Bug 3679

Summary:	S4 resume: kacpid uses 99% CPU
Product:	ACPI	Reporter:	Florian Petri (flop)
Component:	Power-Sleep-Wake	Assignee:	Shaohua (shaohua.li)
Status:	REJECTED INSUFFICIENT_DATA
Severity:	normal	CC:	acpi-bugzilla, bunk, mnencia
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.8 & 2.6.7 (Debian kernel source package)	Subsystem:
Regression:	---	Bisected commit-id:
Attachments:	.config

Description Florian Petri 2004-11-01 05:50:59 UTC

Distribution: Debian
 
Hardware Environment: HP compaq nx5000

Software Environment: Debian unstable, dist-upgrade today (problem occurs also
with older versions)

Problem Description: When the system resumes the kacpid kernel process takes 99%
of my CPU.

Steps to reproduce:
Suspend/resume notebook (with the kernel swsuspend) 

If you search the net you'll find some other people having also problems that
kacpid blocks there CPUs. But the have different ways to trigger that bug... 


cu Floh

Comment 1 Rhett Kipps 2004-11-02 00:36:08 UTC

I also experience the same difficulties presented here.

I have a Dell Inspiron 2650 laptop/notebook, running 2.6.9.  I've run random 
versions between 2.6.5-2.6.9 and have experienced kacpid hitting 99-100% cpu 
usage whenever the lid is shut.

Someone has suggested the problem doesn't exist in 2.6.4 (at least for the 
Inspiron 2650 lid).  I've as yet not tried to confirm this.

It seems others are having the same problem:
http://www.ussg.iu.edu/hypermail/linux/kernel/0405.1/1720.html
http://www.ussg.iu.edu/hypermail/linux/kernel/0406.2/0052.html

Cheers.

Comment 2 Len Brown 2004-11-04 01:14:51 UTC

Rhett, you too see this only after resume?

does /proc/interrupts show the line with acpi taking events?
if you kill the user land acpid and cat /proc/acpi/event
does it show any clues?

Comment 3 Marco Nenciarini 2004-11-13 02:04:00 UTC

I'm experiencing the same bug on an ACER1703SM_2.6. I have tried with debian
kernel 2.6.7-2.6.9 and after some minutes of normal activity (I supose when the
themperature reach a certain value) fan starts at maximun speed, kacpid eats
80%/99% of cpu and the system becomes unusable (the only chance is to renice the
kacpid process).

If you need it I can post the DSDT (that I've modified to make ACPI battery
working).

Best Regards

Comment 4 Marco Nenciarini 2004-11-13 02:35:41 UTC

I've missed this: The kernel 2.6.0 (debian version) work correctly (but have
some other problem with acpi...)

Comment 5 Florian Petri 2004-11-16 00:46:25 UTC

Hi Len,

after reporting this bug I compiled the 2.6.9 kernel and now the error is no
more reproducable via suspend and resume (even with the old kernels). Before it
was. After working (and suspending/resuming) some hours with my computer kacpid
started again using 99% of my cpu.

I don't think this helps you:

/proc/interrupts:

           CPU0
  0:   22369595    IO-APIC-edge  timer
  1:      47510    IO-APIC-edge  i8042
  7:          2    IO-APIC-edge  parport0
  8:    1978654    IO-APIC-edge  rtc
  9:      16006   IO-APIC-level  acpi
 10:          2    IO-APIC-edge  pc110pad
 12:    1790264    IO-APIC-edge  i8042
 14:     328041    IO-APIC-edge  ide0
 15:       3742    IO-APIC-edge  ide1
 16:          1   IO-APIC-level  uhci_hcd
 17:      41775   IO-APIC-level  Intel 82801DB-ICH4, Intel 82801DB-ICH4 Modem
 18:          0   IO-APIC-level  yenta, yenta, uhci_hcd
 19:          0   IO-APIC-level  uhci_hcd
 20:     377954   IO-APIC-level  eth0
 23:        412   IO-APIC-level  ohci1394, ehci_hcd
NMI:          0
LOC:   22371927
ERR:          0
MIS:          0

/proc/acpi/event:
<empty>

(produced with kernel 2.6.9 ...)

