[PATCH v3 02/13] coresight: etb10: Fix handling of perf mode

On Tue, 31 Jul 2018 at 15:32, Suzuki K Poulose [off-list ref] wrote:

On 07/31/2018 06:07 PM, Mathieu Poirier wrote:

quoted

Hello,

On Thu, Jul 26, 2018 at 01:54:40PM +0100, Suzuki K Poulose wrote:

quoted

If the ETB is already enabled in sysfs mode, the ETB reports
success even if a perf mode is requested. Fix this by checking
the requested mode.

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
  drivers/hwtracing/coresight/coresight-etb10.c | 8 ++++++--
  1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/hwtracing/coresight/coresight-etb10.c b/drivers/hwtracing/coresight/coresight-etb10.c
index 306119e..e5808fd 100644
--- a/drivers/hwtracing/coresight/coresight-etb10.c
+++ b/drivers/hwtracing/coresight/coresight-etb10.c

@@ -148,8 +148,12 @@ static int etb_enable(struct coresight_device *csdev, u32 mode)
             return -EBUSY;

     /* Nothing to do, the tracer is already enabled. */
-    if (val == CS_MODE_SYSFS)
-            goto out;
+    if (val == CS_MODE_SYSFS) {
+            if (mode == CS_MODE_SYSFS)
+                    goto out;
+            else
+                    return -EBUSY;
+    }

How about:

         if (val == CS_MODE_PERF)
                 return -EBUSY;

+       /* Don't let perf disturb sysFS sessions */
+       if (val == CS_MODE_SYSFS && mode = CS_MODE_PERF)
+               return -EBUSY;
+
         /* Nothing to do, the tracer is already enabled. */
         if (val == CS_MODE_SYSFS)
                 goto out;


I just find it more readable that way.

Sure, thats fine by me.

Ok good.  Doing another round just to address this is cumbersome so I
just fixed it in my tree.  Also, I still think the WARN_ON() in patch
5/13 is superfluous but once again not important enough to make a big
deal out of it.  I have applied this set.

Thanks,
Mathieu

Cheers
Suzuki