- OP-TEE - lists.trustedfirmware.org

[PATCH 0/3] optee: async notif with PPI + interrupt provider

by Etienne Carriere

Dear all, This small series aggregates 2 change proposals related to OP-TEE async notif. I've made a single series since the 2 patches hit the same portions of optee driver source file and I think this will help the review. If you prefer having independent post and deal with the conflicts afterward, please tell me. Patch "optee: add per cpu asynchronous notification" aims at allowing optee to use a per-cpu interrupt (PPI on Arm CPUs) for async notif instead of a shared peripheral interrupt. The 2 next patches implement a new feature in OP-TEE, based on optee async notif. The allow optee driver to behave as an interrupt controller, for when a secure OP-TEE event is to be delivered to the Linux kernel as a interrupt event. Regards, Etienne Etienne Carriere (3): optee: add per cpu asynchronous notification dt-bindings: arm: optee: add interrupt controller properties optee core: add irq chip using optee async notification .../arm/firmware/linaro,optee-tz.yaml | 19 +- drivers/tee/optee/optee_private.h | 24 ++ drivers/tee/optee/optee_smc.h | 78 +++++- drivers/tee/optee/smc_abi.c | 249 +++++++++++++++++- 4 files changed, 358 insertions(+), 12 deletions(-) -- 2.25.1

3 years

4
10
0 0

Preparing for OP-TEE 3.20.0

by Jens Wiklander

[BCC all OP-TEE maintainers] Hi OP-TEE maintainers & contributors, OP-TEE v3.20.0 is scheduled to be released on 2023-01-20. So, now is a good time to start testing the master branch on the various platforms and report/fix any bugs. The GitHub pull request for collecting Tested-by tags or any other comments is https://github.com/OP-TEE/optee_os/pull/5751 As usual, we will create a release candidate tag one week before the release date for final testing. In addition to that you can find some additional information related to releases here: https://optee.readthedocs.io/en/latest/general/releases.html Thanks, Jens

3 years

1
1
0 0

optee device tree and tee driver

by 梅建强(禹夜)

Hello expert, I have a few questions about the optee device tree and tee driver. In our environment, optee uses version 3.19 vexpress-fvp and runs on top of hafnium as SPMC. Here's my question. 1. Does the current optee version support adding ACPI table iterm for optee to enable REE to identify the installed optee driver so that TA and CA can establish communication? (I raise this question because I cannot find the corresponding optee node in "/sys/devices/platform" in the current linux environment. While in the QEMU runtime environment, the corresponding optee node is in "/proc/device-tree/firmware", i.e. "linaro,optee-tz".) 2. If I want to transfer dtb file which includes a device tree node written by optee to linux, How to configure CFG_DT, CFG_DT_ADDR, and CFG_EXTERNAL_DTB_OVERLAY? (I understand that this method should be independent from the method in question 1. If it is not, look forward your explain.) Looking forward to your reply. Thanks. regards, yuye

