From mboxrd@z Thu Jan 1 00:00:00 1970 From: Russell King - ARM Linux Subject: Re: [PATCH v5 01/11] staging: imx-drm-core: Use OF graph to find components and connections between encoder and crtcs Date: Wed, 5 Mar 2014 10:05:17 +0000 Message-ID: <20140305100516.GV21483@n2100.arm.linux.org.uk> References: <1394011262-16849-1-git-send-email-p.zabel@pengutronix.de> <1394011262-16849-2-git-send-email-p.zabel@pengutronix.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <1394011262-16849-2-git-send-email-p.zabel@pengutronix.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: driverdev-devel-bounces@linuxdriverproject.org Sender: driverdev-devel-bounces@linuxdriverproject.org To: Philipp Zabel Cc: devel@driverdev.osuosl.org, devicetree@vger.kernel.org, Philipp Zabel , David Airlie , Greg Kroah-Hartman , dri-devel@lists.freedesktop.org, kernel@pengutronix.de, Grant Likely , Shawn Guo , linux-arm-kernel@lists.infradead.org List-Id: devicetree@vger.kernel.org On Wed, Mar 05, 2014 at 10:20:52AM +0100, Philipp Zabel wrote: > +struct imx_drm_component { > + struct device_node *of_node; > + struct list_head list; > +}; > + The only thing this structure appears to be doing is ensuring that a single component doesn't get added twice - is that correct? If so, (and the troublesome problem with the IPU crtcs is now gone) we can modify the core component code such that it does this: if (c->master && c->master != master) continue; if (compare(c->dev, compare_data)) { if (!c->master) component_attach_master(master, c); ret = 0; break; } which will mean that you don't need to build this list anymore to track what will be added - though I'd like to think a little more about that before making that change. Please confirm whether this will eliminate your list generation. Thanks. -- FTTC broadband for 0.8mile line: now at 9.7Mbps down 460kbps up... slowly improving, and getting towards what was expected from it. From mboxrd@z Thu Jan 1 00:00:00 1970 From: linux@arm.linux.org.uk (Russell King - ARM Linux) Date: Wed, 5 Mar 2014 10:05:17 +0000 Subject: [PATCH v5 01/11] staging: imx-drm-core: Use OF graph to find components and connections between encoder and crtcs In-Reply-To: <1394011262-16849-2-git-send-email-p.zabel@pengutronix.de> References: <1394011262-16849-1-git-send-email-p.zabel@pengutronix.de> <1394011262-16849-2-git-send-email-p.zabel@pengutronix.de> Message-ID: <20140305100516.GV21483@n2100.arm.linux.org.uk> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Wed, Mar 05, 2014 at 10:20:52AM +0100, Philipp Zabel wrote: > +struct imx_drm_component { > + struct device_node *of_node; > + struct list_head list; > +}; > + The only thing this structure appears to be doing is ensuring that a single component doesn't get added twice - is that correct? If so, (and the troublesome problem with the IPU crtcs is now gone) we can modify the core component code such that it does this: if (c->master && c->master != master) continue; if (compare(c->dev, compare_data)) { if (!c->master) component_attach_master(master, c); ret = 0; break; } which will mean that you don't need to build this list anymore to track what will be added - though I'd like to think a little more about that before making that change. Please confirm whether this will eliminate your list generation. Thanks. -- FTTC broadband for 0.8mile line: now at 9.7Mbps down 460kbps up... slowly improving, and getting towards what was expected from it.