The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

nous · 2011-12-25 22:50:17

aakos23 wrote:

:: Starting Open Sound System                                                                                           [BUSY] Relinking OSS kernel modules for "3.1-pf SMP preempt mod_unload modversions 686 "
This may take few moments - please stand by...
Relinking OSS kernel modules finished

Error: OSS driver modules do not match the current kernel

Kernel vermagic:  3.1-pf SMP preempt mod_unload modversions 686
OSS    vermagic:  3.1-pf SMP preempt mod_unload modversions ATOM

The most likely reason is that the kernel development package
installed in the system is wrong. Please refer to the documentation
of your Linux distribution for more info about setting up the
kernel/driver development environment properly.

                                                                                                                        [FAIL]

pfkernel/linux-pf-atom 3.1.6-1 (base)
pfkernel/linux-pf-core2 3.1.6-1 (base)

OSS not working.

Didn't try the previous version, but I uploaded the latest at my repo and it works fine on 3.1.6.

eXine · 2011-12-26 20:03:13

gee wrote:

gee wrote:
When building aufs3-util against latest kernel I get this error:
In file included from ver.c:19:0:
/lib/modules/3.1-pf/build/include/linux/aufs_type.h:170:27: error: expected ‘:’, ‘,’, ‘;’, ‘}’ or ‘__attribute__’ before ‘*’ token

When searching for it online, I found this:
http://comments.gmane.org/gmane.linux.f … .user/3241
which makes me think that there was an issue when the kernel sources were patched or something like that.
Anyone else getting this?
Noone else getting this or using aufs?

I have it too. It's very strange and I haven't fixed it yet, but at least you aren't alone

post-factum · 2011-12-31 19:08:24

aufs is not a part of pf-kernel.

gee · 2011-12-31 21:41:44

eXine wrote:

I have it too. It's very strange and I haven't fixed it yet, but at least you aren't alone

Phlogiston · 2012-01-08 22:31:43

nous wrote:

kalpik wrote:
I cannot resume tuxonice on 3.1 I remember reading tuxonice was now built as a module? How do I load it in order to get it to work?
MODULES="... tuxonice_swap tuxonice_userui" in /etc/mkinitcpio.conf (userui being optional). Also, add 'resume' in HOOKS before 'filesystems'
[EDIT] Don't forget to run 'mkinitcpio linux-pf' afterwards.

==> ERROR: Hook 'tuxonice_swap' can not be found.
==> ERROR: Hook 'tuxonice_userui' can not be found.

I'm having the same issues, no resume.

[ 1.254313] PM: Checking hibernation image partition /dev/sda2
[ 1.254315] PM: Hibernation image not present or could not be loaded.
[ 2.438078] PM: Starting manual resume from disk
[ 2.438081] PM: Hibernation image partition 8:2 present
[ 2.438082] PM: Looking for hibernation image.
[ 2.438254] PM: Image not found (code -22)
[ 2.438256] PM: Hibernation image not present or could not be loaded.

Last edited by Phlogiston (2012-01-08 22:37:03)

nous · 2012-01-09 16:26:25

Phlogiston wrote:

nous wrote:
kalpik wrote:
I cannot resume tuxonice on 3.1 I remember reading tuxonice was now built as a module? How do I load it in order to get it to work?
MODULES="... tuxonice_swap tuxonice_userui" in /etc/mkinitcpio.conf (userui being optional). Also, add 'resume' in HOOKS before 'filesystems'
[EDIT] Don't forget to run 'mkinitcpio linux-pf' afterwards.
==> ERROR: Hook 'tuxonice_swap' can not be found.
==> ERROR: Hook 'tuxonice_userui' can not be found.

I'm having the same issues, no resume.

[ 1.254313] PM: Checking hibernation image partition /dev/sda2
[ 1.254315] PM: Hibernation image not present or could not be loaded.
[ 2.438078] PM: Starting manual resume from disk
[ 2.438081] PM: Hibernation image partition 8:2 present
[ 2.438082] PM: Looking for hibernation image.
[ 2.438254] PM: Image not found (code -22)
[ 2.438256] PM: Hibernation image not present or could not be loaded.

