linux-kernel - Re: [PATCH 2/5] firmware: arm_scmi: fix transfer missing re-initialization

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: <CAN5uoS9EGrzoaJ7tA__A=dPx2Y1Cp=v2THMm9YJTrOLUpkaggg@mail.gmail.com>
Date:   Fri, 9 Oct 2020 14:38:16 +0200
From:   Etienne Carriere <etienne.carriere@...aro.org>
To:     Sudeep Holla <sudeep.holla@....com>
Cc:     linux-kernel@...r.kernel.org,
        "moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE" 
        <linux-arm-kernel@...ts.infradead.org>,
        Cristian Marussi <cristian.marussi@....com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Souvik Chakravarty <Souvik.Chakravarty@....com>
Subject: Re: [PATCH 2/5] firmware: arm_scmi: fix transfer missing re-initialization

On Thu, 8 Oct 2020 at 23:18, Sudeep Holla <sudeep.holla@....com> wrote:
>
> On Thu, Oct 08, 2020 at 04:37:19PM +0200, Etienne Carriere wrote:
> > Implement helper function scmi_do_xfer_again() to process consecutive
> > transfers that are initialized only once with scmi_xfer_get_init()
> > and hence get the pool completion and responses message length not
> > reloaded regarding last completed transfer.
> >
> > Signed-off-by: Etienne Carriere <etienne.carriere@...aro.org>
> > ---
> >  drivers/firmware/arm_scmi/base.c    |  2 +-
> >  drivers/firmware/arm_scmi/clock.c   |  2 +-
> >  drivers/firmware/arm_scmi/common.h  |  2 ++
> >  drivers/firmware/arm_scmi/driver.c  | 10 ++++++++++
> >  drivers/firmware/arm_scmi/perf.c    |  2 +-
> >  drivers/firmware/arm_scmi/sensors.c |  2 +-
> >  6 files changed, 16 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/firmware/arm_scmi/base.c b/drivers/firmware/arm_scmi/base.c
> > index 9853bd3c4d45..508f214baa1b 100644
> > --- a/drivers/firmware/arm_scmi/base.c
> > +++ b/drivers/firmware/arm_scmi/base.c
> > @@ -183,7 +183,7 @@ static int scmi_base_implementation_list_get(const struct scmi_handle *handle,
> >               /* Set the number of protocols to be skipped/already read */
> >               *num_skip = cpu_to_le32(tot_num_ret);
> >
> > -             ret = scmi_do_xfer(handle, t);
> > +             ret = scmi_do_xfer_again(handle, t);
> >               if (ret)
> >                       break;
> >
> > diff --git a/drivers/firmware/arm_scmi/clock.c b/drivers/firmware/arm_scmi/clock.c
> > index c1cfe3ee3d55..9bb54c1a8d55 100644
> > --- a/drivers/firmware/arm_scmi/clock.c
> > +++ b/drivers/firmware/arm_scmi/clock.c
> > @@ -161,7 +161,7 @@ scmi_clock_describe_rates_get(const struct scmi_handle *handle, u32 clk_id,
> >               /* Set the number of rates to be skipped/already read */
> >               clk_desc->rate_index = cpu_to_le32(tot_rate_cnt);
> >
> > -             ret = scmi_do_xfer(handle, t);
> > +             ret = scmi_do_xfer_again(handle, t);
> >               if (ret)
> >                       goto err;
> >
> > diff --git a/drivers/firmware/arm_scmi/common.h b/drivers/firmware/arm_scmi/common.h
> > index 37fb583f1bf5..6d4eea7b0f3e 100644
> > --- a/drivers/firmware/arm_scmi/common.h
> > +++ b/drivers/firmware/arm_scmi/common.h
> > @@ -143,6 +143,8 @@ struct scmi_xfer {
> >
> >  void scmi_xfer_put(const struct scmi_handle *h, struct scmi_xfer *xfer);
> >  int scmi_do_xfer(const struct scmi_handle *h, struct scmi_xfer *xfer);
> > +int scmi_do_xfer_again(const struct scmi_handle *handle,
> > +                    struct scmi_xfer *xfer);
> >  int scmi_do_xfer_with_response(const struct scmi_handle *h,
> >                              struct scmi_xfer *xfer);
> >  int scmi_xfer_get_init(const struct scmi_handle *h, u8 msg_id, u8 prot_id,
> > diff --git a/drivers/firmware/arm_scmi/driver.c b/drivers/firmware/arm_scmi/driver.c
> > index c5dea87edf8f..887cb8249db0 100644
> > --- a/drivers/firmware/arm_scmi/driver.c
> > +++ b/drivers/firmware/arm_scmi/driver.c
> > @@ -402,6 +402,16 @@ int scmi_do_xfer(const struct scmi_handle *handle, struct scmi_xfer *xfer)
> >       return ret;
> >  }
> >
> > +int scmi_do_xfer_again(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;
>
> I am tempted to just have helper for above and use it where needed.
> Or may be I just don't like the name of the function, how about
> scmi_do_xfer_rxlen_reinit or anything else. Suggestions ?

I think a smoother way would be that scmi_do_xfer() initializes
both
  xfer->rx.len = info->desc->max_msg_size
and
  xfer->hdr.poll_completion = false
instead of doing that from scmi_xfer_get_init().

>
> > +     xfer->hdr.poll_completion = false;
>
> Do we really need the above ?

I think so. Once a transfer is completed, poll_completion is true. But
the next transfer we expect should start assuming completion is not
yet done, hence this reset to false.

Regards,
Etienne

>
> --
> Regards,
> Sudeep