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;
MIME-Version: 1.0
References: <20220422013215.2301793-1-tweek@google.com> <YmK9lhEENKFnr6sd@bombadil.infradead.org>
In-Reply-To: <YmK9lhEENKFnr6sd@bombadil.infradead.org>
From:   =?UTF-8?Q?Thi=C3=A9baud_Weksteen?= <tweek@google.com>
Date:   Tue, 26 Apr 2022 14:18:59 +1000
Message-ID: <CA+zpnLfS4nTjXYYgE_BNcjQPF1vWynOe_118YgTZa9S5G9dVdg@mail.gmail.com>
Subject: Re: [PATCH v2] firmware_loader: use kernel credentials when reading firmware
To:     Luis Chamberlain <mcgrof@kernel.org>
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Jeffrey Vander Stoep <jeffv@google.com>,
        Saravana Kannan <saravanak@google.com>,
        Alistair Delva <adelva@google.com>,
        Adam Shih <adamshih@google.com>,
        SElinux list <selinux@vger.kernel.org>,
        linux-kernel <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

> Can you elaborate on the last sentence? It's unclear how what you
> describe is used exactly to allow driver to use direct filesystem
> firmware loading.

I realize my use of the word "device" here was unfortunate. I meant devices as
Android devices/systems. This may have contributed to the confusion.

Previously, Android systems were not setting up the firmware_class.path
command line argument. It means that the userspace fallback was always
kicking-in when a driver called request_firmware. This was handled by the
ueventd process on Android, which is generally given access to all firmware
files.

Now that more devices are setting up firmware_class.path, the call to
request_firmware will end up using kernel_read_file_from_path_initns, which
would have used the current process credentials.

>
> And, given the feedback from Android it would seem this is a fix
> which likely may be desirable to backport to some stable kernels?

Yes, that's right.

Thanks