[PATCH 0/6] coresight: next v4.3-rc4

[PATCH 0/6] coresight: next v4.3-rc4

[Mathieu Poirier]

Good morning,

Please consider the following patches for merging in the 4.4 cycle.

Thanks,
Mathieu

Chunyan Zhang (1):
  Coresight: ETMv4: Prevent TRCRSCTLR0&1 from being accessed

Mathieu Poirier (5):
  coresight: fixing typographical error
  coresight: adding comments to remove ambiguity
  coresight: etm3x: adding cpu affinity to sysFS interface
  coresight: etm3x: breaking down sysFS status interface
  coresight: etm3x: making error message unambiguous

 .../ABI/testing/sysfs-bus-coresight-devices-etm3x  |  77 +++++++++++++--
 drivers/hwtracing/coresight/coresight-etm3x.c      | 107 ++++++++++++---------
 drivers/hwtracing/coresight/coresight-etm4x.c      |  22 +++--
 drivers/hwtracing/coresight/coresight.c            |   5 +
 include/linux/coresight.h                          |   2 +-
 5 files changed, 154 insertions(+), 59 deletions(-)

-- 
1.9.1

[PATCH 1/6] Coresight: ETMv4: Prevent TRCRSCTLR0&1 from being accessed

[Mathieu Poirier]

From: Chunyan Zhang <zhang.chunyan@linaro.org>

1. TRCRSCTLRn - Resource Selection Control Registers n=0~1 are reserved,
   we shouldn't access them.
2. The max number of 'n' here is defined in TRCIDR4.NUMRSPAIR whoes value
   indicates the number of resource selection *pairs*, and 0 indicates
   one resource selection pair, 1 indicates two pairs, and so on ...

So, the total number of resource selection control registers which we can
access is (TRCIDR4.NUMRSPAIR * 2)

Signed-off-by: Chunyan Zhang <zhang.chunyan@linaro.org>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
 drivers/hwtracing/coresight/coresight-etm4x.c | 22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/hwtracing/coresight/coresight-etm4x.c b/drivers/hwtracing/coresight/coresight-etm4x.c
index 254a81a4e6f4..a6707642bb23 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x.c
+++ b/drivers/hwtracing/coresight/coresight-etm4x.c
@@ -136,7 +136,9 @@ static void etm4_enable_hw(void *info)
 		writel_relaxed(drvdata->cntr_val[i],
 			       drvdata->base + TRCCNTVRn(i));
 	}
-	for (i = 0; i < drvdata->nr_resource; i++)
+
+	/* Resource selector pair 0 is always implemented and reserved */
+	for (i = 2; i < drvdata->nr_resource * 2; i++)
 		writel_relaxed(drvdata->res_ctrl[i],
 			       drvdata->base + TRCRSCTLRn(i));
 
@@ -489,8 +491,9 @@ static ssize_t reset_store(struct device *dev,
 		drvdata->cntr_val[i] = 0x0;
 	}
 
-	drvdata->res_idx = 0x0;
-	for (i = 0; i < drvdata->nr_resource; i++)
+	/* Resource selector pair 0 is always implemented and reserved */
+	drvdata->res_idx = 0x2;
+	for (i = 2; i < drvdata->nr_resource * 2; i++)
 		drvdata->res_ctrl[i] = 0x0;
 
 	for (i = 0; i < drvdata->nr_ss_cmp; i++) {
@@ -1732,7 +1735,7 @@ static ssize_t res_idx_store(struct device *dev,
 	if (kstrtoul(buf, 16, &val))
 		return -EINVAL;
 	/* Resource selector pair 0 is always implemented and reserved */
-	if ((val == 0) || (val >= drvdata->nr_resource))
+	if (val < 2 || val >= drvdata->nr_resource * 2)
 		return -EINVAL;
 
 	/*
@@ -2416,8 +2419,13 @@ static void etm4_init_arch_data(void *info)
 	drvdata->nr_addr_cmp = BMVAL(etmidr4, 0, 3);
 	/* NUMPC, bits[15:12] number of PE comparator inputs for tracing */
 	drvdata->nr_pe_cmp = BMVAL(etmidr4, 12, 15);
-	/* NUMRSPAIR, bits[19:16] the number of resource pairs for tracing */
-	drvdata->nr_resource = BMVAL(etmidr4, 16, 19);
+	/*
+	 * NUMRSPAIR, bits[19:16]
+	 * The number of resource pairs conveyed by the HW starts at 0, i.e a
+	 * value of 0x0 indicate 1 resource pair, 0x1 indicate two and so on.
+	 * As such add 1 to the value of NUMRSPAIR for a better representation.
+	 */
+	drvdata->nr_resource = BMVAL(etmidr4, 16, 19) + 1;
 	/*
 	 * NUMSSCC, bits[23:20] the number of single-shot
 	 * comparator control for tracing
@@ -2504,6 +2512,8 @@ static void etm4_init_default_data(struct etmv4_drvdata *drvdata)
 		drvdata->cntr_val[i] = 0x0;
 	}
 
