Re: [PATCH] tee: ACPI support for optee driver

[Jon Masters]

On 09/22/2017 05:37 AM, Lorenzo Pieralisi wrote:
> On Thu, Sep 21, 2017 at 03:45:28PM +0800, Hanjun Guo wrote:
>> On 2017/9/21 15:12, Mayuresh Chitale wrote:
>>> This patch modifies the optee driver to add support for parsing
>>> the conduit method from an ACPI node.
>>
>> Sorry I didn't involve this earlier, but I think this is a wrong
>> approach, in ACPI 5.1+ spec, there is a bit in FADT table which
>> indicates PSCI using SMC or HVC, please see ACPI 6.2 [1],
>> Table 5-37 Fixed ACPI Description Table ARM Boot Architecture Flags.
>>
>> Can we just use that to get the conduit method for optee driver too?
>>
>> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf
> 
> It is just not a matter of conduit method but also to define how OPTEE
> should be detected. It is up to Linaro (who owns OP-TEE) to put forward
> a binding at ACPI (ARM) spec level.
> 
> We do not define ACPI bindings on a kernel mailing list.
> 
> NAK on this patch.

Is this actively being tracked by Linaro? If not, can Applied folks ping
me off-list and I will proxy your request into Linaro.

Jon.


-- 
Computer Architect | Sent from my Fedora powered laptop

Re: [PATCH] tee: ACPI support for optee driver

[Jon Masters]

On 09/22/2017 05:37 AM, Lorenzo Pieralisi wrote:
> On Thu, Sep 21, 2017 at 03:45:28PM +0800, Hanjun Guo wrote:
>> On 2017/9/21 15:12, Mayuresh Chitale wrote:
>>> This patch modifies the optee driver to add support for parsing
>>> the conduit method from an ACPI node.
>>
>> Sorry I didn't involve this earlier, but I think this is a wrong
>> approach, in ACPI 5.1+ spec, there is a bit in FADT table which
>> indicates PSCI using SMC or HVC, please see ACPI 6.2 [1],
>> Table 5-37 Fixed ACPI Description Table ARM Boot Architecture Flags.
>>
>> Can we just use that to get the conduit method for optee driver too?
>>
>> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf
> 
> It is just not a matter of conduit method but also to define how OPTEE
> should be detected. It is up to Linaro (who owns OP-TEE) to put forward
> a binding at ACPI (ARM) spec level.
> 
> We do not define ACPI bindings on a kernel mailing list.
> 
> NAK on this patch.

Is this actively being tracked by Linaro? If not, can Applied folks ping
me off-list and I will proxy your request into Linaro.

Jon.


-- 
Computer Architect | Sent from my Fedora powered laptop

Re: [PATCH] tee: ACPI support for optee driver

[Lorenzo Pieralisi]

On Thu, Sep 21, 2017 at 03:45:28PM +0800, Hanjun Guo wrote:
> On 2017/9/21 15:12, Mayuresh Chitale wrote:
> > This patch modifies the optee driver to add support for parsing
> > the conduit method from an ACPI node.
> 
> Sorry I didn't involve this earlier, but I think this is a wrong
> approach, in ACPI 5.1+ spec, there is a bit in FADT table which
> indicates PSCI using SMC or HVC, please see ACPI 6.2 [1],
> Table 5-37 Fixed ACPI Description Table ARM Boot Architecture Flags.
> 
> Can we just use that to get the conduit method for optee driver too?
> 
> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf

It is just not a matter of conduit method but also to define how OPTEE
should be detected. It is up to Linaro (who owns OP-TEE) to put forward
a binding at ACPI (ARM) spec level.

We do not define ACPI bindings on a kernel mailing list.

NAK on this patch.

Lorenzo

Re: [PATCH] tee: ACPI support for optee driver

[Lorenzo Pieralisi]

On Thu, Sep 21, 2017 at 03:45:28PM +0800, Hanjun Guo wrote:
> On 2017/9/21 15:12, Mayuresh Chitale wrote:
> > This patch modifies the optee driver to add support for parsing
> > the conduit method from an ACPI node.
> 
> Sorry I didn't involve this earlier, but I think this is a wrong
> approach, in ACPI 5.1+ spec, there is a bit in FADT table which
> indicates PSCI using SMC or HVC, please see ACPI 6.2 [1],
> Table 5-37 Fixed ACPI Description Table ARM Boot Architecture Flags.
> 
> Can we just use that to get the conduit method for optee driver too?
> 
> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf

It is just not a matter of conduit method but also to define how OPTEE
should be detected. It is up to Linaro (who owns OP-TEE) to put forward
a binding at ACPI (ARM) spec level.

We do not define ACPI bindings on a kernel mailing list.

NAK on this patch.

