Hi,
(I'm not sure this is the right list, so please point me in the right
direction if it's not.)
I've been using Guido's pmdisk patch with...a fair amount of success,
but it occasionally oopses on resume (even with
USB/dmasound/therm_adt746x modules unloaded), and fairly often takes a
looooooooong time to suspend (more then 3 minutes), sitting at the
"[nosave XXXX]" stage. Just now, though, it took a veeeery long time to
suspend, and when I came back the suspend had failed and dmesg had the
following error:
Stopping tasks: =========================================================================================================================================|
Freeing memory: ........|
eth0: suspending, WakeOnLan disabled
radeonfb: suspending to state: 3...
PM: Attempting to suspend to disk.
PM: snapshotting memory.
pmdisk: Counting pages to copy.
[nosave c02e3000]
pmdisk: (pages needed: 62434 + 512 free: 68637)
pwrctl-wrapper: page allocation failure. order:8, mode:0x120
Call trace:
[c000bd58] dump_stack+0x18/0x28
[c0045594] __alloc_pages+0x324/0x388
[c0045624] __get_free_pages+0x2c/0x60
[c003d7d8] alloc_pagedir+0x28/0x74
[c003da30] pmdisk_suspend+0x88/0x11c
[c000ff9c] pmdisk_arch_suspend+0xfc/0x110
[c003ca04] pm_suspend_disk+0x68/0x100
[c003bbc8] enter_state+0xd4/0xdc
[c003bd14] state_store+0xa4/0xb8
[c00a5114] subsys_attr_store+0x44/0x4c
[c00a5464] flush_write_buffer+0x38/0x48
[c00a54cc] sysfs_write_file+0x58/0x88
[c00634d0] vfs_write+0xdc/0x128
[c00635d0] sys_write+0x40/0x74
[c0007d00] ret_from_syscall+0x0/0x44
pmdisk: Allocating pagedir failed.
cpufreq: resume failed to assert current frequency is what timing core thinks it is.
PM: snapshot memory failed !
eth0: resuming
PHY ID: 4061e4, addr: 0
hda: Enabling Ultra DMA 5
hdc: MDMA, cycleTime: 120, accessTime: 90, recTime: 30
hdc: Set MDMA timing for mode 2, reg: 0x00011d26
hdc: Enabling MultiWord DMA 2
Restarting tasks... done
</quote>
My .config is attached, and yes, I do have CONFIG_PREEMPT enabled. I
have 512MB of RAM and 1GB of swap enabled, but < 1MB of swap was in use
when I triggered suspend. Oh yes, I'm using kernel.org 2.6.7 with
guido's 2.6.7-rc2 pmdisk patch and the cpufreq patch. DRI is disabled,
too.
--
Words of the day: mindwar CDC Blowfish TWA defense information warfare
On Thu, Jul 01, 2004 at 02:17:03PM +1000, Rob Weir wrote:
I've been using Guido's pmdisk patch with...a fair amount of success,
but it occasionally oopses on resume (even with
USB/dmasound/therm_adt746x modules unloaded), and fairly often takes a
Hmm...I've never seen this. I'll have a look.
looooooooong time to suspend (more then 3 minutes), sitting at the
I think this is a general problem of pmdisk, not powerpc specific, not
nice though.
Cheers,
-- Guido
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
On Fri, Jul 16, 2004 at 09:41:52AM +0200, Guido Guenther said
On Thu, Jul 01, 2004 at 02:17:03PM +1000, Rob Weir wrote:
quoted
I've been using Guido's pmdisk patch with...a fair amount
of success, but it occasionally oopses on resume (even with
USB/dmasound/therm_adt746x modules unloaded), and fairly often takes
Hmm...I've never seen this. I'll have a look.
Everytime I've resumed in the past week I've got an oops, then a series
of "out of memory" errors while things like hotplug restart...would it
be useful I wrote this oops down by hand?
quoted
a looooooooong time to suspend (more then 3 minutes), sitting at the
I think this is a general problem of pmdisk, not powerpc specific, not
nice though.
Hm, but *sometimes* (soon after I installed it, mostly) it suspends very
quickly.
--
Words of the day: Mantis UK Hacker encryption New World Order eavesdropping
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
On Fri, Jul 16, 2004 at 06:23:10PM +1000, Rob Weir wrote:
On Fri, Jul 16, 2004 at 09:41:52AM +0200, Guido Guenther said
quoted
On Thu, Jul 01, 2004 at 02:17:03PM +1000, Rob Weir wrote:
quoted
I've been using Guido's pmdisk patch with...a fair amount
of success, but it occasionally oopses on resume (even with
USB/dmasound/therm_adt746x modules unloaded), and fairly often
Hmm...I've never seen this. I'll have a look.
Everytime I've resumed in the past week I've got an oops, then
a series of "out of memory" errors while things like hotplug
restart...would it be useful I wrote this oops down by hand?
It wouldn't hurt. Are they all the same? Since when exactly does this
happen? Did you change anything since then?
Cheers,
-- Guido
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
On Fri, Jul 16, 2004 at 11:20:04AM +0200, Guido Guenther said
On Fri, Jul 16, 2004 at 06:23:10PM +1000, Rob Weir wrote:
quoted
On Fri, Jul 16, 2004 at 09:41:52AM +0200, Guido Guenther said
quoted
On Thu, Jul 01, 2004 at 02:17:03PM +1000, Rob Weir wrote:
quoted
I've been using Guido's pmdisk patch with...a fair amount
of success, but it occasionally oopses on resume (even with
USB/dmasound/therm_adt746x modules unloaded), and fairly often
Hmm...I've never seen this. I'll have a look.
Everytime I've resumed in the past week I've got an oops, then
a series of "out of memory" errors while things like hotplug
restart...would it be useful I wrote this oops down by hand?
It wouldn't hurt. Are they all the same?
They all...look similar, but I've never written them down so I'm not
sure. It starts with a "order-0 allocation failure" or so, then
programs start failing to allocate memory, too. I'll write it down next
time it happens (I've taken to just shutting down when moving it, since
this happens so often).
Since when exactly does this happen?
On most resumes.
Did you change anything since then?
I do a dist-upgrade against sid every few days, but I haven't changed
the kernel in weeks. Hm...I'm using your 2.6.7-rc3 pmdisk patch against
2.6.7 itself, maybe I should just update to your 2.6.7 one and 2.6.8-rcX
before bothering you further...
--
Words of the day: diwn clones CDC SSL AMEMB 22nd SAS colonel cypher=
punk
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
On Fri, Jul 16, 2004 at 07:40:49PM +1000, Rob Weir wrote:
I do a dist-upgrade against sid every few days, but I haven't changed
the kernel in weeks. Hm...I'm using your 2.6.7-rc3 pmdisk patch
against 2.6.7 itself, maybe I should just update to your 2.6.7 one and
2.6.8-rcX before bothering you further...
Just for completeness: we tracked this down do be CONFIG_PREEMPT
related. Without that it works fine on 2.6.8-rc2.
Cheers,
-- Guido
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/