+	/* Resource selector pair 0 is always implemented and reserved */
+	drvdata->res_idx = 0x2;
 	for (i = 2; i < drvdata->nr_resource * 2; i++)
 		drvdata->res_ctrl[i] = 0x0;
 
-- 
1.9.1

[PATCH 2/6] coresight: fixing typographical error

[Mathieu Poirier]

Tracing gets enabled _for_ a source rather than _from_ a source.

Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
 include/linux/coresight.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/coresight.h b/include/linux/coresight.h
index c69e1b932809..a7cabfa23b55 100644
--- a/include/linux/coresight.h
+++ b/include/linux/coresight.h
@@ -207,7 +207,7 @@ struct coresight_ops_link {
  * Operations available for sources.
  * @trace_id:	returns the value of the component's trace ID as known
 		to the HW.
- * @enable:	enables tracing from a source.
+ * @enable:	enables tracing for a source.
  * @disable:	disables tracing for a source.
  */
 struct coresight_ops_source {
-- 
1.9.1

[PATCH 3/6] coresight: adding comments to remove ambiguity

[Mathieu Poirier]

Add comment to function coresight_enable_path() to make
sure there is no misunderstanding about what the code does.

Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
 drivers/hwtracing/coresight/coresight.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/hwtracing/coresight/coresight.c b/drivers/hwtracing/coresight/coresight.c
index 894531d315b8..e25492137d8b 100644
--- a/drivers/hwtracing/coresight/coresight.c
+++ b/drivers/hwtracing/coresight/coresight.c
@@ -240,6 +240,11 @@ static int coresight_enable_path(struct list_head *path)
 	int ret = 0;
 	struct coresight_device *cd;
 
+	/*
+	 * At this point we have a full @path, from source to sink.  The
+	 * sink is the first entry and the source the last one.  Go through
+	 * all the components and enable them one by one.
+	 */
 	list_for_each_entry(cd, path, path_link) {
 		if (cd == list_first_entry(path, struct coresight_device,
 					   path_link)) {
-- 
1.9.1

[PATCH 4/6] coresight: etm3x: adding cpu affinity to sysFS interface

[Mathieu Poirier]

Without access to the device tree, it is impossible to know
what CPU a tracer is affined to.  As such adding a new sysFS
interface to convey the information.

Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
 drivers/hwtracing/coresight/coresight-etm3x.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index bf2476ed9356..4797f56e7f13 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -1545,6 +1545,18 @@ static ssize_t status_show(struct device *dev,
 }
 static DEVICE_ATTR_RO(status);
 
+static ssize_t cpu_show(struct device *dev,
+			struct device_attribute *attr, char *buf)
+{
+	int val;
+	struct etm_drvdata *drvdata = dev_get_drvdata(dev->parent);
+
+	val = drvdata->cpu;
+	return scnprintf(buf, PAGE_SIZE, "%d\n", val);
+
+}
+static DEVICE_ATTR_RO(cpu);
+
 static ssize_t traceid_show(struct device *dev,
 			    struct device_attribute *attr, char *buf)
 {
@@ -1621,6 +1633,7 @@ static struct attribute *coresight_etm_attrs[] = {
 	&dev_attr_timestamp_event.attr,
 	&dev_attr_status.attr,
 	&dev_attr_traceid.attr,
+	&dev_attr_cpu.attr,
 	NULL,
 };
 ATTRIBUTE_GROUPS(coresight_etm);
-- 
1.9.1

[PATCH 5/6] coresight: etm3x: breaking down sysFS status interface

[Mathieu Poirier]

SysFS rules stipulate that only one value can be conveyed per
file.  As such splitting the "status" interface in individual files.

This is also useful for user space applications - that way they can
probe each file individually rather than having to parse a list of entries.

Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
 .../ABI/testing/sysfs-bus-coresight-devices-etm3x  |  77 ++++++++++++++--
 drivers/hwtracing/coresight/coresight-etm3x.c      | 100 +++++++++++----------
 2 files changed, 121 insertions(+), 56 deletions(-)

diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x
index d72ca1736ba4..67e3557bc4bf 100644
--- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x
+++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x
@@ -8,13 +8,6 @@ Description:	(RW) Enable/disable tracing on this specific trace entiry.
 		of coresight components linking the source to the sink is
 		configured and managed automatically by the coresight framework.
 
