Re: [Bug #13306] hibernate slow on _second_ run

From: Johannes Berg
Date: Thu Jun 04 2009 - 01:56:22 EST


On Wed, 2009-06-03 at 23:51 +0200, Rafael J.
> > > Can you please send me /proc/interrupts from the affected box?
> >
> > $ cat /proc/interrupts
> > CPU0 CPU1
> > 0: 1820687 0 IO-APIC-edge timer
> > 7: 1 0 IO-APIC-edge
> > 9: 38721 0 IO-APIC-fasteoi acpi
> > 16: 940 0 IO-APIC-fasteoi HDA Intel
> > 17: 517 0 IO-APIC-fasteoi ohci_hcd:usb4
> > 18: 289829 0 IO-APIC-fasteoi ohci_hcd:usb3
> > 19: 2035671 0 IO-APIC-fasteoi ehci_hcd:usb2
> > 20: 32 0 IO-APIC-fasteoi ehci_hcd:usb1
> > 25: 515569 0 PCI-MSI-edge ahci
> > 26: 0 0 PCI-MSI-edge eth0
> > NMI: 0 0 Non-maskable interrupts
> > LOC: 152936 638337 Local timer interrupts
> > SPU: 0 0 Spurious interrupts
> > RES: 317238 581858 Rescheduling interrupts
> > CAL: 43 128 Function call interrupts
> > TLB: 60645 53553 TLB shootdowns
> > TRM: 0 0 Thermal event interrupts
> > THR: 0 0 Threshold APIC interrupts
> > ERR: 1
> > MIS: 0
>
> Hmm. The patch shouldn't matter for this box, except for one thing:
> check_irq_resend() on resume. Interesting.

Yeah, well, turns out it didn't help anyway:

[30490.846160] ACPI: Waking up from system sleep state S4
[30491.371450] HDA Intel 0000:00:08.0: restoring config space at offset 0x1 (was 0xb00006, writing 0xb00002)
[30491.379411] forcedeth 0000:00:0a.0: restoring config space at offset 0x1 (was 0xb00003, writing 0xb00007)
[30491.387352] ahci 0000:00:0b.0: restoring config space at offset 0x1 (was 0xb00403, writing 0xb00407)
[30511.970606] pci 0000:00:03.2: PME# disabled

The problem doesn't always happen. But now that it happens here again,
maybe I can try ftrace again...

johannes

Attachment: signature.asc
Description: This is a digitally signed message part