lkml.org 
[lkml]   [2010]   [Nov]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [patch] x86: avoid unnecessary smp alternatives switch during suspend/resume
On Sun, Nov 21, 2010 at 08:27:41PM +1100, Nigel Cunningham wrote:
> Hi.
>
> We have a few others things that want to modify their behaviour
> according to whether we're doing the atomic copy/restore. Perhaps it
> would be an idea to just use a single flag, perhaps a value for
> system_state?

I agree, it would be nicer if we don't introduce a special flag
just for that. Other than that, I like all sane code that speeds up
suspend/resume :). I've attached before/after dmesg excerpts on my
system with the patch ontop of v2.6.37-rc2-181-gb86db47. We end up
saving 11601 µsecs according to CONFIG_PRINTK_TIME but hey, the code is
simple enough :).

Tested-by: Borislav Petkov <bp@alien8.de>

Thanks.

--
Regards/Gruss,
Boris.
[ 122.004099] [drm] Resetting GPU
[ 122.014614] PM: Syncing filesystems ... done.
[ 122.027334] Freezing user space processes ... (elapsed 0.01 seconds) done.
[ 122.039644] Freezing remaining freezable tasks ... (elapsed 0.01 seconds) done.
[ 122.050629] PM: Preallocating image memory... done (allocated 147655 pages)
[ 122.324258] PM: Allocated 590620 kbytes in 0.27 seconds (2187.48 MB/s)
[ 122.324298] Suspending console(s) (use no_console_suspend to debug)
[ 122.325991] sd 3:0:0:0: [sdb] Synchronizing SCSI cache
[ 122.326104] sd 1:0:0:0: [sda] Synchronizing SCSI cache
[ 122.330976] serial 00:0b: disabled
[ 122.331077] serial 00:0b: wake-up capability disabled by ACPI
[ 122.331161] i8042 kbd 00:0a: wake-up capability enabled by ACPI
[ 122.331560] pci 0000:01:00.0: PCI INT A disabled
[ 122.331967] ATIIXP_IDE 0000:00:14.1: PCI INT A disabled
[ 122.332736] ahci 0000:00:11.0: PCI INT A disabled
[ 122.432770] HDA Intel 0000:01:00.1: PCI INT B disabled
[ 122.432932] ACPI handle has no context!
[ 122.433187] HDA Intel 0000:00:14.2: PCI INT A disabled
[ 122.444799] PM: freeze of devices complete after 119.816 msecs
[ 122.448594] PM: late freeze of devices complete after 3.785 msecs
[ 122.448628] Disabling non-boot CPUs ...
[ 122.471872] CPU 1 is now offline
[ 122.489261] CPU 2 is now offline
[ 122.498177] CPU 3 is now offline
[ 122.498183] lockdep: fixing up alternatives.
[ 122.498228] SMP alternatives: switching to UP code
[ 122.509784] Extended CMOS year: 2000
[ 122.510008] PM: Creating hibernation image:
[ 122.510008] PM: Need to copy 146817 pages
[ 72.337143] [drm] Resetting GPU
[ 72.347054] PM: Syncing filesystems ... done.
[ 72.377683] Freezing user space processes ... (elapsed 0.01 seconds) done.
[ 72.390714] Freezing remaining freezable tasks ... (elapsed 0.01 seconds) done.
[ 72.401705] PM: Preallocating image memory... done (allocated 147611 pages)
[ 72.690885] PM: Allocated 590444 kbytes in 0.28 seconds (2108.72 MB/s)
[ 72.690925] Suspending console(s) (use no_console_suspend to debug)
[ 72.692422] sd 3:0:0:0: [sdb] Synchronizing SCSI cache
[ 72.692432] sd 1:0:0:0: [sda] Synchronizing SCSI cache
[ 72.704185] serial 00:0b: disabled
[ 72.704275] serial 00:0b: wake-up capability disabled by ACPI
[ 72.704356] i8042 kbd 00:0a: wake-up capability enabled by ACPI
[ 72.704742] pci 0000:01:00.0: PCI INT A disabled
[ 72.705258] ATIIXP_IDE 0000:00:14.1: PCI INT A disabled
[ 72.705915] ahci 0000:00:11.0: PCI INT A disabled
[ 72.806818] HDA Intel 0000:01:00.1: PCI INT B disabled
[ 72.806831] HDA Intel 0000:00:14.2: PCI INT A disabled
[ 72.807141] ACPI handle has no context!
[ 72.819836] PM: freeze of devices complete after 128.198 msecs
[ 72.823408] PM: late freeze of devices complete after 3.565 msecs
[ 72.823443] Disabling non-boot CPUs ...
[ 72.840884] CPU 1 is now offline
[ 72.866229] CPU 2 is now offline
[ 72.884239] CPU 3 is now offline
[ 72.886941] Extended CMOS year: 2000
[ 72.887016] PM: Creating hibernation image:
[ 72.887016] PM: Need to copy 146787 pages

\
 
 \ /
  Last update: 2010-11-21 11:05    [W:0.106 / U:0.452 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site