Tee-dev September 2019

tee-dev@lists.linaro.org

9 participants
7 discussions

[PATCH] optee: allow to work without static shared memory

by Volodymyr Babchuk

From: Volodymyr Babchuk <vlad.babchuk(a)gmail.com> On virtualized systems it is possible that OP-TEE will provide only dynamic shared memory support. So it is fine to boot without static SHM enabled if dymanic one is supported. Signed-off-by: Volodymyr Babchuk <vlad.babchuk(a)gmail.com> --- drivers/tee/optee/core.c | 83 ++++++++++++++++++++++++++++-------------------- 1 file changed, 49 insertions(+), 34 deletions(-) diff --git a/drivers/tee/optee/core.c b/drivers/tee/optee/core.c index 4a2c420..d80da29 100644 --- a/drivers/tee/optee/core.c +++ b/drivers/tee/optee/core.c @@ -420,9 +420,35 @@ static bool optee_msg_exchange_capabilities(optee_invoke_fn *invoke_fn, return true; } +static struct tee_shm_pool *optee_config_dyn_shm(void) +{ + struct tee_shm_pool_mgr *priv_mgr; + struct tee_shm_pool_mgr *dmabuf_mgr; + void *rc; + + rc = optee_shm_pool_alloc_pages(); + if (IS_ERR(rc)) + return rc; + priv_mgr = rc; + + rc = optee_shm_pool_alloc_pages(); + if (IS_ERR(rc)) { + tee_shm_pool_mgr_destroy(priv_mgr); + return rc; + } + dmabuf_mgr = rc; + + rc = tee_shm_pool_alloc(priv_mgr, dmabuf_mgr); + if (IS_ERR(rc)) { + tee_shm_pool_mgr_destroy(priv_mgr); + tee_shm_pool_mgr_destroy(dmabuf_mgr); + } + + return rc; +} + static struct tee_shm_pool * -optee_config_shm_memremap(optee_invoke_fn *invoke_fn, void **memremaped_shm, - u32 sec_caps) +optee_config_shm_memremap(optee_invoke_fn *invoke_fn, void **memremaped_shm) { union { struct arm_smccc_res smccc; @@ -437,10 +463,11 @@ optee_config_shm_memremap(optee_invoke_fn *invoke_fn, void **memremaped_shm, struct tee_shm_pool_mgr *priv_mgr; struct tee_shm_pool_mgr *dmabuf_mgr; void *rc; + const int sz = OPTEE_SHM_NUM_PRIV_PAGES * PAGE_SIZE; invoke_fn(OPTEE_SMC_GET_SHM_CONFIG, 0, 0, 0, 0, 0, 0, 0, &res.smccc); if (res.result.status != OPTEE_SMC_RETURN_OK) { - pr_info("shm service not available\n"); + pr_err("static shm service not available\n"); return ERR_PTR(-ENOENT); } @@ -466,28 +493,15 @@ optee_config_shm_memremap(optee_invoke_fn *invoke_fn, void **memremaped_shm, } vaddr = (unsigned long)va; - /* - * If OP-TEE can work with unregistered SHM, we will use own pool - * for private shm - */ - if (sec_caps & OPTEE_SMC_SEC_CAP_DYNAMIC_SHM) { - rc = optee_shm_pool_alloc_pages(); - if (IS_ERR(rc)) - goto err_memunmap; - priv_mgr = rc; - } else { - const size_t sz = OPTEE_SHM_NUM_PRIV_PAGES * PAGE_SIZE; - - rc = tee_shm_pool_mgr_alloc_res_mem(vaddr, paddr, sz, - 3 /* 8 bytes aligned */); - if (IS_ERR(rc)) - goto err_memunmap; - priv_mgr = rc; - - vaddr += sz; - paddr += sz; - size -= sz; - } + rc = tee_shm_pool_mgr_alloc_res_mem(vaddr, paddr, sz, + 3 /* 8 bytes aligned */); + if (IS_ERR(rc)) + goto err_memunmap; + priv_mgr = rc; + + vaddr += sz; + paddr += sz; + size -= sz; rc = tee_shm_pool_mgr_alloc_res_mem(vaddr, paddr, size, PAGE_SHIFT); if (IS_ERR(rc)) @@ -553,7 +567,7 @@ static optee_invoke_fn *get_invoke_func(struct device_node *np) static struct optee *optee_probe(struct device_node *np) { optee_invoke_fn *invoke_fn; - struct tee_shm_pool *pool; + struct tee_shm_pool *pool = ERR_PTR(-EINVAL); struct optee *optee = NULL; void *memremaped_shm = NULL; struct tee_device *teedev; @@ -582,13 +596,17 @@ static struct optee *optee_probe(struct device_node *np) } /* - * We have no other option for shared memory, if secure world - * doesn't have any reserved memory we can use we can't continue. + * Try to use dynamic shared memory if possible */ - if (!(sec_caps & OPTEE_SMC_SEC_CAP_HAVE_RESERVED_SHM)) - return ERR_PTR(-EINVAL); + if (sec_caps & OPTEE_SMC_SEC_CAP_DYNAMIC_SHM) + pool = optee_config_dyn_shm(); + + /* + * If dynamic shared memory is not available or failed - try static one + */ + if (IS_ERR(pool) && (sec_caps & OPTEE_SMC_SEC_CAP_HAVE_RESERVED_SHM)) + pool = optee_config_shm_memremap(invoke_fn, &memremaped_shm); - pool = optee_config_shm_memremap(invoke_fn, &memremaped_shm, sec_caps); if (IS_ERR(pool)) return (void *)pool; @@ -632,9 +650,6 @@ static struct optee *optee_probe(struct device_node *np) optee_enable_shm_cache(optee); - if (optee->sec_caps & OPTEE_SMC_SEC_CAP_DYNAMIC_SHM) - pr_info("dynamic shared memory is enabled\n"); - pr_info("initialized driver\n"); return optee; err: -- 2.7.4

