coresight: tmc-etr: Handle enable failure in tmc_read_unprepare_etr

[PATCH] coresight: tmc-etr: Handle enable failure in tmc_read_unprepare_etr

Posted by Yabin Cui 2 years, 12 months ago

It's similar to what we did in tmc_read_unprepare_etb.

Signed-off-by: Yabin Cui <yabinc@google.com>
---
 drivers/hwtracing/coresight/coresight-tmc-etr.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
index 918d461fcf4a..b04f12079efd 100644
--- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
+++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
@@ -1763,6 +1763,7 @@ int tmc_read_unprepare_etr(struct tmc_drvdata *drvdata)
 {
 	unsigned long flags;
 	struct etr_buf *sysfs_buf = NULL;
+	int rc = 0;
 
 	/* config types are set a boot time and never change */
 	if (WARN_ON_ONCE(drvdata->config_type != TMC_CONFIG_TYPE_ETR))
@@ -1777,7 +1778,11 @@ int tmc_read_unprepare_etr(struct tmc_drvdata *drvdata)
 		 * buffer. Since the tracer is still enabled drvdata::buf can't
 		 * be NULL.
 		 */
-		__tmc_etr_enable_hw(drvdata);
+		rc = __tmc_etr_enable_hw(drvdata);
+		if (rc) {
+			spin_unlock_irqrestore(&drvdata->spinlock, flags);
+			return rc;
+		}
 	} else {
 		/*
 		 * The ETR is not tracing and the buffer was just read.
-- 
2.39.1.581.gbfd45094c4-goog

Re: [PATCH] coresight: tmc-etr: Handle enable failure in tmc_read_unprepare_etr

Posted by Yabin Cui 2 years, 11 months ago

Ping for review?

On Fri, Feb 10, 2023 at 11:43 PM Yabin Cui <yabinc@google.com> wrote:
>
> It's similar to what we did in tmc_read_unprepare_etb.
>
> Signed-off-by: Yabin Cui <yabinc@google.com>
> ---
>  drivers/hwtracing/coresight/coresight-tmc-etr.c | 7 ++++++-
>  1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
> index 918d461fcf4a..b04f12079efd 100644
> --- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
> +++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
> @@ -1763,6 +1763,7 @@ int tmc_read_unprepare_etr(struct tmc_drvdata *drvdata)
>  {
>         unsigned long flags;
>         struct etr_buf *sysfs_buf = NULL;
> +       int rc = 0;
>
>         /* config types are set a boot time and never change */
>         if (WARN_ON_ONCE(drvdata->config_type != TMC_CONFIG_TYPE_ETR))
> @@ -1777,7 +1778,11 @@ int tmc_read_unprepare_etr(struct tmc_drvdata *drvdata)
>                  * buffer. Since the tracer is still enabled drvdata::buf can't
>                  * be NULL.
>                  */
> -               __tmc_etr_enable_hw(drvdata);
> +               rc = __tmc_etr_enable_hw(drvdata);
> +               if (rc) {
> +                       spin_unlock_irqrestore(&drvdata->spinlock, flags);
> +                       return rc;
> +               }
>         } else {
>                 /*
>                  * The ETR is not tracing and the buffer was just read.
> --
> 2.39.1.581.gbfd45094c4-goog
>

Re: [PATCH] coresight: tmc-etr: Handle enable failure in tmc_read_unprepare_etr

Posted by Suzuki K Poulose 2 years, 11 months ago

On 21/02/2023 18:38, Yabin Cui wrote:
> Ping for review?
> 
> On Fri, Feb 10, 2023 at 11:43 PM Yabin Cui <yabinc@google.com> wrote:
>>
>> It's similar to what we did in tmc_read_unprepare_etb.
>>
>> Signed-off-by: Yabin Cui <yabinc@google.com>
>> ---

Thanks Yabin for the patch, will queue this at rc1

Suzuki

>>   drivers/hwtracing/coresight/coresight-tmc-etr.c | 7 ++++++-
>>   1 file changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
>> index 918d461fcf4a..b04f12079efd 100644
>> --- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
>> +++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
>> @@ -1763,6 +1763,7 @@ int tmc_read_unprepare_etr(struct tmc_drvdata *drvdata)
>>   {
>>          unsigned long flags;
>>          struct etr_buf *sysfs_buf = NULL;
>> +       int rc = 0;
>>
>>          /* config types are set a boot time and never change */
>>          if (WARN_ON_ONCE(drvdata->config_type != TMC_CONFIG_TYPE_ETR))
>> @@ -1777,7 +1778,11 @@ int tmc_read_unprepare_etr(struct tmc_drvdata *drvdata)
>>                   * buffer. Since the tracer is still enabled drvdata::buf can't
>>                   * be NULL.
>>                   */
>> -               __tmc_etr_enable_hw(drvdata);
>> +               rc = __tmc_etr_enable_hw(drvdata);
>> +               if (rc) {
>> +                       spin_unlock_irqrestore(&drvdata->spinlock, flags);
>> +                       return rc;
>> +               }
>>          } else {
>>                  /*
>>                   * The ETR is not tracing and the buffer was just read.
>> --
>> 2.39.1.581.gbfd45094c4-goog
>>