Skip to contentSkip to conversation

hald in status D with 2.6.16-rc4

DISCUSSION Hesse, Christian 3 participants 12 messages Feb 20, 2006 – Feb 28, 2006 ACPI

lore
Share & tools
lore
#0 lore raw

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

Signature · 2 lines
Regards,
Christian
#1 lore raw
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Thanks.

#2 lore raw
Quoted 8 lines · Andrew Morton
Quoted 5 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

hald D E0B50480 0 7791 1 7797 10654 7609 (NOTLB)
cc6cbccc e0b50480 000f4428 e0b50480 000f4428 c6c9605c c14dce00 c14dce00

       e0b50480 000f4428 cc3df530 dff6e5c0 cc3df530 00000296 dff6e5c8 c046f202
       00000001 cc3df530 c0115680 d7ea1ce0 dff6e5c8 00000003 00000001 00000000
Call Trace:
 [<c046f202>] __down+0x62/0xc0
 [<c0115680>] default_wake_function+0x0/0x20
 [<c046dd3f>] __down_failed+0x7/0xc
 [<c0288bbf>] .text.lock.osl+0x13/0x3c
 [<c0292678>] acpi_ex_system_wait_semaphore+0x34/0x48
 [<c028d5da>] acpi_ev_acquire_global_lock+0x67/0x6c
 [<c0293fdc>] acpi_ex_acquire_global_lock+0x14/0x3b
 [<c028f578>] acpi_ex_read_data_from_field+0x114/0x14b
 [<c029475b>] acpi_ex_resolve_node_to_value+0x123/0x1ac
 [<c028fdc2>] acpi_ex_resolve_to_value+0x5e/0x69
 [<c02923df>] acpi_ex_resolve_operands+0x277/0x4dc
 [<c028a4fd>] acpi_ds_exec_end_op+0xab/0x36e
 [<c0298fe6>] acpi_ps_parse_loop+0x5ba/0x8bc
 [<c0298881>] acpi_ps_parse_aml+0x4e/0x1f9
 [<c02998fb>] acpi_ps_execute_pass+0x72/0x83
 [<c0299824>] acpi_ps_execute_method+0x54/0x7d
 [<c0296c5f>] acpi_ns_execute_control_method+0x5a/0x67
 [<c0296bee>] acpi_ns_evaluate_by_handle+0x73/0x8a
 [<c0296aee>] acpi_ns_evaluate_relative+0xaa/0xc6
 [<c0296375>] acpi_evaluate_object+0x139/0x1fb
 [<c024e2e2>] copy_to_user+0x42/0x60
 [<c02a0107>] acpi_battery_get_status+0x6b/0x11c
 [<c02a056b>] acpi_battery_read_state+0x52/0x185
 [<c01832d8>] seq_read+0xe8/0x2f0
 [<c0162dba>] vfs_read+0xaa/0x1a0
 [<c01631d1>] sys_read+0x51/0x80
 [<c0102b5f>] sysenter_past_esp+0x54/0x75

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Signature · 1 line
Christian
#3 lore raw
Quoted 51 lines · Christian
Quoted 8 lines · Andrew Morton
Quoted 5 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

hald D E0B50480 0 7791 1 7797 10654 7609 (NOTLB)
cc6cbccc e0b50480 000f4428 e0b50480 000f4428 c6c9605c c14dce00 c14dce00

