* [PATCH 01/28] coresight: Rename of_coresight to coresight-platform
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:52 ` [PATCH 02/28] coresight: platform: Make memory allocation helper generic Mathieu Poirier
` (27 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
Rename the firmware handling file to a more generic
name, in preparation for adding ACPI support. Right now
we only support DT and we have all the platform handling
code in of_coresight.c. Let us rename the file to
coresight-platform.c in order to keep the platform handling
in a single place for DT and the upcoming ACPI support.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/Makefile | 3 +--
.../coresight/{of_coresight.c => coresight-platform.c} | 3 ++-
2 files changed, 3 insertions(+), 3 deletions(-)
rename drivers/hwtracing/coresight/{of_coresight.c => coresight-platform.c} (99%)
diff --git a/drivers/hwtracing/coresight/Makefile b/drivers/hwtracing/coresight/Makefile
index 3b435aa42af5..3c0ac421e211 100644
--- a/drivers/hwtracing/coresight/Makefile
+++ b/drivers/hwtracing/coresight/Makefile
@@ -2,8 +2,7 @@
#
# Makefile for CoreSight drivers.
#
-obj-$(CONFIG_CORESIGHT) += coresight.o coresight-etm-perf.o
-obj-$(CONFIG_OF) += of_coresight.o
+obj-$(CONFIG_CORESIGHT) += coresight.o coresight-etm-perf.o coresight-platform.o
obj-$(CONFIG_CORESIGHT_LINK_AND_SINK_TMC) += coresight-tmc.o \
coresight-tmc-etf.o \
coresight-tmc-etr.o
diff --git a/drivers/hwtracing/coresight/of_coresight.c b/drivers/hwtracing/coresight/coresight-platform.c
similarity index 99%
rename from drivers/hwtracing/coresight/of_coresight.c
rename to drivers/hwtracing/coresight/coresight-platform.c
index 7045930fc958..514cc2b67c36 100644
--- a/drivers/hwtracing/coresight/of_coresight.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -17,7 +17,7 @@
#include <linux/cpumask.h>
#include <asm/smp_plat.h>
-
+#ifdef CONFIG_OF
static int of_dev_node_match(struct device *dev, void *data)
{
return dev->of_node == data;
@@ -295,3 +295,4 @@ of_get_coresight_platform_data(struct device *dev,
return pdata;
}
EXPORT_SYMBOL_GPL(of_get_coresight_platform_data);
+#endif
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 02/28] coresight: platform: Make memory allocation helper generic
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
2019-06-19 19:52 ` [PATCH 01/28] coresight: Rename of_coresight to coresight-platform Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:52 ` [PATCH 03/28] coresight: Make sure device uses DT for obsolete compatible check Mathieu Poirier
` (26 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
Rename the of_coresight_alloc_memory() => coresight_alloc_conns()
as it is independent of the underlying firmware type. This is in
preparation for the ACPI support.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
.../hwtracing/coresight/coresight-platform.c | 34 +++++++++++--------
1 file changed, 19 insertions(+), 15 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index 514cc2b67c36..4c31299607cf 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -17,6 +17,24 @@
#include <linux/cpumask.h>
#include <asm/smp_plat.h>
+/*
+ * coresight_alloc_conns: Allocate connections record for each output
+ * port from the device.
+ */
+static int coresight_alloc_conns(struct device *dev,
+ struct coresight_platform_data *pdata)
+{
+ if (pdata->nr_outport) {
+ pdata->conns = devm_kzalloc(dev, pdata->nr_outport *
+ sizeof(*pdata->conns),
+ GFP_KERNEL);
+ if (!pdata->conns)
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
#ifdef CONFIG_OF
static int of_dev_node_match(struct device *dev, void *data)
{
@@ -133,20 +151,6 @@ static void of_coresight_get_ports(const struct device_node *node,
}
}
-static int of_coresight_alloc_memory(struct device *dev,
- struct coresight_platform_data *pdata)
-{
- if (pdata->nr_outport) {
- pdata->conns = devm_kzalloc(dev, pdata->nr_outport *
- sizeof(*pdata->conns),
- GFP_KERNEL);
- if (!pdata->conns)
- return -ENOMEM;
- }
-
- return 0;
-}
-
int of_coresight_get_cpu(const struct device_node *node)
{
int cpu;
@@ -252,7 +256,7 @@ of_get_coresight_platform_data(struct device *dev,
if (!pdata->nr_outport)
return pdata;
- ret = of_coresight_alloc_memory(dev, pdata);
+ ret = coresight_alloc_conns(dev, pdata);
if (ret)
return ERR_PTR(ret);
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 03/28] coresight: Make sure device uses DT for obsolete compatible check
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
2019-06-19 19:52 ` [PATCH 01/28] coresight: Rename of_coresight to coresight-platform Mathieu Poirier
2019-06-19 19:52 ` [PATCH 02/28] coresight: platform: Make memory allocation helper generic Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:52 ` [PATCH 04/28] coresight: Introduce generic platform data helper Mathieu Poirier
` (25 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
As we prepare to add support for ACPI bindings, let us make sure we do
the compatible check only if we are sure we are dealing with a DT based
system.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-funnel.c | 3 ++-
drivers/hwtracing/coresight/coresight-replicator.c | 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-funnel.c b/drivers/hwtracing/coresight/coresight-funnel.c
index 6236a847bb0b..3423042e7a52 100644
--- a/drivers/hwtracing/coresight/coresight-funnel.c
+++ b/drivers/hwtracing/coresight/coresight-funnel.c
@@ -196,7 +196,8 @@ static int funnel_probe(struct device *dev, struct resource *res)
dev->platform_data = pdata;
}
- if (of_device_is_compatible(np, "arm,coresight-funnel"))
+ if (is_of_node(dev_fwnode(dev)) &&
+ of_device_is_compatible(dev->of_node, "arm,coresight-funnel"))
pr_warn_once("Uses OBSOLETE CoreSight funnel binding\n");
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
diff --git a/drivers/hwtracing/coresight/coresight-replicator.c b/drivers/hwtracing/coresight/coresight-replicator.c
index ee6ad34061a5..7e0514557920 100644
--- a/drivers/hwtracing/coresight/coresight-replicator.c
+++ b/drivers/hwtracing/coresight/coresight-replicator.c
@@ -187,7 +187,8 @@ static int replicator_probe(struct device *dev, struct resource *res)
dev->platform_data = pdata;
}
- if (of_device_is_compatible(np, "arm,coresight-replicator"))
+ if (is_of_node(dev_fwnode(dev)) &&
+ of_device_is_compatible(dev->of_node, "arm,coresight-replicator"))
pr_warn_once("Uses OBSOLETE CoreSight replicator binding\n");
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 04/28] coresight: Introduce generic platform data helper
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (2 preceding siblings ...)
