You are not logged in.
After updating to the latest kernel26 package (2.6.34.2) on my Acer Extensa 5235 laptop, the boot process crashes almost immediately after grub has run. The funny thing is that the fallback kernel works, but seems to show the same messages in 'dmesg' as a warning:
WARNING: at mm/page_alloc.c:1828 __alloc_pages_nodemask+0x47e/0x580()
Hardware name: Extensa 5235
Modules linked in:
Pid: 1, comm: swapper Not tainted 2.6.34-ARCH #1
Call Trace:
[<c10430fd>] warn_slowpath_common+0x6d/0xa0
[<c10c214e>] ? __alloc_pages_nodemask+0x47e/0x580
[<c10c214e>] ? __alloc_pages_nodemask+0x47e/0x580
[<c1043145>] warn_slowpath_null+0x15/0x20
[<c10c214e>] __alloc_pages_nodemask+0x47e/0x580
[<c12cac1c>] ? acpi_os_map_memory+0x11/0x25
[<c11d748b>] ? acpi_ex_system_memory_space_handler+0x10b/0x1fd
[<c10c2267>] __get_free_pages+0x17/0x30
[<c10e9ac5>] __kmalloc+0xf5/0x1a0
[<c11d31fd>] ? acpi_ex_region_read+0x25/0x41
[<c11d3464>] acpi_ex_load_op+0x14b/0x287
[<c11d5298>] acpi_ex_opcode_1A_1T_0R+0x20/0x44
[<c11cde1d>] acpi_ds_exec_end_op+0xd2/0x371
[<c11de506>] acpi_ps_parse_loop+0x6d3/0x831
[<c11dda33>] acpi_ps_parse_aml+0x82/0x25f
[<c11dec33>] acpi_ps_execute_method+0x18b/0x220
[<c11da89c>] acpi_ns_evaluate+0xc0/0x176
[<c11da3b8>] acpi_evaluate_object+0x167/0x256
[<c11c77fc>] acpi_processor_set_pdc+0x25b/0x27e
[<c11c7a58>] early_init_pdc+0xaa/0xae
[<c11dbfa6>] acpi_ns_walk_namespace+0xb0/0x17a
[<c11c79ae>] ? early_init_pdc+0x0/0xae
[<c11da227>] acpi_walk_namespace+0x65/0x8f
[<c11c79ae>] ? early_init_pdc+0x0/0xae
[<c142ea28>] acpi_early_processor_set_pdc+0x28/0x2d
[<c11c79ae>] ? early_init_pdc+0x0/0xae
[<c142e663>] acpi_init+0xfb/0x2c1
[<c100120d>] do_one_initcall+0x2d/0x190
[<c142e568>] ? acpi_init+0x0/0x2c1
[<c140b960>] kernel_init+0x132/0x1cd
[<c140b82e>] ? kernel_init+0x0/0x1cd
[<c1003d3e>] kernel_thread_helper+0x6/0x18
These last 10 lines are pretty much what I see when I boot the normal kernel (I cannot scroll up or do anything else at that point). I have changed nothing in mkinitcpio, so I'm pretty baffled by this. I'd appreciate any pointers on how to solve this.
Offline
i got the same problem...
acpi=off in grub boot command solve this problem temporary.
Offline
it would be nice to get help in what to do now, cause i need a full running system for work.
Last edited by lenzy (2010-08-05 18:39:00)
Offline
Doesn't the fallback option work for you?
Offline
I've played around a bit with the HOOKS setting in mkinitcpio.conf:
HOOKS="base udev sata filesystems" --> WORKS
HOOKS="base udev sata" --> CRASH
HOOKS="base udev sata autodetect filesystems" --> CRASH
I don't understand this. How can a missing module from 'filesystems' in the initial ramdisk cause a crash? Especially since the above crash log seems to be related to ACPI?
Anyway, these are the filesystem modules added from autodetect: mbcache, jbd2, crc16, ext4.
I could now add the other modules from 'filesystems' one by one and see where it stops crashing, but this would take ages... maybe someone around here has a pointer which one might be the culprit.
Offline
my fallback option doesn't work for me..