3 years

2
4
0 0

[PATCH] tee: optee: ffa_abi: Fix use-after-free in optee_ffa_open

by Yoochan Lee

A race condition may occur if the user physically removes the ffa_abi device while calling open(). This is a race condition between optee_open() function and the optee_ffa_remove() function, which may lead to Use-After-Free. Therefore, add a refcount check to optee_ffa_remove() function to free the "optee" structure after the device is close()d. ---------------CPU 0--------------------CPU 1----------------- optee_open() | optee_ffa_remove() -------------------------------------------------------------- struct optee *optee = tee_get_| drvdata(teedev); — (1) | | struct optee *optee = ffa_dev_ | get_drvdata(ffa_dev); | ... | kfree(optee); — (2) if (teedev == optee->supp_ | teedev) { — (3) | Signed-off-by: Yoochan Lee <yoochan1026(a)gmail.com> --- drivers/tee/optee/ffa_abi.c | 49 +++++++++++++++++++++++++------ drivers/tee/optee/optee_private.h | 1 + 2 files changed, 41 insertions(+), 9 deletions(-) diff --git a/drivers/tee/optee/ffa_abi.c b/drivers/tee/optee/ffa_abi.c index 0828240f27e6..ea76d7532419 100644 --- a/drivers/tee/optee/ffa_abi.c +++ b/drivers/tee/optee/ffa_abi.c @@ -726,15 +726,52 @@ static void optee_ffa_get_version(struct tee_device *teedev, *vers = v; } +static void optee_ffa_delete(struct kref *kref) +{ + struct optee *optee = container_of(kref, struct optee, refcnt); + + optee_remove_common(optee); + + mutex_destroy(&optee->ffa.mutex); + rhashtable_free_and_destroy(&optee->ffa.global_ids, rh_free_fn, NULL); + + kfree(optee); + +} + +static void optee_ffa_release(struct tee_context *ctx) +{ + struct optee *optee = tee_get_drvdata(teedev); + + optee_release_helper(ctx, optee_close_session_helper); + kref_put(&optee->refcnt, optee_ffa_delete); +} + +void optee_ffa_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); + kref_put(&optee->refcnt, optee_ffa_delete); +} + static int optee_ffa_open(struct tee_context *ctx) { + struct optee *optee = tee_get_drvdata(teedev); + kref_get(&optee->refcnt); + return optee_open(ctx, true); } static const struct tee_driver_ops optee_ffa_clnt_ops = { .get_version = optee_ffa_get_version, .open = optee_ffa_open, - .release = optee_release, + .release = optee_ffa_release, .open_session = optee_open_session, .close_session = optee_close_session, .invoke_func = optee_invoke_func, @@ -752,7 +789,7 @@ static const struct tee_desc optee_ffa_clnt_desc = { static const struct tee_driver_ops optee_ffa_supp_ops = { .get_version = optee_ffa_get_version, .open = optee_ffa_open, - .release = optee_release_supp, + .release = optee_ffa_release_supp, .supp_recv = optee_supp_recv, .supp_send = optee_supp_send, .shm_register = optee_ffa_shm_register, /* same as for clnt ops */ @@ -775,13 +812,7 @@ static const struct optee_ops optee_ffa_ops = { static void optee_ffa_remove(struct ffa_device *ffa_dev) { struct optee *optee = ffa_dev_get_drvdata(ffa_dev); - - optee_remove_common(optee); - - mutex_destroy(&optee->ffa.mutex); - rhashtable_free_and_destroy(&optee->ffa.global_ids, rh_free_fn, NULL); - - kfree(optee); + kref_put(&optee->refcnt, optee_ffa_delete); } static int optee_ffa_probe(struct ffa_device *ffa_dev) diff --git a/drivers/tee/optee/optee_private.h b/drivers/tee/optee/optee_private.h index 04ae58892608..f52b1cf20eab 100644 --- a/drivers/tee/optee/optee_private.h +++ b/drivers/tee/optee/optee_private.h @@ -175,6 +175,7 @@ struct optee { bool scan_bus_done; struct workqueue_struct *scan_bus_wq; struct work_struct scan_bus_work; + struct kref refcnt; }; struct optee_session { -- 2.39.0

