[v2] qcom_tzmem: Enhance Error Handling for shmbridge

[PATCH v2 1/2] firmware: qcom: scm: Return -EOPNOTSUPP for unsupported SHM bridge enabling

Posted by Kuldeep Singh 1 month, 1 week ago

From: Qingqing Zhou <quic_qqzhou@quicinc.com>

When enabling SHM bridge, QTEE returns 0 and sets error 4 in result to
qcom_scm for unsupported platforms. Currently, tzmem interprets this as
an unknown error rather than recognizing it as an unsupported platform.

Error log:
[    0.177224] qcom_scm firmware:scm: error (____ptrval____): Failed to enable the TrustZone memory allocator
[    0.177244] qcom_scm firmware:scm: probe with driver qcom_scm failed with error 4

To address this, modify the function call qcom_scm_shm_bridge_enable()
to remap result to indicate an unsupported error. This way, tzmem will
correctly identify it as an unsupported platform case instead of
reporting it as an error.

Fixes: 178e19c0df1b ("firmware: qcom: scm: add support for SHM bridge operations")
Signed-off-by: Qingqing Zhou <quic_qqzhou@quicinc.com>
Co-developed-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
Signed-off-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
---
 drivers/firmware/qcom/qcom_scm.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/firmware/qcom/qcom_scm.c b/drivers/firmware/qcom/qcom_scm.c
index 10986cb11ec0..0df81a9ed438 100644
--- a/drivers/firmware/qcom/qcom_scm.c
+++ b/drivers/firmware/qcom/qcom_scm.c
@@ -112,6 +112,7 @@ enum qcom_scm_qseecom_tz_cmd_info {
 };
 
 #define QSEECOM_MAX_APP_NAME_SIZE		64
+#define SHMBRIDGE_RESULT_NOTSUPP		4
 
 /* Each bit configures cold/warm boot address for one of the 4 CPUs */
 static const u8 qcom_scm_cpu_cold_bits[QCOM_SCM_BOOT_MAX_CPUS] = {
@@ -1361,6 +1362,8 @@ EXPORT_SYMBOL_GPL(qcom_scm_lmh_dcvsh_available);
 
 int qcom_scm_shm_bridge_enable(void)
 {
+	int ret;
+
 	struct qcom_scm_desc desc = {
 		.svc = QCOM_SCM_SVC_MP,
 		.cmd = QCOM_SCM_MP_SHM_BRIDGE_ENABLE,
@@ -1373,7 +1376,15 @@ int qcom_scm_shm_bridge_enable(void)
 					  QCOM_SCM_MP_SHM_BRIDGE_ENABLE))
 		return -EOPNOTSUPP;
 
-	return qcom_scm_call(__scm->dev, &desc, &res) ?: res.result[0];
+	ret = qcom_scm_call(__scm->dev, &desc, &res);
+
+	if (ret)
+		return ret;
+
+	if (res.result[0] == SHMBRIDGE_RESULT_NOTSUPP)
+		return -EOPNOTSUPP;
+
+	return res.result[0];
 }
 EXPORT_SYMBOL_GPL(qcom_scm_shm_bridge_enable);
 
-- 
2.34.1

Re: [PATCH v2 1/2] firmware: qcom: scm: Return -EOPNOTSUPP for unsupported SHM bridge enabling

Posted by Kuldeep Singh 1 month ago


On 10/14/2024 4:45 PM, Kuldeep Singh wrote:
> From: Qingqing Zhou <quic_qqzhou@quicinc.com>
> 
> When enabling SHM bridge, QTEE returns 0 and sets error 4 in result to
> qcom_scm for unsupported platforms. Currently, tzmem interprets this as
> an unknown error rather than recognizing it as an unsupported platform.
> 
> Error log:
> [    0.177224] qcom_scm firmware:scm: error (____ptrval____): Failed to enable the TrustZone memory allocator
> [    0.177244] qcom_scm firmware:scm: probe with driver qcom_scm failed with error 4
> 
> To address this, modify the function call qcom_scm_shm_bridge_enable()
> to remap result to indicate an unsupported error. This way, tzmem will
> correctly identify it as an unsupported platform case instead of
> reporting it as an error.
> 
> Fixes: 178e19c0df1b ("firmware: qcom: scm: add support for SHM bridge operations")
> Signed-off-by: Qingqing Zhou <quic_qqzhou@quicinc.com>
> Co-developed-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
> Signed-off-by: Kuldeep Singh <quic_kuldsing@quicinc.com>

Hi Bjorn,

Can you please apply this fix I'll split this one and send separately.
This fixes qcom_scm probe failure and is needed for other patches to
build upon. Thank you.

