Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20200525115235.5405-1-maxim.uvarov@linaro.org>
 <20200525115235.5405-2-maxim.uvarov@linaro.org> <096698e5493ffe714357b0d6926b900418f71dd0.camel@linux.intel.com>
In-Reply-To: <096698e5493ffe714357b0d6926b900418f71dd0.camel@linux.intel.com>
From:   Maxim Uvarov <maxim.uvarov@linaro.org>
Date:   Thu, 28 May 2020 11:26:35 +0300
Message-ID: <CAD8XO3a7fyi70bnNF5P2h-qT7ek_V8myMyiJh0LfVFzEQ4OKmQ@mail.gmail.com>
Subject: Re: [PATCHv3 1/3] optee: do drivers initialization before and after
 tee-supplicant run
To:     Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Cc:     Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        "tee-dev @ lists . linaro . org" <tee-dev@lists.linaro.org>,
        peterhuewe@gmx.de, Jason Gunthorpe <jgg@ziepe.ca>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Jens Wiklander <jens.wiklander@linaro.org>,
        linux-integrity@vger.kernel.org, Arnd Bergmann <arnd@linaro.org>,
        Sumit Garg <sumit.garg@linaro.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, 27 May 2020 at 22:56, Jarkko Sakkinen
<jarkko.sakkinen@linux.intel.com> wrote:
>
> On Mon, 2020-05-25 at 14:52 +0300, Maxim Uvarov wrote:
> > Some drivers (like ftpm) can operate only after tee-supplicant
> > runs becase of tee-supplicant provides things like storage
> > services.  This patch splits probe of non tee-supplicant dependable
> > drivers to early stage, and after tee-supplicant run probe other
> > drivers.
>
> Fix the typos and remove extra spaces (one space after '.').
>
> Why "tee-supplicant" and not "TEE-supplicant"?
>
> >
> > Signed-off-by: Maxim Uvarov <maxim.uvarov@linaro.org>
> > Suggested-by: Sumit Garg <sumit.garg@linaro.org>
> > Suggested-by: Arnd Bergmann <arnd@linaro.org>
> > ---
> >  drivers/tee/optee/core.c          | 28 +++++++++++++++++++++++++---
> >  drivers/tee/optee/device.c        | 17 +++++++++++------
> >  drivers/tee/optee/optee_private.h | 10 +++++++++-
> >  3 files changed, 45 insertions(+), 10 deletions(-)
> >
> > diff --git a/drivers/tee/optee/core.c b/drivers/tee/optee/core.c
> > index 99698b8a3a74..d059e3ac491c 100644
> > --- a/drivers/tee/optee/core.c
> > +++ b/drivers/tee/optee/core.c
> > @@ -17,6 +17,7 @@
> >  #include <linux/tee_drv.h>
> >  #include <linux/types.h>
> >  #include <linux/uaccess.h>
> > +#include <linux/workqueue.h>
> >  #include "optee_private.h"
> >  #include "optee_smc.h"
> >  #include "shm_pool.h"
> > @@ -218,6 +219,15 @@ static void optee_get_version(struct tee_device *teedev,
> >       *vers = v;
> >  }
> >
> > +static void optee_bus_scan(struct work_struct *work)
> > +{
> > +     int rc;
> > +
> > +     rc = optee_enumerate_devices(PTA_CMD_GET_DEVICES_SUPP);
> > +     if (rc)
> > +             pr_err("optee_enumerate_devices failed %d\n", rc);
> > +}
>
> Is pr_err() required really here? Why not just pr_info()?
>
> Why have it all and not just ftrace when needed? There is series going
> on to add CONFIG_FTRACE_FUNC_PROTOTYPE to that to get return values and
> arguments.
>

Actually if  optee_enumerate_devices() fail then I would say there is
something wrong with memory or Secure OS and more likely we should
stop execution.
I think something line WARN_ON might be more suitable here. What do you think?

> > +
> >  static int optee_open(struct tee_context *ctx)
> >  {
> >       struct optee_context_data *ctxdata;
> > @@ -241,8 +251,18 @@ static int optee_open(struct tee_context *ctx)
> >                       kfree(ctxdata);
> >                       return -EBUSY;
> >               }
> > -     }
> >
> > +             if (!optee->scan_bus_done) {
> > +                     INIT_WORK(&optee->scan_bus_work, optee_bus_scan);
> > +                     optee->scan_bus_wq = create_workqueue("optee_bus_scan");
> > +                     if (!optee->scan_bus_wq) {
> > +                             pr_err("optee: couldn't create workqueue\n");
>
> Neither get pr_err() here. Please remove this one completely.
>

What is the benefit of removing print here? Kernel binary size?