2019-06-19 19:52 ` [PATCH 03/28] coresight: Make sure device uses DT for obsolete compatible check Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:52 ` [PATCH 05/28] coresight: Make device to CPU mapping generic Mathieu Poirier
` (24 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
So far we have hard coded the DT platform parsing code in
every driver. Introduce generic helper to parse the information
provided by the firmware in a platform agnostic manner, in preparation
for the ACPI support.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-catu.c | 13 ++---
drivers/hwtracing/coresight/coresight-etb10.c | 11 ++--
drivers/hwtracing/coresight/coresight-etm3x.c | 12 ++--
drivers/hwtracing/coresight/coresight-etm4x.c | 11 ++--
.../hwtracing/coresight/coresight-funnel.c | 11 ++--
.../hwtracing/coresight/coresight-platform.c | 58 ++++++++++++++-----
.../coresight/coresight-replicator.c | 11 ++--
drivers/hwtracing/coresight/coresight-stm.c | 11 ++--
drivers/hwtracing/coresight/coresight-tmc.c | 13 ++---
drivers/hwtracing/coresight/coresight-tpiu.c | 11 ++--
include/linux/coresight.h | 7 +--
11 files changed, 83 insertions(+), 86 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-catu.c b/drivers/hwtracing/coresight/coresight-catu.c
index 63109c98765c..799ba1dd877e 100644
--- a/drivers/hwtracing/coresight/coresight-catu.c
+++ b/drivers/hwtracing/coresight/coresight-catu.c
@@ -503,17 +503,14 @@ static int catu_probe(struct amba_device *adev, const struct amba_id *id)
struct coresight_desc catu_desc;
struct coresight_platform_data *pdata = NULL;
struct device *dev = &adev->dev;
- struct device_node *np = dev->of_node;
void __iomem *base;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata)) {
- ret = PTR_ERR(pdata);
- goto out;
- }
- dev->platform_data = pdata;
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto out;
}
+ dev->platform_data = pdata;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata) {
diff --git a/drivers/hwtracing/coresight/coresight-etb10.c b/drivers/hwtracing/coresight/coresight-etb10.c
index 0c9161fe4233..cb8cb03e0cdb 100644
--- a/drivers/hwtracing/coresight/coresight-etb10.c
+++ b/drivers/hwtracing/coresight/coresight-etb10.c
@@ -732,14 +732,11 @@ static int etb_probe(struct amba_device *adev, const struct amba_id *id)
struct etb_drvdata *drvdata;
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
- struct device_node *np = adev->dev.of_node;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- adev->dev.platform_data = pdata;
- }
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ adev->dev.platform_data = pdata;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index fa2f1417cafb..fa2164ff07c2 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -790,20 +790,16 @@ static int etm_probe(struct amba_device *adev, const struct amba_id *id)
struct etm_drvdata *drvdata;
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
- struct device_node *np = adev->dev.of_node;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
-
- adev->dev.platform_data = pdata;
- }
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ adev->dev.platform_data = pdata;
drvdata->use_cp14 = fwnode_property_read_bool(dev->fwnode, "arm,cp14");
dev_set_drvdata(dev, drvdata);
diff --git a/drivers/hwtracing/coresight/coresight-etm4x.c b/drivers/hwtracing/coresight/coresight-etm4x.c
index 77d1d837da52..4355b2e8c308 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x.c
+++ b/drivers/hwtracing/coresight/coresight-etm4x.c
@@ -1084,18 +1084,15 @@ static int etm4_probe(struct amba_device *adev, const struct amba_id *id)
struct etmv4_drvdata *drvdata;
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
- struct device_node *np = adev->dev.of_node;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- adev->dev.platform_data = pdata;
- }
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ adev->dev.platform_data = pdata;
dev_set_drvdata(dev, drvdata);
diff --git a/drivers/hwtracing/coresight/coresight-funnel.c b/drivers/hwtracing/coresight/coresight-funnel.c
index 3423042e7a52..fc033fdb6cd5 100644
--- a/drivers/hwtracing/coresight/coresight-funnel.c
+++ b/drivers/hwtracing/coresight/coresight-funnel.c
@@ -187,14 +187,11 @@ static int funnel_probe(struct device *dev, struct resource *res)
struct coresight_platform_data *pdata = NULL;
struct funnel_drvdata *drvdata;
struct coresight_desc desc = { 0 };
- struct device_node *np = dev->of_node;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- dev->platform_data = pdata;
- }
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ dev->platform_data = pdata;
if (is_of_node(dev_fwnode(dev)) &&
of_device_is_compatible(dev->of_node, "arm,coresight-funnel"))
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index 4c31299607cf..5d78f4fbd97d 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -230,23 +230,16 @@ static int of_coresight_parse_endpoint(struct device *dev,
return ret;
}
-struct coresight_platform_data *
-of_get_coresight_platform_data(struct device *dev,
- const struct device_node *node)
+static int of_get_coresight_platform_data(struct device *dev,
+ struct coresight_platform_data *pdata)
{
int ret = 0;
- struct coresight_platform_data *pdata;
struct coresight_connection *conn;
struct device_node *ep = NULL;
const struct device_node *parent = NULL;
bool legacy_binding = false;
+ struct device_node *node = dev->of_node;
- pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL);
- if (!pdata)
- return ERR_PTR(-ENOMEM);
-