5 years

[PATCH v3 0/3] arch/arm: optee: fix TODOs and change status to "Tech Preview"

by Volodymyr Babchuk

This is the third version of maturing the OP-TEE mediator patches. Changes also can be pulled from [3]. Changes from v2: - The following 3 patches were commited: xen/arm: optee: impose limit on shared buffer size xen/arm: optee: check for preemption while freeing shared buffers xen/arm: optee: limit number of shared buffers - Other changes are described in the corresponding patches Changes from v1: - Added patch that updates SUPPORT.md - Instead of removing "experimental" status I changed it to "Tech Preview" - Other changes are described in the corresponding patches Cover letter for v1: This patch series fixes various unfinished items in the OP-TEE mediator. Mostly this is about limiting resources that guest can consume. This includes both memory and time - how many buffers guest can share with OP-TEE (this uses Xen memory) and when mediator should preempt itself, to make sure that guest does not stress scheduling. Apart from this, there were one case, when mediator's actions might lead to memory leak in a good-behaving guest. To fix this issue I had to extend mediator logic, so now it can issue RPC requests to guest in the same way, as OP-TEE does this. This is useful feature, because it allows to preempt mediator during long operations. So, in the future it will be possible to remove shared buffer size limitation, because mediator can preempt self during buffer translation. This patch series can be pulled from [1]. [1] https://github.com/lorc/xen/tree/optee3_v1 [2] https://github.com/lorc/xen/tree/optee3_v2 [3] https://github.com/lorc/xen/tree/optee3_v3 Volodymyr Babchuk (3): xen/arm: optee: handle shared buffer translation error SUPPORT.md: Describe OP-TEE mediator xen/arm: optee: update description in Kconfig SUPPORT.md | 4 + xen/arch/arm/tee/Kconfig | 9 +- xen/arch/arm/tee/optee.c | 173 ++++++++++++++++++++++++++++++++------- 3 files changed, 151 insertions(+), 35 deletions(-) -- 2.23.0

5 years, 1 month

[PATCH v2 0/6] arch/arm: optee: fix TODOs and change status to "Tech Preview"

by Volodymyr Babchuk

Hello, This is the second version for maturing the OP-TEE mediator. Changes also can be pulled from [2]. Changes from v1: - Added patch that updates SUPPORT.md - Instead of removing "experimental" status I changed it to "Tech Preview" - Other changes are described in the corresponding patches Cover letter for v1: This patch series fixes various unfinished items in the OP-TEE mediator. Mostly this is about limiting resources that guest can consume. This includes both memory and time - how many buffers guest can share with OP-TEE (this uses Xen memory) and when mediator should preempt itself, to make sure that guest does not stress scheduling. Apart from this, there were one case, when mediator's actions might lead to memory leak in a good-behaving guest. To fix this issue I had to extend mediator logic, so now it can issue RPC requests to guest in the same way, as OP-TEE does this. This is useful feature, because it allows to preempt mediator during long operations. So, in the future it will be possible to remove shared buffer size limitation, because mediator can preempt self during buffer translation. This patch series can be pulled from [1]. [1] https://github.com/lorc/xen/tree/optee3_v1 [2] https://github.com/lorc/xen/tree/optee3_v2 Volodymyr Babchuk (6): xen/arm: optee: impose limit on shared buffer size xen/arm: optee: check for preemption while freeing shared buffers xen/arm: optee: limit number of shared buffers xen/arm: optee: handle shared buffer translation error SUPPORT.md: Describe OP-TEE mediator xen/arm: optee: update description in Kconfig SUPPORT.md | 4 + xen/arch/arm/tee/Kconfig | 12 +- xen/arch/arm/tee/optee.c | 259 ++++++++++++++++++++++++++++++--------- 3 files changed, 213 insertions(+), 62 deletions(-) -- 2.22.0

5 years, 2 months

[PATCH v7 0/5] TEE mediator (and OP-TEE) support in XEN

by Volodymyr Babchuk

Hello community, Please find new version of OP-TEE patch series. This is the kind of follow-up for the previous version, as most of the patches of the previous version were commited. This series includes leftovers of the prev. version and three new patches. One of the new patches adds a way to detect if OP-TEE were build with virtualization support and two others bear cosmetic changes to Kconfig files. This patches also can be pulled from [4] == Changes in v7: - 8 of 10 patches were commited - New patch "xen/arm: tee: place OP-TEE Kconfig option right after TEE" places options in menuconfig in more natural way - New patch "xen/arm: optee: check if OP-TEE is virtualization-aware" ensues that OP-TEE is virtualization-aware - New patch "xen/arm: optee: document OPTEE option in tee/Kconfig" add short description of OP-TEE mediator - Documentation in "tools/arm: tee: add "tee" option for xl.cfg" was updated === Changes in v6: - Series rebased to staging branch instead of master one. - OP-TEE protocol headers was taken from OP-TEE tree instead of Linux one - Added acked-by tags - Fixed (and tested) issue when XEN would not boot if it is build with CONFIG_TEE=n ==== Changes in v5: - Substantial rework of OP-TEE mediator. Now it tries to return meaningful error codes back to the guest. - OP-TEE mediator does not use struct cpu_user_regs as a storage for parameters and return values when calling OP-TEE. This makes it compatbile with requirement from SMCCC. - tee=native option replaced with tee=optee - Authorship and s-o-b tag reset to my EPAM mail address ==== Changes in v4: - Patch "arm: add tee_enabled flag to xen_arch_domainconfig" was squashed into "xen/arm: add generic TEE mediator framework" - I implemented more elaborate error repoting to a guest. Now guest will get meaningful error codes instead of generic ARM_SMCCC_ERR_UNKNOWN_FUNCTION. ==== Changes in v3: - Use domain flags insted of domctl interface to enable optee for guests - Remove patch "libxc: add xc_dom_tee_enable(...) function" because of previous change - Mediator now stores own context in arch part of struct domain, so I removed patch "optee: add domain contexts" ==== Changes in v2: This is v2 of patch series for OP-TEE mediator support in XEN. Changes from v1: - Added domctl interface, so now xl decides what domain should work with TEE - Removed XSM support due to change described above - Patch with OP-TEE mediator was splited to 7 separate patches - Removed patch with call_smccc() function. Now this series depend on Julien Grall's series "xen/arm: SMCCC fixup and improvement" [3] ===== v1: This is follow for patch series [1]. There was lots of discussions for that series and I tried to address all of them in this new patchset. Currently, I had a working solution for OP-TEE virtualization and it is being upstreamed right now ([2]). So, I think it is a good time to introduce support in XEN as well. This series include generic TEE mediator framework and full-scale OP-TEE mediator which is working with mentioned chages in OP-TEE. So, multiple domains can work simultaneously with OP-TEE. I added XSM support, so now it is possible to control which domains can work with TEEs. Also I changed way how TEE discovery is done. Now it is very generic and should support any platform. [1] https://lists.xenproject.org/archives/html/xen-devel/2017-10/msg01451.html [2] https://github.com/OP-TEE/optee_os/pull/2370 [3] https://lists.xenproject.org/archives/html/xen-devel/2018-08/msg02138.html [4] https://github.com/lorc/xen/tree/optee_v7 Volodymyr Babchuk (5): tools/arm: tee: add "tee" option for xl.cfg tools/arm: optee: create optee firmware node in DT if tee=optee xen/arm: tee: place OP-TEE Kconfig option right after TEE xen/arm: optee: check if OP-TEE is virtualization-aware xen/arm: optee: document OPTEE option in tee/Kconfig docs/man/xl.cfg.5.pod.in | 29 +++++++++++++++++++++++++ tools/libxl/libxl.h | 5 +++++ tools/libxl/libxl_arm.c | 42 +++++++++++++++++++++++++++++++++++++ tools/libxl/libxl_types.idl | 6 ++++++ tools/xl/xl_parse.c | 9 ++++++++ xen/arch/arm/Kconfig | 4 ++-- xen/arch/arm/tee/Kconfig | 5 +++++ xen/arch/arm/tee/optee.c | 10 +++++++++ 8 files changed, 108 insertions(+), 2 deletions(-) -- 2.21.0

