > Author: Kay Sievers <kay.sievers@vrfy.org>
> > Date:   Tue Dec 2 10:31:50 2008 -0800
> >
> >    bdi: register sysfs bdi device only once per queue
> 
> > Below is the extract from dmesg:-
> >
> > ------------[ cut here ]------------
> > Badness at mm/backing-dev.c:179
> 
> That's a known issue, and will not cause any harm besides printing
> this warning. Andrew added the warning to catch drivers who use one
> single request queue for multiple devices. Currently known are mtd and
> floppy (triggers only if you have more than one device per driver).
> The drivers need to be changed to use a separate queue per device, to
> get rid of this warning.

I guess this one (introduced after 2.6.28-rc7) is an incarnation of the same
issue?

WARNING: at mm/backing-dev.c:179 0x2069cb()
Modules linked in:
Call Trace: [<000273ee>] warn_on_slowpath+0x44/0x60
 [<001be5b0>] mutex_lock+0x0/0x24
 [<00075d1e>] ifind+0x16/0x5c
 [<000a0844>] sysfs_ilookup_test+0x0/0x18
 [<00075d92>] ilookup5+0x2e/0x38
 [<000a0e70>] sysfs_addrm_finish+0x1c/0x200
 [<000a047e>] sysfs_create_file+0x0/0x44
 [<000a0432>] sysfs_add_file_mode+0x5a/0x84
 [<000efc5a>] kobject_uevent_env+0x46/0x30c
 [<000a047e>] sysfs_create_file+0x0/0x44
 [<000a0478>] sysfs_add_file+0x1c/0x22
 [<000eff30>] kobject_uevent+0x10/0x16
 [<000e1332>] elv_register_queue+0x64/0x6e
 [<000eff20>] kobject_uevent+0x0/0x16
 [<000e450e>] blk_register_queue+0x56/0x94
 [<00051684>] bdi_register+0x54/0x5e
 [<000f4040>] sprintf+0x0/0x26
 [<000516b4>] bdi_register_dev+0x26/0x2c
 [<000e7e70>] add_disk+0xb8/0x12c
 [<000e7db8>] add_disk+0x0/0x12c
 [<0027e424>] atari_floppy_init+0x14c/0x6e4
 [<0027e2d8>] atari_floppy_init+0x0/0x6e4
 [<00002008>] do_one_initcall+0x0/0x19a
 [<002781ce>] __alloc_bootmem+0x0/0x1a
 [<00002024>] do_one_initcall+0x1c/0x19a
 [<001be5b0>] mutex_lock+0x0/0x24
 [<00075d1e>] ifind+0x16/0x5c
 [<000a0844>] sysfs_ilookup_test+0x0/0x18
 [<00075d92>] ilookup5+0x2e/0x38
 [<000a0e70>] sysfs_addrm_finish+0x1c/0x200
 [<002781ce>] __alloc_bootmem+0x0/0x1a
 [<000a0432>] sysfs_add_file_mode+0x5a/0x84
 [<001244aa>] sysdev_class_create_file+0x0/0x16
 [<000a0478>] sysfs_add_file+0x1c/0x22
 [<000a04bc>] sysfs_create_file+0x3e/0x44
 [<001244bc>] sysdev_class_create_file+0x12/0x16
 [<0027e102>] cpu_dev_init+0x4e/0x6c
 [<0026eaa8>] kernel_init+0x6e/0xc2
 [<0027e2d8>] atari_floppy_init+0x0/0x6e4
 [<0026ea3a>] kernel_init+0x0/0xc2
 [<00027902>] printk+0x0/0x1a
 [<00002ada>] kernel_thread+0x3a/0x4e

---[ end trace 7d0e1e0e92952ce0 ]---

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: Badness at mm/backing-dev.c:179

[Kay Sievers]

On Mon, Dec 8, 2008 at 09:16, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> On Wed, 3 Dec 2008, Kay Sievers wrote:
>> On Wed, Dec 3, 2008 at 13:07, Mark Jackson <mpfj@mimc.co.uk> wrote:
>> > The following commit causes a "Badness at mm/backing-dev.c:179":-
>> >
>> > f1d0b063d993527754f062c589b73f125024d216 is first bad commit
>> > commit f1d0b063d993527754f062c589b73f125024d216
>> > Author: Kay Sievers <kay.sievers@vrfy.org>
>> > Date:   Tue Dec 2 10:31:50 2008 -0800
>> >
>> >    bdi: register sysfs bdi device only once per queue
>>
>> > Below is the extract from dmesg:-
>> >
>> > ------------[ cut here ]------------
>> > Badness at mm/backing-dev.c:179
>>
>> That's a known issue, and will not cause any harm besides printing
>> this warning. Andrew added the warning to catch drivers who use one
>> single request queue for multiple devices. Currently known are mtd and
>> floppy (triggers only if you have more than one device per driver).
>> The drivers need to be changed to use a separate queue per device, to
>> get rid of this warning.
>
> I guess this one (introduced after 2.6.28-rc7) is an incarnation of the same
> issue?

You have more than one floppy device registered at the same time,
right? Then it's very likely the same issue, that both devices share
the same request queue, like the x86 floppy.c.

Kay

Re: Badness at mm/backing-dev.c:179

[Geert Uytterhoeven]

On Mon, 8 Dec 2008, Kay Sievers wrote:
> On Mon, Dec 8, 2008 at 09:16, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> > On Wed, 3 Dec 2008, Kay Sievers wrote:
> >> On Wed, Dec 3, 2008 at 13:07, Mark Jackson <mpfj@mimc.co.uk> wrote:
> >> > The following commit causes a "Badness at mm/backing-dev.c:179":-
> >> >
> >> > f1d0b063d993527754f062c589b73f125024d216 is first bad commit
> >> > commit f1d0b063d993527754f062c589b73f125024d216
> >> > Author: Kay Sievers <kay.sievers@vrfy.org>
> >> > Date:   Tue Dec 2 10:31:50 2008 -0800
> >> >
> >> >    bdi: register sysfs bdi device only once per queue
> >>
> >> > Below is the extract from dmesg:-
> >> >
> >> > ------------[ cut here ]------------
> >> > Badness at mm/backing-dev.c:179
> >>
> >> That's a known issue, and will not cause any harm besides printing
> >> this warning. Andrew added the warning to catch drivers who use one
> >> single request queue for multiple devices. Currently known are mtd and
> >> floppy (triggers only if you have more than one device per driver).
> >> The drivers need to be changed to use a separate queue per device, to
> >> get rid of this warning.
> >
> > I guess this one (introduced after 2.6.28-rc7) is an incarnation of the same
> > issue?
> 
> You have more than one floppy device registered at the same time,
> right? Then it's very likely the same issue, that both devices share
> the same request queue, like the x86 floppy.c.

Yep, ataflop.c (and amiflop.c) forked from floppy.c a long time ago...

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds