linux-kernel - Re: [PATCH 3/3] drm/rockchip: Add support for Rockchip Soc LVDS

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1b115a5e-417e-5f35-806b-e9af9780e01a@rock-chips.com>
Date:   Mon, 14 Aug 2017 11:59:49 +0800
From:   Sandy Huang <hjc@...k-chips.com>
To:     Sean Paul <seanpaul@...omium.org>,
        Sandy Huang <sandy.huang@...k-chips.com>
Cc:     Heiko Stuebner <heiko@...ech.de>, David Airlie <airlied@...ux.ie>,
        dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org,
        linux-rockchip@...ts.infradead.org,
        linux-arm-kernel@...ts.infradead.org,
        Mark Yao <mark.yao@...k-chips.com>
Subject: Re: [PATCH 3/3] drm/rockchip: Add support for Rockchip Soc LVDS



在 2017/8/11 22:44, Sean Paul 写道:
> On Fri, Aug 11, 2017 at 10:15:16AM +0800, Sandy Huang wrote:
>>
>>
>> 在 2017/8/11 2:05, Sean Paul 写道:
>>> On Thu, Aug 10, 2017 at 05:35:52PM +0800, Sandy Huang wrote:
>>>> Hi Sean Paul,
>>>>       Thanks for your review.
>>>>
>>>> 在 2017/8/10 3:58, Sean Paul 写道:
>>>>> On Wed, Aug 09, 2017 at 06:00:59PM +0800, Sandy Huang wrote:
>>>>>> This adds support for Rockchip soc lvds found on rk3288
>>>>>> Based on the patches from Mark yao and Heiko Stuebner
>>>>>>
>>>>>> Signed-off-by: Sandy Huang <hjc@...k-chips.com>
>>>>>> Signed-off-by: Mark yao <mark.yao@...k-chips.com>
>>>>>> Signed-off-by: Heiko Stuebner <heiko@...ech.de>
>>>>>> ---
>>>>>>     drivers/gpu/drm/rockchip/Kconfig         |   9 +
>>>>>>     drivers/gpu/drm/rockchip/Makefile        |   1 +
>>>>>>     drivers/gpu/drm/rockchip/rockchip_lvds.c | 734 +++++++++++++++++++++++++++++++
>>>>>>     drivers/gpu/drm/rockchip/rockchip_lvds.h | 112 +++++
>>>>>>     4 files changed, 856 insertions(+)
>>>>>>     create mode 100644 drivers/gpu/drm/rockchip/rockchip_lvds.c
>>>>>>     create mode 100644 drivers/gpu/drm/rockchip/rockchip_lvds.h
>>>>>>
>>>
>>> <snip />
>>>
>>>>>> diff --git a/drivers/gpu/drm/rockchip/rockchip_lvds.c b/drivers/gpu/drm/rockchip/rockchip_lvds.c
>>>>>> new file mode 100644
>>>>>> index 0000000..a4ad3f0
>>>>>> --- /dev/null
>>>>>> +++ b/drivers/gpu/drm/rockchip/rockchip_lvds.c
>>>
>>> <snip />
>>>
>>>>>> +	lvds->drm_dev = drm_dev;
>>>>>> +	port = of_graph_get_port_by_id(dev->of_node, 1);
>>>>>> +	if (!port) {
>>>>>> +		dev_err(dev, "can't found port point, please init lvds panel port!\n");
>>>>>> +		return -EINVAL;
>>>>>> +	}
>>>>>> +
>>>>>> +	for_each_child_of_node(port, endpoint) {
>>>>>> +		remote = of_graph_get_remote_port_parent(endpoint);
>>>>>> +		if (!remote) {
>>>>>> +			dev_err(dev, "can't found panel node, please init!\n");
>>>>>> +			ret = -EINVAL;
>>>>>> +			goto err_put_port;
>>>>>> +		}
>>>>>> +		if (!of_device_is_available(remote)) {
>>>>>> +			of_node_put(remote);
>>>>>> +			remote = NULL;
>>>>>> +			continue;
>>>>>> +		}
>>>>>> +		break;
>>>>>> +	}
>>>>>> +	if (!remote) {
>>>>>> +		dev_err(dev, "can't found remote node, please init!\n");
>>>>>> +		ret = -EINVAL;
>>>>>> +		goto err_put_port;
>>>>>> +	}
>>>>>> +
>>>>>> +	lvds->panel = of_drm_find_panel(remote);
>>>>>> +	if (!lvds->panel)
>>>>>> +		lvds->bridge = of_drm_find_bridge(remote);
>>>>>
>>>>> drm_of_find_panel_or_bridge()
>>>>>
>>>>
>>>> because the lvds ports maybe connect to lvds-panel or connect to
>>>> rk1000(which is convert RGB to CVBS output), so i have to get the remote
>>>> port parent and check the status, and final get the active remote point.
>>>>
>>>
>>>
>>>
>>>> lvds_panel: lvds-panel {
>>>> 	status = "disabled";
>>>> 	ports {
>>>> 		panel_in_lvds: endpoint {
>>>> 			remote-endpoint = <&lvds_out_panel>;
>>>> 		};
>>>> 	};
>>>> };
>>>>
>>>> rk1000: rk1000@...f000000 {
>>>> 	status = "okay";
>>>> 	ports {
>>>> 		rk1000_in_lvds: endpoint {
>>>> 			remote-endpoint = <&lvds_out_panel>;
>>>> 		};
>>>> 	};
>>>> };
>>>>
>>>> &lvds {
>>>> 	status = "okay";
>>>> 	ports {
>>>> 		lvds_out: port@1 {
>>>> 			reg = <1>;
>>>> 			lvds_out_panel: endpoint@0 {
>>>> 				reg = <0>;
>>>> 				remote-endpoint = <&panel_in_lvds>;
>>>> 			};
>>>> 			lvds_out_rk1000: endpoint@1 {
>>>> 				reg = <1>;
>>>> 				remote-endpoint = <&rk1000_in_lvds>;
>>>> 			};
>>>> 		};
>>>>
>>>> 	};
>>>> };
>>>
>>> Hi Sandy,
>>> Forgive me, this is probably a stupid question. I don't see how this usecase is
>>> unique from the other users of drm_of_find_panel_or_bridge. Couldn't you change
>>> your devicetree bindings to conform to something drm_of_find_panel_or_bridge()
>>> can work with?
>>>
>>> Thank you,
>>>
>> Hi sean,
>>      Maybe i can use the following method to use
>> drm_of_find_panel_or_bridge() and no need to change my DT, but there is
>> another question:
>>      The LVDS output format(rockchip,output、rockchip,data-mapping etc.)
>> depend on different panel, so it should be put under remote panel point.
>> If use drm_of_find_panel_or_bridge（), this just return panel or bridge, so
>> i have to back to get remote panel point and get the output format.
> 
> This should be easy since you can grab dev->of_node from panel or bridge once
> it's found.
> 
ok, thanks.
>>
>> ret = drm_of_find_panel_or_bridge(dev->of_node, 1, 0, &lvds->panel,
>>                                    &lvds->bridge);
>> if (ret)
>>      ret = drm_of_find_panel_or_bridge(dev->of_node, 1, 1, &lvds->panel,
>>                                        &lvds->bridge);
> 
> Would be easier to read in a for loop.
> 
> Sean
>
ok, thanks.

>> if (ret) {
>>      DRM_DEV_ERROR(dev, "failed to find panel and bridge node\n");
>>      ret  = -EPROBE_DEFER;
>>      goto err_put_remote;
>> }
>> 	
>>
>