On 22/03/16 20:23, Mathieu Poirier wrote:
Moving tmc_drvdata::enable to a local_t mode. That way the
sink interface is aware of it's orgin and the foundation for
mutual exclusion between the sysFS and Perf interface can be
laid out.
Signed-off-by: Mathieu Poirier <mathieu.poir...@linaro.org>
static int tmc_enable_etf_sink(struct coresight_device *csdev, u32 mode)
{
+ u32 val;
bool allocated = false;
char *buf = NULL;
unsigned long flags;
@@ -125,6 +126,15 @@ static int tmc_enable_etf_sink(struct coresight_device
*csdev, u32 mode)
return -EBUSY;
}
+ val = local_cmpxchg(&drvdata->mode, CS_MODE_DISABLED, mode);
Since we don't support PERF mode yet, should we check that before setting the
mode ?
static int tmc_enable_etr_sink(struct coresight_device *csdev, u32 mode)
{
+ u32 val;
bool allocated = false;
unsigned long flags;
void __iomem *vaddr;
@@ -107,6 +108,15 @@ static int tmc_enable_etr_sink(struct coresight_device
*csdev, u32 mode)
return -EBUSY;
}
+ val = local_cmpxchg(&drvdata->mode, CS_MODE_DISABLED, mode);
+ /*
+ * In sysFS mode we can have multiple writers per sink. Since this
+ * sink is already enabled no memory is needed and the HW need not be
+ * touched.
+ */
+ if (val == CS_MODE_SYSFS)
+ goto out;
+
Should we make sure 'mode' is not PERF ? Since we assume that the mode is
could be SYSFS in disable below ?
diff --git a/drivers/hwtracing/coresight/coresight-tmc.h
b/drivers/hwtracing/coresight/coresight-tmc.h
index 6b11caf77ad1..6dbd70861b17 100644
--- a/drivers/hwtracing/coresight/coresight-tmc.h
+++ b/drivers/hwtracing/coresight/coresight-tmc.h
@@ -18,6 +18,7 @@
#ifndef _CORESIGHT_TMC_H
#define _CORESIGHT_TMC_H
+#include <linux/io.h>
Why ?
Suzuki