Lorenzo

Re: [PATCH] tee: ACPI support for optee driver

[Mayuresh Chitale]

On Thu, Sep 21, 2017 at 1:15 PM, Hanjun Guo  wrote:
> On 2017/9/21 15:12, Mayuresh Chitale wrote:
>> This patch modifies the optee driver to add support for parsing
>> the conduit method from an ACPI node.
>
> Sorry I didn't involve this earlier, but I think this is a wrong
> approach, in ACPI 5.1+ spec, there is a bit in FADT table which
> indicates PSCI using SMC or HVC, please see ACPI 6.2 [1],
> Table 5-37 Fixed ACPI Description Table ARM Boot Architecture Flags.
>
> Can we just use that to get the conduit method for optee driver too?
Thanks for the comment. I will check it out.
>
> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf
>
> Thanks
> Hanjun
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] tee: ACPI support for optee driver

[Mayuresh Chitale]

On Thu, Sep 21, 2017 at 1:15 PM, Hanjun Guo  wrote:
> On 2017/9/21 15:12, Mayuresh Chitale wrote:
>> This patch modifies the optee driver to add support for parsing
>> the conduit method from an ACPI node.
>
> Sorry I didn't involve this earlier, but I think this is a wrong
> approach, in ACPI 5.1+ spec, there is a bit in FADT table which
> indicates PSCI using SMC or HVC, please see ACPI 6.2 [1],
> Table 5-37 Fixed ACPI Description Table ARM Boot Architecture Flags.
>
> Can we just use that to get the conduit method for optee driver too?
Thanks for the comment. I will check it out.
>
> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf
>
> Thanks
> Hanjun
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] tee: ACPI support for optee driver

[Hanjun Guo]

On 2017/9/21 15:12, Mayuresh Chitale wrote:
> This patch modifies the optee driver to add support for parsing
> the conduit method from an ACPI node.

Sorry I didn't involve this earlier, but I think this is a wrong
approach, in ACPI 5.1+ spec, there is a bit in FADT table which
indicates PSCI using SMC or HVC, please see ACPI 6.2 [1],
Table 5-37 Fixed ACPI Description Table ARM Boot Architecture Flags.

Can we just use that to get the conduit method for optee driver too?

[1]: http://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf

Thanks
Hanjun

Re: [PATCH] tee: ACPI support for optee driver

[Hanjun Guo]

On 2017/9/21 15:12, Mayuresh Chitale wrote:
> This patch modifies the optee driver to add support for parsing
> the conduit method from an ACPI node.

Sorry I didn't involve this earlier, but I think this is a wrong
approach, in ACPI 5.1+ spec, there is a bit in FADT table which
indicates PSCI using SMC or HVC, please see ACPI 6.2 [1],
Table 5-37 Fixed ACPI Description Table ARM Boot Architecture Flags.

Can we just use that to get the conduit method for optee driver too?

[1]: http://www.uefi.org/sites/default/files/resources/ACPI_6_2.pdf

Thanks
Hanjun

[PATCH] tee: ACPI support for optee driver

[Mayuresh Chitale]

This patch modifies the optee driver to add support for parsing
the conduit method from an ACPI node.

Signed-off-by: Mayuresh Chitale 
---
 drivers/tee/optee/core.c | 112 ---
 1 file changed, 77 insertions(+), 35 deletions(-)

diff --git a/drivers/tee/optee/core.c b/drivers/tee/optee/core.c
index 58169e5..8b15c49 100644
--- a/drivers/tee/optee/core.c
+++ b/drivers/tee/optee/core.c
@@ -14,6 +14,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include 
 #include 
 #include 
 #include 
@@ -30,6 +31,7 @@
 #include "optee_smc.h"
 
 #define DRIVER_NAME "optee"
+#define OPTEE_DEVICE "\\_SB.OPTE"
 
 #define OPTEE_SHM_NUM_PRIV_PAGES   1
 
@@ -425,29 +427,87 @@ static void optee_smccc_hvc(unsigned long a0, unsigned 
long a1,
arm_smccc_hvc(a0, a1, a2, a3, a4, a5, a6, a7, res);
 }
 
-static optee_invoke_fn *get_invoke_func(struct device_node *np)
+static const struct of_device_id optee_match[] = {
+   { .compatible = "linaro,optee-tz" },
+   {},
+};
+
+#ifdef CONFIG_ACPI
+static const char *get_invoke_str_acpi(void)
 {
-   const char *method;
+   struct acpi_device_info *info;
+   const union acpi_object *obj;
+   struct acpi_device *adev;
+   acpi_status status;
+   acpi_handle handle;
 
-   pr_info("probing for conduit method from DT.\n");
+   status = acpi_get_handle(ACPI_ROOT_OBJECT, OPTEE_DEVICE, );
+   if (ACPI_FAILURE(status))
+   return NULL;
 
