[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #38 from James Ettle 2009-04-03 10:34:51 --- This seems to be a Tickless issue. With nohz=off, I don't seen an excessive change in wakeups (roughly 1000 both before and after suspend/resume). -- Configure bugmail: http://bugzill

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #37 from James Ettle 2009-03-31 17:39:02 --- Additional info: this seems to be only a sleep-to-RAM problem, things seem OK when I resume from hibernation. -- Configure bugmail: http://bugzilla.kernel.org/userprefs.cgi?tab=email

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #36 from James Ettle 2009-03-30 16:24:33 --- (In reply to comment #35) > Hi, James > How about this issue if only one cpu is used? (by adding the boot option > of > "maxcpus=1") > Thanks. The excessive wakeups still exis

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #35 from ykzhao 2009-03-30 05:45:07 --- Hi, James How about this issue if only one cpu is used? (by adding the boot option of "maxcpus=1") Thanks. -- Configure bugmail: http://bugzilla.kernel.org/userprefs.cgi?tab=email -

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #34 from James Ettle 2009-03-29 15:17:08 --- (In reply to comment #33) > Hi, James > Thanks for the test. It seems that the wakeups per second become normal > after resuming if the boot option of "idle=nomwait" is used. >

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 ykzhao changed: What|Removed |Added CC||yakui.z...@intel.com --- Comment #33 from ykz

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #32 from James Ettle 2009-03-26 09:22:20 --- (In reply to comment #29) > Hi, James > Will you please add the boot option of "idle=nomwait" on the latest kernel > and see whether this issue still exists after resume? > Than

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #31 from James Ettle 2009-03-25 07:59:20 --- (In reply to comment #30) > did you try windows in the system? No, I don't have a Windows license for this system. -- Configure bugmail: http://bugzilla.kernel.org/userprefs.cgi?tab=

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 Shaohua changed: What|Removed |Added CC||shaohua...@intel.com Regression|---

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #29 from yakui.z...@intel.com 2009-03-23 01:30 --- Hi, James Will you please add the boot option of "idle=nomwait" on the latest kernel and see whether this issue still exists after resume? Thanks. -- Configure bu

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #28 from theholyet...@googlemail.com 2009-03-20 01:19 --- (In reply to comment #27) > please unload uhci-hcd/ehci-hcd/ohci-hcd before suspend, > does the problem still exist after resume? Yes. > does the problem still exis

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #27 from rui.zh...@intel.com 2009-03-17 01:21 --- please unload uhci-hcd/ehci-hcd/ohci-hcd before suspend, does the problem still exist after resume? does the problem still exist with uhci-hcd/ehci-hcd/ohci-hcd reloaded afte

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #26 from theholyet...@googlemail.com 2009-03-11 15:35 --- (In reply to comment #25) > Created an attachment (id=20487) --> (http://bugzilla.kernel.org/attachment.cgi?id=20487&action=view) [details] > debug patch > > please

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #25 from rui.zh...@intel.com 2009-03-10 22:54 --- Created an attachment (id=20487) --> (http://bugzilla.kernel.org/attachment.cgi?id=20487&action=view) debug patch please try this debug patch and see if the problem still e

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #24 from theholyet...@googlemail.com 2009-03-10 02:02 --- (In reply to comment #23) > make sure CONFIG_PM_DEBUG is set, > run "echo core > /sys/power/pm_test" and "echo mem > /sys/power/state". > is the problem reproducible

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 rui.zh...@intel.com changed: What|Removed |Added Status|NEW |NEEDINFO --- Comment #23 from

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #22 from theholyet...@googlemail.com 2009-03-09 14:42 --- (In reply to comment #20) > As this seem to be acpi related, offlining the CPUs might not reproduce it: > echo 0 >/sys/devices/system/cpu/cpu1/online > echo 0 >/sys/d

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #21 from tr...@suse.de 2009-03-09 05:02 --- Ehh and re-enable them again of course, this is what suspend also does with your CPUs. -- Configure bugmail: http://bugzilla.kernel.org/userprefs.cgi?tab=email --- You are r

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #20 from tr...@suse.de 2009-03-09 05:01 --- As this seem to be acpi related, offlining the CPUs might not reproduce it: echo 0 >/sys/devices/system/cpu/cpu1/online echo 0 >/sys/devices/system/cpu/cpu2/online ... Still it's a

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #16 from theholyet...@googlemail.com 2009-03-09 00:36 --- Created an attachment (id=20459) --> (http://bugzilla.kernel.org/attachment.cgi?id=20459&action=view) cpu0ist -- Configure bugmail: http://bugzilla.kernel.org/use

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #17 from theholyet...@googlemail.com 2009-03-09 00:36 --- Created an attachment (id=20460) --> (http://bugzilla.kernel.org/attachment.cgi?id=20460&action=view) cpu1cst -- Configure bugmail: http://bugzilla.kernel.org/use

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #15 from theholyet...@googlemail.com 2009-03-09 00:35 --- Created an attachment (id=20458) --> (http://bugzilla.kernel.org/attachment.cgi?id=20458&action=view) cpu0cst -- Configure bugmail: http://bugzilla.kernel.org/use

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #19 from theholyet...@googlemail.com 2009-03-09 00:37 --- The ACPI interrupt seems normal (in /proc/interrupts) after suspend/resume, I can't see any evidence of it machine-gunning. -- Configure bugmail: http://bugzilla.k

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #18 from theholyet...@googlemail.com 2009-03-09 00:36 --- Created an attachment (id=20461) --> (http://bugzilla.kernel.org/attachment.cgi?id=20461&action=view) cpu1ist -- Configure bugmail: http://bugzilla.kernel.org/use

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #13 from theholyet...@googlemail.com 2009-03-09 00:35 --- Created an attachment (id=20456) --> (http://bugzilla.kernel.org/attachment.cgi?id=20456&action=view) grep . /sys/firmware/acpi/interrupts/* before suspend/resume

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 --- Comment #14 from theholyet...@googlemail.com 2009-03-09 00:35 --- Created an attachment (id=20457) --> (http://bugzilla.kernel.org/attachment.cgi?id=20457&action=view) grep . /sys/firmware/acpi/interrupts/* after suspend/resume -

[Bug 12788] After suspend/resume, processor spends more time in C0 state, C3 hardly used

http://bugzilla.kernel.org/show_bug.cgi?id=12788 yakui.z...@intel.com changed: What|Removed |Added CC||acpi- |