From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexandre Courbot Subject: Re: [PATCH v3 0/3] Add support for Tegra Activity Monitor Date: Tue, 9 Dec 2014 23:18:49 +0900 Message-ID: References: <1417709696-29281-1-git-send-email-tomeu.vizoso@collabora.com> <548703C4.1030300@collabora.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Return-path: In-Reply-To: <548703C4.1030300-ZGY8ohtN/8qB+jHODAdFcQ@public.gmane.org> Sender: linux-tegra-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Tomeu Vizoso Cc: "linux-pm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , Javier Martinez Canillas , "devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org" , Linux Kernel Mailing List , "linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" List-Id: devicetree@vger.kernel.org On Tue, Dec 9, 2014 at 11:14 PM, Tomeu Vizoso wrote: > On 12/09/2014 06:38 AM, Alexandre Courbot wrote: >> On Fri, Dec 5, 2014 at 1:14 AM, Tomeu Vizoso wrote: >>> Hello, >>> >>> this v3 addresses the comments that the devfreq implementation got, namely: >>> >>> * Address misc. style issues found by Thierry and Alexander >>> * Added helpers for register i/o >>> * Further documented the structs >>> * Enable the ACTMON after the IRQ handler has been installed >>> * Disable the ACTMON before removing the IRQ handler >>> * Add governor in a subsys initcall >>> >>> There's an open question on whether some functionality currently in this >>> devfreq driver should be moved into the devfreq framework, but without knowing >>> of other SoC family that would benefit from it, I'm reticent. It would be >>> great to hear from the devfreq maintainers if they have any plans regarding >>> this, or if they have any suggestion. >> >> I cannot make a thorough review because of the problem I mentioned in >> patch 3/3, but I am guessing this series is converging towards what we >> want. Now the main question will be how we can leverage Arto's >> watermark series for this one. I am ready to bet that doing so can >> reduce quite a lot of code. >> >> Since you are likely to be the first user of the watermarking feature, >> could you comment on its potential shortcomings and whatever needs to >> be fixed to best implement ACTMON support using it? I will try to push >> it myself, but you are obviously in a better position to understand >> what is needed. > > Sure, I'm still playing with the idea, but I have sent a few questions > to that thread already. Seen that, thanks! After seeing your comments I come to thing that maybe we should merge ACTMON support first and then adapt it to use watermarking afterwards. I'm all for incremental changes and after 4 iterations this series should be close to completion now - would be sad to go back to the design board at this stage. Reviewing your v4 is on my schedule for tomorrow, thanks for your patience!