Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Feedback-ID: i56a14606:Fastmail
User-Agent: Cyrus-JMAP/3.9.0-alpha0-499-gf27bbf33e2-fm-20230619.001-gf27bbf33
Mime-Version: 1.0
Message-Id: <227ff023-a0fb-476d-ae5c-0733cc6223eb@app.fastmail.com>
In-Reply-To: <9c52ce87-f69c-e69c-ec43-6b2cc3b37c4a@linaro.org>
References: <20230622101813.3453772-1-arnd@kernel.org>
 <286cf2d8-2da8-7203-ee53-bf08332fa519@linaro.org>
 <f28f78fc-84ce-4990-9126-4075f3d88e01@app.fastmail.com>
 <9c52ce87-f69c-e69c-ec43-6b2cc3b37c4a@linaro.org>
Date:   Thu, 22 Jun 2023 17:57:36 +0200
From:   "Arnd Bergmann" <arnd@arndb.de>
To:     "Neil Armstrong" <neil.armstrong@linaro.org>,
        "Arnd Bergmann" <arnd@kernel.org>,
        "Heikki Krogerus" <heikki.krogerus@linux.intel.com>,
        "Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
        "Dmitry Baryshkov" <dmitry.baryshkov@linaro.org>
Cc:     "Bjorn Andersson" <andersson@kernel.org>,
        linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] usb: typec: nb7vpq904m: add CONFIG_DRM dependency
Content-Type: text/plain
Precedence: bulk

On Thu, Jun 22, 2023, at 17:39, Neil Armstrong wrote:
> On 22/06/2023 13:24, Arnd Bergmann wrote:
>> On Thu, Jun 22, 2023, at 12:39, Neil Armstrong wrote:
>>>>    
>>>>    config TYPEC_MUX_NB7VPQ904M
>>>>    	tristate "On Semiconductor NB7VPQ904M Type-C redriver driver"
>>>> -	depends on I2C
>>>> +	depends on I2C && DRM
>>>>    	select REGMAP_I2C
>>>>    	help
>>>>    	  Say Y or M if your system has a On Semiconductor NB7VPQ904M Type-C
>>>
>>> I think it could be :
>>>
>>> +	depends on DRM || DRM=n
>>> +	select DRM_PANEL_BRIDGE if DRM
>> 
>> As far as I can tell, this would only avoid the link error
>> against devm_drm_of_get_bridge(), but not the one against
>> devm_drm_bridge_add(), which is defined in drm.ko through
>> drivers/gpu/drm/drm_bridge.c.
>
> I'm trying to reproduce such situation, but so fail I fail.
>
> In the driver there's a guard to avoid calling into DRM functions
> when disabled:
> #if IS_ENABLED(CONFIG_OF) && IS_ENABLED(CONFIG_DRM_PANEL_BRIDGE)

Ah, you are right. I got confused because the check is in
header file for devm_drm_of_get_bridge(), but not for
devm_drm_bridge_add(), which has the check in the source
file as you point out.

> so I wonder which kind on config leaded to that since
> CONFIG_DRM_PANEL_BRIDGE is only enabled when DRM_PANEL and DRM are.

I only saw the original issue with

CONFIG_DRM=m
CONFIG_DRM_PANEL=y
CONFIG_DRM_BRIDGE=y
CONFIG_DRM_PANEL_BRIDGE=y
CONFIG_TYPEC_MUX_NB7VPQ904M=y

and since CONFIG_DRM_PANEL_BRIDGE already depends on CONFIG_DRM,
I think that is the only one that can go wrong, so your
suggestion of

   depends on DRM || DRM=n

should be sufficient. I see that DRM_PANEL, DRM_BRIDGE and
DRM_PANEL_BRIDGE are now always =y whenever DRM is enabled,
so I don't think the 'select CONFIG_DRM_PANEL_BRIDGE' serves
any purpose any more, but it's also harmless if you think it
helps for clarification.

Can you send the updated patch, or should I?


     Arnd