Mailinglist Archive: opensuse-bugs (13462 mails)
| < Previous | Next > |
[Bug 333733] [suspend] Machine hangs in ksoftirqd/1 and events/ 1 after resume
- From: bugzilla_noreply@xxxxxxxxxx
- Date: Fri, 19 Oct 2007 13:44:26 -0600 (MDT)
- Message-id: <20071019194426.5EE8ACC7AC@xxxxxxxxxxxxxxxxxxxxxx>
https://bugzilla.novell.com/show_bug.cgi?id=333733#c1
--- Comment #1 from r myster <rmyster@xxxxxxxxx> 2007-10-19 13:44:25 MST ---
Created an attachment (id=179521)
--> (https://bugzilla.novell.com/attachment.cgi?id=179521)
suspend-resume logs for before & after using "nohz=off" boot option
openSuSE 10.3 2.6.22.9-0.4-default (latest kernel from update)
Similar story here. After a suspend & resume, the ksoftirqd/0 and migration/0
jumps to 9999% cpu use(shown under top). Unlike the case above, initially
after the resume, everything seems fine but as soon some other application is
opened, the usage jumps. The cpu usage jump also happens after a couple
minutes with no activity. All I could find on ksoftirqd was this blip from the
manual:
ksoftirqd is a per-cpu kernel thread that runs when the machine is under heavy
soft-interrupt load. Soft interrupts are normally serviced on return from a
hard interrupt, but it's possible for soft interrupts to be triggered more
quickly than they can be serviced. If a soft interrupt is triggered for a
second time while soft interrupts are being handled, the ksoftirq daemon is
triggered to handle the soft interrupts in process context. If ksoftirqd is
taking more than a tiny percentage of CPU time, this indicates the machine is
under heavy soft interrupt load.
I did run across another thread discussing high cpu useage on a virtual
machine(http://marc.info/?l=vbox-users&m=119012208118392&w=2) and the solution
there was to use the "nohz=off" boot option(something to do with the tickless
kernel) which seems to fix the issue. Whether it has introduced other side
effects is yet unknown.
The attached gzip filepm-suspend-logs.tar.gz) contains pm-suspend logs for both
the high cpu result after resume case and one for a suspend-resume cycle after
a reboot using the "nohz=off" boot option.
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
--- Comment #1 from r myster <rmyster@xxxxxxxxx> 2007-10-19 13:44:25 MST ---
Created an attachment (id=179521)
--> (https://bugzilla.novell.com/attachment.cgi?id=179521)
suspend-resume logs for before & after using "nohz=off" boot option
openSuSE 10.3 2.6.22.9-0.4-default (latest kernel from update)
Similar story here. After a suspend & resume, the ksoftirqd/0 and migration/0
jumps to 9999% cpu use(shown under top). Unlike the case above, initially
after the resume, everything seems fine but as soon some other application is
opened, the usage jumps. The cpu usage jump also happens after a couple
minutes with no activity. All I could find on ksoftirqd was this blip from the
manual:
ksoftirqd is a per-cpu kernel thread that runs when the machine is under heavy
soft-interrupt load. Soft interrupts are normally serviced on return from a
hard interrupt, but it's possible for soft interrupts to be triggered more
quickly than they can be serviced. If a soft interrupt is triggered for a
second time while soft interrupts are being handled, the ksoftirq daemon is
triggered to handle the soft interrupts in process context. If ksoftirqd is
taking more than a tiny percentage of CPU time, this indicates the machine is
under heavy soft interrupt load.
I did run across another thread discussing high cpu useage on a virtual
machine(http://marc.info/?l=vbox-users&m=119012208118392&w=2) and the solution
there was to use the "nohz=off" boot option(something to do with the tickless
kernel) which seems to fix the issue. Whether it has introduced other side
effects is yet unknown.
The attached gzip filepm-suspend-logs.tar.gz) contains pm-suspend logs for both
the high cpu result after resume case and one for a suspend-resume cycle after
a reboot using the "nohz=off" boot option.
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
| < Previous | Next > |