5 years, 2 months

[PATCH 0/5] arch/arm: optee: fix TODOs and remove "experimental" status

by Volodymyr Babchuk

Hello, This patch series fixes various unfinished items in the OP-TEE mediator. Mostly this is about limiting resources that guest can consume. This includes both memory and time - how many buffers guest can share with OP-TEE (this uses Xen memory) and when mediator should preempt itself, to make sure that guest does not stress scheduling. Apart from this, there were one case, when mediator's actions might lead to memory leak in a good-behaving guest. To fix this issue I had to extend mediator logic, so now it can issue RPC requests to guest in the same way, as OP-TEE does this. This is useful feature, because it allows to preempt mediator during long operations. So, in the future it will be possible to remove shared buffer size limitation, because mediator can preempt self during buffer translation. This patch series can be pulled from [1]. [1] https://github.com/lorc/xen/tree/optee3_v1 Volodymyr Babchuk (5): xen/arm: optee: impose limit on shared buffer size xen/arm: optee: check for preemption while freeing shared buffers xen/arm: optee: limit number of shared buffers xen/arm: optee: handle share buffer translation error xen/arm: optee: remove experimental status xen/arch/arm/Kconfig | 2 +- xen/arch/arm/tee/Kconfig | 2 +- xen/arch/arm/tee/optee.c | 237 ++++++++++++++++++++++++++++++--------- 3 files changed, 184 insertions(+), 57 deletions(-) -- 2.22.0