3 years

3
4
0 0

[PATCH] tee: optee: smc_abi: Fix use-after-free in optee_smc_open

by Yoochan Lee

A race condition may occur if the user physically removes the smc_abi device while calling open(). This is a race condition between optee_smc_open() function and the optee_smc_remove() function, which may lead to Use-After-Free. Therefore, add a refcount check to optee_smc_remove() function to free the "optee" structure after the device is close()d. ---------------CPU 0--------------------CPU 1----------------- optee_smc_open() | optee_smc_remove() -------------------------------------------------------------- struct optee *optee = tee_get_| drvdata(ctx->teedev); — (1) | | struct optee *optee = platform_ | get_drvdata(pdev); | ... | kfree(optee); — (2) u32 sec_caps = optee->smc.sec_| caps; — (3) Signed-off-by: Yoochan Lee <yoochan1026(a)gmail.com> --- drivers/tee/optee/optee_private.h | 1 + drivers/tee/optee/smc_abi.c | 66 ++++++++++++++++++++++--------- 2 files changed, 48 insertions(+), 19 deletions(-) diff --git a/drivers/tee/optee/optee_private.h b/drivers/tee/optee/optee_private.h index 04ae58892608..f52b1cf20eab 100644 --- a/drivers/tee/optee/optee_private.h +++ b/drivers/tee/optee/optee_private.h @@ -175,6 +175,7 @@ struct optee { bool scan_bus_done; struct workqueue_struct *scan_bus_wq; struct work_struct scan_bus_work; + struct kref refcnt; }; struct optee_session { diff --git a/drivers/tee/optee/smc_abi.c b/drivers/tee/optee/smc_abi.c index a1c1fa1a9c28..4fbec2acc255 100644 --- a/drivers/tee/optee/smc_abi.c +++ b/drivers/tee/optee/smc_abi.c @@ -1077,18 +1077,61 @@ static void optee_get_version(struct tee_device *teedev, *vers = v; } +static void optee_smc_delete(struct kref *kref) +{ + struct optee *optee = container_of(kref, struct optee, refcnt); + + /* + * Ask OP-TEE to free all cached shared memory objects to decrease + * reference counters and also avoid wild pointers in secure world + * into the old shared memory range. + */ + if (!optee->rpc_param_count) + optee_disable_shm_cache(optee); + + optee_smc_notif_uninit_irq(optee); + + optee_remove_common(optee); + + if (optee->smc.memremaped_shm) + memunmap(optee->smc.memremaped_shm); + + kfree(optee); +} + +static void optee_smc_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); + kref_put(&optee->refcnt, optee_smc_delete); +} + +static void optee_smc_release(struct tee_context *ctx) +{ + struct optee *optee = tee_get_drvdata(ctx->teedev); + + optee_release_helper(ctx, optee_close_session_helper); + kref_put(&optee->refcnt, optee_smc_delete); +} + static int optee_smc_open(struct tee_context *ctx) { struct optee *optee = tee_get_drvdata(ctx->teedev); u32 sec_caps = optee->smc.sec_caps; - + kref_get(&optee->refcnt); return optee_open(ctx, sec_caps & OPTEE_SMC_SEC_CAP_MEMREF_NULL); } static const struct tee_driver_ops optee_clnt_ops = { .get_version = optee_get_version, .open = optee_smc_open, - .release = optee_release, + .release = optee_smc_release, .open_session = optee_open_session, .close_session = optee_close_session, .invoke_func = optee_invoke_func, @@ -1106,7 +1149,7 @@ static const struct tee_desc optee_clnt_desc = { static const struct tee_driver_ops optee_supp_ops = { .get_version = optee_get_version, .open = optee_smc_open, - .release = optee_release_supp, + .release = optee_smc_release_supp, .supp_recv = optee_supp_recv, .supp_send = optee_supp_send, .shm_register = optee_shm_register_supp, @@ -1319,22 +1362,7 @@ static int optee_smc_remove(struct platform_device *pdev) { struct optee *optee = platform_get_drvdata(pdev); - /* - * Ask OP-TEE to free all cached shared memory objects to decrease - * reference counters and also avoid wild pointers in secure world - * into the old shared memory range. - */ - if (!optee->rpc_param_count) - optee_disable_shm_cache(optee); - - optee_smc_notif_uninit_irq(optee); - - optee_remove_common(optee); - - if (optee->smc.memremaped_shm) - memunmap(optee->smc.memremaped_shm); - - kfree(optee); + kref_put(&optee->refcnt, optee_smc_delete); return 0; } -- 2.39.0

3 years

3
4
0 0

Linaro OP-TEE Contribution (LOC) monthly meeting December 27

by Jens Wiklander

Hi, The LOC meeting this month is in the middle of the holidays. Let's cancel it since we don't have any urgent topics. Regards, Jens on behalf of the OP-TEE team.