- /* Use device name as sysfs handle */
- pdata->name = dev_name(dev);
pdata->cpu = of_coresight_get_cpu(node);
/* Get the number of input and output port for this component */
@@ -254,11 +247,11 @@ of_get_coresight_platform_data(struct device *dev,
/* If there are no output connections, we are done */
if (!pdata->nr_outport)
- return pdata;
+ return 0;
ret = coresight_alloc_conns(dev, pdata);
if (ret)
- return ERR_PTR(ret);
+ return ret;
parent = of_coresight_get_output_ports_node(node);
/*
@@ -292,11 +285,46 @@ of_get_coresight_platform_data(struct device *dev,
case 0:
break;
default:
- return ERR_PTR(ret);
+ return ret;
}
}
- return pdata;
+ return 0;
+}
+#else
+static inline int
+of_get_coresight_platform_data(struct device *dev,
+ struct coresight_platform_data *pdata)
+{
+ return -ENOENT;
}
-EXPORT_SYMBOL_GPL(of_get_coresight_platform_data);
#endif
+
+struct coresight_platform_data *
+coresight_get_platform_data(struct device *dev)
+{
+ int ret = -ENOENT;
+ struct coresight_platform_data *pdata;
+ struct fwnode_handle *fwnode = dev_fwnode(dev);
+
+ if (IS_ERR_OR_NULL(fwnode))
+ goto error;
+
+ pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL);
+ if (!pdata) {
+ ret = -ENOMEM;
+ goto error;
+ }
+
+ /* Use device name as sysfs handle */
+ pdata->name = dev_name(dev);
+
+ if (is_of_node(fwnode))
+ ret = of_get_coresight_platform_data(dev, pdata);
+
+ if (!ret)
+ return pdata;
+error:
+ return ERR_PTR(ret);
+}
+EXPORT_SYMBOL_GPL(coresight_get_platform_data);
diff --git a/drivers/hwtracing/coresight/coresight-replicator.c b/drivers/hwtracing/coresight/coresight-replicator.c
index 7e0514557920..054b33554695 100644
--- a/drivers/hwtracing/coresight/coresight-replicator.c
+++ b/drivers/hwtracing/coresight/coresight-replicator.c
@@ -177,15 +177,12 @@ static int replicator_probe(struct device *dev, struct resource *res)
struct coresight_platform_data *pdata = NULL;
struct replicator_drvdata *drvdata;
struct coresight_desc desc = { 0 };
- struct device_node *np = dev->of_node;
void __iomem *base;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- dev->platform_data = pdata;
- }
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ dev->platform_data = pdata;
if (is_of_node(dev_fwnode(dev)) &&
of_device_is_compatible(dev->of_node, "arm,coresight-replicator"))
diff --git a/drivers/hwtracing/coresight/coresight-stm.c b/drivers/hwtracing/coresight/coresight-stm.c
index 3992a35e34e4..9faa1ed01500 100644
--- a/drivers/hwtracing/coresight/coresight-stm.c
+++ b/drivers/hwtracing/coresight/coresight-stm.c
@@ -809,14 +809,11 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id)
struct resource ch_res;
size_t bitmap_size;
struct coresight_desc desc = { 0 };
- struct device_node *np = adev->dev.of_node;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- adev->dev.platform_data = pdata;
- }
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ adev->dev.platform_data = pdata;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
diff --git a/drivers/hwtracing/coresight/coresight-tmc.c b/drivers/hwtracing/coresight/coresight-tmc.c
index 9c5e615c1486..be0bd98ca8c1 100644
--- a/drivers/hwtracing/coresight/coresight-tmc.c
+++ b/drivers/hwtracing/coresight/coresight-tmc.c
@@ -397,16 +397,13 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
struct tmc_drvdata *drvdata;
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
- struct device_node *np = adev->dev.of_node;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata)) {
- ret = PTR_ERR(pdata);
- goto out;
- }
- adev->dev.platform_data = pdata;
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto out;
}
+ adev->dev.platform_data = pdata;
ret = -ENOMEM;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
diff --git a/drivers/hwtracing/coresight/coresight-tpiu.c b/drivers/hwtracing/coresight/coresight-tpiu.c
index 4dd3e7f63050..aec0ed7bf924 100644
--- a/drivers/hwtracing/coresight/coresight-tpiu.c
+++ b/drivers/hwtracing/coresight/coresight-tpiu.c
@@ -124,14 +124,11 @@ static int tpiu_probe(struct amba_device *adev, const struct amba_id *id)
struct tpiu_drvdata *drvdata;
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
- struct device_node *np = adev->dev.of_node;
- if (np) {
- pdata = of_get_coresight_platform_data(dev, np);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- adev->dev.platform_data = pdata;
- }
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ dev->platform_data = pdata;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
diff --git a/include/linux/coresight.h b/include/linux/coresight.h
index 62a520df8add..e2b95e05e0be 100644
--- a/include/linux/coresight.h
+++ b/include/linux/coresight.h
@@ -294,14 +294,11 @@ static inline void coresight_disclaim_device_unlocked(void __iomem *base) {}
#ifdef CONFIG_OF
extern int of_coresight_get_cpu(const struct device_node *node);
-extern struct coresight_platform_data *
-of_get_coresight_platform_data(struct device *dev,
- const struct device_node *node);
#else
static inline int of_coresight_get_cpu(const struct device_node *node)
{ return 0; }
-static inline struct coresight_platform_data *of_get_coresight_platform_data(
- struct device *dev, const struct device_node *node) { return NULL; }
#endif
+struct coresight_platform_data *coresight_get_platform_data(struct device *dev);
+
#endif
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 05/28] coresight: Make device to CPU mapping generic
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (3 preceding siblings ...)