-   if (of_property_read_string(np, "method", )) {
+   status = acpi_get_object_info(handle, );
+   if (ACPI_FAILURE(status) || !(info->valid & ACPI_VALID_HID))
+   return NULL;
+
+   if (acpi_bus_get_device(handle, ))
+   return NULL;
+
+   if (acpi_dev_get_property(adev, "method",
+   ACPI_TYPE_ANY, ))
+   return NULL;
+
+   return obj->string.pointer;
+}
+#endif
+
+static const char *get_invoke_str_of(void)
+{
+   struct device_node *fw_np;
+   const char *method = NULL;
+   struct device_node *np;
+
+   /* Node is supposed to be below /firmware */
+   fw_np = of_find_node_by_name(NULL, "firmware");
+   if (!fw_np)
+   return NULL;
+
+   np = of_find_matching_node(fw_np, optee_match);
+   of_node_put(fw_np);
+   if (!np)
+   return NULL;
+
+   pr_info("probing for conduit method from DT.\n");
+   if (of_property_read_string(np, "method", ))
pr_warn("missing \"method\" property\n");
-   return ERR_PTR(-ENXIO);
-   }
 
-   if (!strcmp("hvc", method))
-   return optee_smccc_hvc;
-   else if (!strcmp("smc", method))
-   return optee_smccc_smc;
+   of_node_put(np);
+   return method;
+}
+
+static optee_invoke_fn *get_invoke_func(void)
+{
+   const char *method;
+
+#ifdef CONFIG_ACPI
+   if (!acpi_disabled)
+   method = get_invoke_str_acpi();
+   else
+#endif
+   method = get_invoke_str_of();
+
+   if (method) {
+   if (!strcmp("hvc", method))
+   return optee_smccc_hvc;
+   else if (!strcmp("smc", method))
+   return optee_smccc_smc;
+   }
 
pr_warn("invalid \"method\" property: %s\n", method);
return ERR_PTR(-EINVAL);
 }
 
-static struct optee *optee_probe(struct device_node *np)
+static struct optee *optee_probe(optee_invoke_fn *invoke_fn)
 {
-   optee_invoke_fn *invoke_fn;
struct tee_shm_pool *pool;
struct optee *optee = NULL;
void *memremaped_shm = NULL;
@@ -455,10 +515,6 @@ static struct optee *optee_probe(struct device_node *np)
u32 sec_caps;
int rc;
 
-   invoke_fn = get_invoke_func(np);
-   if (IS_ERR(invoke_fn))
-   return (void *)invoke_fn;
-
if (!optee_msg_api_uid_is_optee_api(invoke_fn)) {
pr_warn("api uid mismatch\n");
return ERR_PTR(-EINVAL);
@@ -570,32 +626,18 @@ static void optee_remove(struct optee *optee)
kfree(optee);
 }
 
-static const struct of_device_id optee_match[] = {
-   { .compatible = "linaro,optee-tz" },
-   {},
-};
-
 static struct optee *optee_svc;
 
 static int __init optee_driver_init(void)
 {
-   struct device_node *fw_np;
-   struct device_node *np;
+   optee_invoke_fn *invoke_fn;
struct optee *optee;
 
-   /* Node is supposed to be below /firmware */
-   fw_np = of_find_node_by_name(NULL, "firmware");
-   if (!fw_np)
-   return -ENODEV;
-
-   np = of_find_matching_node(fw_np, optee_match);
-   of_node_put(fw_np);
-   if (!np)
-   return -ENODEV;
-
-   optee = optee_probe(np);
-   of_node_put(np);
+   invoke_fn = get_invoke_func();
+   if (IS_ERR(invoke_fn))
+   return PTR_ERR(invoke_fn);
 
+   optee = optee_probe(invoke_fn);
if 

[PATCH] tee: ACPI support for optee driver

[Mayuresh Chitale]

This patch modifies the optee driver to add support for parsing
the conduit method from an ACPI node.

Signed-off-by: Mayuresh Chitale 
---
 drivers/tee/optee/core.c | 112 ---
 1 file changed, 77 insertions(+), 35 deletions(-)

diff --git a/drivers/tee/optee/core.c b/drivers/tee/optee/core.c
index 58169e5..8b15c49 100644
--- a/drivers/tee/optee/core.c
+++ b/drivers/tee/optee/core.c
@@ -14,6 +14,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include 
 #include 
 #include 
 #include 
@@ -30,6 +31,7 @@
 #include "optee_smc.h"
 
 #define DRIVER_NAME "optee"
+#define OPTEE_DEVICE "\\_SB.OPTE"
 
