On Fri, Sep 20, 2024 at 11:01:40AM GMT, Elliot Berman wrote:
> Some devicetrees representing Qualcomm Technologies, Inc. SoCs are
> missing the SCM node. Users of the SCM device assume the device is
> present and the driver also assumes it has probed. This can lead to
> unanticipated crashes when there isn't an SCM device. All Qualcomm
> Technologies, Inc. SoCs use SCM to communicate with firmware, so create
> the platform device if it's not present in the devicetree.
Which devicetrees? I assume that this mostly concerns arm32 machines,
but I don't see if you have tested this on any of them. Also on some of
those machines SCM require additional clocks, I don't see that being
handled in the patch.
If we are to manually instantiate SCM node, I'd prefer for it to be
explicit, e.g. MSM8x60, create SCM device, using this-and-that clock.
> Tested that SCM node still probes on:
> - sm8650-qrd with the SCM DT node still present
> - sm845-mtp with the SCM DT node still present
> - sm845-mtp with the node removed
>
> Fixes: 449d0d84bcd8 ("firmware: qcom: scm: smc: switch to using the SCM allocator")
> Reported-by: Rudraksha Gupta <guptarud@xxxxxxxxx>
> Closes: https://lore.kernel.org/lkml/692cfe9a-8c05-4ce4-813e-82b3f310019a@xxxxxxxxx/
> Link: https://lore.kernel.org/all/CAA8EJpqSKbKJ=y0LAigGdj7_uk+5mezDgnzV5XEzwbxRJgpN1w@xxxxxxxxxxxxxx/
> Suggested-by: Bartosz Golaszewski <brgl@xxxxxxxx>
> Signed-off-by: Elliot Berman <quic_eberman@xxxxxxxxxxx>
> ---
> drivers/firmware/qcom/qcom_scm.c | 75 +++++++++++++++++++++++++++++++++++-----
> 1 file changed, 66 insertions(+), 9 deletions(-)
--
With best wishes
Dmitry