5 years, 2 months

[PATCH] firmware: broadcom: add OP-TEE based BNXT f/w manager

by Sheetal Tigadoli

From: Vikas Gupta <vikas.gupta(a)broadcom.com> This driver registers on TEE bus to interact with OP-TEE based BNXT firmware management modules Signed-off-by: Vikas Gupta <vikas.gupta(a)broadcom.com> Signed-off-by: Sheetal Tigadoli <sheetal.tigadoli(a)broadcom.com> --- drivers/firmware/broadcom/Kconfig | 8 + drivers/firmware/broadcom/Makefile | 1 + drivers/firmware/broadcom/tee_bnxt_fw.c | 447 ++++++++++++++++++++++++++ include/linux/firmware/broadcom/tee_bnxt_fw.h | 17 + 4 files changed, 473 insertions(+) create mode 100644 drivers/firmware/broadcom/tee_bnxt_fw.c create mode 100644 include/linux/firmware/broadcom/tee_bnxt_fw.h diff --git a/drivers/firmware/broadcom/Kconfig b/drivers/firmware/broadcom/Kconfig index 6468082..a846a21 100644 --- a/drivers/firmware/broadcom/Kconfig +++ b/drivers/firmware/broadcom/Kconfig @@ -22,3 +22,11 @@ config BCM47XX_SPROM In case of SoC devices SPROM content is stored on a flash used by bootloader firmware CFE. This driver provides method to ssb and bcma drivers to read SPROM on SoC. + +config TEE_BNXT_FW + bool "Broadcom BNXT firmware manager" + depends on ARCH_BCM_IPROC && OPTEE + default ARCH_BCM_IPROC + help + This module help to manage firmware on Broadcom BNXT device. The module + registers on tee bus and invoke calls to manage firmware on BNXT device. diff --git a/drivers/firmware/broadcom/Makefile b/drivers/firmware/broadcom/Makefile index 72c7fdc..17c5061 100644 --- a/drivers/firmware/broadcom/Makefile +++ b/drivers/firmware/broadcom/Makefile @@ -1,3 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only obj-$(CONFIG_BCM47XX_NVRAM) += bcm47xx_nvram.o obj-$(CONFIG_BCM47XX_SPROM) += bcm47xx_sprom.o +obj-$(CONFIG_TEE_BNXT_FW) += tee_bnxt_fw.o diff --git a/drivers/firmware/broadcom/tee_bnxt_fw.c b/drivers/firmware/broadcom/tee_bnxt_fw.c new file mode 100644 index 00000000..89a48fd --- /dev/null +++ b/drivers/firmware/broadcom/tee_bnxt_fw.c @@ -0,0 +1,447 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright 2019 Broadcom. + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/tee_drv.h> +#include <linux/uuid.h> + +#include <linux/firmware/broadcom/tee_bnxt_fw.h> + +#define DRIVER_NAME "tee-bnxt-fw" + +#define MAX_SHM_MEM_SZ SZ_4M + +#define MAX_TEE_PARAM_ARRY_MEMB 4 + +enum ta_cmd { +/* + * TA_CMD_BNXT_FASTBOOT - boot bnxt device by copying f/w into sram + * + * param[0] unused + * param[1] unused + * param[2] unused + * param[3] unused + * + * Result: + * TEE_SUCCESS - Invoke command success + * TEE_ERROR_ITEM_NOT_FOUND - Corrupt f/w image found on memory + */ + TA_CMD_BNXT_FASTBOOT = 0, + +/* + * TA_CMD_BNXT_HEALTH_STATUS - to check health of bnxt device + * + * param[0] (out value) - value.a: health status + * param[1] unused + * param[2] unused + * param[3] unused + * + * Result: + * TEE_SUCCESS - Invoke command success + * TEE_ERROR_BAD_PARAMETERS - Incorrect input param + */ + TA_CMD_BNXT_HEALTH_STATUS, + +/* + * TA_CMD_BNXT_HANDSHAKE - to check bnxt device is booted + * + * param[0] (in value) - value.a: max timeout value + * param[0] (out value) - value.a: boot status + * param[1] unused + * param[2] unused + * param[3] unused + * + * Result: + * TEE_SUCCESS - Invoke command success + * TEE_ERROR_BAD_PARAMETERS - Incorrect input param + */ + TA_CMD_BNXT_HANDSHAKE, + +/* + * TA_CMD_BNXT_COPY_COREDUMP - copy the core dump into shm + * + * param[0] (in value) - value.a: offset at which data to be copied from + * value.b: size of the data + * param[1] unused + * param[2] unused + * param[3] unused + * + * Result: + * TEE_SUCCESS - Invoke command success + * TEE_ERROR_BAD_PARAMETERS - Incorrect input param + * TEE_ERROR_ITEM_NOT_FOUND - Corrupt core dump + */ + TA_CMD_BNXT_COPY_COREDUMP, + +/* + * TA_CMD_BNXT_FW_UPGRADE - upgrade the bnxt firmware + * + * param[0] (in value) - value.a: size of the f/w image + * param[1] unused + * param[2] unused + * param[3] unused + * + * Result: + * TEE_SUCCESS - Invoke command success + * TEE_ERROR_BAD_PARAMETERS - Incorrect input param + */ + TA_CMD_BNXT_FW_UPGRADE, +}; + +/** + * struct tee_bnxt_fw_private - OP-TEE bnxt private data + * @dev: OP-TEE based bnxt device. + * @ctx: OP-TEE context handler. + * @session_id: TA session identifier. + */ +struct tee_bnxt_fw_private { + struct device *dev; + struct tee_context *ctx; + u32 session_id; + struct tee_shm *fw_shm_pool; +}; + +static struct tee_bnxt_fw_private pvt_data; + +static inline void prepare_args(int cmd, + struct tee_ioctl_invoke_arg *inv_arg, + struct tee_param *param) +{ + memset(inv_arg, 