 #define OPTEE_SHM_NUM_PRIV_PAGES   1
 
@@ -425,29 +427,87 @@ static void optee_smccc_hvc(unsigned long a0, unsigned 
long a1,
arm_smccc_hvc(a0, a1, a2, a3, a4, a5, a6, a7, res);
 }
 
-static optee_invoke_fn *get_invoke_func(struct device_node *np)
+static const struct of_device_id optee_match[] = {
+   { .compatible = "linaro,optee-tz" },
+   {},
+};
+
+#ifdef CONFIG_ACPI
+static const char *get_invoke_str_acpi(void)
 {
-   const char *method;
+   struct acpi_device_info *info;
+   const union acpi_object *obj;
+   struct acpi_device *adev;
+   acpi_status status;
+   acpi_handle handle;
 
-   pr_info("probing for conduit method from DT.\n");
+   status = acpi_get_handle(ACPI_ROOT_OBJECT, OPTEE_DEVICE, );
+   if (ACPI_FAILURE(status))
+   return NULL;
 
-   if (of_property_read_string(np, "method", )) {
+   status = acpi_get_object_info(handle, );
+   if (ACPI_FAILURE(status) || !(info->valid & ACPI_VALID_HID))
+   return NULL;
+
+   if (acpi_bus_get_device(handle, ))
+   return NULL;
+
+   if (acpi_dev_get_property(adev, "method",
+   ACPI_TYPE_ANY, ))
+   return NULL;
+
+   return obj->string.pointer;
+}
+#endif
+
+static const char *get_invoke_str_of(void)
+{
+   struct device_node *fw_np;
+   const char *method = NULL;
+   struct device_node *np;
+
+   /* Node is supposed to be below /firmware */
+   fw_np = of_find_node_by_name(NULL, "firmware");
+   if (!fw_np)
+   return NULL;
+
+   np = of_find_matching_node(fw_np, optee_match);
+   of_node_put(fw_np);
+   if (!np)
+   return NULL;
+
+   pr_info("probing for conduit method from DT.\n");
+   if (of_property_read_string(np, "method", ))
pr_warn("missing \"method\" property\n");
-   return ERR_PTR(-ENXIO);
-   }
 
-   if (!strcmp("hvc", method))
-   return optee_smccc_hvc;
-   else if (!strcmp("smc", method))
-   return optee_smccc_smc;
+   of_node_put(np);
+   return method;
+}
+
+static optee_invoke_fn *get_invoke_func(void)
+{
+   const char *method;
+
+#ifdef CONFIG_ACPI
+   if (!acpi_disabled)
+   method = get_invoke_str_acpi();
+   else
+#endif
+   method = get_invoke_str_of();
+
+   if (method) {
+   if (!strcmp("hvc", method))
+   return optee_smccc_hvc;
+   else if (!strcmp("smc", method))
+   return optee_smccc_smc;
+   }
 
pr_warn("invalid \"method\" property: %s\n", method);
return ERR_PTR(-EINVAL);
 }
 
-static struct optee *optee_probe(struct device_node *np)
+static struct optee *optee_probe(optee_invoke_fn *invoke_fn)
 {
-   optee_invoke_fn *invoke_fn;
struct tee_shm_pool *pool;
struct optee *optee = NULL;
void *memremaped_shm = NULL;
@@ -455,10 +515,6 @@ static struct optee *optee_probe(struct device_node *np)
u32 sec_caps;
int rc;
 
-   invoke_fn = get_invoke_func(np);
-   if (IS_ERR(invoke_fn))
-   return (void *)invoke_fn;
-
if (!optee_msg_api_uid_is_optee_api(invoke_fn)) {
pr_warn("api uid mismatch\n");
return ERR_PTR(-EINVAL);
@@ -570,32 +626,18 @@ static void optee_remove(struct optee *optee)
kfree(optee);
 }
 
-static const struct of_device_id optee_match[] = {
-   { .compatible = "linaro,optee-tz" },
-   {},
-};
-
 static struct optee *optee_svc;
 
 static int __init optee_driver_init(void)
 {
-   struct device_node *fw_np;
-   struct device_node *np;
+   optee_invoke_fn *invoke_fn;
struct optee *optee;
 
-   /* Node is supposed to be below /firmware */
-   fw_np = of_find_node_by_name(NULL, "firmware");
-   if (!fw_np)
-   return -ENODEV;
-
-   np = of_find_matching_node(fw_np, optee_match);
-   of_node_put(fw_np);
-   if (!np)
-   return -ENODEV;
-
-   optee = optee_probe(np);
-   of_node_put(np);
+   invoke_fn = get_invoke_func();
+   if (IS_ERR(invoke_fn))
+   return PTR_ERR(invoke_fn);
 
+   optee = optee_probe(invoke_fn);
if (IS_ERR(optee))