The error message indicates you put tuxonice_* in the HOOKS array, not MODULES?

sxe · 2012-01-23 13:10:23

Hey guys,
i tried mounting my tablet and got this error lately (it worked before):

sxe@main ~ sudo mount /dev/sdc test/
mount: unknown filesystem type 'vfat'

 sxe@main ~ $ sudo modprobe vfat
FATAL: Module vfat not found.

Any idea?

I had this before on the stock kernel and a reboot fixed it but not now.

nous · 2012-01-23 20:13:04

sxe wrote:

Hey guys,
i tried mounting my tablet and got this error lately (it worked before):
sxe@main ~ sudo mount /dev/sdc test/
mount: unknown filesystem type 'vfat'

 sxe@main ~ $ sudo modprobe vfat
FATAL: Module vfat not found.
Any idea?
I had this before on the stock kernel and a reboot fixed it but not now.

Which arch/version/cpu? Here on x86_64/3.2.1-4/AMD-K8 all is well:

hyperion:[root]:~# lsmod|grep vfat
hyperion:[root]:~# modprobe vfat
hyperion:[root]:~# lsmod|grep vfat
vfat                   10055  0 
fat                    48962  1 vfat
hyperion:[root]:~# uname -a
Linux hyperion 3.2-pf #1 SMP PREEMPT Fri Jan 20 00:44:38 EET 2012 x86_64 AMD Athlon(tm) II X2 250 Processor AuthenticAMD GNU/Linux
hyperion:[root]:~# pacman -Qo /boot/vmlinuz-linux-pf 
/boot/vmlinuz-linux-pf belongs to linux-pf-k8 3.2.1-4

nous · 2012-01-24 07:19:42

hyperion:[nous]:/tmp% ./mempodipper 
===============================
= Mempodipper =
= by zx2c4 =
= Jan 21, 2012 =
===============================

[+] Opening socketpair.
[+] Executing child from child fork.
[+] Waiting for transferred fd in parent.
[+] Opening parent mem /proc/23140/mem in child.
[+] Sending fd 5 to parent.
[+] Received fd at 5.
[+] Assigning fd 5 to stderr.
[+] Reading gpasswd for exit@plt.
[+] Resolved exit@plt to 0x4018a0.
[+] Calculating gpasswd padding.
[+] Seeking to offset 0x401890.
[+] Executing gpasswd with shellcode.
sh-4.2#

Please, everyone UPDATE to 3.2.1-5

http://blog.zx2c4.com/749

nous · 2012-01-26 18:07:01

Just a little show-off from my laptop, running kernel26-pf-2.6.39.3 (yes, the one where tuxonice/lzo last worked). The numbers would have been higher, had I not tried a more recent, broken linux-pf.

94 hibernation cycles.

orion:[nous]:/temp/pkgbuild% dmesg | tail -27
PM: restore of devices complete after 616.143 msecs
...20%...40%...60%...80%
Waited for i/o due to readahead not ready 1373 times.
Waited for i/o due to throughput_throttle 1284 times.
Restarting all filesystems ...
Restarting tasks ... 
usb 2-3: USB disconnect, device number 14
done.
TuxOnIce debugging info:
- TuxOnIce core  : 3.2
- Kernel Version : 2.6.39-pf
- Compiler vers. : 4.6
- Attempt number : 94
- Parameters     : 0 667656 0 1 -2 5
- Overall expected compression percentage: 0.
- Compressor is 'lzo'.
  Compressed 2027749376 bytes into 740040618 (63 percent compression).
- Block I/O active.
  Used 182561 pages from swap on /dev/sda8.
- Max outstanding reads 1391. Max writes 1561.
  Memory_needed: 1024 x (4096 + 336 + 104) = 4644864 bytes.
  Free mem throttle point reached 0.
- Swap Allocator enabled.
  Swap available for image: 634558 pages.
- I/O speed: Write 123 MB/s, Read 146 MB/s.
- Extra pages    : 890 used/7500.
- Result         : Succeeded.
orion:[nous]:/temp/pkgbuild% uptime; uname -a
 20:05:53 up 102 days, 5 min,  6 users,  load average: 0.54, 0.60, 0.64
Linux orion 2.6.39-pf #1 SMP PREEMPT Sat Oct 1 19:41:38 EEST 2011 x86_64 Intel(R) Core(TM)2 Duo CPU T7250 @ 2.00GHz GenuineIntel GNU/Linux

nous · 2012-01-30 16:18:17

FYI, tuxonice w/ lzo still croaks 'error -22' on resume.

kokoko3k · 2012-01-30 17:34:06

Anybody using standard (not toi) s2disk method?
With stock arch kernel it works, while with pf it fails 1 on every 3..4 attempts, it starts to write the image but then hangs at random percentages, while writing.
Last tested with linux-pf-atom 3.2.1-4, stock arch kernel is at the same version.