3 years, 1 month

1
0
0 0

Re: [PATCH 2/4] tee: Remove vmalloc page support

by Sumit Garg

Hi Phil, Please don't top-post in the OSS mailing list. On Wed, 5 Oct 2022 at 08:59, Phil Chang (張世勳) <Phil.Chang(a)mediatek.com> wrote: > > Hi Sumit > > Thanks for mentioning that, in fact, our product is low memory devices, and continuous pages are extremely valuable. > Although our driver is not upstream yet but highly dependent on tee shm vmalloc support, Sorry but you need to get your driver mainline in order to support vmalloc interface. As otherwise it's a maintenance nightmare to support interfaces in the mainline for out-of-tree drivers. -Sumit > some scenarios are driver alloc high order pages but system memory is fragmentation so that alloc failed. > In this situation, vmalloc support is important and gives flexible usage to user. > > > -----Original Message----- > From: Sumit Garg <sumit.garg(a)linaro.org> > Sent: Monday, October 3, 2022 2:57 PM > To: ira.weiny(a)intel.com > Cc: Jens Wiklander <jens.wiklander(a)linaro.org>; Andrew Morton <akpm(a)linux-foundation.org>; Al Viro <viro(a)zeniv.linux.org.uk>; Fabio M. De Francesco <fmdefrancesco(a)gmail.com>; Christoph Hellwig <hch(a)lst.de>; Linus Torvalds <torvalds(a)linux-foundation.org>; op-tee(a)lists.trustedfirmware.org; linux-kernel(a)vger.kernel.org; linux-mm(a)kvack.org; Phil Chang (張世勳) <Phil.Chang(a)mediatek.com> > Subject: Re: [PATCH 2/4] tee: Remove vmalloc page support > > + Phil > > Hi Ira, > > On Sun, 2 Oct 2022 at 05:53, <ira.weiny(a)intel.com> wrote: > > > > From: Ira Weiny <ira.weiny(a)intel.com> > > > > The kernel pages used by shm_get_kernel_pages() are allocated using > > GFP_KERNEL through the following call stack: > > > > trusted_instantiate() > > trusted_payload_alloc() -> GFP_KERNEL > > <trusted key op> > > tee_shm_register_kernel_buf() > > register_shm_helper() > > shm_get_kernel_pages() > > > > Where <trusted key op> is one of: > > > > trusted_key_unseal() > > trusted_key_get_random() > > trusted_key_seal() > > > > Remove the vmalloc page support from shm_get_kernel_pages(). Replace > > with a warn on once. > > > > Cc: Jens Wiklander <jens.wiklander(a)linaro.org> > > Cc: Al Viro <viro(a)zeniv.linux.org.uk> > > Cc: "Fabio M. De Francesco" <fmdefrancesco(a)gmail.com> > > Cc: Christoph Hellwig <hch(a)lst.de> > > Cc: Linus Torvalds <torvalds(a)linux-foundation.org> > > Signed-off-by: Ira Weiny <ira.weiny(a)intel.com> > > > > --- > > Jens I went with the suggestion from Linus and Christoph and rejected > > vmalloc addresses. I did not hear back from you regarding Linus' > > question if the vmalloc page support was required by an up coming > > patch set or not. So I assumed it was something out of tree. > > It looks like I wasn't CC'd to that conversation. IIRC, support for vmalloc addresses was added recently by Phil here [1]. So I would like to give him a chance if he is planning to post a corresponding kernel driver upstream. > > [1] https://urldefense.com/v3/__https://lists.trustedfirmware.org/archives/list… > > -Sumit > > > --- > > drivers/tee/tee_shm.c | 36 ++++++++++++------------------------ > > 1 file changed, 12 insertions(+), 24 deletions(-) > > > > diff --git a/drivers/tee/tee_shm.c b/drivers/tee/tee_shm.c index > > 27295bda3e0b..527a6eabc03e 100644 > > --- a/drivers/tee/tee_shm.c > > +++ b/drivers/tee/tee_shm.c > > @@ -24,37 +24,25 @@ static void shm_put_kernel_pages(struct page > > **pages, size_t page_count) static int shm_get_kernel_pages(unsigned long start, size_t page_count, > > struct page **pages) { > > + struct kvec *kiov; > > size_t n; > > int rc; > > > > - if (is_vmalloc_addr((void *)start)) { > > - struct page *page; > > - > > - for (n = 0; n < page_count; n++) { > > - page = vmalloc_to_page((void *)(start + PAGE_SIZE * n)); > > - if (!page) > > - return -ENOMEM; > > - > > - get_page(page); > > - pages[n] = page; > > - } > > - rc = page_count; > > - } else { > > - struct kvec *kiov; > > - > > - kiov = kcalloc(page_count, sizeof(*kiov), GFP_KERNEL); > > - if (!kiov) > > - return -ENOMEM; > > + if (WARN_ON_ONCE(is_vmalloc_addr((void *)start))) > > + return -EINVAL; > > > > - for (n = 0; n < page_count; n++) { > > - kiov[n].iov_base = (void *)(start + n * PAGE_SIZE); > > - kiov[n].iov_len = PAGE_SIZE; > > - } > > + kiov = kcalloc(page_count, sizeof(*kiov), GFP_KERNEL); > > + if (!kiov) > > + return -ENOMEM; > > > > - rc = get_kernel_pages(kiov, page_count, 0, pages); > > - kfree(kiov); > > + for (n = 0; n < page_count; n++) { > > + kiov[n].iov_base = (void *)(start + n * PAGE_SIZE); > > + kiov[n].iov_len = PAGE_SIZE; > > } > > > > + rc = get_kernel_pages(kiov, page_count, 0, pages); > > + kfree(kiov); > > + > > return rc; > > } > > > > -- > > 2.37.2 > >

