linux-kernel - Re: [PATCH] misc: habanalabs: avoid HWMON dependency

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: <CAFCwf132vLS4jkc6QriNmV8bf4LzRGCyKXkdxQteJPcwuJBjUg@mail.gmail.com>
Date:   Mon, 4 Mar 2019 22:21:24 +0200
From:   Oded Gabbay <oded.gabbay@...il.com>
To:     Arnd Bergmann <arnd@...db.de>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Mike Rapoport <rppt@...ux.ibm.com>,
        "Linux-Kernel@...r. Kernel. Org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] misc: habanalabs: avoid HWMON dependency

On Mon, Mar 4, 2019 at 10:04 PM Arnd Bergmann <arnd@...db.de> wrote:
>
> When CONFIG_HWMON is disabled, we currently get a link error for this driver:
>
> ERROR: "hwmon_device_unregister" [drivers/misc/habanalabs/habanalabs.ko] undefined!
> ERROR: "hwmon_device_register_with_info" [drivers/misc/habanalabs/habanalabs.ko] undefined!
>
> As the hardware monitoring is not a required part of the functionality,
> make it possible to compile without HWMON by ignoring all that functionality
> depending on the configuration.
>
> Signed-off-by: Arnd Bergmann <arnd@...db.de>
> ---
>  drivers/misc/habanalabs/hwmon.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/misc/habanalabs/hwmon.c b/drivers/misc/habanalabs/hwmon.c
> index 77facd25c4a2..933746639831 100644
> --- a/drivers/misc/habanalabs/hwmon.c
> +++ b/drivers/misc/habanalabs/hwmon.c
> @@ -423,7 +423,7 @@ int hl_hwmon_init(struct hl_device *hdev)
>         struct device *dev = hdev->pdev ? &hdev->pdev->dev : hdev->dev;
>         int rc;
>
> -       if ((hdev->hwmon_initialized) || !(hdev->fw_loading))
> +       if (!IS_ENABLED(CONFIG_HWMON) || (hdev->hwmon_initialized) || !(hdev->fw_loading))
>                 return 0;
>
>         if (hdev->hl_chip_info->info) {
> @@ -451,7 +451,7 @@ int hl_hwmon_init(struct hl_device *hdev)
>
>  void hl_hwmon_fini(struct hl_device *hdev)
>  {
> -       if (!hdev->hwmon_initialized)
> +       if (!IS_ENABLED(CONFIG_HWMON) || !hdev->hwmon_initialized)
>                 return;
>
>         hwmon_device_unregister(hdev->hwmon_dev);
> --
> 2.20.0
>

Hi Arnd,
I sent a patch to gkh to fix this last week (and he merged that into
his tree), but I took a different approach of selecting HWMON in
kconfig (with an added depends on of HAS_IOMEM to prevent dependency
breakage).
I don't really want the driver to load without HWMON support. I took
the example from AMD's amdgpu driver which does the same thing.

Thanks,
Oded