       e0b50480 000f4428 cc3df530 dff6e5c0 cc3df530 00000296 dff6e5c8 c046f202
       00000001 cc3df530 c0115680 d7ea1ce0 dff6e5c8 00000003 00000001 00000000
Call Trace:
 [<c046f202>] __down+0x62/0xc0
 [<c0115680>] default_wake_function+0x0/0x20
 [<c046dd3f>] __down_failed+0x7/0xc
 [<c0288bbf>] .text.lock.osl+0x13/0x3c
 [<c0292678>] acpi_ex_system_wait_semaphore+0x34/0x48
 [<c028d5da>] acpi_ev_acquire_global_lock+0x67/0x6c
 [<c0293fdc>] acpi_ex_acquire_global_lock+0x14/0x3b
 [<c028f578>] acpi_ex_read_data_from_field+0x114/0x14b
 [<c029475b>] acpi_ex_resolve_node_to_value+0x123/0x1ac
 [<c028fdc2>] acpi_ex_resolve_to_value+0x5e/0x69
 [<c02923df>] acpi_ex_resolve_operands+0x277/0x4dc
 [<c028a4fd>] acpi_ds_exec_end_op+0xab/0x36e
 [<c0298fe6>] acpi_ps_parse_loop+0x5ba/0x8bc
 [<c0298881>] acpi_ps_parse_aml+0x4e/0x1f9
 [<c02998fb>] acpi_ps_execute_pass+0x72/0x83
 [<c0299824>] acpi_ps_execute_method+0x54/0x7d
 [<c0296c5f>] acpi_ns_execute_control_method+0x5a/0x67
 [<c0296bee>] acpi_ns_evaluate_by_handle+0x73/0x8a
 [<c0296aee>] acpi_ns_evaluate_relative+0xaa/0xc6
 [<c0296375>] acpi_evaluate_object+0x139/0x1fb
 [<c024e2e2>] copy_to_user+0x42/0x60
 [<c02a0107>] acpi_battery_get_status+0x6b/0x11c
 [<c02a056b>] acpi_battery_read_state+0x52/0x185
 [<c01832d8>] seq_read+0xe8/0x2f0
 [<c0162dba>] vfs_read+0xaa/0x1a0
 [<c01631d1>] sys_read+0x51/0x80
 [<c0102b5f>] sysenter_past_esp+0x54/0x75
Quoted 1 line

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Could the ACPI team please take a look at fixing this regression?

#4 lore raw
Quoted 19 lines · Andrew Morton
Quoted 18 lines · Christian
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

Quoted 8 lines
Quoted 3 lines

[ trace snipped ]

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Could the ACPI team please take a look at fixing this regression?

I did two cycles with mainline suspend now and did not hit the problem... I will keep an eye on it.

Signature · 1 line
Christian
#5 lore raw
Quoted 24 lines · Christian
Quoted 23 lines · Andrew Morton
Quoted 21 lines · Christian
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

[ trace snipped ]

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Quoted 7 lines
Quoted 4 lines

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Could the ACPI team please take a look at fixing this regression?

I did two cycles with mainline suspend now and did not hit the problem... I will keep an eye on it.

It is independent from suspend version, hald hangs with mainline suspend as well.

Signature · 1 line
Christian
#9 lore raw
In reply to Hesse, Christian

Hi.

Quoted 24 lines · Christian
Quoted 23 lines · Andrew Morton
Quoted 21 lines · Christian
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

[ trace snipped ]

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Quoted 7 lines
Quoted 4 lines

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Could the ACPI team please take a look at fixing this regression?

I did two cycles with mainline suspend now and did not hit the problem... I will keep an eye on it.

Could you let me know how you go? I didn't make any changes between 2.2 and 2.2.0.1 that I think could cause this, but if you can't reproduce it otherwise, I'll happily look again.

Regards,

Nigel

#10 lore raw
Quoted 1 line · Nigel Cunningham

Hi.

Hi Nigel,

Quoted 28 lines · Christian
Quoted 27 lines · Andrew Morton
Quoted 26 lines
Quoted 13 lines · Christian
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 8 lines
Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

[ trace snipped ]

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Quoted 8 lines
Quoted 4 lines
Quoted 1 line

Could the ACPI team please take a look at fixing this regression?

I did two cycles with mainline suspend now and did not hit the problem... I will keep an eye on it.

Could you let me know how you go? I didn't make any changes between 2.2 and 2.2.0.1 that I think could cause this, but if you can't reproduce it otherwise, I'll happily look again.