3 years, 1 month

6
10
0 0

register_ddr() with larger than physical RAM size

by Jeffrey Kardatzke

Hello, New to this list, I'm working on integrating Op-Tee into ChromeOS ARM devices for the purpose of Widevine DRM. Does anybody know if Op-Tee creates mappings in the MMU that go beyond the registered TZ memory region and the static non-secure shared memory region? TL;DR Does anyone know if there would be problems associated with invoking register_ddr()[1] with a size larger than the physical RAM size? The reason I want to do this is because we have multiple different SKUs for the same reference board, and for ChromeOS we build per-board images and not per-SKU images. We could use a DT to do this, but that's running into other issues...so I'm trying to determine if this easy fix will solve the problem w/out creating new ones. (it does work fine when I test it) I understand that an exploited normal world could send a PA pointing to a region outside of the physical bounds for dynamic shared memory, and then Op-Tee would try to map that and undesirable things may happen (but I don't think they would lead to secure world exploits since the memory will be mapped as non-secure, so it shouldn't be any worse that the normal world trying to access physical memory out of bounds). One of the main concerns is around whether Op-Tee is creating mappings in the MMU that reference the whole nonsecure DDR size for some other reason. Then the CPU may do speculative execution and try to access the physical memory at that invalid mapping and create problems, which could then lead to failure when someone isn't trying to exploit the system. Does anybody know if Op-Tee creates mappings in the MMU that go beyond the registered TZ memory region and the static non-secure shared memory region? [1] https://github.com/OP-TEE/optee_os/blob/master/core/arch/arm/plat-mediatek/… Thanks, -- Jeffrey Kardatzke jkardatzke(a)google.com Google, Inc.

3 years, 2 months

2
2
0 0

[GIT PULL] Add __init/__exit annotations OP-TEE driver for v6.2

by Jens Wiklander

Hello arm-soc maintainers, Please pull this small patch which adds __init/__exit annotations to the OP-TEE driver module. Thanks, Jens The following changes since commit 4fe89d07dcc2804c8b562f6c7896a45643d34b2f: Linux 6.0 (2022-10-02 14:09:07 -0700) are available in the Git repository at: https://git.linaro.org/people/jens.wiklander/linux-tee.git tags/optee-for-6.2 for you to fetch changes up to bd52407221b4702af690456b2b6006fa6440e893: optee: Add __init/__exit annotations to module init/exit funcs (2022-10-11 11:20:16 +0200) ---------------------------------------------------------------- Add missing __init/__exit annotations to OP-TEE driver ---------------------------------------------------------------- Xiu Jianfeng (1): optee: Add __init/__exit annotations to module init/exit funcs drivers/tee/optee/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

3 years, 2 months

1
0
0 0

[PATCH] tee: optee: fix possible memory leak in optee_register_device()

by Yang Yingliang

If device_register() returns error in optee_register_device(), the name allocated by dev_set_name() need be freed. As comment of device_register() says, it should use put_device() to give up the reference in the error path. So fix this by calling put_device(), then the name can be freed in kobject_cleanup(), and optee_device is freed in optee_release_device(). Fixes: c3fa24af9244 ("tee: optee: add TEE bus device enumeration support") Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/tee/optee/device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tee/optee/device.c b/drivers/tee/optee/device.c index f3947be13e2e..64f0e047c23d 100644 --- a/drivers/tee/optee/device.c +++ b/drivers/tee/optee/device.c @@ -80,7 +80,7 @@ static int optee_register_device(const uuid_t *device_uuid) rc = device_register(&optee_device->dev); if (rc) { pr_err("device registration failed, err: %d\n", rc); - kfree(optee_device); + put_device(&optee_device->dev); } return rc; -- 2.25.1

3 years, 2 months

3
2
0 0

2026

2025

2024

2023

2022

2021

2020

OP-TEE