Offline
Then it seems we don't exactly have the same problem. Fallback works fine for me. You could either try to downgrade the kernel package (see http://wiki.archlinux.org/index.php/Dow … _Packages), or you could try switching to the kernel26-lts package.
Regarding mkinitcpio, I've fiddled a bit more and the filesystem modules have nothing to do with it (wouldn't have made sense anyway). It seems it all depends on the size of the initial ramdisk - if it's too small, the kernel boot crashes, so I'm suspecting some memory problem - maybe a faulty chip or still a kernel bug. Anyway, I'll run memcheck for a few hours and see what happens.
Offline
which bios version do u have'?
Offline
I had 3219, then upgraded to 3311. I had the same crash with both.
Offline
mysterious..
have u tried the testing kernel?
Last edited by lenzy (2010-08-05 21:36:09)
Offline
No. I'll just stick with the fallback kernel for now. I suspect it's some weird ACPI thing; in fact, ACPI has always been a bit unreliable on my Extensa. Sometimes it suddenly looses track of the battery, for example. Also, I sometimes saw messages like this
ACPI Exception: AE_AML_BUFFER_LIMIT, Index (FFFFFFFFFFFFFFFF) is beyond end of object (20100121/exoparg2-445)
when changing screen brightness. Maybe there's some acpi_osi setting which can cure this; at the moment I don't have time to deal with this further.
Offline
A little update: kernel26-lts (2.6.32) is booting fine; maybe the other ACPI issues I've had are also fixed by downgrading to LTS - I'll have to use it a few days to be sure. The only little issue I experienced was that you cannot toggle (mute) the headphone (you can still set its volume to 0, so this is just a little annoyance).
Kernel from testing (2.6.35) shows the same crash as 2.6.34.
So currently, I'd say that Extensa 5235 users should better use the LTS kernel.
Offline
Does it work with 2.6.34.1, or did you upgrade from an earlier version? There have been some patches related to ACPI in .2 (changelog), if the problem persists you should check if one of them introduced the regression. I got a EX 5635, which has the same Bios - haven't upgraded yet, but will report back how it went.
Offline
i got the same problem with version 2.6.35, so the problem still persists for me.
Offline
Yea it would be good to know with which version the problem started...
If 2.6.34.1 worked, you could try compiling the snapshot from this commit with the PKGCONFIG from arch svn, it is right before the first commit concerning ACPI is applied. See here for the shortlog.
Last edited by hokasch (2010-08-07 09:49:45)
Offline
P.S.: lenzy, do you have the same/similar hardware as the OP? Another thing - do you have ext4?
Offline
I downgraded to 2.6.34.1 and it's booting fine; the problem started with 2.6.34.2. I tried to bisect this issue and I think it starts with this commit, which was introduced to fix this bug. The Extensa 5235 is single core and does not support cpufreq, so I'm a bit puzzled why this change would crash my machine.
Offline
I verified that removing that patch from the 2.6.34.2 kernel makes it boot again on my machine. I posted a comment in the above thread in the kernel bugtracker.
Offline
does it work now with testing or core?
Offline
still the same problem with 2.6.35.4-1 from core.
Offline
My laptop is booting again with 2.6.36rc3 (available via AUR, package kernel26-mainline; binaries available at miffe.org). Kernel boots, but some AE_NO_MEMORY errors are now appearing, which still seem to be important, so use this kernel with care. More details in this thread: https://bugzilla.kernel.org/show_bug.cgi?id=16548.
Offline
Seems to work for me too with lattest Kernel package (2.6.36.3-1). I have not noticed anything strange so far.
Offline
sorry, wrong thread
Last edited by luuuciano (2011-01-12 14:58:43)
I arch, you arch, he arch, she arch, we arch, they arch...
Offline
I can confirm this. altough it started to happen since kernel26 2.6.36.3-1 update yesterday. But the most weird thing is that lately those problems went away, altough i didnt do anything, my system now boots fine with no errors. anyone know why this has happened?
Offline
this looks related to this one?
https://bbs.archlinux.org/viewtopic.php?pid=876909
or that one to this... (?)
I arch, you arch, he arch, she arch, we arch, they arch...
Offline