-What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/status
-Date:		November 2014
-KernelVersion:	3.19
-Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
-Description:	(R) List various control and status registers.  The specific
-		layout and content is driver specific.
-
 What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/addr_idx
 Date:		November 2014
 KernelVersion:	3.19
@@ -251,3 +244,73 @@ Date:		November 2014
 KernelVersion:	3.19
 Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
 Description: 	(RW) Define the event that controls the trigger.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmccr
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM Configuration Code register
+		(0x004).  The value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmccer
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM Configuration Code Extension
+		register (0x1e8).  The value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmscr
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM System Configuration
+		register (0x014).  The value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmidr
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM ID register (0x1e4).  The
+		value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmcr
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM Main Control register (0x000).
+		The value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmtraceidr
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM Trace ID register (0x200).
+		The value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmteevr
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM Trace Enable Event register
+		(0x020). The value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmtsscr
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM Trace Start/Stop Conrol
+		register (0x018). The value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmtecr1
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM Enable Conrol #1
+		register (0x024). The value is read directly from the HW.
+
+What:		/sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmtecr2
+Date:		September 2015
+KernelVersion:	4.4
+Contact:	Mathieu Poirier <mathieu.poirier@linaro.org>
+Description: 	(RO) Print the content of the ETM Enable Conrol #2
+		register (0x01c). The value is read directly from the HW.
diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index 4797f56e7f13..87374b3388ad 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -313,14 +313,6 @@ static void etm_enable_hw(void *info)
 	dev_dbg(drvdata->dev, "cpu: %d enable smp call done\n", drvdata->cpu);
 }
 
-static int etm_trace_id_simple(struct etm_drvdata *drvdata)
-{
-	if (!drvdata->enable)
-		return drvdata->traceid;
-
-	return (etm_readl(drvdata, ETMTRACEIDR) & ETM_TRACEID_MASK);
-}
-
 static int etm_trace_id(struct coresight_device *csdev)
 {
 	struct etm_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent);
@@ -1506,45 +1498,6 @@ static ssize_t timestamp_event_store(struct device *dev,
 }
 static DEVICE_ATTR_RW(timestamp_event);
 
