[tainted,fglrx] page allocation failure: basic meaning of log messages?

[tainted,fglrx] page allocation failure: basic meaning of log messages?

[Andreas K. Huettel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160


Dear developers, 

I've found the following messages in my syslog after a suspend2ram cycle, 
(And similar repeated a few times, all between "freezing userspace" and 
"back to C".) 

I know the system is tainted / running the fglrx module, and (to me) it 
even looks like this might be causing the problem.

So - I'm not asking for a solution, but could cou please please give me a 
rough hint what the error message means???

TIA, best, Andreas 

(Of course I'd be happy to provide more info.)


Apr  3 09:01:30 bellini hal-system-powe: page allocation failure. order:10, mode:0x4020
Apr  3 09:01:30 bellini Pid: 327, comm: hal-system-powe Tainted: PF       2.6.24-gentoo-r3 #1
Apr  3 09:01:30 bellini [<c0140c48>] __alloc_pages+0x2bd/0x2ce
Apr  3 09:01:30 bellini [<c0140c92>] __get_free_pages+0x39/0x47
Apr  3 09:01:30 bellini [<f8adcd9e>] firegl_cmmqs_save_fb+0x8e/0x200 [fglrx]
Apr  3 09:01:30 bellini [<f8ad34d6>] firegl_pm_save_framebuffer+0x66/0x70 [fglrx]
Apr  3 09:01:30 bellini [<f8add198>] firegl_cmmqs_recoverable_surface_info+0x38/0x90 [fglrx]
Apr  3 09:01:30 bellini [<f8ad4be3>] firegl_cail_powerdown+0x93/0x120 [fglrx]
Apr  3 09:01:30 bellini [<f8ac3101>] __ke_remove_wait_queue+0x117/0x2be [fglrx]
Apr  3 09:01:30 bellini [<c01d92a2>] pci_device_suspend+0x18/0x47
Apr  3 09:01:30 bellini [<c023747d>] device_suspend+0xf4/0x1df
Apr  3 09:01:30 bellini [<c011bbae>] printk+0x1b/0x1f
Apr  3 09:01:30 bellini [<c013584b>] suspend_devices_and_enter+0x3d/0xd1
Apr  3 09:01:30 bellini [<c0136092>] freeze_processes+0x71/0x76
Apr  3 09:01:30 bellini [<c01359d7>] enter_state+0xf8/0x162
Apr  3 09:01:30 bellini [<c0135acf>] state_store+0x8e/0xa2
Apr  3 09:01:30 bellini [<c0135a41>] state_store+0x0/0xa2
Apr  3 09:01:30 bellini [<c01851e4>] subsys_attr_store+0x24/0x28
Apr  3 09:01:30 bellini [<c01854bf>] sysfs_write_file+0xa8/0xdb
Apr  3 09:01:30 bellini [<c0185417>] sysfs_write_file+0x0/0xdb
Apr  3 09:01:30 bellini [<c015827e>] vfs_write+0x8a/0x10c
Apr  3 09:01:30 bellini [<c015871d>] sys_write+0x41/0x67
Apr  3 09:01:30 bellini [<c0103b7a>] sysenter_past_esp+0x5f/0x85
Apr  3 09:01:30 bellini =======================
Apr  3 09:01:30 bellini Mem-info:
Apr  3 09:01:30 bellini DMA per-cpu:
Apr  3 09:01:30 bellini CPU    0: Hot: hi:    0, btch:   1 usd:   0   Cold: hi:    0, btch:   1 usd:   0
Apr  3 09:01:30 bellini Normal per-cpu:
Apr  3 09:01:30 bellini CPU    0: Hot: hi:  186, btch:  31 usd:  72   Cold: hi:   62, btch:  15 usd:  18
Apr  3 09:01:30 bellini HighMem per-cpu:
Apr  3 09:01:30 bellini CPU    0: Hot: hi:  186, btch:  31 usd:  70   Cold: hi:   62, btch:  15 usd:  14
Apr  3 09:01:30 bellini Active:247556 inactive:48481 dirty:27 writeback:0 unstable:0
Apr  3 09:01:30 bellini free:47336 slab:5849 mapped:108336 pagetables:1018 bounce:0
Apr  3 09:01:30 bellini DMA free:5772kB min:68kB low:84kB high:100kB active:2224kB inactive:64kB present:16256kB pages_scanned:0 all_unreclaimable? no
Apr  3 09:01:30 bellini lowmem_reserve[]: 0 873 1507 1507
Apr  3 09:01:30 bellini Normal free:182420kB min:3744kB low:4680kB high:5616kB active:449244kB inactive:99460kB present:894080kB pages_scanned:0 all_unreclaimable? no
Apr  3 09:01:30 bellini lowmem_reserve[]: 0 0 5071 5071
Apr  3 09:01:30 bellini HighMem free:1152kB min:512kB low:1192kB high:1872kB active:538756kB inactive:94400kB present:649100kB pages_scanned:0 all_unreclaimable? no
Apr  3 09:01:30 bellini lowmem_reserve[]: 0 0 0 0
Apr  3 09:01:30 bellini DMA: 431*4kB 242*8kB 84*16kB 20*32kB 2*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 5772kB
Apr  3 09:01:30 bellini Normal: 26655*4kB 5841*8kB 949*16kB 94*32kB 68*64kB 43*128kB 2*256kB 1*512kB 0*1024kB 0*2048kB 0*4096kB = 182420kB
Apr  3 09:01:30 bellini HighMem: 0*4kB 0*8kB 0*16kB 4*32kB 6*64kB 1*128kB 2*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1152kB
Apr  3 09:01:30 bellini Swap cache: add 142720, delete 108392, find 40471/51544, race 0+0
Apr  3 09:01:30 bellini Free swap  = 3748932kB
Apr  3 09:01:30 bellini Total swap = 4008176kB
Apr  3 09:01:30 bellini Free swap:       3748932kB
Apr  3 09:01:30 bellini 392928 pages of RAM
Apr  3 09:01:30 bellini 163552 pages of HIGHMEM