2019-06-19 19:52 ` [PATCH 04/28] coresight: Introduce generic platform data helper Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:52 ` [PATCH 06/28] coresight: Remove cpu field from platform data Mathieu Poirier
` (23 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
The CoreSight components ETM and CPU-Debug are always associated
with CPUs. Replace the of_coresight_get_cpu() with a platform
agnostic helper, in preparation to add ACPI support.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
.../hwtracing/coresight/coresight-cpu-debug.c | 3 +--
.../hwtracing/coresight/coresight-platform.c | 18 +++++++++++++-----
include/linux/coresight.h | 7 +------
3 files changed, 15 insertions(+), 13 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-cpu-debug.c b/drivers/hwtracing/coresight/coresight-cpu-debug.c
index e8819d750938..07a1367c733f 100644
--- a/drivers/hwtracing/coresight/coresight-cpu-debug.c
+++ b/drivers/hwtracing/coresight/coresight-cpu-debug.c
@@ -572,14 +572,13 @@ static int debug_probe(struct amba_device *adev, const struct amba_id *id)
struct device *dev = &adev->dev;
struct debug_drvdata *drvdata;
struct resource *res = &adev->res;
- struct device_node *np = adev->dev.of_node;
int ret;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
- drvdata->cpu = np ? of_coresight_get_cpu(np) : 0;
+ drvdata->cpu = coresight_get_cpu(dev);
if (per_cpu(debug_drvdata, drvdata->cpu)) {
dev_err(dev, "CPU%d drvdata has already been initialized\n",
drvdata->cpu);
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index 5d78f4fbd97d..ba8c14635c6b 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -151,12 +151,14 @@ static void of_coresight_get_ports(const struct device_node *node,
}
}
-int of_coresight_get_cpu(const struct device_node *node)
+static int of_coresight_get_cpu(struct device *dev)
{
int cpu;
struct device_node *dn;
- dn = of_parse_phandle(node, "cpu", 0);
+ if (!dev->of_node)
+ return 0;
+ dn = of_parse_phandle(dev->of_node, "cpu", 0);
/* Affinity defaults to CPU0 */
if (!dn)
return 0;
@@ -166,7 +168,6 @@ int of_coresight_get_cpu(const struct device_node *node)
/* Affinity to CPU0 if no cpu nodes are found */
return (cpu < 0) ? 0 : cpu;
}
-EXPORT_SYMBOL_GPL(of_coresight_get_cpu);
/*
* of_coresight_parse_endpoint : Parse the given output endpoint @ep
@@ -240,8 +241,6 @@ static int of_get_coresight_platform_data(struct device *dev,
bool legacy_binding = false;
struct device_node *node = dev->of_node;
- pdata->cpu = of_coresight_get_cpu(node);
-
/* Get the number of input and output port for this component */
of_coresight_get_ports(node, &pdata->nr_inport, &pdata->nr_outport);
@@ -300,6 +299,14 @@ of_get_coresight_platform_data(struct device *dev,
}
#endif
+int coresight_get_cpu(struct device *dev)
+{
+ if (is_of_node(dev->fwnode))
+ return of_coresight_get_cpu(dev);
+ return 0;
+}
+EXPORT_SYMBOL_GPL(coresight_get_cpu);
+
struct coresight_platform_data *
coresight_get_platform_data(struct device *dev)
{
@@ -318,6 +325,7 @@ coresight_get_platform_data(struct device *dev)
/* Use device name as sysfs handle */
pdata->name = dev_name(dev);
+ pdata->cpu = coresight_get_cpu(dev);
if (is_of_node(fwnode))
ret = of_get_coresight_platform_data(dev, pdata);
diff --git a/include/linux/coresight.h b/include/linux/coresight.h
index e2b95e05e0be..98a4440dea3e 100644
--- a/include/linux/coresight.h
+++ b/include/linux/coresight.h
@@ -292,12 +292,7 @@ static inline void coresight_disclaim_device_unlocked(void __iomem *base) {}
#endif
-#ifdef CONFIG_OF
-extern int of_coresight_get_cpu(const struct device_node *node);
-#else
-static inline int of_coresight_get_cpu(const struct device_node *node)
-{ return 0; }
-#endif
+extern int coresight_get_cpu(struct device *dev);
struct coresight_platform_data *coresight_get_platform_data(struct device *dev);
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 06/28] coresight: Remove cpu field from platform data
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (4 preceding siblings ...)
2019-06-19 19:52 ` [PATCH 05/28] coresight: Make device to CPU mapping generic Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:52 ` [PATCH 07/28] coresight: Remove name from platform description Mathieu Poirier
` (22 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
CPU field is only used by ETMs and there is a separate API
for fetching the same. So, let us use that instead of using
the common platform probing helper. Also, remove it from the
platform_data.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-etm3x.c | 2 +-
drivers/hwtracing/coresight/coresight-etm4x.c | 2 +-
drivers/hwtracing/coresight/coresight-platform.c | 1 -
include/linux/coresight.h | 2 --
4 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index fa2164ff07c2..722fab9632ef 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -819,7 +819,7 @@ static int etm_probe(struct amba_device *adev, const struct amba_id *id)
return ret;
}
- drvdata->cpu = pdata ? pdata->cpu : 0;
+ drvdata->cpu = coresight_get_cpu(dev);
cpus_read_lock();
etmdrvdata[drvdata->cpu] = drvdata;
diff --git a/drivers/hwtracing/coresight/coresight-etm4x.c b/drivers/hwtracing/coresight/coresight-etm4x.c
index 4355b2e8c308..03576f3ed22d 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x.c
+++ b/drivers/hwtracing/coresight/coresight-etm4x.c
@@ -1105,7 +1105,7 @@ static int etm4_probe(struct amba_device *adev, const struct amba_id *id)
spin_lock_init(&drvdata->spinlock);
- drvdata->cpu = pdata ? pdata->cpu : 0;
+ drvdata->cpu = coresight_get_cpu(dev);
cpus_read_lock();
etmdrvdata[drvdata->cpu] = drvdata;
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index ba8c14635c6b..541e500a83c2 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -325,7 +325,6 @@ coresight_get_platform_data(struct device *dev)
/* Use device name as sysfs handle */
pdata->name = dev_name(dev);
- pdata->cpu = coresight_get_cpu(dev);
if (is_of_node(fwnode))
ret = of_get_coresight_platform_data(dev, pdata);
diff --git a/include/linux/coresight.h b/include/linux/coresight.h
index 98a4440dea3e..bf241dbf99c5 100644
--- a/include/linux/coresight.h
+++ b/include/linux/coresight.h
@@ -91,14 +91,12 @@ union coresight_dev_subtype {
/**
* struct coresight_platform_data - data harvested from the DT specification
- * @cpu: the CPU a source belongs to. Only applicable for ETM/PTMs.
* @name: name of the component as shown under sysfs.
* @nr_inport: number of input ports for this component.
* @nr_outport: number of output ports for this component.
* @conns: Array of nr_outport connections from this component
*/
struct coresight_platform_data {
- int cpu;
const char *name;
int nr_inport;
int nr_outport;
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 07/28] coresight: Remove name from platform description
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (5 preceding siblings ...)
2019-06-19 19:52 ` [PATCH 06/28] coresight: Remove cpu field from platform data Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:52 ` [PATCH 08/28] coresight: Cleanup coresight_remove_conns Mathieu Poirier
` (21 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
We are about to use a name independent of the parent AMBA device
name. As such, there is no need to have it in the platform description.