> > +                             return -ECHILD;
> > +                     }
> > +                     queue_work(optee->scan_bus_wq, &optee->scan_bus_work);
> > +                     optee->scan_bus_done = true;
> > +             }
> > +     }
> >       mutex_init(&ctxdata->mutex);
> >       INIT_LIST_HEAD(&ctxdata->sess_list);
> >
> > @@ -296,8 +316,10 @@ static void optee_release(struct tee_context *ctx)
> >
> >       ctx->data = NULL;
> >
> > -     if (teedev == optee->supp_teedev)
> > +     if (teedev == optee->supp_teedev) {
> > +             destroy_workqueue(optee->scan_bus_wq);
> >               optee_supp_release(&optee->supp);
> > +     }
> >  }
> >
> >  static const struct tee_driver_ops optee_ops = {
> > @@ -675,7 +697,7 @@ static int optee_probe(struct platform_device *pdev)
> >
> >       platform_set_drvdata(pdev, optee);
> >
> > -     rc = optee_enumerate_devices();
> > +     rc = optee_enumerate_devices(PTA_CMD_GET_DEVICES);
> >       if (rc) {
> >               optee_remove(pdev);
> >               return rc;
> > diff --git a/drivers/tee/optee/device.c b/drivers/tee/optee/device.c
> > index e3a148521ec1..d4931dad07aa 100644
> > --- a/drivers/tee/optee/device.c
> > +++ b/drivers/tee/optee/device.c
> > @@ -21,7 +21,6 @@
> >   * TEE_ERROR_BAD_PARAMETERS - Incorrect input param
> >   * TEE_ERROR_SHORT_BUFFER - Output buffer size less than required
> >   */
> > -#define PTA_CMD_GET_DEVICES          0x0
> >
> >  static int optee_ctx_match(struct tee_ioctl_version_data *ver, const void *data)
> >  {
> > @@ -32,7 +31,8 @@ static int optee_ctx_match(struct tee_ioctl_version_data *ver, const void *data)
> >  }
> >
> >  static int get_devices(struct tee_context *ctx, u32 session,
> > -                    struct tee_shm *device_shm, u32 *shm_size)
> > +                    struct tee_shm *device_shm, u32 *shm_size,
> > +                    u32 func)
> >  {
> >       int ret = 0;
> >       struct tee_ioctl_invoke_arg inv_arg;
> > @@ -42,7 +42,7 @@ static int get_devices(struct tee_context *ctx, u32 session,
> >       memset(&param, 0, sizeof(param));
> >
> >       /* Invoke PTA_CMD_GET_DEVICES function */
> > -     inv_arg.func = PTA_CMD_GET_DEVICES;
> > +     inv_arg.func = func;
> >       inv_arg.session = session;
> >       inv_arg.num_params = 4;
> >
> > @@ -87,7 +87,7 @@ static int optee_register_device(const uuid_t *device_uuid, u32 device_id)
> >       return rc;
> >  }
> >
> > -int optee_enumerate_devices(void)
> > +static int __optee_enumerate_devices(u32 func)
> >  {
> >       const uuid_t pta_uuid =
> >               UUID_INIT(0x7011a688, 0xddde, 0x4053,
> > @@ -118,7 +118,7 @@ int optee_enumerate_devices(void)
> >               goto out_ctx;
> >       }
> >
> > -     rc = get_devices(ctx, sess_arg.session, NULL, &shm_size);
> > +     rc = get_devices(ctx, sess_arg.session, NULL, &shm_size, func);
> >       if (rc < 0 || !shm_size)
> >               goto out_sess;
> >
> > @@ -130,7 +130,7 @@ int optee_enumerate_devices(void)
> >               goto out_sess;
> >       }
> >
> > -     rc = get_devices(ctx, sess_arg.session, device_shm, &shm_size);
> > +     rc = get_devices(ctx, sess_arg.session, device_shm, &shm_size, func);
> >       if (rc < 0)
> >               goto out_shm;
> >
> > @@ -158,3 +158,8 @@ int optee_enumerate_devices(void)
> >
> >       return rc;
> >  }
> > +
> > +int optee_enumerate_devices(u32 func)
> > +{
> > +     return  __optee_enumerate_devices(func);
> > +}
> > diff --git a/drivers/tee/optee/optee_private.h b/drivers/tee/optee/optee_private.h
> > index d9c5037b4e03..8b71839a357e 100644
> > --- a/drivers/tee/optee/optee_private.h
> > +++ b/drivers/tee/optee/optee_private.h
> > @@ -78,6 +78,9 @@ struct optee_supp {
> >   * @memremaped_shm   virtual address of memory in shared memory pool
> >   * @sec_caps:                secure world capabilities defined by
> >   *                   OPTEE_SMC_SEC_CAP_* in optee_smc.h
> > + * @scan_bus_done    flag if device registation was already done.
> > + * @scan_bus_wq              workqueue to scan optee bus and register optee drivers
> > + * @scan_bus_work    workq to scan optee bus and register optee drivers
> >   */
> >  struct optee {
> >       struct tee_device *supp_teedev;
> > @@ -89,6 +92,9 @@ struct optee {
> >       struct tee_shm_pool *pool;
> >       void *memremaped_shm;
> >       u32 sec_caps;
> > +     bool   scan_bus_done;
> > +     struct workqueue_struct *scan_bus_wq;
> > +     struct work_struct scan_bus_work;
> >  };
> >
> >  struct optee_session {
> > @@ -173,7 +179,9 @@ void optee_free_pages_list(void *array, size_t num_entries);
> >  void optee_fill_pages_list(u64 *dst, struct page **pages, int num_pages,
> >                          size_t page_offset);
> >
> > -int optee_enumerate_devices(void);
> > +#define PTA_CMD_GET_DEVICES          0x0
> > +#define PTA_CMD_GET_DEVICES_SUPP     0x1
> > +int optee_enumerate_devices(u32 func);
> >
> >  /*
> >   * Small helpers
>
> No exalanation of "scan bus" neither in the comments nor in the code.
>

Idea was when tee-supplicant opens /dev/tee-priv0 then we run
workqueue (to not hold current open()) and do bus scan for TA
applications which require tee-supplicant run. There is no "scan bus"
as API. The is optee_enumerate_devices() is split up on 2 stages.

Maxim.

> /Jarkko
>