From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthias Brugger Subject: Re: [PATCH v5 05/12] drm: mediatek: Omit warning on probe defers Date: Tue, 20 Nov 2018 11:19:00 +0100 Message-ID: <9a445746-b07d-73e1-68c9-10eb9d5bf771@gmail.com> References: <20181116125449.23581-1-matthias.bgg@kernel.org> <20181116125449.23581-6-matthias.bgg@kernel.org> <1542605939.32082.12.camel@mtksdaap41> <0a02edf9-78da-dde9-4b69-38c4719b711b@gmail.com> <1542686700.9073.15.camel@mtksdaap41> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1542686700.9073.15.camel@mtksdaap41> Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org To: CK Hu Cc: matthias.bgg@kernel.org, robh+dt@kernel.org, mark.rutland@arm.com, p.zabel@pengutronix.de, airlied@linux.ie, mturquette@baylibre.com, sboyd@codeaurora.org, ulrich.hecht+renesas@gmail.com, laurent.pinchart@ideasonboard.com, sean.wang@mediatek.com, sean.wang@kernel.org, rdunlap@infradead.org, wens@csie.org, dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-mediatek@lists.infradead.org, linux-clk@vger.kernel.org, devicetree@vger.kernel.org, Matthias Brugger List-Id: devicetree@vger.kernel.org On 20/11/2018 05:05, CK Hu wrote: > Hi, Matthias: > > On Mon, 2018-11-19 at 10:26 +0100, Matthias Brugger wrote: >> >> On 19/11/2018 06:38, CK Hu wrote: >>> Hi, Matthias: >>> >>> On Fri, 2018-11-16 at 13:54 +0100, matthias.bgg@kernel.org wrote: >>>> From: Matthias Brugger >>>> >>>> It can happen that the mmsys clock drivers aren't probed before the >>>> platform driver gets invoked. The platform driver used to print a warning >>>> that the driver failed to get the clocks. Omit this error on >>>> the defered probe path. >>> >>> This patch looks good to me, but you have not modified the sub driver in >>> HDMI path. We could let HDMI path print the warning and someone send >>> another patch later, or you modify for HDMI path in this patch. >> >> Sure, I'll add this in v6. After inspecting the code, I think we will need to >> also check for not initialized clocks in mtk_mdp_comp_init, as the driver for >> now does not even check if the clocks are present. What do you think? > > Yes, we do really need to consider mdp driver because mmsys clock > include mdp clock. You remind me that mmsys control 4 major function: > drm routing, drm clock, mdp routing, and mdp clock. Your design let the > mmsys device as drm device (control drm routing) and create a sub device > as clock device (control drm clock, mdp clock). If one day mdp device > (may need control drm routing) need to control the register of mdp > routing, would mdp device be a sub device? Or we need not to consider > this because it need not to access mmsys register now? > I think we should for now concentrate to fix the clock probing issue. If in the future we will need to access drm routing from the mdp device, we can have a look into this. Sounds good? Regards, Matthias