Let us move this to coresight description instead.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-catu.c | 2 ++
drivers/hwtracing/coresight/coresight-etb10.c | 3 ++-
drivers/hwtracing/coresight/coresight-etm3x.c | 1 +
drivers/hwtracing/coresight/coresight-etm4x.c | 1 +
drivers/hwtracing/coresight/coresight-funnel.c | 1 +
drivers/hwtracing/coresight/coresight-platform.c | 3 ---
drivers/hwtracing/coresight/coresight-replicator.c | 2 ++
drivers/hwtracing/coresight/coresight-stm.c | 1 +
drivers/hwtracing/coresight/coresight-tmc.c | 5 +++--
drivers/hwtracing/coresight/coresight-tpiu.c | 1 +
drivers/hwtracing/coresight/coresight.c | 2 +-
include/linux/coresight.h | 8 ++++----
12 files changed, 19 insertions(+), 11 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-catu.c b/drivers/hwtracing/coresight/coresight-catu.c
index 799ba1dd877e..05c73045532a 100644
--- a/drivers/hwtracing/coresight/coresight-catu.c
+++ b/drivers/hwtracing/coresight/coresight-catu.c
@@ -551,6 +551,8 @@ static int catu_probe(struct amba_device *adev, const struct amba_id *id)
catu_desc.type = CORESIGHT_DEV_TYPE_HELPER;
catu_desc.subtype.helper_subtype = CORESIGHT_DEV_SUBTYPE_HELPER_CATU;
catu_desc.ops = &catu_ops;
+ catu_desc.name = dev_name(dev);
+
drvdata->csdev = coresight_register(&catu_desc);
if (IS_ERR(drvdata->csdev))
ret = PTR_ERR(drvdata->csdev);
diff --git a/drivers/hwtracing/coresight/coresight-etb10.c b/drivers/hwtracing/coresight/coresight-etb10.c
index cb8cb03e0cdb..8726d6c7663c 100644
--- a/drivers/hwtracing/coresight/coresight-etb10.c
+++ b/drivers/hwtracing/coresight/coresight-etb10.c
@@ -777,12 +777,13 @@ static int etb_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &etb_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
+ desc.name = dev_name(dev);
desc.groups = coresight_etb_groups;
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev))
return PTR_ERR(drvdata->csdev);
- drvdata->miscdev.name = pdata->name;
+ drvdata->miscdev.name = desc.name;
drvdata->miscdev.minor = MISC_DYNAMIC_MINOR;
drvdata->miscdev.fops = &etb_fops;
ret = misc_register(&drvdata->miscdev);
diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index 722fab9632ef..101fb01e20de 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -854,6 +854,7 @@ static int etm_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &etm_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
+ desc.name = dev_name(dev);
desc.groups = coresight_etm_groups;
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
diff --git a/drivers/hwtracing/coresight/coresight-etm4x.c b/drivers/hwtracing/coresight/coresight-etm4x.c
index 03576f3ed22d..8adc1485cd89 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x.c
+++ b/drivers/hwtracing/coresight/coresight-etm4x.c
@@ -1142,6 +1142,7 @@ static int etm4_probe(struct amba_device *adev, const struct amba_id *id)
desc.pdata = pdata;
desc.dev = dev;
desc.groups = coresight_etmv4_groups;
+ desc.name = dev_name(dev);
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
ret = PTR_ERR(drvdata->csdev);
diff --git a/drivers/hwtracing/coresight/coresight-funnel.c b/drivers/hwtracing/coresight/coresight-funnel.c
index fc033fdb6cd5..ded33f5e7d43 100644
--- a/drivers/hwtracing/coresight/coresight-funnel.c
+++ b/drivers/hwtracing/coresight/coresight-funnel.c
@@ -229,6 +229,7 @@ static int funnel_probe(struct device *dev, struct resource *res)
desc.ops = &funnel_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
+ desc.name = dev_name(dev);
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
ret = PTR_ERR(drvdata->csdev);
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index 541e500a83c2..f500de61e7f9 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -323,9 +323,6 @@ coresight_get_platform_data(struct device *dev)
goto error;
}
- /* Use device name as sysfs handle */
- pdata->name = dev_name(dev);
-
if (is_of_node(fwnode))
ret = of_get_coresight_platform_data(dev, pdata);
diff --git a/drivers/hwtracing/coresight/coresight-replicator.c b/drivers/hwtracing/coresight/coresight-replicator.c
index 054b33554695..f28bafd98995 100644
--- a/drivers/hwtracing/coresight/coresight-replicator.c
+++ b/drivers/hwtracing/coresight/coresight-replicator.c
@@ -220,6 +220,8 @@ static int replicator_probe(struct device *dev, struct resource *res)
desc.ops = &replicator_cs_ops;
desc.pdata = dev->platform_data;
desc.dev = dev;
+ desc.name = dev_name(dev);
+
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
ret = PTR_ERR(drvdata->csdev);
diff --git a/drivers/hwtracing/coresight/coresight-stm.c b/drivers/hwtracing/coresight/coresight-stm.c
index 9faa1ed01500..02031d93fb8b 100644
--- a/drivers/hwtracing/coresight/coresight-stm.c
+++ b/drivers/hwtracing/coresight/coresight-stm.c
@@ -871,6 +871,7 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &stm_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
+ desc.name = dev_name(dev);
desc.groups = coresight_stm_groups;
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
diff --git a/drivers/hwtracing/coresight/coresight-tmc.c b/drivers/hwtracing/coresight/coresight-tmc.c
index be0bd98ca8c1..44a571988219 100644
--- a/drivers/hwtracing/coresight/coresight-tmc.c
+++ b/drivers/hwtracing/coresight/coresight-tmc.c
@@ -437,6 +437,7 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
desc.pdata = pdata;
desc.dev = dev;
desc.groups = coresight_tmc_groups;
+ desc.name = dev_name(dev);
switch (drvdata->config_type) {
case TMC_CONFIG_TYPE_ETB:
@@ -461,7 +462,7 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &tmc_etf_cs_ops;
break;
default:
- pr_err("%s: Unsupported TMC config\n", pdata->name);
+ pr_err("%s: Unsupported TMC config\n", desc.name);
ret = -EINVAL;
goto out;
}
@@ -472,7 +473,7 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
goto out;
}
- drvdata->miscdev.name = pdata->name;
+ drvdata->miscdev.name = desc.name;
drvdata->miscdev.minor = MISC_DYNAMIC_MINOR;
drvdata->miscdev.fops = &tmc_fops;
ret = misc_register(&drvdata->miscdev);
diff --git a/drivers/hwtracing/coresight/coresight-tpiu.c b/drivers/hwtracing/coresight/coresight-tpiu.c
index aec0ed7bf924..d8a2e3991c7e 100644
--- a/drivers/hwtracing/coresight/coresight-tpiu.c
+++ b/drivers/hwtracing/coresight/coresight-tpiu.c
@@ -157,6 +157,7 @@ static int tpiu_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &tpiu_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
+ desc.name = dev_name(dev);
drvdata->csdev = coresight_register(&desc);
if (!IS_ERR(drvdata->csdev)) {
diff --git a/drivers/hwtracing/coresight/coresight.c b/drivers/hwtracing/coresight/coresight.c
index 4b130281236a..04b5d3c2bb3a 100644
--- a/drivers/hwtracing/coresight/coresight.c
+++ b/drivers/hwtracing/coresight/coresight.c
@@ -1199,7 +1199,7 @@ struct coresight_device *coresight_register(struct coresight_desc *desc)
csdev->dev.parent = desc->dev;
csdev->dev.release = coresight_device_release;
csdev->dev.bus = &coresight_bustype;
- dev_set_name(&csdev->dev, "%s", desc->pdata->name);
+ dev_set_name(&csdev->dev, "%s", desc->name);
ret = device_register(&csdev->dev);
if (ret) {
diff --git a/include/linux/coresight.h b/include/linux/coresight.h
index bf241dbf99c5..298db20ba8ce 100644
--- a/include/linux/coresight.h
+++ b/include/linux/coresight.h
@@ -91,13 +91,11 @@ union coresight_dev_subtype {
/**
* struct coresight_platform_data - data harvested from the DT specification
- * @name: name of the component as shown under sysfs.