- --
This message is transmitted using 100% recycled electrons.
- --
- ---------------------------------------------------------------------
Dr. Andreas K. Huettel                  tel. +31 15 27 88102  (univ.)
Molecular Electronics and Devices            +31 6 42527466  (mobile)
Kavli Institute of Nanoscience Delft
Delft University of Technology                 A.K.Huettel@tudelft.nl
PO Box 5046, 2600 GA Delft                          mail@akhuettel.de
The Netherlands                     http://www.akhuettel.de/research/
- ---------------------------------------------------------------------
Please use GNUPG or PGP for signed and encrypted email. My public key
can be found at  http://www.akhuettel.de/pgp_key.php

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)

iD8DBQFH9KMU3ao2Zwy3NWoRA0fdAKCfum8VwjSpfkfmfHNCW+bpG5wImwCdFRFL
40gzgzSaMzy6NqrxMDpWXy4=
=+miJ
-----END PGP SIGNATURE-----

Re: [tainted,fglrx] page allocation failure: basic meaning of log messages?

[KOSAKI Motohiro]

Hi

> Apr  3 09:01:30 bellini hal-system-powe: page allocation failure. order:10, mode:0x4020
> Apr  3 09:01:30 bellini Pid: 327, comm: hal-system-powe Tainted: PF       2.6.24-gentoo-r3 #1
> Apr  3 09:01:30 bellini [<c0140c48>] __alloc_pages+0x2bd/0x2ce
> Apr  3 09:01:30 bellini [<c0140c92>] __get_free_pages+0x39/0x47
> Apr  3 09:01:30 bellini [<f8adcd9e>] firegl_cmmqs_save_fb+0x8e/0x200 [fglrx]

this messages doesn't mean bug, merely indicate allocation failure.
if any other bad thind doen't happend, you can ignore it.

Re: [tainted,fglrx] page allocation failure: basic meaning of log messages?