Comment 6 Shaohua 2004-12-20 17:04:04 UTC

I can't reproduce the issue in the nx5000 laptop with latest kernel. My BIOS 
type is F.0C 07/28/2004. Could you please upgrade the BIOS and test again?

Comment 7 Florian Petri 2004-12-23 00:10:17 UTC

I upgraded my bios but today after the third resume the kacpid runned amok again.

You'll receive my kernel config via email, maybe then you can reproduce this issue.

Comment 8 Shaohua 2004-12-23 01:04:30 UTC

No luck still :( (with your config, my /proc/interrupt looks just like yours)
I noticed your /proc/interrupts have great number of RTC interrupts, did you 
used RTC? Can the patch in bug 1320 help any? Thanks.

Comment 9 Florian Petri 2004-12-23 05:38:22 UTC

Created attachment 4296 [details]
.config

Comment 10 Florian Petri 2004-12-23 05:44:07 UTC

I'm currently testing the patch with my config. It's to early to say something
final, but the problem did not occur yet. Now I'm on vacation for two weeks, but
I'll continue this test.

Is this patch included into 2.6.10, when it's released?

Comment 11 Shaohua 2004-12-23 16:55:02 UTC

Good to know the patch has positive effect. Ok, I will wait for you to do a 
complete test.
The patch has been in Len's test tree, so it will be in 2.6.10 if it opens.

Comment 12 Florian Petri 2005-01-14 12:17:07 UTC

The problem still exists with the mentioned patch. I worked with the laptop 2-3h
(and suspended it about 3 times). Then I worked about 8h with it,
suspend/resumed it and the kacpid problem reoccured. 

Is there anyting I can test?

Next week I'll upgrade to 2.6.10.

Comment 13 Shaohua 2005-03-03 17:33:55 UTC

Could you please try the patch in Bug 3967 with latest kernel? It clears the 
event status before enable it. Thanks.

