From mboxrd@z Thu Jan 1 00:00:00 1970 From: CK Hu Subject: Re: [PATCH 3/8] drm/mediatek: mt2701: switch to mfd probing. Date: Thu, 23 Nov 2017 16:47:00 +0800 Message-ID: <1511426820.8489.2.camel@mtksdaap41> References: <20171114214114.15793-1-mbrugger@suse.com> <20171114214114.15793-4-mbrugger@suse.com> <1511416092.20405.4.camel@mtksdaap41> <66be07f1-c084-8db1-c845-86b2057a0d5d@suse.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <66be07f1-c084-8db1-c845-86b2057a0d5d@suse.com> Sender: linux-clk-owner@vger.kernel.org To: Matthias Brugger Cc: Matthias Brugger , mark.rutland@arm.com, sean.wang@mediatek.com, airlied@linux.ie, gregkh@linuxfoundation.org, mturquette@baylibre.com, rdunlap@infradead.org, sboyd@codeaurora.org, linux@armlinux.org.uk, dri-devel@lists.freedesktop.org, linux-clk@vger.kernel.org, pi-cheng.chen@linaro.org, robh+dt@kernel.org, linux-mediatek@lists.infradead.org, laurent.pinchart@ideasonboard.com, p.zabel@pengutronix.de, ulrich.hecht+renesas@gmail.com, mchehab@kernel.org, lee.jones@linaro.org, davem@davemloft.net, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org List-Id: linux-mediatek@lists.infradead.org Hi, Matthias: On Thu, 2017-11-23 at 09:30 +0100, Matthias Brugger wrote: > > On 11/23/2017 06:48 AM, CK Hu wrote: > > Hi, Matthias: > > > > On Tue, 2017-11-14 at 22:41 +0100, Matthias Brugger wrote: > >> With the mtk-mmsys MFD device in place, we switch the probing for > >> mt2701 from device-tree to mfd. > >> > >> Signed-off-by: Matthias Brugger > >> --- > >> drivers/gpu/drm/mediatek/mtk_drm_drv.c | 32 +++++++++++++++++++++++++------- > >> 1 file changed, 25 insertions(+), 7 deletions(-) > >> > >> diff --git a/drivers/gpu/drm/mediatek/mtk_drm_drv.c b/drivers/gpu/drm/mediatek/mtk_drm_drv.c > >> index dd249cf5121e..5a263aa3ab6e 100644 > >> --- a/drivers/gpu/drm/mediatek/mtk_drm_drv.c > >> +++ b/drivers/gpu/drm/mediatek/mtk_drm_drv.c > >> @@ -21,6 +21,7 @@ > >> #include > >> #include > >> #include > >> +#include > >> #include > >> #include > >> #include > >> @@ -392,9 +393,10 @@ static const struct of_device_id mtk_ddp_comp_dt_ids[] = { > >> > >> static int mtk_drm_probe(struct platform_device *pdev) > >> { > >> + struct mmsys_dev *mmsys_private; > >> struct device *dev = &pdev->dev; > >> struct mtk_drm_private *private; > >> - struct device_node *node; > >> + struct device_node *node, *parent_node; > >> struct component_match *match = NULL; > >> int ret; > >> int i; > >> @@ -407,12 +409,23 @@ static int mtk_drm_probe(struct platform_device *pdev) > >> INIT_WORK(&private->commit.work, mtk_atomic_work); > >> private->data = of_device_get_match_data(dev); > >> > >> - private->config_regs = syscon_node_to_regmap(dev->of_node); > >> - if (IS_ERR(private->config_regs)) > >> - return PTR_ERR(private->config_regs); > >> + /* Check if called from mfd */ > >> + if (!dev->of_node) { > >> + mmsys_private = dev_get_drvdata(pdev->dev.parent); > > > > Why do you directly access parent's driver data? You just need the > > device node of mmsys, maybe you could refer to [1]. > > > > [1] > > https://elixir.free-electrons.com/linux/latest/source/drivers/reset/reset-berlin.c#L78 > > > > The difference is, that the driver you mentioned gets probed via device tree > matching, while we get invoked here through the id_table. So there is no device > node (the if actually checkes for pdev->dev.of_node to identify exactly this case). > > Regards, > Matthias Yes, you are right. So Acked-by: CK Hu Regards, CK > > _______________________________________________ > Linux-mediatek mailing list > Linux-mediatek@lists.infradead.org > http://lists.infradead.org/mailman/listinfo/linux-mediatek