-static ssize_t status_show(struct device *dev,
-			   struct device_attribute *attr, char *buf)
-{
-	int ret;
-	unsigned long flags;
-	struct etm_drvdata *drvdata = dev_get_drvdata(dev->parent);
-
-	pm_runtime_get_sync(drvdata->dev);
-	spin_lock_irqsave(&drvdata->spinlock, flags);
-
-	CS_UNLOCK(drvdata->base);
-	ret = sprintf(buf,
-		      "ETMCCR: 0x%08x\n"
-		      "ETMCCER: 0x%08x\n"
-		      "ETMSCR: 0x%08x\n"
-		      "ETMIDR: 0x%08x\n"
-		      "ETMCR: 0x%08x\n"
-		      "ETMTRACEIDR: 0x%08x\n"
-		      "Enable event: 0x%08x\n"
-		      "Enable start/stop: 0x%08x\n"
-		      "Enable control: CR1 0x%08x CR2 0x%08x\n"
-		      "CPU affinity: %d\n",
-		      drvdata->etmccr, drvdata->etmccer,
-		      etm_readl(drvdata, ETMSCR), etm_readl(drvdata, ETMIDR),
-		      etm_readl(drvdata, ETMCR), etm_trace_id_simple(drvdata),
-		      etm_readl(drvdata, ETMTEEVR),
-		      etm_readl(drvdata, ETMTSSCR),
-		      etm_readl(drvdata, ETMTECR1),
-		      etm_readl(drvdata, ETMTECR2),
-		      drvdata->cpu);
-	CS_LOCK(drvdata->base);
-
-	spin_unlock_irqrestore(&drvdata->spinlock, flags);
-	pm_runtime_put(drvdata->dev);
-
-	return ret;
-}
-static DEVICE_ATTR_RO(status);
-
 static ssize_t cpu_show(struct device *dev,
 			struct device_attribute *attr, char *buf)
 {
@@ -1631,12 +1584,61 @@ static struct attribute *coresight_etm_attrs[] = {
 	&dev_attr_ctxid_mask.attr,
 	&dev_attr_sync_freq.attr,
 	&dev_attr_timestamp_event.attr,
-	&dev_attr_status.attr,
 	&dev_attr_traceid.attr,
 	&dev_attr_cpu.attr,
 	NULL,
 };
-ATTRIBUTE_GROUPS(coresight_etm);
+
+#define coresight_simple_func(name, offset)                             \
+static ssize_t name##_show(struct device *_dev,                         \
+			   struct device_attribute *attr, char *buf)    \
+{                                                                       \
+	struct etm_drvdata *drvdata = dev_get_drvdata(_dev->parent);    \
+	return scnprintf(buf, PAGE_SIZE, "0x%x\n",                      \
+			 readl_relaxed(drvdata->base + offset));        \
+}                                                                       \
+DEVICE_ATTR_RO(name)
+
+coresight_simple_func(etmccr, ETMCCR);
+coresight_simple_func(etmccer, ETMCCER);
+coresight_simple_func(etmscr, ETMSCR);
+coresight_simple_func(etmidr, ETMIDR);
+coresight_simple_func(etmcr, ETMCR);
+coresight_simple_func(etmtraceidr, ETMTRACEIDR);
+coresight_simple_func(etmteevr, ETMTEEVR);
+coresight_simple_func(etmtssvr, ETMTSSCR);
+coresight_simple_func(etmtecr1, ETMTECR1);
+coresight_simple_func(etmtecr2, ETMTECR2);
+
+static struct attribute *coresight_etm_mgmt_attrs[] = {
+	&dev_attr_etmccr.attr,
+	&dev_attr_etmccer.attr,
+	&dev_attr_etmscr.attr,
+	&dev_attr_etmidr.attr,
+	&dev_attr_etmcr.attr,
+	&dev_attr_etmtraceidr.attr,
+	&dev_attr_etmteevr.attr,
+	&dev_attr_etmtssvr.attr,
+	&dev_attr_etmtecr1.attr,
+	&dev_attr_etmtecr2.attr,
+	NULL,
+};
+
+static const struct attribute_group coresight_etm_group = {
+	.attrs = coresight_etm_attrs,
+};
+
+
+static const struct attribute_group coresight_etm_mgmt_group = {
+	.attrs = coresight_etm_mgmt_attrs,
+	.name = "mgmt",
+};
+
+static const struct attribute_group *coresight_etm_groups[] = {
+	&coresight_etm_group,
+	&coresight_etm_mgmt_group,
+	NULL,
+};
 
 static int etm_cpu_callback(struct notifier_block *nfb, unsigned long action,
 			    void *hcpu)
-- 
1.9.1

[PATCH 6/6] coresight: etm3x: making error message unambiguous

[Mathieu Poirier]

By adding the function name at the beginning of the error
message there is no doubt as to where the failing condition
occurred.

Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
 drivers/hwtracing/coresight/coresight-etm3x.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index 87374b3388ad..d630b7ece735 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -191,7 +191,8 @@ static void etm_set_prog(struct etm_drvdata *drvdata)
 	isb();
 	if (coresight_timeout_etm(drvdata, ETMSR, ETMSR_PROG_BIT, 1)) {
 		dev_err(drvdata->dev,
-			"timeout observed when probing at offset %#x\n", ETMSR);
+			"%s: timeout observed when probing at offset %#x\n",
+			__func__, ETMSR);
 	}
 }
 
@@ -209,7 +210,8 @@ static void etm_clr_prog(struct etm_drvdata *drvdata)
 	isb();
 	if (coresight_timeout_etm(drvdata, ETMSR, ETMSR_PROG_BIT, 0)) {
 		dev_err(drvdata->dev,
-			"timeout observed when probing at offset %#x\n", ETMSR);
+			"%s: timeout observed when probing at offset %#x\n",
+			__func__, ETMSR);
 	}
 }
 
-- 
1.9.1

[PATCH 4/6] coresight: etm3x: adding cpu affinity to sysFS interface

[Greg KH]

On Wed, Oct 07, 2015 at 09:26:41AM -0600, Mathieu Poirier wrote:
> Without access to the device tree, it is impossible to know
> what CPU a tracer is affined to.  As such adding a new sysFS
> interface to convey the information.
> 
> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>


No documentation for this new attribute?

[PATCH 5/6] coresight: etm3x: breaking down sysFS status interface

[Greg KH]

On Wed, Oct 07, 2015 at 09:26:42AM -0600, Mathieu Poirier wrote:
> SysFS rules stipulate that only one value can be conveyed per
> file.  As such splitting the "status" interface in individual files.
> 
> This is also useful for user space applications - that way they can
> probe each file individually rather than having to parse a list of entries.

Ah, much nicer, thanks for doing this.

greg k-h