Comment 14 Ulrich Scholz 2005-05-31 00:56:21 UTC

I have the same problem with kernel 2.6.11.10 on my Toshiba Port

Comment 15 Julien HENRY 2005-12-10 08:01:28 UTC

kernel 2.6.12-13mdk-i686-up-4GB

I don't know if it's the same issue, but today is the first time I managed to
use S4 resume with my Asus notebook (thanks to new ATI drivers). But kacpid was
using most of CPU.
After inspecting the logs, it's what I found (in /var/log/message):
Dec 10 13:46:40 Orthanc kernel: swsusp: Need to copy 36935 pages
Dec 10 13:46:40 Orthanc kernel: swsusp: Restoring Highmem
Dec 10 13:46:40 Orthanc kernel: ACPI: PCI Interrupt 0000:00:1d.0[A] -> Link
[LNKA] -> GSI 11 (level, low) -> IRQ 11
Dec 10 13:46:40 Orthanc kernel: ACPI: PCI Interrupt 0000:00:1d.1[B] -> Link
[LNKD] -> GSI 5 (level, low) -> IRQ 5
Dec 10 13:46:40 Orthanc kernel: ACPI: PCI Interrupt 0000:00:1d.2[C] -> Link
[LNKC] -> GSI 10 (level, low) -> IRQ 10
Dec 10 13:46:44 Orthanc kernel: ACPI: PCI Interrupt 0000:00:1d.7[D] -> Link
[LNKH] -> GSI 10 (level, low) -> IRQ 10
Dec 10 13:46:45 Orthanc kernel: ehci_hcd 0000:00:1d.7: debug port 1
Dec 10 13:46:45 Orthanc kernel: ehci_hcd 0000:00:1d.7: USB 2.0 restarted, EHCI
1.00, driver 10 Dec 2004
Dec 10 13:46:45 Orthanc kernel: ACPI: PCI Interrupt 0000:00:1f.1[A] -> Link
[LNKC] -> GSI 10 (level, low) -> IRQ 10
Dec 10 13:46:45 Orthanc kernel: ACPI: PCI Interrupt 0000:00:1f.5[B] -> Link
[LNKB] -> GSI 4 (level, low) -> IRQ 4
Dec 10 13:46:45 Orthanc kernel: ACPI: PCI Interrupt 0000:01:00.0[A] -> Link
[LNKA] -> GSI 11 (level, low) -> IRQ 11
Dec 10 13:46:45 Orthanc kernel: tg3: eth1: Link is down.
Dec 10 13:46:45 Orthanc kernel: ACPI: PCI Interrupt 0000:02:01.2[C] -> Link
[LNKC] -> GSI 10 (level, low) -> IRQ 10
Dec 10 13:46:45 Orthanc kernel: eth2: Coming out of suspend...
Dec 10 13:46:45 Orthanc kernel: ACPI: PCI Interrupt 0000:02:02.0[A] -> Link
[LNKB] -> GSI 4 (level, low) -> IRQ 4
Dec 10 13:46:45 Orthanc kernel: usbhid 2-2:1.0: PM: resume from 3, parent 2-2
still 3
Dec 10 13:46:45 Orthanc kernel: Restarting tasks... done
Dec 10 13:46:45 Orthanc kernel: usb 2-2: USB disconnect, address 4
Dec 10 13:46:45 Orthanc kernel: tg3: eth1: Link is up at 10 Mbps, half duplex.
Dec 10 13:46:45 Orthanc kernel: tg3: eth1: Flow control is off for TX and off
for RX.
Dec 10 13:46:45 Orthanc kernel: usb 2-2: new low speed USB device using uhci_hcd
and address 5
Dec 10 13:46:45 Orthanc kernel: input: USB HID v1.10 Mouse [Logitech Optical USB
Mouse] on usb-0000:00:1d.1-2
Dec 10 13:46:57 Orthanc pam_console.dev[16138]: Restoring console permissions
for /dev/input/mouse1 
Dec 10 13:46:57 Orthanc pam_console.dev[16132]: Restoring console permissions
for /dev/input/event2 
Dec 10 13:46:57 Orthanc pam_console.dev[16140]: Restoring console permissions
for /dev/input/ts1 
Dec 10 13:47:10 Orthanc kernel:     ACPI-0352: *** Error: Looking up [REDO] in
namespace, AE_NOT_FOUND
Dec 10 13:47:10 Orthanc kernel: search_node c14ed240 start_node c14ed240
return_node 00000000
Dec 10 13:47:10 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_TZ_.TCHG] (Node c14ed240), AE_NOT_FOUND
Dec 10 13:47:10 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_GPE._L00] (Node c14ede80), AE_NOT_FOUND
Dec 10 13:47:10 Orthanc kernel:     ACPI-0552: *** Error: AE_NOT_FOUND while
evaluating method [_L00] for GPE[ 0]
Dec 10 13:47:39 Orthanc kernel:     ACPI-0352: *** Error: Looking up [REDO] in
namespace, AE_NOT_FOUND
Dec 10 13:47:39 Orthanc kernel: search_node c14ed240 start_node c14ed240
return_node 00000000
Dec 10 13:47:39 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_TZ_.TCHG] (Node c14ed240), AE_NOT_FOUND
