NAK: [Utopic][SRU][PATCH 0/3] Fix for LP: #1470404
Chris J Arges
chris.j.arges at canonical.com
Thu Jul 2 15:32:45 UTC 2015
Forgot to add a patch.
On 07/02/2015 10:29 AM, Chris J Arges wrote:
> BugLink: http://bugs.launchpad.net/bugs/1470404
>
> SRU Justification:
> [Impact]
> Certain workloads can exhibit a large variance in behavior due to how how cpus
> are idled on power8 systems.
>
> [Fix]
>
> 74aa51b5ccd3975392e30d11820dc073c5f2cd32
> 92c83ff5b42b109c94fdeee53cb31f674f776d75
> 70734a786acfd1998e47d40df19cba5c29469bdf
> 78eaa10f027cf69f9bd409e64eaff902172b2327
>
> [Test Case]
> Set the system with the SMT8 mode and scaling_governor=performance or ondemand.
> Run the workload 100 times.
>
> --
>
> Preeti U Murthy (2):
> cpuidle: powernv: Read target_residency value of idle states from DT
> if available
> cpuidle: powernv: Avoid endianness conversions while parsing DT
>
> Preeti U. Murthy (1):
> cpuidle: powernv: Populate cpuidle state details by querying the
> device-tree
>
> drivers/cpuidle/cpuidle-powernv.c | 71 ++++++++++++++++++++++++++++-----------
> 1 file changed, 51 insertions(+), 20 deletions(-)
>
More information about the kernel-team
mailing list