0, sizeof(*inv_arg)); + memset(param, 0, (MAX_TEE_PARAM_ARRY_MEMB * sizeof(*param))); + + inv_arg->func = cmd; + inv_arg->session = pvt_data.session_id; + inv_arg->num_params = MAX_TEE_PARAM_ARRY_MEMB; + + /* Fill invoke cmd params */ + switch (cmd) { + case TA_CMD_BNXT_HEALTH_STATUS: + param[0].attr = TEE_IOCTL_PARAM_ATTR_TYPE_VALUE_OUTPUT; + break; + case TA_CMD_BNXT_HANDSHAKE: + param[0].attr = TEE_IOCTL_PARAM_ATTR_TYPE_VALUE_INOUT; + break; + case TA_CMD_BNXT_COPY_COREDUMP: + case TA_CMD_BNXT_FW_UPGRADE: + param[0].attr = TEE_IOCTL_PARAM_ATTR_TYPE_MEMREF_INOUT; + param[0].u.memref.shm = pvt_data.fw_shm_pool; + param[0].u.memref.size = MAX_SHM_MEM_SZ; + param[0].u.memref.shm_offs = 0; + param[1].attr = TEE_IOCTL_PARAM_ATTR_TYPE_VALUE_INPUT; + break; + case TA_CMD_BNXT_FASTBOOT: + default: + /* Nothing to do */ + break; + } +} + +/** + * tee_bnxt_fw_load() - Load the bnxt firmware + * Uses an OP-TEE call to start a secure + * boot process. + * Returns 0 on success, negative errno otherwise. + */ +int tee_bnxt_fw_load(void) +{ + int ret = 0; + struct tee_ioctl_invoke_arg inv_arg; + struct tee_param param[MAX_TEE_PARAM_ARRY_MEMB]; + + if (!pvt_data.ctx) + return -ENODEV; + + prepare_args(TA_CMD_BNXT_FASTBOOT, &inv_arg, param); + + ret = tee_client_invoke_func(pvt_data.ctx, &inv_arg, param); + if ((ret < 0) || (inv_arg.ret != 0)) { + dev_err(pvt_data.dev, "TA_CMD_BNXT_LOAD invoke err: %x\n", + (ret < 0) ? ret : inv_arg.ret); + return -EINVAL; + } + + return 0; +} +EXPORT_SYMBOL(tee_bnxt_fw_load); + +/** + * tee_bnxt_health_status() - Get the health status + * Uses an OP-TEE call to get health + * status of bnxt device. + * @status: status is returned on this pointer + * Returns 0 on success, negative errno otherwise. + */ +int tee_bnxt_health_status(u32 *status) +{ + int ret = 0; + struct tee_ioctl_invoke_arg inv_arg; + struct tee_param param[MAX_TEE_PARAM_ARRY_MEMB]; + + if (!pvt_data.ctx) + return -ENODEV; + + prepare_args(TA_CMD_BNXT_HEALTH_STATUS, &inv_arg, param); + + ret = tee_client_invoke_func(pvt_data.ctx, &inv_arg, param); + if ((ret < 0) || (inv_arg.ret != 0)) { + dev_err(pvt_data.dev, "TA_CMD_BNXT_HEALTH_STATUS invoke err: %x\n", + (ret < 0) ? ret : inv_arg.ret); + return -EINVAL; + } + + *status = param[0].u.value.a; + + return 0; +} +EXPORT_SYMBOL(tee_bnxt_health_status); + +/** + * tee_bnxt_handshake() - Get the handshake status + * Uses an OP-TEE call to get handshake + * status after bnxt device`s boot process. + * @timeout: max timeout to wait for handshake + * @status: status is populated + * Returns 0 on success, negative errno otherwise. + */ +int tee_bnxt_handshake(u32 timeout, u32 *status) +{ + int ret = 0; + struct tee_ioctl_invoke_arg inv_arg; + struct tee_param param[MAX_TEE_PARAM_ARRY_MEMB]; + + if (!pvt_data.ctx) + return -ENODEV; + + prepare_args(TA_CMD_BNXT_HANDSHAKE, &inv_arg, param); + + /* Fill additional invoke cmd params */ + param[0].u.value.a = timeout; + + ret = tee_client_invoke_func(pvt_data.ctx, &inv_arg, param); + if ((ret < 0) || (inv_arg.ret != 0)) { + dev_err(pvt_data.dev, "TA_CMD_BNXT_HANDSHAKE invoke err: %x\n", + (ret < 0) ? ret : inv_arg.ret); + return -EINVAL; + } + + *status = param[0].u.value.a; + + return 0; +} +EXPORT_SYMBOL(tee_bnxt_handshake); + +/** + * tee_bnxt_copy_coredump() - Copy coredump from the allocated memory + * Uses an OP-TEE call to copy coredump + * @buf: desintation buffer where core dump is copied into + * @offset: offset from the base address of core dump area + * @size: size of the dump + * + * Returns 0 on success, negative errno otherwise. + */ +int tee_bnxt_copy_coredump(void *buf, u32 offset, u32 size) +{ + struct tee_ioctl_invoke_arg inv_arg; + struct tee_param param[MAX_TEE_PARAM_ARRY_MEMB]; + void *core_data; + u32 rbytes = size; + u32 nbytes = 0; + int ret = 0; + + if (!pvt_data.ctx) + return -ENODEV; + + if (!buf) + return -EINVAL; + + prepare_args(TA_CMD_BNXT_COPY_COREDUMP, &inv_arg, param); + + while (rbytes) { + nbytes = rbytes; + + nbytes = min_t(u32, rbytes, param[0].u.memref.size); + + /* Fill additional invoke cmd params */ + param[1].u.value.a = offset; + param[1].u.value.b = nbytes; + + ret = tee_client_invoke_func(pvt_data.ctx, &inv_arg, param); + if ((ret < 0) || (inv_arg.ret != 0)) { + dev_err(pvt_data.dev, + "TA_CMD_BNXT_COPY_COREDUMP invoke err: %x\n", + (ret < 0) ? ret : inv_arg.ret); + return -EINVAL; + } + + core_data = tee_shm_get_va(pvt_data.fw_shm_pool, 0); + if (IS_ERR(core_data)) { + dev_err(pvt_data.dev, "tee_shm_get_va failed\n"); + return PTR_ERR(core_data); + } + + memcpy(buf, core_data, nbytes); + + rbytes -= nbytes; + buf += nbytes; + offset += nbytes; + } + + return 0; +} +EXPORT_SYMBOL(tee_bnxt_copy_coredump); + +/** + * bnxt_fw_upgrade() - Upgrade the bnxt firmware and configuration of + * bnxt device into the flash device. + * Uses an OP-TEE call to upgrade firmware. + * @buf: source buffer of firmware image + * @size: size of the image + * + * Returns 0 on success, negative errno otherwise. + */ +int bnxt_fw_upgrade(void *buf, u32 size) +{ + struct tee_ioctl_invoke_arg inv_arg; + struct tee_param param[MAX_TEE_PARAM_ARRY_MEMB]; + void *fw_image_dst; + int ret = 0; + + if (!pvt_data.ctx) + return -ENODEV; + + /* we do not expect firmware size more than allocated tee shm size */ + if (!buf || size > MAX_SHM_MEM_SZ) + return -EINVAL; + + prepare_args(TA_CMD_BNXT_FW_UPGRADE, &inv_arg, param); + + /* Fill additional invoke cmd params */ + param[1].u.value.a = size; + + fw_image_dst = tee_shm_get_va(pvt_data.fw_shm_pool, 0); + if (IS_ERR(fw_image_dst)) { + dev_err(pvt_data.dev, "tee_shm_get_va failed\n"); + return PTR_ERR(fw_image_dst); + } + + memcpy(fw_image_dst, buf, size); + + ret = tee_client_invoke_func(pvt_data.ctx, &inv_arg, param); + if ((ret < 0) || (inv_arg.ret != 0)) { + dev_err(pvt_data.dev, "TA_CMD_BNXT_FW_UPGRADE invoke err: %x\n", + (ret < 0) ? ret : inv_arg.ret); + return -EINVAL; + } + + return 0; +} +EXPORT_SYMBOL(bnxt_fw_upgrade); + +static int optee_ctx_match(struct tee_ioctl_version_data *ver, const void *data) +{ + if (ver->impl_id == TEE_IMPL_ID_OPTEE) + return 1; + else + return 0; +} + +static int tee_bnxt_fw_probe(struct device *dev) +{ + struct tee_client_device *bnxt_device = to_tee_client_device(dev); + int ret, err = -ENODEV; + struct tee_ioctl_open_session_arg sess_arg; + struct tee_shm *fw_shm_pool; + + memset(&sess_arg, 0, sizeof(sess_arg)); + + /* Open context with TEE driver */ + pvt_data.ctx = tee_client_open_context(NULL, optee_ctx_match, NULL, + NULL); + if (IS_ERR(pvt_data.ctx)) + return -ENODEV; + + /* Open session with Bnxt load Trusted App */ + memcpy(sess_arg.uuid, bnxt_device->id.uuid.b, TEE_IOCTL_UUID_LEN); + sess_arg.clnt_login = TEE_IOCTL_LOGIN_PUBLIC; + sess_arg.num_params = 0; + + ret = tee_client_open_session(pvt_data.ctx, &sess_arg, NULL); + if ((ret < 0) || (sess_arg.ret != 0)) { + dev_err(dev, "tee_client_open_session failed, err: %x\n", + sess_arg.ret); + err = -EINVAL; + goto out_ctx; + } + pvt_data.session_id = sess_arg.session; + + pvt_data.dev = dev; + + fw_shm_pool = tee_shm_alloc(pvt_data.ctx, MAX_SHM_MEM_SZ, + TEE_SHM_MAPPED | TEE_SHM_DMA_BUF); + if (IS_ERR(fw_shm_pool)) { + tee_client_close_context(pvt_data.ctx); + dev_err(pvt_data.dev, "tee_shm_alloc failed\n"); + err = PTR_ERR(fw_shm_pool); + goto out_sess; + } + + pvt_data.fw_shm_pool = fw_shm_pool; + + return 0; + +out_sess: + tee_client_close_session(pvt_data.ctx, pvt_data.session_id); +out_ctx: + tee_client_close_context(pvt_data.ctx); + + return err; +} + +static int tee_bnxt_fw_remove(struct device *dev) +{ + tee_client_close_session(pvt_data.ctx, pvt_data.session_id); + tee_client_close_context(pvt_data.ctx); + pvt_data.ctx = NULL; + + return 0; +} + +static const struct tee_client_device_id tee_bnxt_fw_id_table[] = { + {UUID_INIT(0x6272636D, 0x2019, 0x0716, + 0x42, 0x43, 0x4D, 0x5F, 0x53, 0x43, 0x48, 0x49)}, + {} +}; + +MODULE_DEVICE_TABLE(tee, tee_bnxt_fw_id_table); + +static struct tee_client_driver tee_bnxt_fw_driver = { + .id_table = tee_bnxt_fw_id_table, + .driver = { + .name = DRIVER_NAME, + .bus = &tee_bus_type, + .probe = tee_bnxt_fw_probe, + .remove = tee_bnxt_fw_remove, + }, +}; + +static int __init tee_bnxt_fw_mod_init(void) +{ + return driver_register(&tee_bnxt_fw_driver.driver); +} + +static void __exit tee_bnxt_fw_mod_exit(void) +{ + driver_unregister(&tee_bnxt_fw_driver.driver); +} + +module_init(tee_bnxt_fw_mod_init); +module_exit(tee_bnxt_fw_mod_exit); + +MODULE_AUTHOR("Broadcom"); +MODULE_DESCRIPTION("Broadcom bnxt firmware manager"); +MODULE_LICENSE("GPL v2"); diff --git a/include/linux/firmware/broadcom/tee_bnxt_fw.h b/include/linux/firmware/broadcom/tee_bnxt_fw.h new file mode 100644 index 00000000..d3b7206 --- /dev/null +++ b/include/linux/firmware/broadcom/tee_bnxt_fw.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: BSD-2-Clause */ +/* + * Copyright 2019 Broadcom. + */ + +#ifndef _BROADCOM_TEE_BNXT_FW_H +#define _BROADCOM_TEE_BNXT_FW_H + +#include <linux/types.h> + +int tee_bnxt_fw_load(void); +int tee_bnxt_health_status(u32 *status); +int tee_bnxt_handshake(u32 timeout, u32 *status); +int tee_bnxt_fw_update(void *buf, u32 size); +int tee_bnxt_copy_coredump(void *buf, u32 offset, u32 size); + +#endif /* _BROADCOM_TEE_BNXT_FW_H */ -- 1.9.1