Dec 10 13:47:39 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_GPE._L00] (Node c14ede80), AE_NOT_FOUND
Dec 10 13:47:39 Orthanc kernel:     ACPI-0552: *** Error: AE_NOT_FOUND while
evaluating method [_L00] for GPE[ 0]
Dec 10 13:48:08 Orthanc kernel:     ACPI-0352: *** Error: Looking up [REDO] in
namespace, AE_NOT_FOUND
Dec 10 13:48:08 Orthanc kernel: search_node c14ed240 start_node c14ed240
return_node 00000000
Dec 10 13:48:08 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_TZ_.TCHG] (Node c14ed240), AE_NOT_FOUND
Dec 10 13:48:08 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_GPE._L00] (Node c14ede80), AE_NOT_FOUND
Dec 10 13:48:08 Orthanc kernel:     ACPI-0552: *** Error: AE_NOT_FOUND while
evaluating method [_L00] for GPE[ 0]
Dec 10 13:48:37 Orthanc kernel:     ACPI-0352: *** Error: Looking up [REDO] in
namespace, AE_NOT_FOUND
Dec 10 13:48:37 Orthanc kernel: search_node c14ed240 start_node c14ed240
return_node 00000000
Dec 10 13:48:37 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_TZ_.TCHG] (Node c14ed240), AE_NOT_FOUND
Dec 10 13:48:37 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_GPE._L00] (Node c14ede80), AE_NOT_FOUND
Dec 10 13:48:37 Orthanc kernel:     ACPI-0552: *** Error: AE_NOT_FOUND while
evaluating method [_L00] for GPE[ 0]
Dec 10 13:48:43 Orthanc dhclient: DHCPREQUEST on eth1 to 192.168.0.1 port 67
Dec 10 13:48:43 Orthanc dhclient: DHCPACK from 192.168.0.1
Dec 10 13:48:44 Orthanc NET: /sbin/dhclient-script : updated /etc/resolv.conf
Dec 10 13:48:45 Orthanc dhclient: bound to 192.168.0.133 -- renewal in 255 seconds.
Dec 10 13:49:07 Orthanc kernel:     ACPI-0352: *** Error: Looking up [REDO] in
namespace, AE_NOT_FOUND
Dec 10 13:49:07 Orthanc kernel: search_node c14ed240 start_node c14ed240
return_node 00000000
Dec 10 13:49:07 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_TZ_.TCHG] (Node c14ed240), AE_NOT_FOUND
Dec 10 13:49:07 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_GPE._L00] (Node c14ede80), AE_NOT_FOUND
Dec 10 13:49:07 Orthanc kernel:     ACPI-0552: *** Error: AE_NOT_FOUND while
evaluating method [_L00] for GPE[ 0]
Dec 10 13:49:36 Orthanc kernel:     ACPI-0352: *** Error: Looking up [REDO] in
namespace, AE_NOT_FOUND
Dec 10 13:49:36 Orthanc kernel: search_node c14ed240 start_node c14ed240
return_node 00000000
Dec 10 13:49:36 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_TZ_.TCHG] (Node c14ed240), AE_NOT_FOUND
Dec 10 13:49:36 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_GPE._L00] (Node c14ede80), AE_NOT_FOUND
Dec 10 13:49:36 Orthanc kernel:     ACPI-0552: *** Error: AE_NOT_FOUND while
evaluating method [_L00] for GPE[ 0]
Dec 10 13:50:05 Orthanc kernel:     ACPI-0352: *** Error: Looking up [REDO] in
namespace, AE_NOT_FOUND
Dec 10 13:50:05 Orthanc kernel: search_node c14ed240 start_node c14ed240
return_node 00000000
Dec 10 13:50:05 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_TZ_.TCHG] (Node c14ed240), AE_NOT_FOUND
Dec 10 13:50:05 Orthanc kernel:     ACPI-1138: *** Error: Method execution
failed [\_GPE._L00] (Node c14ede80), AE_NOT_FOUND
Dec 10 13:50:05 Orthanc kernel:     ACPI-0552: *** Error: AE_NOT_FOUND while
evaluating method [_L00] for GPE[ 0]
Dec 10 13:50:35 Orthanc kernel:     ACPI-0352: *** Error: Looking up [REDO] in
namespace, AE_NOT_FOUND
...

And this was repeated as long as I reboot (Ctrl+Sys+B) my laptop.

Comment 16 Ilya Gavrilov 2006-03-14 09:02:22 UTC

I have the same problem with kernel 2.6.15 on my HP Compaq nc6220

Comment 17 Alexey Starikovskiy 2007-06-04 12:14:54 UTC

Please check if the problem still present in 2.6.21

Comment 18 Adrian Bunk 2007-07-13 20:59:34 UTC

Please reopen this bug if it's still present with kernel 2.6.22.