[Andreas K. Huettel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160


Hi, 

#]> Apr  3 09:01:30 bellini hal-system-powe: page allocation failure. order:10, mode:0x4020
#]> Apr  3 09:01:30 bellini Pid: 327, comm: hal-system-powe Tainted: PF       2.6.24-gentoo-r3 #1
#]> Apr  3 09:01:30 bellini [<c0140c48>] __alloc_pages+0x2bd/0x2ce
#]> Apr  3 09:01:30 bellini [<c0140c92>] __get_free_pages+0x39/0x47
#]> Apr  3 09:01:30 bellini [<f8adcd9e>] firegl_cmmqs_save_fb+0x8e/0x200 [fglrx]
#]
#]this messages doesn't mean bug, merely indicate allocation failure.
#]if any other bad thind doen't happend, you can ignore it.

Swapping like mad without any obvious cause (500M swap in use, 1.5G RAM 
- - which is normally not such a big problem). This however stopped when 
I closed some memory-intensive applications (firefox, vmware).

Since then the machine is running without problems. 

Thanks for the info, 
Andreas



- --
This message is transmitted using 100% recycled electrons.
- --
- ---------------------------------------------------------------------
Dr. Andreas K. Huettel                  tel. +31 15 27 88102  (univ.)
Molecular Electronics and Devices            +31 6 42527466  (mobile)
Kavli Institute of Nanoscience Delft
Delft University of Technology                 A.K.Huettel@tudelft.nl
PO Box 5046, 2600 GA Delft                          mail@akhuettel.de
The Netherlands                     http://www.akhuettel.de/research/
- ---------------------------------------------------------------------
Please use GNUPG or PGP for signed and encrypted email. My public key
can be found at  http://www.akhuettel.de/pgp_key.php

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)

iD8DBQFH9L4I3ao2Zwy3NWoRA90bAKCVUgen8y3P6xBKcAPIZSARE1VudQCgoklk
X+V4eXmg7Tk5pikhui/zCeI=
=pEk1
-----END PGP SIGNATURE-----

Re: [tainted,fglrx] page allocation failure: basic meaning of log messages?

[Lennart Sorensen]

On Thu, Apr 03, 2008 at 01:22:40PM +0200, Andreas K. Huettel wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: RIPEMD160
> 
> 
> Hi, 
> 
> #]> Apr  3 09:01:30 bellini hal-system-powe: page allocation failure. order:10, mode:0x4020
> #]> Apr  3 09:01:30 bellini Pid: 327, comm: hal-system-powe Tainted: PF       2.6.24-gentoo-r3 #1
> #]> Apr  3 09:01:30 bellini [<c0140c48>] __alloc_pages+0x2bd/0x2ce
> #]> Apr  3 09:01:30 bellini [<c0140c92>] __get_free_pages+0x39/0x47
> #]> Apr  3 09:01:30 bellini [<f8adcd9e>] firegl_cmmqs_save_fb+0x8e/0x200 [fglrx]
> #]
> #]this messages doesn't mean bug, merely indicate allocation failure.
> #]if any other bad thind doen't happend, you can ignore it.
> 
> Swapping like mad without any obvious cause (500M swap in use, 1.5G RAM 
> - - which is normally not such a big problem). This however stopped when 
> I closed some memory-intensive applications (firefox, vmware).
> 
> Since then the machine is running without problems. 

firefox 2.x tends to leak memory like crazy in my experience, so it may
be perfectly normal.  If you have adobe's flash plugin firefox can leak
very quickly too.

-- 
Len Sorensen

Re: [tainted,fglrx] page allocation failure: basic meaning of log messages?

[Peter Zijlstra]

On Thu, 2008-04-03 at 10:48 -0400, Lennart Sorensen wrote:
> On Thu, Apr 03, 2008 at 01:22:40PM +0200, Andreas K. Huettel wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: RIPEMD160
> > 
> > 
> > Hi, 
> > 
> > #]> Apr  3 09:01:30 bellini hal-system-powe: page allocation failure. order:10, mode:0x4020
> > #]> Apr  3 09:01:30 bellini Pid: 327, comm: hal-system-powe Tainted: PF       2.6.24-gentoo-r3 #1
> > #]> Apr  3 09:01:30 bellini [<c0140c48>] __alloc_pages+0x2bd/0x2ce
> > #]> Apr  3 09:01:30 bellini [<c0140c92>] __get_free_pages+0x39/0x47
> > #]> Apr  3 09:01:30 bellini [<f8adcd9e>] firegl_cmmqs_save_fb+0x8e/0x200 [fglrx]
> > #]
> > #]this messages doesn't mean bug, merely indicate allocation failure.
> > #]if any other bad thind doen't happend, you can ignore it.
> > 
> > Swapping like mad without any obvious cause (500M swap in use, 1.5G RAM 
> > - - which is normally not such a big problem). This however stopped when 
> > I closed some memory-intensive applications (firefox, vmware).
> > 
> > Since then the machine is running without problems. 
> 
> firefox 2.x tends to leak memory like crazy in my experience, so it may
> be perfectly normal.  If you have adobe's flash plugin firefox can leak
> very quickly too.

