On Fri, Apr 22, 2022 at 11:32:15AM +1000, Thiébaud Weksteen wrote:
> Device drivers may decide to not load firmware when probed to avoid
> slowing down the boot process should the firmware filesystem not be
> available yet. In this case, the firmware loading request may be done
> when a device file associated with the driver is first accessed. The
> credentials of the userspace process accessing the device file may be
> used to validate access to the firmware files requested by the driver.
> Ensure that the kernel assumes the responsibility of reading the
> firmware.
>
> This was observed on Android for a graphic driver loading their firmware
> when the device file (e.g. /dev/mali0) was first opened by userspace
> (i.e. surfaceflinger). The security context of surfaceflinger was used
> to validate the access to the firmware file (e.g.
> /vendor/firmware/mali.bin).
>
> Because previous configurations were relying on the userspace fallback
> mechanism, the security context of the userspace daemon (i.e. ueventd)
> was consistently used to read firmware files. More devices are found to
> use the command line argument firmware_class.path which gives the kernel
> the opportunity to read the firmware directly, hence surfacing this
> misattribution.
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.
And, given the feedback from Android it would seem this is a fix
which likely may be desirable to backport to some stable kernels?
Otherwise looks good
Reviewed-by: Luis Chamberlain <mcgrof@xxxxxxxxxx>
> Signed-off-by: Thiébaud Weksteen <tweek@xxxxxxxxxx>
> ---
> v2: Add comment
>
> drivers/base/firmware_loader/main.c | 16 ++++++++++++++++
> 1 file changed, 16 insertions(+)
>
> diff --git a/drivers/base/firmware_loader/main.c b/drivers/base/firmware_loader/main.c
> index 94d1789a233e..8f3c2b2cfc61 100644
> --- a/drivers/base/firmware_loader/main.c
> +++ b/drivers/base/firmware_loader/main.c
> @@ -735,6 +735,8 @@ _request_firmware(const struct firmware **firmware_p, const char *name,
> size_t offset, u32 opt_flags)
> {
> struct firmware *fw = NULL;
> + struct cred *kern_cred = NULL;
> + const struct cred *old_cred;
> bool nondirect = false;
> int ret;
>
> @@ -751,6 +753,18 @@ _request_firmware(const struct firmware **firmware_p, const char *name,
> if (ret <= 0) /* error or already assigned */
> goto out;
>
> + /*
> + * We are about to try to access the firmware file. Because we may have been
> + * called by a driver when serving an unrelated request from userland, we use
> + * the kernel credentials to read the file.
> + */
> + kern_cred = prepare_kernel_cred(NULL);
> + if (!kern_cred) {
> + ret = -ENOMEM;
> + goto out;
> + }
> + old_cred = override_creds(kern_cred);
> +
> ret = fw_get_filesystem_firmware(device, fw->priv, "", NULL);
>
> /* Only full reads can support decompression, platform, and sysfs. */
> @@ -776,6 +790,8 @@ _request_firmware(const struct firmware **firmware_p, const char *name,
> } else
> ret = assign_fw(fw, device);
>
> + revert_creds(old_cred);
> +
> out:
> if (ret < 0) {
> fw_abort_batch_reqs(fw);
> --
> 2.36.0.rc2.479.g8af0fa9b8e-goog
>