5 years, 2 months

[RFC/RFT v4 0/5] Add generic trusted keys framework/subsystem

by Sumit Garg

This patch-set is an outcome of discussion here [1]. It has evolved very much since v1 to create, consolidate and generalize trusted keys subsystem. This framework has been tested with trusted keys support provided via TEE but I wasn't able to test it with a TPM device as I don't possess one. It would be really helpful if others could test this patch-set using a TPM device. [1] https://www.mail-archive.com/linux-doc@vger.kernel.org/msg30591.html Changes in v4: 1. Separate patch for export of tpm_buf code to include/linux/tpm.h 2. Change TPM1.x trusted keys code to use common tpm_buf 3. Keep module name as trusted.ko only Changes in v3: Move TPM2 trusted keys code to trusted keys subsystem. Changes in v2: Split trusted keys abstraction patch for ease of review. Sumit Garg (5): tpm: move tpm_buf code to include/linux/ KEYS: trusted: use common tpm_buf for TPM1.x code KEYS: trusted: create trusted keys subsystem KEYS: trusted: move tpm2 trusted keys code KEYS: trusted: Add generic trusted keys framework crypto/asymmetric_keys/asym_tpm.c | 2 +- drivers/char/tpm/tpm-chip.c | 1 + drivers/char/tpm/tpm-interface.c | 56 --- drivers/char/tpm/tpm.h | 230 ----------- drivers/char/tpm/tpm2-cmd.c | 308 +-------------- include/keys/trusted-type.h | 45 +++ include/keys/{trusted.h => trusted_tpm.h} | 61 +-- include/linux/tpm.h | 270 ++++++++++++- security/keys/Makefile | 2 +- security/keys/trusted-keys/Makefile | 9 + security/keys/trusted-keys/trusted-common.c | 343 ++++++++++++++++ .../keys/{trusted.c => trusted-keys/trusted-tpm.c} | 437 +++++---------------- security/keys/trusted-keys/trusted-tpm2.c | 378 ++++++++++++++++++ 13 files changed, 1141 insertions(+), 1001 deletions(-) rename include/keys/{trusted.h => trusted_tpm.h} (64%) create mode 100644 security/keys/trusted-keys/Makefile create mode 100644 security/keys/trusted-keys/trusted-common.c rename security/keys/{trusted.c => trusted-keys/trusted-tpm.c} (72%) create mode 100644 security/keys/trusted-keys/trusted-tpm2.c -- 2.7.4

5 years, 2 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

Tee-dev September 2019