-- 
Regards
Kuldeep

Re: [PATCH v2 1/2] firmware: qcom: scm: Return -EOPNOTSUPP for unsupported SHM bridge enabling

Posted by Mukesh Ojha 1 month, 1 week ago

On Mon, Oct 14, 2024 at 04:45:26PM +0530, Kuldeep Singh wrote:
> From: Qingqing Zhou <quic_qqzhou@quicinc.com>
> 
> When enabling SHM bridge, QTEE returns 0 and sets error 4 in result to
> qcom_scm for unsupported platforms. Currently, tzmem interprets this as
> an unknown error rather than recognizing it as an unsupported platform.
> 
> Error log:
> [    0.177224] qcom_scm firmware:scm: error (____ptrval____): Failed to enable the TrustZone memory allocator
> [    0.177244] qcom_scm firmware:scm: probe with driver qcom_scm failed with error 4
> 
> To address this, modify the function call qcom_scm_shm_bridge_enable()
> to remap result to indicate an unsupported error. This way, tzmem will
> correctly identify it as an unsupported platform case instead of
> reporting it as an error.
> 
> Fixes: 178e19c0df1b ("firmware: qcom: scm: add support for SHM bridge operations")
> Signed-off-by: Qingqing Zhou <quic_qqzhou@quicinc.com>
> Co-developed-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
> Signed-off-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
> ---
>  drivers/firmware/qcom/qcom_scm.c | 13 ++++++++++++-
>  1 file changed, 12 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/firmware/qcom/qcom_scm.c b/drivers/firmware/qcom/qcom_scm.c
> index 10986cb11ec0..0df81a9ed438 100644
> --- a/drivers/firmware/qcom/qcom_scm.c
> +++ b/drivers/firmware/qcom/qcom_scm.c
> @@ -112,6 +112,7 @@ enum qcom_scm_qseecom_tz_cmd_info {
>  };
>  
>  #define QSEECOM_MAX_APP_NAME_SIZE		64
> +#define SHMBRIDGE_RESULT_NOTSUPP		4
>  
>  /* Each bit configures cold/warm boot address for one of the 4 CPUs */
>  static const u8 qcom_scm_cpu_cold_bits[QCOM_SCM_BOOT_MAX_CPUS] = {
> @@ -1361,6 +1362,8 @@ EXPORT_SYMBOL_GPL(qcom_scm_lmh_dcvsh_available);
>  
>  int qcom_scm_shm_bridge_enable(void)
>  {
> +	int ret;
> +
>  	struct qcom_scm_desc desc = {
>  		.svc = QCOM_SCM_SVC_MP,
>  		.cmd = QCOM_SCM_MP_SHM_BRIDGE_ENABLE,
> @@ -1373,7 +1376,15 @@ int qcom_scm_shm_bridge_enable(void)
>  					  QCOM_SCM_MP_SHM_BRIDGE_ENABLE))
>  		return -EOPNOTSUPP;
>  
> -	return qcom_scm_call(__scm->dev, &desc, &res) ?: res.result[0];
> +	ret = qcom_scm_call(__scm->dev, &desc, &res);
> +
> +	if (ret)
> +		return ret;
> +
> +	if (res.result[0] == SHMBRIDGE_RESULT_NOTSUPP)
> +		return -EOPNOTSUPP;
> +
> +	return res.result[0];
>  }
>  EXPORT_SYMBOL_GPL(qcom_scm_shm_bridge_enable);

Reviewed-by: Mukesh Ojha <quic_mojha@quicinc.com>

-Mukesh

Re: [PATCH v2 1/2] firmware: qcom: scm: Return -EOPNOTSUPP for unsupported SHM bridge enabling