* @nr_inport: number of input ports for this component.
* @nr_outport: number of output ports for this component.
* @conns: Array of nr_outport connections from this component
*/
struct coresight_platform_data {
- const char *name;
int nr_inport;
int nr_outport;
struct coresight_connection *conns;
@@ -108,11 +106,12 @@ struct coresight_platform_data {
* @type: as defined by @coresight_dev_type.
* @subtype: as defined by @coresight_dev_subtype.
* @ops: generic operations for this component, as defined
- by @coresight_ops.
+ * by @coresight_ops.
* @pdata: platform data collected from DT.
* @dev: The device entity associated to this component.
* @groups: operations specific to this component. These will end up
- in the component's sysfs sub-directory.
+ * in the component's sysfs sub-directory.
+ * @name: name for the coresight device, also shown under sysfs.
*/
struct coresight_desc {
enum coresight_dev_type type;
@@ -121,6 +120,7 @@ struct coresight_desc {
struct coresight_platform_data *pdata;
struct device *dev;
const struct attribute_group **groups;
+ const char *name;
};
/**
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 08/28] coresight: Cleanup coresight_remove_conns
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (6 preceding siblings ...)
2019-06-19 19:52 ` [PATCH 07/28] coresight: Remove name from platform description Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:52 ` [PATCH 09/28] coresight: Reuse platform data structure for connection tracking Mathieu Poirier
` (20 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
When a device is unregistered, we remove all connection
references to it, by searching the connection records of
all devices in the coresight bus, via coresight_remove_conns.
We could avoid searching if this device doesn't have an input
port (e.g, a source). Also document the purpose of the function.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight.c | 15 +++++++++++++--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight.c b/drivers/hwtracing/coresight/coresight.c
index 04b5d3c2bb3a..068bd2fc0985 100644
--- a/drivers/hwtracing/coresight/coresight.c
+++ b/drivers/hwtracing/coresight/coresight.c
@@ -1096,10 +1096,21 @@ static int coresight_remove_match(struct device *dev, void *data)
return 0;
}
+/*
+ * coresight_remove_conns - Remove references to this given devices
+ * from the connections of other devices.
+ */
static void coresight_remove_conns(struct coresight_device *csdev)
{
- bus_for_each_dev(&coresight_bustype, NULL,
- csdev, coresight_remove_match);
+ /*
+ * Another device will point to this device only if there is
+ * an output port connected to this one. i.e, if the device
+ * doesn't have at least one input port, there is no point
+ * in searching all the devices.
+ */
+ if (csdev->nr_inport)
+ bus_for_each_dev(&coresight_bustype, NULL,
+ csdev, coresight_remove_match);
}
/**
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 09/28] coresight: Reuse platform data structure for connection tracking
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (7 preceding siblings ...)
