Messages in this thread | | | From | "Rafael J. Wysocki" <> | Subject | Re: [PATCH 7/9] PM: Measure device suspend and resume times | Date | Sat, 12 Sep 2009 00:18:23 +0200 |
| |
On Friday 11 September 2009, Ben Gamari wrote: > On Thu, Sep 10, 2009 at 01:39:21AM +0200, Rafael J. Wysocki wrote: > > From: Rafael J. Wysocki <rjw@sisk.pl> > > > > Measure and print the time of suspending and resuming all devices. > > > > Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> > > Do you have any rough numbers describing actual suspend/resume times > before and after this patch set? Of what order of magnitude is the > improvement on standard x86 hardware? Just curious.
Depending on the system on which it's run, it gives up to 15% speedup, but please note that only PCI and ACPI devices are now allowed to suspend/resume asynchronously, while on my test systems the most significant stalls are from USB, serio devices and (SATA) drives which are still suspended/resumed synchronously.
Thanks, Rafael
| |