| 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
| ||
|
Message-ID: <CADQ0-X-k0w67sKw3XMQsgtc96KcZTbaCX-pEuZ_YCmp9Mf9zxQ@mail.gmail.com>
Date: Mon, 31 Jul 2023 15:21:22 +0900
From: Masahisa Kojima <masahisa.kojima@...aro.org>
To: Sumit Garg <sumit.garg@...aro.org>
Cc: op-tee@...ts.trustedfirmware.org, jens.wiklander@...aro.org,
jan.kiszka@...mens.com, arnd@...aro.org, ardb@...nel.org,
jerome.forissier@...aro.org, ilias.apalodimas@...aro.org,
maxim.uvarov@...aro.org, jarkko.sakkinen@...ux.intel.com,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] tee: optee: Fix supplicant based device enumeration
On Fri, 28 Jul 2023 at 22:48, Sumit Garg <sumit.garg@...aro.org> wrote:
>
> Currently supplicant dependent optee device enumeration only registers
> devices whenever tee-supplicant is invoked for the first time. But it
> forgets to remove devices when tee-supplicant daemon stops running and
> closes its context gracefully. This leads to following error for fTPM
> driver during reboot/shutdown:
>
> [ 73.466791] tpm tpm0: ftpm_tee_tpm_op_send: SUBMIT_COMMAND invoke error: 0xffff3024
>
> Fix this by separating supplicant dependent devices so that the
> user-space service can detach supplicant devices before closing the
> supplicant. While at it use the global system workqueue for OP-TEE bus
> scanning work rather than our own custom one.
>
> Reported-by: Jan Kiszka <jan.kiszka@...mens.com>
> Link: https://github.com/OP-TEE/optee_os/issues/6094
> Fixes: 5f178bb71e3a ("optee: enable support for multi-stage bus enumeration")
> Signed-off-by: Sumit Garg <sumit.garg@...aro.org>
> ---
>
> Changes in v2:
>
> Apologies for taking it too long push this v2. Actually I did brainstorm
> how to best fix this tee-supplicant dependent device probing. Its hard
> to predict the lifetime of user-space daemon from kernel space. So
> following is the least intrusive approach:
>
> - Use device names to seperate out tee-supplicant dependent devices via
> this patch.
> - Since user-space service is aware about tee-supplicant lifespan, so
> allow the user-space service to unbind tee-supplicant dependent
> devices before killing the supplicant. Following command has to be
> added to the tee-supplicant service file.
>
> $ for dev in /sys/bus/tee/devices/*; do if [[ "$dev" == *"optee-ta-supp-"* ]]; \
> then echo $(basename "$dev") > $dev/driver/unbind; fi done
>
> drivers/tee/optee/core.c | 13 ++-----------
> drivers/tee/optee/device.c | 13 ++++++++++---
> drivers/tee/optee/optee_private.h | 2 --
> 3 files changed, 12 insertions(+), 16 deletions(-)
>
> diff --git a/drivers/tee/optee/core.c b/drivers/tee/optee/core.c
> index d01ca47f7bde..8ee3c71bd989 100644
> --- a/drivers/tee/optee/core.c
> +++ b/drivers/tee/optee/core.c
> @@ -15,7 +15,6 @@
> #include <linux/string.h>
> #include <linux/tee_drv.h>
> #include <linux/types.h>
> -#include <linux/workqueue.h>
> #include "optee_private.h"
>
> int optee_pool_op_alloc_helper(struct tee_shm_pool *pool, struct tee_shm *shm,
> @@ -110,12 +109,7 @@ int optee_open(struct tee_context *ctx, bool cap_memref_null)
>
> 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) {
> - kfree(ctxdata);
> - return -ECHILD;
> - }
> - queue_work(optee->scan_bus_wq, &optee->scan_bus_work);
> + schedule_work(&optee->scan_bus_work);
> optee->scan_bus_done = true;
> }
> }
> @@ -159,10 +153,7 @@ void optee_release_supp(struct tee_context *ctx)
> struct optee *optee = tee_get_drvdata(ctx->teedev);
>
> optee_release_helper(ctx, optee_close_session_helper);
> - if (optee->scan_bus_wq) {
> - destroy_workqueue(optee->scan_bus_wq);
> - optee->scan_bus_wq = NULL;
> - }
> +
> optee_supp_release(&optee->supp);
> }
>
> diff --git a/drivers/tee/optee/device.c b/drivers/tee/optee/device.c
> index 64f0e047c23d..78fc0a15c463 100644
> --- a/drivers/tee/optee/device.c
> +++ b/drivers/tee/optee/device.c
> @@ -60,9 +60,10 @@ static void optee_release_device(struct device *dev)
> kfree(optee_device);
> }
>
> -static int optee_register_device(const uuid_t *device_uuid)
> +static int optee_register_device(const uuid_t *device_uuid, u32 func)
> {
> struct tee_client_device *optee_device = NULL;
> + const char *dev_name_fmt = NULL;
> int rc;
>
> optee_device = kzalloc(sizeof(*optee_device), GFP_KERNEL);
> @@ -71,7 +72,13 @@ static int optee_register_device(const uuid_t *device_uuid)
>
> optee_device->dev.bus = &tee_bus_type;
> optee_device->dev.release = optee_release_device;
> - if (dev_set_name(&optee_device->dev, "optee-ta-%pUb", device_uuid)) {
> +
> + if (func == PTA_CMD_GET_DEVICES_SUPP)
> + dev_name_fmt = "optee-ta-supp-%pUb";
> + else
> + dev_name_fmt = "optee-ta-%pUb";
> +
> + if (dev_set_name(&optee_device->dev, dev_name_fmt, device_uuid)) {
> kfree(optee_device);
> return -ENOMEM;
> }
> @@ -142,7 +149,7 @@ static int __optee_enumerate_devices(u32 func)
> num_devices = shm_size / sizeof(uuid_t);
>
> for (idx = 0; idx < num_devices; idx++) {
> - rc = optee_register_device(&device_uuid[idx]);
> + rc = optee_register_device(&device_uuid[idx], func);
> if (rc)
> goto out_shm;
> }
> diff --git a/drivers/tee/optee/optee_private.h b/drivers/tee/optee/optee_private.h
> index 6dcecb83c893..af4aa266c3fb 100644
> --- a/drivers/tee/optee/optee_private.h
> +++ b/drivers/tee/optee/optee_private.h
> @@ -193,7 +193,6 @@ struct optee_ops {
> * @pool: shared memory pool
> * @rpc_param_count: If > 0 number of RPC parameters to make room for
> * @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 {
> @@ -212,7 +211,6 @@ struct optee {
> struct tee_shm_pool *pool;
> unsigned int rpc_param_count;
> bool scan_bus_done;
> - struct workqueue_struct *scan_bus_wq;
> struct work_struct scan_bus_work;
> };
>
It works for me, I tested with a 6.4.0-based kernel on the Developerbox.
Both fTPM and tee-based EFI variable service are shutdown successfully.
Tested-by: Masahisa Kojima <masahisa.kojima@...aro.org>
Thanks,
Masahisa Kojima
> --
> 2.34.1
>
Powered by blists - more mailing lists