2019-06-19 19:52 ` [PATCH 08/28] coresight: Cleanup coresight_remove_conns Mathieu Poirier
@ 2019-06-19 19:52 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 10/28] coresight: Rearrange platform data probing Mathieu Poirier
` (19 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:52 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
The platform specific information describes the connections and
the ports of a given coresigh device. This information is also
recorded in the coresight device as separate fields. Let us reuse
the original platform description to streamline the handling
of the data.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
.../hwtracing/coresight/coresight-tmc-etr.c | 4 +-
drivers/hwtracing/coresight/coresight.c | 46 +++++++++----------
include/linux/coresight.h | 8 +---
3 files changed, 27 insertions(+), 31 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
index 709448cf2dff..ce0114a5435c 100644
--- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
+++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
@@ -756,8 +756,8 @@ tmc_etr_get_catu_device(struct tmc_drvdata *drvdata)
if (!IS_ENABLED(CONFIG_CORESIGHT_CATU))
return NULL;
- for (i = 0; i < etr->nr_outport; i++) {
- tmp = etr->conns[i].child_dev;
+ for (i = 0; i < etr->pdata->nr_outport; i++) {
+ tmp = etr->pdata->conns[i].child_dev;
if (tmp && coresight_is_catu_device(tmp))
return tmp;
}
diff --git a/drivers/hwtracing/coresight/coresight.c b/drivers/hwtracing/coresight/coresight.c
index 068bd2fc0985..96e15154a566 100644
--- a/drivers/hwtracing/coresight/coresight.c
+++ b/drivers/hwtracing/coresight/coresight.c
@@ -100,8 +100,8 @@ static int coresight_find_link_inport(struct coresight_device *csdev,
int i;
struct coresight_connection *conn;
- for (i = 0; i < parent->nr_outport; i++) {
- conn = &parent->conns[i];
+ for (i = 0; i < parent->pdata->nr_outport; i++) {
+ conn = &parent->pdata->conns[i];
if (conn->child_dev == csdev)
return conn->child_port;
}
@@ -118,8 +118,8 @@ static int coresight_find_link_outport(struct coresight_device *csdev,
int i;
struct coresight_connection *conn;
- for (i = 0; i < csdev->nr_outport; i++) {
- conn = &csdev->conns[i];
+ for (i = 0; i < csdev->pdata->nr_outport; i++) {
+ conn = &csdev->pdata->conns[i];
if (conn->child_dev == child)
return conn->outport;
}
@@ -306,10 +306,10 @@ static void coresight_disable_link(struct coresight_device *csdev,
if (link_subtype == CORESIGHT_DEV_SUBTYPE_LINK_MERG) {
refport = inport;
- nr_conns = csdev->nr_inport;
+ nr_conns = csdev->pdata->nr_inport;
} else if (link_subtype == CORESIGHT_DEV_SUBTYPE_LINK_SPLIT) {
refport = outport;
- nr_conns = csdev->nr_outport;
+ nr_conns = csdev->pdata->nr_outport;
} else {
refport = 0;
nr_conns = 1;
@@ -595,9 +595,10 @@ static void coresight_grab_device(struct coresight_device *csdev)
{
int i;
- for (i = 0; i < csdev->nr_outport; i++) {
- struct coresight_device *child = csdev->conns[i].child_dev;
+ for (i = 0; i < csdev->pdata->nr_outport; i++) {
+ struct coresight_device *child;
+ child = csdev->pdata->conns[i].child_dev;
if (child && child->type == CORESIGHT_DEV_TYPE_HELPER)
pm_runtime_get_sync(child->dev.parent);
}
@@ -613,9 +614,10 @@ static void coresight_drop_device(struct coresight_device *csdev)
int i;
pm_runtime_put(csdev->dev.parent);
- for (i = 0; i < csdev->nr_outport; i++) {
- struct coresight_device *child = csdev->conns[i].child_dev;
+ for (i = 0; i < csdev->pdata->nr_outport; i++) {
+ struct coresight_device *child;
+ child = csdev->pdata->conns[i].child_dev;
if (child && child->type == CORESIGHT_DEV_TYPE_HELPER)
pm_runtime_put(child->dev.parent);
}
@@ -645,9 +647,10 @@ static int _coresight_build_path(struct coresight_device *csdev,
goto out;
/* Not a sink - recursively explore each port found on this element */
- for (i = 0; i < csdev->nr_outport; i++) {
- struct coresight_device *child_dev = csdev->conns[i].child_dev;
+ for (i = 0; i < csdev->pdata->nr_outport; i++) {
+ struct coresight_device *child_dev;
+ child_dev = csdev->pdata->conns[i].child_dev;
if (child_dev &&
_coresight_build_path(child_dev, sink, path) == 0) {
found = true;
@@ -1000,8 +1003,8 @@ static int coresight_orphan_match(struct device *dev, void *data)
* Circle throuch all the connection of that component. If we find
* an orphan connection whose name matches @csdev, link it.
*/
- for (i = 0; i < i_csdev->nr_outport; i++) {
- conn = &i_csdev->conns[i];
+ for (i = 0; i < i_csdev->pdata->nr_outport; i++) {
+ conn = &i_csdev->pdata->conns[i];
/* We have found at least one orphan connection */
if (conn->child_dev == NULL) {
@@ -1040,8 +1043,8 @@ static void coresight_fixup_device_conns(struct coresight_device *csdev)
{
int i;
- for (i = 0; i < csdev->nr_outport; i++) {
- struct coresight_connection *conn = &csdev->conns[i];
+ for (i = 0; i < csdev->pdata->nr_outport; i++) {
+ struct coresight_connection *conn = &csdev->pdata->conns[i];
struct device *dev = NULL;
if (conn->child_name)
@@ -1075,8 +1078,8 @@ static int coresight_remove_match(struct device *dev, void *data)
* Circle throuch all the connection of that component. If we find
* a connection whose name matches @csdev, remove it.
*/
- for (i = 0; i < iterator->nr_outport; i++) {
- conn = &iterator->conns[i];
+ for (i = 0; i < iterator->pdata->nr_outport; i++) {
+ conn = &iterator->pdata->conns[i];
if (conn->child_dev == NULL)
continue;
@@ -1108,7 +1111,7 @@ static void coresight_remove_conns(struct coresight_device *csdev)
* doesn't have at least one input port, there is no point
* in searching all the devices.
*/
- if (csdev->nr_inport)
+ if (csdev->pdata->nr_inport)
bus_for_each_dev(&coresight_bustype, NULL,
csdev, coresight_remove_match);
}
@@ -1195,10 +1198,7 @@ struct coresight_device *coresight_register(struct coresight_desc *desc)
csdev->refcnt = refcnts;
- csdev->nr_inport = desc->pdata->nr_inport;
- csdev->nr_outport = desc->pdata->nr_outport;
-
- csdev->conns = desc->pdata->conns;
+ csdev->pdata = desc->pdata;
csdev->type = desc->type;
csdev->subtype = desc->subtype;
diff --git a/include/linux/coresight.h b/include/linux/coresight.h
index 298db20ba8ce..b67d5074ece0 100644
--- a/include/linux/coresight.h
+++ b/include/linux/coresight.h
@@ -140,9 +140,7 @@ struct coresight_connection {
/**
* struct coresight_device - representation of a device as used by the framework
- * @conns: array of coresight_connections associated to this component.
- * @nr_inport: number of input port associated to this component.
- * @nr_outport: number of output port associated to this component.
+ * @pdata: Platform data with device connections associated to this device.
* @type: as defined by @coresight_dev_type.
* @subtype: as defined by @coresight_dev_subtype.
* @ops: generic operations for this component, as defined
@@ -157,9 +155,7 @@ struct coresight_connection {
* @ea: Device attribute for sink representation under PMU directory.
*/
struct coresight_device {
- struct coresight_connection *conns;
- int nr_inport;
- int nr_outport;
+ struct coresight_platform_data *pdata;
enum coresight_dev_type type;
union coresight_dev_subtype subtype;
const struct coresight_ops *ops;
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 10/28] coresight: Rearrange platform data probing
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (8 preceding siblings ...)
