[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160125053804.GA5599@in.ibm.com>
Date: Mon, 25 Jan 2016 11:08:05 +0530
From: Gautham R Shenoy <ego@...ux.vnet.ibm.com>
To: Balbir Singh <bsingharora@...il.com>
Cc: Shilpasri G Bhat <shilpa.bhat@...ux.vnet.ibm.com>,
linuxppc-dev@...abs.org, linux-kernel@...r.kernel.org,
rjw@...ysocki.net, viresh.kumar@...aro.org,
linux-pm@...r.kernel.org, pc@...ibm.com, anton@...ba.org,
ego@...ux.vnet.ibm.com, shreyas@...ux.vnet.ibm.com
Subject: Re: [PATCH v6 2/5] cpufreq: powernv: Remove cpu_to_chip_id() from
hot-path
Hello Balbir,
On Sat, Jan 23, 2016 at 07:59:20PM +1100, Balbir Singh wrote:
> On Fri, 22 Jan 2016 12:49:02 +0530
> Shilpasri G Bhat <shilpa.bhat@...ux.vnet.ibm.com> wrote:
>
> > cpu_to_chip_id() does a DT walk through to find out the chip id by
> > taking a contended device tree lock. This adds an unnecessary overhead
> > in a hot path. So instead of calling cpu_to_chip_id() everytime cache
> > the chip ids for all cores in the array 'core_to_chip_map' and use it
> > in the hotpath.
> >
> > Reported-by: Anton Blanchard <anton@...ba.org>
> > Signed-off-by: Shilpasri G Bhat <shilpa.bhat@...ux.vnet.ibm.com>
> > Reviewed-by: Gautham R. Shenoy <ego@...ux.vnet.ibm.com>
>
> snip
>
> Does the core_to_chip_map need to be updated/refreshed on/after/ a
> cpu (core) hotplug? I presume id's don't change
No, the id's don't change on cpu/core hotplug. The core_to_chip_map is
initialized in init_chip_info() where we use for_each_possible_cpu().
Thanks for the review!
>
> Balbir
>
--
Thanks and Regards
gautham.
Powered by blists - more mailing lists