linux-kernel - Re: [PATCH 1/2] HID: trivial devm conversion for special hid drivers

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: <CAHp75Vcr1MP7eeKHK3pwh49R5Yy0-guQHBJjZ+WXyiR5MT+ZBg@mail.gmail.com>
Date:	Wed, 24 Jul 2013 18:37:45 +0300
From:	Andy Shevchenko <andy.shevchenko@...il.com>
To:	Benjamin Tissoires <benjamin.tissoires@...hat.com>
Cc:	Benjamin Tissoires <benjamin.tissoires@...il.com>,
	Henrik Rydberg <rydberg@...omail.se>,
	Jiri Kosina <jkosina@...e.cz>,
	Stephane Chatty <chatty@...c.fr>,
	linux-input <linux-input@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 1/2] HID: trivial devm conversion for special hid drivers

On Wed, Jul 24, 2013 at 6:29 PM, Benjamin Tissoires
<benjamin.tissoires@...hat.com> wrote:
> It is safe to use devres allocation within the hid subsystem:
> - the devres release is called _after_ the call to .remove(), meaning
>   that no freed pointers will exists while removing the device
> - if a .probe() fails, devres releases all the allocated ressources
>   before going to the next driver: there will not be ghost ressources
>   attached to a hid device if several drivers are probed.
>
> Given that, we can clean up a little some of the HID drivers. These ones
> are trivial:
> - there is only one kzalloc in the driver
> - the .remove() callback contains only one kfree on top of hid_hw_stop()
> - the error path in the probe is easy enough to be manually checked

Thanks for the patch! I'm sorry I didn't find time to do what I was
talking about last time.

Few comments below.

> --- a/drivers/hid/hid-a4tech.c
> +++ b/drivers/hid/hid-a4tech.c

> @@ -104,29 +103,16 @@ static int a4_probe(struct hid_device *hdev, const struct hid_device_id *id)
>         ret = hid_parse(hdev);
>         if (ret) {
>                 hid_err(hdev, "parse failed\n");
> -               goto err_free;
> +               return ret;
>         }
>
>         ret = hid_hw_start(hdev, HID_CONNECT_DEFAULT);
> -       if (ret) {
> +       if (ret)
>                 hid_err(hdev, "hw start failed\n");
> -               goto err_free;
> -       }
>
> -       return 0;

Isn't it better to leave explicit return 0? I think it would be fool
proof in case someone wants to add anything in the middle.

> -err_free:
> -       kfree(a4);
>         return ret;
>  }

> -static void a4_remove(struct hid_device *hdev)
> -{
> -       struct a4tech_sc *a4 = hid_get_drvdata(hdev);
> -
> -       hid_hw_stop(hdev);

Is it safe to remove this call?
This question is the same for all patched drivers.

-- 
With Best Regards,
Andy Shevchenko
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/