Re: [PATCH] firmware: arm_scmi: Add missing Rx size re-initialisation

[Sudeep Holla]

On Tue, Oct 13, 2020 at 11:58:09AM +0200, Etienne Carriere wrote:
> On Mon, 12 Oct 2020 at 16:17, Sudeep Holla <sudeep.holla@xxxxxxx> wrote:
> >
> > Few commands provide the list of description partially and require
> > to be called consecutively until all the descriptors are fetched
> > completely. In such cases, we don't release the buffers and reuse
> > them for consecutive transmits.
> >
> > However, currently we don't reset the Rx size which will be set as
> > per the response for the last transmit. This may result in incorrect
> > response size being interpretted as the firmware may repond with size
> > greater than the one set but we read only upto the size set by previous
> > response.
> >
> > Let us reset the receive buffer size to max possible in such cases as
> > we don't know the exact size of the response.
> >
> > Fixes: b6f20ff8bd94 ("firmware: arm_scmi: add common infrastructure and support for base protocol")
> > Reported-by: Etienne Carriere <etienne.carriere@xxxxxxxxxx>
> > Signed-off-by: Sudeep Holla <sudeep.holla@xxxxxxx>
> > ---
> >  drivers/firmware/arm_scmi/base.c    | 2 ++
> >  drivers/firmware/arm_scmi/clock.c   | 2 ++
> >  drivers/firmware/arm_scmi/common.h  | 8 ++++++++
> >  drivers/firmware/arm_scmi/perf.c    | 2 ++
> >  drivers/firmware/arm_scmi/sensors.c | 2 ++
> >  5 files changed, 16 insertions(+)
> >
> > Hi Etienne,
> >
> > I reworked this in a different way and hence dropped your authorship and added
> > reported by. If you prefer I can attribute you as author. I want to push
> > 2,4,5/5 as fixes and hence the rush.
> 
> Hi Sudeep,
> 
> Tags are fine like that.
> As for the content, it looks good to me.

Thanks.

> When trying to apply this, I failed, but I guess I'm not testing over
> the same kernel tree/branch as you.

Ah OK. I wasn't sure if we had touch code around these recently.
Anyway, I have all these on my for-next/scmi[1][2]

--
Regards,
Sudeep

[1] https://git.kernel.org/sudeep.holla/linux/h/for-next/scmi
[2] git://git.kernel.org/pub/scm/linux/kernel/git/sudeep.holla/linux.git for-next/scmi

> All in one, I am really fine with this change, I think it does the job
> 
> Regards,
> Etienne
> 
> >
> > Regards,
> > Sudeep
> >
> > diff --git a/drivers/firmware/arm_scmi/base.c b/drivers/firmware/arm_scmi/base.c
> > index 9853bd3c4d45..017e5d8bd869 100644
> > --- a/drivers/firmware/arm_scmi/base.c
> > +++ b/drivers/firmware/arm_scmi/base.c
> > @@ -197,6 +197,8 @@ static int scmi_base_implementation_list_get(const struct scmi_handle *handle,
> >                         protocols_imp[tot_num_ret + loop] = *(list + loop);
> >
> >                 tot_num_ret += loop_num_ret;
> > +
> > +               scmi_reset_rx_to_maxsz(handle, t);
> >         } while (loop_num_ret);
> >
> >         scmi_xfer_put(handle, t);
> > diff --git a/drivers/firmware/arm_scmi/clock.c b/drivers/firmware/arm_scmi/clock.c
> > index c1cfe3ee3d55..4645677d86f1 100644
> > --- a/drivers/firmware/arm_scmi/clock.c
> > +++ b/drivers/firmware/arm_scmi/clock.c
> > @@ -192,6 +192,8 @@ scmi_clock_describe_rates_get(const struct scmi_handle *handle, u32 clk_id,
> >                 }
> >
> >                 tot_rate_cnt += num_returned;
> > +
> > +               scmi_reset_rx_to_maxsz(handle, t);
> >                 /*
> >                  * check for both returned and remaining to avoid infinite
> >                  * loop due to buggy firmware
> > diff --git a/drivers/firmware/arm_scmi/common.h b/drivers/firmware/arm_scmi/common.h
> > index 37fb583f1bf5..a3f1bc44b1de 100644
> > --- a/drivers/firmware/arm_scmi/common.h
> > +++ b/drivers/firmware/arm_scmi/common.h
> > @@ -245,6 +245,14 @@ extern const struct scmi_desc scmi_mailbox_desc;
> >  extern const struct scmi_desc scmi_smc_desc;
> >  #endif
> >
> > +static inline void scmi_reset_rx_to_maxsz(const struct scmi_handle *handle,
> > +                                         struct scmi_xfer *xfer)
> > +{
> > +       struct scmi_info *info = handle_to_scmi_info(handle);
> > +
> > +       xfer->rx.len = info->desc->max_msg_size;
> > +}
> > +
> >  void scmi_rx_callback(struct scmi_chan_info *cinfo, u32 msg_hdr);
> >  void scmi_free_channel(struct scmi_chan_info *cinfo, struct idr *idr, int id);
> >
> > diff --git a/drivers/firmware/arm_scmi/perf.c b/drivers/firmware/arm_scmi/perf.c
> > index ed475b40bd08..82fb3babff72 100644
> > --- a/drivers/firmware/arm_scmi/perf.c
> > +++ b/drivers/firmware/arm_scmi/perf.c
> > @@ -304,6 +304,8 @@ scmi_perf_describe_levels_get(const struct scmi_handle *handle, u32 domain,
> >                 }
> >
> >                 tot_opp_cnt += num_returned;
> > +
> > +               scmi_reset_rx_to_maxsz(handle, t);
> >                 /*
> >                  * check for both returned and remaining to avoid infinite
> >                  * loop due to buggy firmware
> > diff --git a/drivers/firmware/arm_scmi/sensors.c b/drivers/firmware/arm_scmi/sensors.c
> > index 9703cf6356a0..b4232d611033 100644
> > --- a/drivers/firmware/arm_scmi/sensors.c
> > +++ b/drivers/firmware/arm_scmi/sensors.c
> > @@ -166,6 +166,8 @@ static int scmi_sensor_description_get(const struct scmi_handle *handle,
> >                 }
> >
> >                 desc_index += num_returned;
> > +
> > +               scmi_reset_rx_to_maxsz(handle, t);
> >                 /*
> >                  * check for both returned and remaining to avoid infinite
> >                  * loop due to buggy firmware
> > --
> > 2.17.1
> >