An order 10 alloc (4 mb contiguous) isn't likely to succeed by any
measure. Lumpy reclaim will try very hard and likely cause that swap
storm you taked about, but relying on order 10 allocs is insane.

Re: [tainted,fglrx] page allocation failure: basic meaning of log messages?

[Andreas K. Huettel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160


#]> > #]> Apr  3 09:01:30 bellini hal-system-powe: page allocation failure. order:10, mode:0x4020
#]> > #]> Apr  3 09:01:30 bellini Pid: 327, comm: hal-system-powe Tainted: PF       2.6.24-gentoo-r3 #1
#]> > #]> Apr  3 09:01:30 bellini [<c0140c48>] __alloc_pages+0x2bd/0x2ce
#]> > #]> Apr  3 09:01:30 bellini [<c0140c92>] __get_free_pages+0x39/0x47
#]> > #]> Apr  3 09:01:30 bellini [<f8adcd9e>] firegl_cmmqs_save_fb+0x8e/0x200 [fglrx]
#]> > #]
#]> > #]this messages doesn't mean bug, merely indicate allocation failure.
#]> > #]if any other bad thind doen't happend, you can ignore it.
#]> > 
#]> > Swapping like mad without any obvious cause (500M swap in use, 1.5G RAM 
#]> > - - which is normally not such a big problem). This however stopped when 
#]> > I closed some memory-intensive applications (firefox, vmware).
#]> > 
#]> > Since then the machine is running without problems. 
(...)
#]
#]An order 10 alloc (4 mb contiguous) isn't likely to succeed by any
#]measure. Lumpy reclaim will try very hard and likely cause that swap
#]storm you taked about, but relying on order 10 allocs is insane.


Question - this seems to me not to be a kernel issue (correct me if I'm 
wrong) and thus I'd like to take it to the proper place...
Where should I file this as bug? Is this a hald problem or something more 
complicated?

The swap storm is perfectly reproducible with X/fglrx, compiz and vmware 
workstation running - after resume from suspend2ram it starts. No compiz - 
no problem. No vmware - no problem... Closing vmware -> swapping stops.

If you think this is hopeless since too much closed source software is 
involved, tell me... :-]


Thanks, Andreas




- --
This message is transmitted using 100% recycled electrons.
- --
- ---------------------------------------------------------------------
Dr. Andreas K. Huettel                  tel. +31 15 27 88102  (univ.)
Molecular Electronics and Devices            +31 6 42527466  (mobile)
Kavli Institute of Nanoscience Delft
Delft University of Technology                 A.K.Huettel@tudelft.nl
PO Box 5046, 2600 GA Delft                          mail@akhuettel.de
The Netherlands                     http://www.akhuettel.de/research/
- ---------------------------------------------------------------------
Please use GNUPG or PGP for signed and encrypted email. My public key
can be found at  http://www.akhuettel.de/pgp_key.php

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)

iD8DBQFH+z1f3ao2Zwy3NWoRA8/iAJ0bLbIFJ9FfHolzuw4eOsNalMhMQACfTBxm
IEKmTD381DwK2Atl4mJB/24=
=jvjl
-----END PGP SIGNATURE-----

Re: [tainted,fglrx] page allocation failure: basic meaning of log messages?

[Andi Kleen]

"Andreas K. Huettel" <A.K.Huettel@tudelft.nl> writes:

> Question - this seems to me not to be a kernel issue (correct me if I'm 
> wrong) and thus I'd like to take it to the proper place...
> Where should I file this as bug? Is this a hald problem or something more 
> complicated?

It is a firegl bug. You should report it to ATI

-Andi