You just missed my last mail. It happens with mainline suspend as well, so it is not your fault.

Regards,

Signature · 1 line
Christian
#11 lore raw

Hi.

Quoted 34 lines · Christian
Quoted 33 lines · Nigel Cunningham
Quoted 31 lines · Christian
Quoted 19 lines · Andrew Morton
Quoted 18 lines · Christian
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

Quoted 8 lines
Quoted 3 lines

[ trace snipped ]

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Could the ACPI team please take a look at fixing this regression?

I did two cycles with mainline suspend now and did not hit the problem... I will keep an eye on it.

Quoted 6 lines
Quoted 3 lines

Could you let me know how you go? I didn't make any changes between 2.2 and 2.2.0.1 that I think could cause this, but if you can't reproduce it otherwise, I'll happily look again.

You just missed my last mail. It happens with mainline suspend as well, so it is not your fault.

Ok. Thanks! :)

Nigel

#6 lore raw
In reply to Hesse, Christian
Quoted 28 lines · Christian
Quoted 27 lines · Christian
Quoted 26 lines · Andrew Morton
Quoted 13 lines · Christian
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 8 lines
Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

[ trace snipped ]

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

Could the ACPI team please take a look at fixing this regression?

I did two cycles with mainline suspend now and did not hit the problem... I will keep an eye on it.

It is independent from suspend version, hald hangs with mainline suspend as well.

I think this is a big regression since 2.5.15. Is anybody working on it? I did not get a reply for nearly a week now...
If anybody needs more information let me know.

Signature · 2 lines
Regards,
Christian
#7 lore raw

Hi.

Quoted 34 lines · Christian
Quoted 33 lines · Christian
Quoted 31 lines · Christian
Quoted 19 lines · Andrew Morton
Quoted 18 lines · Christian
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

Quoted 8 lines
Quoted 3 lines

[ trace snipped ]

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Could the ACPI team please take a look at fixing this regression?

I did two cycles with mainline suspend now and did not hit the problem... I will keep an eye on it.

Quoted 6 lines
Quoted 2 lines

It is independent from suspend version, hald hangs with mainline suspend as well.

I think this is a big regression since 2.5.15. Is anybody working on it? I did not get a reply for nearly a week now...
If anybody needs more information let me know.

I would suggest bugging the acpi guys :)

Regards,

Nigel

#8 lore raw
Quoted 35 lines · Nigel Cunningham
Quoted 34 lines · Christian
Quoted 33 lines · Christian
Quoted 31 lines · Christian
Quoted 19 lines · Andrew Morton
Quoted 18 lines · Christian
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Christian

Hello everybody,

since using kernel version 2.6.16-rc4 the hal daemon is in status D after resume. I use suspend2 2.2.0.1 for 2.6.16-rc3. Any hints what could be the problem? It worked perfectly with 2.6.15.x and suspend2 2.2.

a) Look in the logs for any oopses, other nasties

Nothing.

Quoted 2 lines

b) Do echo t > /proc/sysrq-trigger', dmesg -s 1000000 > foo' then find the trace for hald' in foo', send it to this list.

Ok, here it is:

Quoted 8 lines
Quoted 3 lines

[ trace snipped ]

This is with 2.6.16-rc4-git1 + suspend2 2.2.0.1.

Hopefully suspend2 isn't involved. People would feel more comfortable if you could test a vanilla mainline tree..

Could the ACPI team please take a look at fixing this regression?

I did two cycles with mainline suspend now and did not hit the problem... I will keep an eye on it.

Quoted 8 lines
Quoted 6 lines
Quoted 2 lines

It is independent from suspend version, hald hangs with mainline suspend as well.

I think this is a big regression since 2.5.15. Is anybody working on it? I did not get a reply for nearly a week now...
If anybody needs more information let me know.

I would suggest bugging the acpi guys :)

They are in cc, but nobody cares. :(

Signature · 1 line
Christian