Any advices? hardware is an eeepc 1005ha (i915 chipset)

nous · 2012-01-30 17:53:21

kokoko3k wrote:

Anybody using standard (not toi) s2disk method?
With stock arch kernel it works, while with pf it fails 1 on every 3..4 attempts, it starts to write the image but then hangs at random percentages, while writing.
Last tested with linux-pf-atom 3.2.1-4, stock arch kernel is at the same version.
Any advices? hardware is an eeepc 1005ha (i915 chipset)

I'm an obligatory user of the mainline kernel s2disk on a box with less swap than RAM, where compression is mandatory. It works quite reliably (I have to close memory hogs first though, otherwise s2disk fails and kills Xorg) but I prefer tuxonice over uswsup any given day.

kokoko3k · 2012-02-05 17:11:12

...Maybe i was unlucky, but after reading your message, today i tried toi on 3.2.2-pf-atom.
It worked a couple of times, next it failed to resume with a beautiful segfault.
I cant use any hibernation method with *-pf, how bad
Edit: 3.2.3, another one:

filed:
http://bugzilla.tuxonice.net/show_bug.cgi?id=482

Last edited by kokoko3k (2012-02-05 18:21:50)

nous · 2012-02-06 22:49:36

kokoko3k wrote:

...Maybe i was unlucky, but after reading your message, today i tried toi on 3.2.2-pf-atom.
It worked a couple of times, next it failed to resume with a beautiful segfault.
I cant use any hibernation method with *-pf, how bad
Edit: 3.2.3, another one:
http://ompldr.org/tY24yeQ
filed:
http://bugzilla.tuxonice.net/show_bug.cgi?id=482

It sucks, I know. Even the builtin hibernation broke with 3.2; my home box fails to hibernate 1 out of 5 times or so.

gee · 2012-02-19 21:47:12

.

Last edited by gee (2012-02-21 03:40:05)

Lockheed · 2012-02-19 22:30:30

This situation sucks. I cannot use normal hibernation, because it crashes on resume and no-one knows why.
I cannot use Tux hibernation because LZO support is broken, LZF doesn't work either, and hibernating without compression is too much of a waste.

When can we expect some solution to the LZO problem?

If I want to compile my own kernel 3.x with TuxOnIce support, how do I do that? All I can see is patches for kernels 2.6.x

Last edited by Lockheed (2012-02-19 22:46:40)

Lockheed · 2012-02-20 07:24:14

I found this a s a possible solution to this problem:
http://lists.tuxonice.net/pipermail/tux … 07041.html

Out of interest, I've tried to get tuxonice and Con Kolivas' BFS patch
running (linux-3.1-rc2). As long as compression is not used, everything
works fine. As soon as I turn on compression (lzo or lzf), I get some
errors -22 on resuming (or some byte count mismatch with lzf).
I can get tuxonice + compression + bfs to work correctly when I revert
the following two commits (and leave no_flusher_thread = 1):
3eaa98fda1f9d6727ab2862bf83c9bcfd8c83789: Fix progress update with
flusher thread.
c2a116d1236344e6ce2556bbc674489ee5d2f703: Start/stop I/O worker threads
outside of freezer
Actually, it's c2a116d which causes the problem. I think the commit
should fix the hangs when using no_flusher_thread=0?

Unfortunately, I know too little about kernels to understand what exactly is being said there.
If this actually can be a solution, can someone explain what should be done to apply it?

nous · 2012-02-20 07:43:23

Lockheed wrote:

This situation sucks. I cannot use normal hibernation, because it crashes on resume and no-one knows why.
I cannot use Tux hibernation because LZO support is broken, LZF doesn't work either, and hibernating without compression is too much of a waste.
When can we expect some solution to the LZO problem?
If I want to compile my own kernel 3.x with TuxOnIce support, how do I do that? All I can see is patches for kernels 2.6.x

As far as I know, BFS and TOI don't work together with LZO since 2.6.39-pf4. Me and postfactum tried (well, not too hard, admittedly) to pinpoint the guilty commit but failed miserably. I use kernel26-pf for my laptop with most excellent results (see https://bbs.archlinux.org/viewtopic.php … 8#p1047938), you can still find it precompiled at the [pfkernel] repo.