Posted by Bartosz Golaszewski 1 month, 1 week ago

On Mon, Oct 14, 2024 at 1:19 PM Kuldeep Singh <quic_kuldsing@quicinc.com> wrote:
>
> From: Qingqing Zhou <quic_qqzhou@quicinc.com>
>
> When enabling SHM bridge, QTEE returns 0 and sets error 4 in result to
> qcom_scm for unsupported platforms. Currently, tzmem interprets this as
> an unknown error rather than recognizing it as an unsupported platform.
>
> Error log:
> [    0.177224] qcom_scm firmware:scm: error (____ptrval____): Failed to enable the TrustZone memory allocator
> [    0.177244] qcom_scm firmware:scm: probe with driver qcom_scm failed with error 4
>
> To address this, modify the function call qcom_scm_shm_bridge_enable()
> to remap result to indicate an unsupported error. This way, tzmem will
> correctly identify it as an unsupported platform case instead of
> reporting it as an error.
>
> Fixes: 178e19c0df1b ("firmware: qcom: scm: add support for SHM bridge operations")
> Signed-off-by: Qingqing Zhou <quic_qqzhou@quicinc.com>
> Co-developed-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
> Signed-off-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
> ---
>  drivers/firmware/qcom/qcom_scm.c | 13 ++++++++++++-
>  1 file changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/firmware/qcom/qcom_scm.c b/drivers/firmware/qcom/qcom_scm.c
> index 10986cb11ec0..0df81a9ed438 100644
> --- a/drivers/firmware/qcom/qcom_scm.c
> +++ b/drivers/firmware/qcom/qcom_scm.c
> @@ -112,6 +112,7 @@ enum qcom_scm_qseecom_tz_cmd_info {
>  };
>
>  #define QSEECOM_MAX_APP_NAME_SIZE              64
> +#define SHMBRIDGE_RESULT_NOTSUPP               4
>
>  /* Each bit configures cold/warm boot address for one of the 4 CPUs */
>  static const u8 qcom_scm_cpu_cold_bits[QCOM_SCM_BOOT_MAX_CPUS] = {
> @@ -1361,6 +1362,8 @@ EXPORT_SYMBOL_GPL(qcom_scm_lmh_dcvsh_available);
>
>  int qcom_scm_shm_bridge_enable(void)
>  {
> +       int ret;
> +
>         struct qcom_scm_desc desc = {
>                 .svc = QCOM_SCM_SVC_MP,
>                 .cmd = QCOM_SCM_MP_SHM_BRIDGE_ENABLE,
> @@ -1373,7 +1376,15 @@ int qcom_scm_shm_bridge_enable(void)
>                                           QCOM_SCM_MP_SHM_BRIDGE_ENABLE))
>                 return -EOPNOTSUPP;
>
> -       return qcom_scm_call(__scm->dev, &desc, &res) ?: res.result[0];
> +       ret = qcom_scm_call(__scm->dev, &desc, &res);
> +
> +       if (ret)
> +               return ret;
> +
> +       if (res.result[0] == SHMBRIDGE_RESULT_NOTSUPP)
> +               return -EOPNOTSUPP;
> +
> +       return res.result[0];
>  }
>  EXPORT_SYMBOL_GPL(qcom_scm_shm_bridge_enable);
>
> --
> 2.34.1
>
>

The patch looks correct to me and like something that should go upstream.

Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

That being said, this doesn't seem to address any of the issues that
we saw with SHM Bridge and it still leads to a crash on sc8180x. :(

Bart

Re: [PATCH v2 1/2] firmware: qcom: scm: Return -EOPNOTSUPP for unsupported SHM bridge enabling

