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: <CAFA6WYNj_onTOtETKTLTGG6=GYHMDvty90KftLkfEz_v7nZx9w@mail.gmail.com>
Date:   Mon, 14 Mar 2022 14:33:26 +0530
From:   Sumit Garg <sumit.garg@...aro.org>
To:     Jens Wiklander <jens.wiklander@...aro.org>
Cc:     linux-kernel@...r.kernel.org, op-tee@...ts.trustedfirmware.org
Subject: Re: [PATCH 2/3] optee: add FF-A capability OPTEE_FFA_SEC_CAP_ARG_OFFSET

On Wed, 2 Mar 2022 at 01:18, Jens Wiklander <jens.wiklander@...aro.org> wrote:
>
> Adds the secure capability OPTEE_FFA_SEC_CAP_ARG_OFFSET to indicate that
> OP-TEE with FF-A can support an argument struct at a non-zero offset into
> a passed shared memory object.
>

It isn't clear to me why FF-A ABI requires this capability.
shm->offset should be honoured by default, if it isn't then it's a
bug. AFAIK, FF-A is currently still in its early stages so it
shouldn't be that hard to fix bugs in the ABI.

-Sumit

> Signed-off-by: Jens Wiklander <jens.wiklander@...aro.org>
> ---
>  drivers/tee/optee/ffa_abi.c   | 17 +++++++++++++++--
>  drivers/tee/optee/optee_ffa.h | 12 +++++++++++-
>  2 files changed, 26 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/tee/optee/ffa_abi.c b/drivers/tee/optee/ffa_abi.c
> index 7686f7020616..cc863aaefcd9 100644
> --- a/drivers/tee/optee/ffa_abi.c
> +++ b/drivers/tee/optee/ffa_abi.c
> @@ -615,12 +615,21 @@ static int optee_ffa_do_call_with_arg(struct tee_context *ctx,
>                 .data0 = OPTEE_FFA_YIELDING_CALL_WITH_ARG,
>                 .data1 = (u32)shm->sec_world_id,
>                 .data2 = (u32)(shm->sec_world_id >> 32),
> -               .data3 = shm->offset,
> +               .data3 = 0,
>         };
>         struct optee_msg_arg *arg;
>         unsigned int rpc_arg_offs;
>         struct optee_msg_arg *rpc_arg;
>
> +       /*
> +        * The shared memory object has to start on a page when passed as
> +        * an argument struct. This is also what the shm pool allocator
> +        * returns, but check this before calling secure world to catch
> +        * eventual errors early in case something changes.
> +        */
> +       if (shm->offset)
> +               return -EINVAL;
> +
>         arg = tee_shm_get_va(shm, 0);
>         if (IS_ERR(arg))
>                 return PTR_ERR(arg);
> @@ -678,6 +687,7 @@ static bool optee_ffa_api_is_compatbile(struct ffa_device *ffa_dev,
>
>  static bool optee_ffa_exchange_caps(struct ffa_device *ffa_dev,
>                                     const struct ffa_dev_ops *ops,
> +                                   u32 *sec_caps,
>                                     unsigned int *rpc_param_count)
>  {
>         struct ffa_send_direct_data data = { OPTEE_FFA_EXCHANGE_CAPABILITIES };
> @@ -694,6 +704,7 @@ static bool optee_ffa_exchange_caps(struct ffa_device *ffa_dev,
>         }
>
>         *rpc_param_count = (u8)data.data1;
> +       *sec_caps = data.data2;
>
>         return true;
>  }
> @@ -777,6 +788,7 @@ static int optee_ffa_probe(struct ffa_device *ffa_dev)
>         struct tee_device *teedev;
>         struct tee_context *ctx;
>         struct optee *optee;
> +       u32 sec_caps;
>         int rc;
>
>         ffa_ops = ffa_dev_ops_get(ffa_dev);
> @@ -788,7 +800,8 @@ static int optee_ffa_probe(struct ffa_device *ffa_dev)
>         if (!optee_ffa_api_is_compatbile(ffa_dev, ffa_ops))
>                 return -EINVAL;
>
> -       if (!optee_ffa_exchange_caps(ffa_dev, ffa_ops, &rpc_param_count))
> +       if (!optee_ffa_exchange_caps(ffa_dev, ffa_ops, &sec_caps,
> +                                    &rpc_param_count))
>                 return -EINVAL;
>
>         optee = kzalloc(sizeof(*optee), GFP_KERNEL);
> diff --git a/drivers/tee/optee/optee_ffa.h b/drivers/tee/optee/optee_ffa.h
> index ee3a03fc392c..97266243deaa 100644
> --- a/drivers/tee/optee/optee_ffa.h
> +++ b/drivers/tee/optee/optee_ffa.h
> @@ -81,8 +81,16 @@
>   *                   as the second MSG arg struct for
>   *                   OPTEE_FFA_YIELDING_CALL_WITH_ARG.
>   *        Bit[31:8]: Reserved (MBZ)
> - * w5-w7: Note used (MBZ)
> + * w5:   Bitfield of secure world capabilities OPTEE_FFA_SEC_CAP_* below,
> + *       unused bits MBZ.
> + * w6-w7: Not used (MBZ)
> + */
> +/*
> + * Secure world supports giving an offset into the argument shared memory
> + * object, see also OPTEE_FFA_YIELDING_CALL_WITH_ARG
>   */
> +#define OPTEE_FFA_SEC_CAP_ARG_OFFSET   BIT(0)
> +
>  #define OPTEE_FFA_EXCHANGE_CAPABILITIES OPTEE_FFA_BLOCKING_CALL(2)
>
>  /*
> @@ -112,6 +120,8 @@
>   *       OPTEE_MSG_GET_ARG_SIZE(num_params) follows a struct optee_msg_arg
>   *       for RPC, this struct has reserved space for the number of RPC
>   *       parameters as returned by OPTEE_FFA_EXCHANGE_CAPABILITIES.
> + *       MBZ unless the bit OPTEE_FFA_SEC_CAP_ARG_OFFSET is received with
> + *       OPTEE_FFA_EXCHANGE_CAPABILITIES.
>   * w7:    Not used (MBZ)
>   * Resume from RPC. Register usage:
>   * w3:    Service ID, OPTEE_FFA_YIELDING_CALL_RESUME
> --
> 2.31.1
>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