If you absolutely want to use the latest linux-pf and BFS/LZO, you can easily so: just use your favourite AUR helper to get the PKGBUILD and, when prompted, select the menuconfig option, General setup ---> [ * ] BFS cpu scheduler (deselect it), exit saving configuration.

Last edited by nous (2012-02-20 07:44:40)

nous · 2012-02-20 07:47:33

Lockheed wrote:

I found this a s a possible solution to this problem:
http://lists.tuxonice.net/pipermail/tux … 07041.html
Out of interest, I've tried to get tuxonice and Con Kolivas' BFS patch
running (linux-3.1-rc2). As long as compression is not used, everything
works fine. As soon as I turn on compression (lzo or lzf), I get some
errors -22 on resuming (or some byte count mismatch with lzf).
I can get tuxonice + compression + bfs to work correctly when I revert
the following two commits (and leave no_flusher_thread = 1):
3eaa98fda1f9d6727ab2862bf83c9bcfd8c83789: Fix progress update with
flusher thread.
c2a116d1236344e6ce2556bbc674489ee5d2f703: Start/stop I/O worker threads
outside of freezer
Actually, it's c2a116d which causes the problem. I think the commit
should fix the hangs when using no_flusher_thread=0?
Unfortunately, I know too little about kernels to understand what exactly is being said there.
If this actually can be a solution, can someone explain what should be done to apply it?

I'll give it a try as soon as I get back from work. Oleksandr, have you tried this?

Last edited by nous (2012-02-20 07:48:09)

post-factum · 2012-02-20 09:10:12

nous wrote:

I'll give it a try as soon as I get back from work. Oleksandr, have you tried this?

I'll try it ASAP and before next pf-kernel release.

Lockheed · 2012-02-20 09:53:12

Good to hear it might work.
If you find it works, will I have to wait till the next pf-kernel release to take advantage of it?

nous wrote:

If you absolutely want to use the latest linux-pf and BFS/LZO, you can easily so: just use your favourite AUR helper to get the PKGBUILD and, when prompted, select the menuconfig option, General setup ---> [ * ] BFS cpu scheduler (deselect it), exit saving configuration.

If I do that, and then install newer linux-pf with working BFS/TOI by copying kernel config from existing kernel, will it be enough just to change this setting back to ON, or does switching it OFF affect some other settings that will need to be reverted?

kokoko3k · 2012-02-20 10:15:07

Is that supposed to fix even this: https://bbs.archlinux.org/viewtopic.php … 0#p1052960
or just the compresison?

alecmg · 2012-02-20 10:39:50

weird, I haven't had problems hibernating

then again I am not sure if I even use compression... where do I check?

Lockheed · 2012-02-20 11:09:05

alecmg wrote:

weird, I haven't had problems hibernating
then again I am not sure if I even use compression... where do I check?

/etc/hibernate/tuxonice.conf
/etc/hibernate/hibernate.conf

Arch Linux

#301 2011-12-25 22:50:17

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#302 2011-12-26 20:03:13

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#303 2011-12-31 19:08:24

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#304 2011-12-31 21:41:44

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#305 2012-01-08 22:31:43

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#306 2012-01-09 16:26:25

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#307 2012-01-23 13:10:23

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#308 2012-01-23 20:13:04

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#309 2012-01-24 07:19:42

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#310 2012-01-26 18:07:01

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#311 2012-01-30 16:18:17

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#312 2012-01-30 17:34:06

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#313 2012-01-30 17:53:21

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#314 2012-02-05 17:11:12

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#315 2012-02-06 22:49:36

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#316 2012-02-19 21:47:12

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#317 2012-02-19 22:30:30

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#318 2012-02-20 07:24:14

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#319 2012-02-20 07:43:23

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#320 2012-02-20 07:47:33

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#321 2012-02-20 09:10:12

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#322 2012-02-20 09:53:12

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#323 2012-02-20 10:15:07

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#324 2012-02-20 10:39:50

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#325 2012-02-20 11:09:05

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

Board footer