Posted by Kuldeep Singh 1 month, 1 week ago

On 10/14/2024 6:35 PM, Bartosz Golaszewski wrote:
> On Mon, Oct 14, 2024 at 1:19 PM Kuldeep Singh <quic_kuldsing@quicinc.com> wrote:
>>
>> From: Qingqing Zhou <quic_qqzhou@quicinc.com>
>>
>> When enabling SHM bridge, QTEE returns 0 and sets error 4 in result to
>> qcom_scm for unsupported platforms. Currently, tzmem interprets this as
>> an unknown error rather than recognizing it as an unsupported platform.
>>
>> Error log:
>> [    0.177224] qcom_scm firmware:scm: error (____ptrval____): Failed to enable the TrustZone memory allocator
>> [    0.177244] qcom_scm firmware:scm: probe with driver qcom_scm failed with error 4
>>
>> To address this, modify the function call qcom_scm_shm_bridge_enable()
>> to remap result to indicate an unsupported error. This way, tzmem will
>> correctly identify it as an unsupported platform case instead of
>> reporting it as an error.
>>
>> Fixes: 178e19c0df1b ("firmware: qcom: scm: add support for SHM bridge operations")
>> Signed-off-by: Qingqing Zhou <quic_qqzhou@quicinc.com>
>> Co-developed-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
>> Signed-off-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
>> ---
>>  drivers/firmware/qcom/qcom_scm.c | 13 ++++++++++++-
>>  1 file changed, 12 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/firmware/qcom/qcom_scm.c b/drivers/firmware/qcom/qcom_scm.c
>> index 10986cb11ec0..0df81a9ed438 100644
>> --- a/drivers/firmware/qcom/qcom_scm.c
>> +++ b/drivers/firmware/qcom/qcom_scm.c
>> @@ -112,6 +112,7 @@ enum qcom_scm_qseecom_tz_cmd_info {
>>  };
>>
>>  #define QSEECOM_MAX_APP_NAME_SIZE              64
>> +#define SHMBRIDGE_RESULT_NOTSUPP               4
>>
>>  /* Each bit configures cold/warm boot address for one of the 4 CPUs */
>>  static const u8 qcom_scm_cpu_cold_bits[QCOM_SCM_BOOT_MAX_CPUS] = {
>> @@ -1361,6 +1362,8 @@ EXPORT_SYMBOL_GPL(qcom_scm_lmh_dcvsh_available);
>>
>>  int qcom_scm_shm_bridge_enable(void)
>>  {
>> +       int ret;
>> +
>>         struct qcom_scm_desc desc = {
>>                 .svc = QCOM_SCM_SVC_MP,
>>                 .cmd = QCOM_SCM_MP_SHM_BRIDGE_ENABLE,
>> @@ -1373,7 +1376,15 @@ int qcom_scm_shm_bridge_enable(void)
>>                                           QCOM_SCM_MP_SHM_BRIDGE_ENABLE))
>>                 return -EOPNOTSUPP;
>>
>> -       return qcom_scm_call(__scm->dev, &desc, &res) ?: res.result[0];
>> +       ret = qcom_scm_call(__scm->dev, &desc, &res);
>> +
>> +       if (ret)
>> +               return ret;
>> +
>> +       if (res.result[0] == SHMBRIDGE_RESULT_NOTSUPP)
>> +               return -EOPNOTSUPP;
>> +
>> +       return res.result[0];
>>  }
>>  EXPORT_SYMBOL_GPL(qcom_scm_shm_bridge_enable);
>>
>> --
>> 2.34.1
>>
>>
> 
> The patch looks correct to me and like something that should go upstream.
It's upstream mailing list only. Please see :)

> 
> Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>>
> That being said, this doesn't seem to address any of the issues that
> we saw with SHM Bridge and it still leads to a crash on sc8180x. :(

Verified on qcs615, and qcs9100 by explicitly making shmbridge
unsupported in QTEE and patch worked there.
Sc,8180x is something different it seems as there's no scm driver probe
failure and instead screen stuck is observed.
I'd like to check this behavior and get rid off it from blacklist
platform list.

-- 
Regards
Kuldeep

Re: [PATCH v2 1/2] firmware: qcom: scm: Return -EOPNOTSUPP for unsupported SHM bridge enabling

