From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Warren Subject: Re: [PATCH v4 3/3] memory: tegra124-emc: Add EMC driver Date: Wed, 27 Aug 2014 10:27:53 -0600 Message-ID: <53FE0709.8070908@wwwdotorg.org> References: <1409058728-13347-1-git-send-email-tomeu.vizoso@collabora.com> <1409058728-13347-4-git-send-email-tomeu.vizoso@collabora.com> <53FCC139.6020703@wwwdotorg.org> <53FD8975.3050706@collabora.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <53FD8975.3050706-ZGY8ohtN/8qB+jHODAdFcQ@public.gmane.org> Sender: linux-tegra-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Tomeu Vizoso , Mikko Perttunen , Peter De Schrijver , Tuomas Tynkkynen Cc: linux-pm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Thierry Reding , =?windows-1252?Q?Terje_Bergstr=F6m?= , "Rafael J. Wysocki" , Pavel Machek , Len Brown , linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Javier Martinez Canillas List-Id: linux-tegra@vger.kernel.org On 08/27/2014 01:32 AM, Tomeu Vizoso wrote: > On 08/26/2014 07:17 PM, Stephen Warren wrote: >> On 08/26/2014 07:12 AM, Tomeu Vizoso wrote: >>> Sets the EMC clock rate based on the bandwidth requirements >>> registered by >>> memory clients through the PM_QOS_MEMORY_BANDWIDTH class. >>> >>> Note: this is just an example and not a proper driver for a external >>> memory >>> controller. Its only purpose is to illustrate how such a driver would >>> set the >>> frequency of the external memory clock based on the bandwidth >>> requirements of >>> memory clients. >> >>> diff --git a/arch/arm/mach-tegra/tegra.c b/arch/arm/mach-tegra/tegra.c >> >>> @@ -112,6 +117,11 @@ static void __init tegra_dt_init(void) >>> parent = soc_device_to_device(soc_dev); >>> >>> /* >>> + * HACK: register a platform device to probe the driver >>> + */ >>> + platform_device_register(&tegra_emc); >> >> I don't think this is a hack, except for the bug: That should only >> happen on Tegra124 not on all Tegra SoCs. >> >> Do you intend all 3 patches in this series to be merged? You'd mentioned >> you didn't when asked about this for a previous version. I'm not sure if >> that's changed? > > Yeah, I don't want 3/3 merged because we don't have a functional EMC > clock yet on T124, and because I don't know yet where will be the best > place to have that code in. That depends on Mikko's work on the real EMC > driver which is in a bit of flux right now. > > I have kept posting it just because I think it complements nicely the > explanation in the cover letter, but maybe confuses more than helps. OK. If you make this one a [PATCH RFC], or add a note to that effect to the commit description (perhaps underneath the --- line), that'd be helpful for future postings.