mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Guenter Roeck <linux@roeck-us.net>
To: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: Camel Guo <Camel.Guo@axis.com>, Jean Delvare <jdelvare@suse.com>,
	"linux-hwmon@vger.kernel.org" <linux-hwmon@vger.kernel.org>,
	kernel <kernel@axis.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2] hwmon: (tmp401) Add of_match_table
Date: Mon, 2 May 2022 09:16:23 -0700	[thread overview]
Message-ID: <25e2dc90-ab01-ab19-ffde-1b709dbebad4@roeck-us.net> (raw)
In-Reply-To: <20220502145800.GA21695@axis.com>

On 5/2/22 07:58, Vincent Whitchurch wrote:
> On Mon, May 02, 2022 at 03:57:50PM +0200, Guenter Roeck wrote:
>> On 5/2/22 02:19, Camel Guo wrote:
>>> When tmp401 is built as kernel module, it won't be automatically loaded
>>> even through there is a device node in the devicetree. e.g:
>>>       i2c {
>>>         #address-cells = <1>;
>>>         #size-cells = <0>;
>>>
>>>         sensor@4c {
>>>           compatible = "ti,tmp401";
>>>           reg = <0x4c>;
>>>         };
>>>       };
>>> In order to make sure it is loaded automatically, this commit adds
>>> of_match_table for tmp401.
>>>
>>
>> As mentioned before, historically i2c devices would instantiate based
>> on the i2c match table. You are claiming that this is no longer the case.
> 
> Note that while the commit message in the first version of the patch did
> wrongly claim that probe would not work without the of_match_table, this
> corrected description in v2 does mention the actual problem: that the
> module will not be automatically loaded without the of_match_table.  (If
> the module is loaded manually or the driver is built-in to the kernel,
> there is no problem.)
> 

No, it doesn't. None of the information you provided below is mentioned
in the description, but is essential to understand your patch and the
reason for it.

> See commit 72fc64c68decf119466 ("hwmon: (tmp103) Add OF device ID
> table") or commit 98b16a09861aa85d6 ("hwmon: (max31785) Add OF device ID
> table") for similar changes to other hwmon drivers.
> 

Those commits provide a valid and acceptable explanation.

> The potential future change mentioned in the commit messages of
> 72fc64c68decf119466 and 98b16a09861aa85d6 happened in commit
> af503716ac1444db61d80 ("i2c: core: report OF style module alias for
> devices registered via OF").  The commit message of
> af503716ac1444db61d80 has a lot of details about the change being made,
> and while it says that all in-tree drivers had been converted, it looks
> like some of them, like tmp401, were missed.
> 

And this is the missing link. If you provide that information
in the commit log I have no problems. Please also provide a Fixes:
tag.

Fixes: af503716ac14 ("i2c: core: report OF style module alias for devices registered via OF")

Thanks,
Guenter

>> The above is no evidence; that would require a log output on an affected
>> system showing that the sensors are not or no longer longer instantiated.
> 
> A log would simply show nothing happening so that's probably not going
> to be that useful, but here is what the modaliases look like.  As you
> can see, the modalias of the device in sysfs does not match any of the
> alias patterns of the module without this patch:
> 
> $ cat /sys/bus/i2c/devices/4-004c/modalias
> of:Ntemperature-sensorT<NULL>Cti,tmp431
> 
> modinfo without this patch:
> 
> $ modinfo ./modules/lib/modules/5.18.0-rc1/kernel/drivers/hwmon/tmp401.ko
> filename:       /storage2/femfyra/linux-2.6/.roadtest/./modules/lib/modules/5.18.0-rc1/kernel/drivers/hwmon/tmp401.ko
> license:        GPL
> description:    Texas Instruments TMP401 temperature sensor driver
> author:         Hans de Goede <hdegoede@redhat.com>
> alias:          i2c:tmp435
> alias:          i2c:tmp432
> alias:          i2c:tmp431
> alias:          i2c:tmp411
> alias:          i2c:tmp401
> depends:
> intree:         Y
> name:           tmp401
> vermagic:       5.18.0-rc1 mod_unload
> 
> modinfo after this patch:
> 
> $ modinfo ./modules/lib/modules/5.18.0-rc1/kernel/drivers/hwmon/tmp401.ko
> filename:       /storage2/femfyra/linux-2.6/./.roadtest/modules/lib/modules/5.18.0-rc1/kernel/drivers/hwmon/tmp401.ko
> license:        GPL
> description:    Texas Instruments TMP401 temperature sensor driver
> author:         Hans de Goede <hdegoede@redhat.com>
> alias:          i2c:tmp435
> alias:          i2c:tmp432
> alias:          i2c:tmp431
> alias:          i2c:tmp411
> alias:          i2c:tmp401
> alias:          of:N*T*Cti,tmp435C*
> alias:          of:N*T*Cti,tmp435
> alias:          of:N*T*Cti,tmp432C*
> alias:          of:N*T*Cti,tmp432
> alias:          of:N*T*Cti,tmp431C*
> alias:          of:N*T*Cti,tmp431
> alias:          of:N*T*Cti,tmp411C*
> alias:          of:N*T*Cti,tmp411
> alias:          of:N*T*Cti,tmp401C*
> alias:          of:N*T*Cti,tmp401
> depends:
> intree:         Y
> name:           tmp401
> vermagic:       5.18.0-rc1 mod_unload


  reply	other threads:[~2022-05-02 16:16 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-02  9:19 Camel Guo
2022-05-02 13:57 ` Guenter Roeck
2022-05-02 14:58   ` Vincent Whitchurch
2022-05-02 16:16     ` Guenter Roeck [this message]
2022-05-03  5:35       ` Camel Guo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=25e2dc90-ab01-ab19-ffde-1b709dbebad4@roeck-us.net \
    --to=linux@roeck-us.net \
    --cc=Camel.Guo@axis.com \
    --cc=jdelvare@suse.com \
    --cc=kernel@axis.com \
    --cc=linux-hwmon@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=vincent.whitchurch@axis.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®