Posted by Bartosz Golaszewski 1 month, 1 week ago

On Mon, Oct 14, 2024 at 5:48 PM Kuldeep Singh <quic_kuldsing@quicinc.com> wrote:
>
>
> On 10/14/2024 6:35 PM, Bartosz Golaszewski wrote:
> > On Mon, Oct 14, 2024 at 1:19 PM Kuldeep Singh <quic_kuldsing@quicinc.com> wrote:
> >>
> >> From: Qingqing Zhou <quic_qqzhou@quicinc.com>
> >>
> >> When enabling SHM bridge, QTEE returns 0 and sets error 4 in result to
> >> qcom_scm for unsupported platforms. Currently, tzmem interprets this as
> >> an unknown error rather than recognizing it as an unsupported platform.
> >>
> >> Error log:
> >> [    0.177224] qcom_scm firmware:scm: error (____ptrval____): Failed to enable the TrustZone memory allocator
> >> [    0.177244] qcom_scm firmware:scm: probe with driver qcom_scm failed with error 4
> >>
> >> To address this, modify the function call qcom_scm_shm_bridge_enable()
> >> to remap result to indicate an unsupported error. This way, tzmem will
> >> correctly identify it as an unsupported platform case instead of
> >> reporting it as an error.
> >>
> >> Fixes: 178e19c0df1b ("firmware: qcom: scm: add support for SHM bridge operations")
> >> Signed-off-by: Qingqing Zhou <quic_qqzhou@quicinc.com>
> >> Co-developed-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
> >> Signed-off-by: Kuldeep Singh <quic_kuldsing@quicinc.com>
> >> ---
> >>  drivers/firmware/qcom/qcom_scm.c | 13 ++++++++++++-
> >>  1 file changed, 12 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/drivers/firmware/qcom/qcom_scm.c b/drivers/firmware/qcom/qcom_scm.c
> >> index 10986cb11ec0..0df81a9ed438 100644
> >> --- a/drivers/firmware/qcom/qcom_scm.c
> >> +++ b/drivers/firmware/qcom/qcom_scm.c
> >> @@ -112,6 +112,7 @@ enum qcom_scm_qseecom_tz_cmd_info {
> >>  };
> >>
> >>  #define QSEECOM_MAX_APP_NAME_SIZE              64
> >> +#define SHMBRIDGE_RESULT_NOTSUPP               4
> >>
> >>  /* Each bit configures cold/warm boot address for one of the 4 CPUs */
> >>  static const u8 qcom_scm_cpu_cold_bits[QCOM_SCM_BOOT_MAX_CPUS] = {
> >> @@ -1361,6 +1362,8 @@ EXPORT_SYMBOL_GPL(qcom_scm_lmh_dcvsh_available);
> >>
> >>  int qcom_scm_shm_bridge_enable(void)
> >>  {
> >> +       int ret;
> >> +
> >>         struct qcom_scm_desc desc = {
> >>                 .svc = QCOM_SCM_SVC_MP,
> >>                 .cmd = QCOM_SCM_MP_SHM_BRIDGE_ENABLE,
> >> @@ -1373,7 +1376,15 @@ int qcom_scm_shm_bridge_enable(void)
> >>                                           QCOM_SCM_MP_SHM_BRIDGE_ENABLE))
> >>                 return -EOPNOTSUPP;
> >>
> >> -       return qcom_scm_call(__scm->dev, &desc, &res) ?: res.result[0];
> >> +       ret = qcom_scm_call(__scm->dev, &desc, &res);
> >> +
> >> +       if (ret)
> >> +               return ret;
> >> +
> >> +       if (res.result[0] == SHMBRIDGE_RESULT_NOTSUPP)
> >> +               return -EOPNOTSUPP;
> >> +
> >> +       return res.result[0];
> >>  }
> >>  EXPORT_SYMBOL_GPL(qcom_scm_shm_bridge_enable);
> >>
> >> --
> >> 2.34.1
> >>
> >>
> >
> > The patch looks correct to me and like something that should go upstream.
> It's upstream mailing list only. Please see :)
>

Yes I know, by saying "go upstream" I mean "merged into mainline".

> >
> > Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>>
> > That being said, this doesn't seem to address any of the issues that
> > we saw with SHM Bridge and it still leads to a crash on sc8180x. :(
>
> Verified on qcs615, and qcs9100 by explicitly making shmbridge
> unsupported in QTEE and patch worked there.
> Sc,8180x is something different it seems as there's no scm driver probe
> failure and instead screen stuck is observed.
> I'd like to check this behavior and get rid off it from blacklist
> platform list.
>

Ah then my review is only more confident then. :)

Thanks,
Bartosz