[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160113095124.186ff487@yairi>
Date: Wed, 13 Jan 2016 09:51:24 -0800
From: Jacob Pan <jacob.jun.pan@...ux.intel.com>
To: Borislav Petkov <bp@...en8.de>
Cc: Thomas Gleixner <tglx@...utronix.de>,
LKML <linux-kernel@...r.kernel.org>,
Linux PM <linux-pm@...r.kernel.org>,
Rafael Wysocki <rafael.j.wysocki@...el.com>,
"H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...hat.com>,
X86 Kernel <x86@...nel.org>,
Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>,
Peter Zijlstra <peterz@...radead.org>,
jacob.jun.pan@...ux.intel.com
Subject: Re: [PATCH v2 2/2] powercap/rapl: reduce ipi calls
On Wed, 13 Jan 2016 17:36:10 +0100
Borislav Petkov <bp@...en8.de> wrote:
> On Wed, Jan 13, 2016 at 08:21:13AM -0800, Jacob Pan wrote:
> > Thanks for bring this out. I didn't mean to ignore. I thought my
> > point was stated in the commit message there was no point of going
> > back and forth. Read-Modify-Write is quite common, not just for
> > RAPL could be used by future code. Sorry if I wasn't clear.
>
> But it also shows that it doesn't suffice for all your needs. So why
> add it?
>
> You can much better define your own functions which do all the MSR
> handling you require and call them with smp_call_function_*.
>
yeah, that is what I did in the original patch.
https://lkml.org/lkml/2015/12/7/1090
Then i was suggested to add a rmw msr api for the common good :), I
think it is a good idea since such operation is not limited to RAPL
driver. Other register access APIs such as regmap have more complete
selections.
Jacob
Powered by blists - more mailing lists