2019-06-19 19:52 ` [PATCH 09/28] coresight: Reuse platform data structure for connection tracking Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 11/28] coresight: Add support for releasing platform specific data Mathieu Poirier
` (18 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
We are about to introduce methods to clean up the platform data
as we switch to tracking the device reference from "name" to "fwnode
handles" for device connections. This requires us to drop the fwnode
handle references when the data is no longer required - i.e, when
the device probe fails or the device gets unregistered.
In order to consolidate the invocation of the cleanup, we delay the
platform probing to the very last minute, possibly before invoking
the coresight_register. Then, we leave the coresight core code to
do the clean up. i.e, if the coresight_register fails, it takes
care of freeing the data. Otherwise, coresight_unregister will
do the necessary operations.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-catu.c | 14 +++++++-------
drivers/hwtracing/coresight/coresight-etb10.c | 10 +++++-----
drivers/hwtracing/coresight/coresight-etm3x.c | 12 +++++++-----
drivers/hwtracing/coresight/coresight-etm4x.c | 12 +++++++-----
drivers/hwtracing/coresight/coresight-funnel.c | 12 +++++++-----
.../hwtracing/coresight/coresight-replicator.c | 12 +++++++-----
drivers/hwtracing/coresight/coresight-stm.c | 11 +++++++----
drivers/hwtracing/coresight/coresight-tmc.c | 16 ++++++++--------
drivers/hwtracing/coresight/coresight-tpiu.c | 10 +++++-----
9 files changed, 60 insertions(+), 49 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-catu.c b/drivers/hwtracing/coresight/coresight-catu.c
index 05c73045532a..1c1ad1268b9d 100644
--- a/drivers/hwtracing/coresight/coresight-catu.c
+++ b/drivers/hwtracing/coresight/coresight-catu.c
@@ -505,13 +505,6 @@ static int catu_probe(struct amba_device *adev, const struct amba_id *id)
struct device *dev = &adev->dev;
void __iomem *base;
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata)) {
- ret = PTR_ERR(pdata);
- goto out;
- }
- dev->platform_data = pdata;
-
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata) {
ret = -ENOMEM;
@@ -544,6 +537,13 @@ static int catu_probe(struct amba_device *adev, const struct amba_id *id)
if (ret)
goto out;
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto out;
+ }
+ dev->platform_data = pdata;
+
drvdata->base = base;
catu_desc.pdata = pdata;
catu_desc.dev = dev;
diff --git a/drivers/hwtracing/coresight/coresight-etb10.c b/drivers/hwtracing/coresight/coresight-etb10.c
index 8726d6c7663c..fffaac3e3677 100644
--- a/drivers/hwtracing/coresight/coresight-etb10.c
+++ b/drivers/hwtracing/coresight/coresight-etb10.c
@@ -733,11 +733,6 @@ static int etb_probe(struct amba_device *adev, const struct amba_id *id)
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- adev->dev.platform_data = pdata;
-
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
@@ -772,6 +767,11 @@ static int etb_probe(struct amba_device *adev, const struct amba_id *id)
/* This device is not associated with a session */
drvdata->pid = -1;
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ adev->dev.platform_data = pdata;
+
desc.type = CORESIGHT_DEV_TYPE_SINK;
desc.subtype.sink_subtype = CORESIGHT_DEV_SUBTYPE_SINK_BUFFER;
desc.ops = &etb_cs_ops;
diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index 101fb01e20de..f2d461610a2f 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -795,11 +795,6 @@ static int etm_probe(struct amba_device *adev, const struct amba_id *id)
if (!drvdata)
return -ENOMEM;
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
-
- adev->dev.platform_data = pdata;
drvdata->use_cp14 = fwnode_property_read_bool(dev->fwnode, "arm,cp14");
dev_set_drvdata(dev, drvdata);
@@ -849,6 +844,13 @@ static int etm_probe(struct amba_device *adev, const struct amba_id *id)
etm_init_trace_id(drvdata);
etm_set_default(&drvdata->config);
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto err_arch_supported;
+ }
+ adev->dev.platform_data = pdata;
+
desc.type = CORESIGHT_DEV_TYPE_SOURCE;
desc.subtype.source_subtype = CORESIGHT_DEV_SUBTYPE_SOURCE_PROC;
desc.ops = &etm_cs_ops;
diff --git a/drivers/hwtracing/coresight/coresight-etm4x.c b/drivers/hwtracing/coresight/coresight-etm4x.c
index 8adc1485cd89..1609da1eaf83 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x.c
+++ b/drivers/hwtracing/coresight/coresight-etm4x.c
@@ -1089,11 +1089,6 @@ static int etm4_probe(struct amba_device *adev, const struct amba_id *id)
if (!drvdata)
return -ENOMEM;
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- adev->dev.platform_data = pdata;
-
dev_set_drvdata(dev, drvdata);
/* Validity for the resource is already checked by the AMBA core */
@@ -1136,6 +1131,13 @@ static int etm4_probe(struct amba_device *adev, const struct amba_id *id)
etm4_init_trace_id(drvdata);
etm4_set_default(&drvdata->config);
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto err_arch_supported;
+ }
+ adev->dev.platform_data = pdata;
+
desc.type = CORESIGHT_DEV_TYPE_SOURCE;
desc.subtype.source_subtype = CORESIGHT_DEV_SUBTYPE_SOURCE_PROC;
desc.ops = &etm4_cs_ops;
diff --git a/drivers/hwtracing/coresight/coresight-funnel.c b/drivers/hwtracing/coresight/coresight-funnel.c
index ded33f5e7d43..75fa2d3ad9b4 100644
--- a/drivers/hwtracing/coresight/coresight-funnel.c
+++ b/drivers/hwtracing/coresight/coresight-funnel.c
@@ -188,11 +188,6 @@ static int funnel_probe(struct device *dev, struct resource *res)
struct funnel_drvdata *drvdata;
struct coresight_desc desc = { 0 };
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- dev->platform_data = pdata;
-
if (is_of_node(dev_fwnode(dev)) &&
of_device_is_compatible(dev->of_node, "arm,coresight-funnel"))
pr_warn_once("Uses OBSOLETE CoreSight funnel binding\n");
@@ -224,6 +219,13 @@ static int funnel_probe(struct device *dev, struct resource *res)
dev_set_drvdata(dev, drvdata);
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto out_disable_clk;
+ }
+ dev->platform_data = pdata;
+
desc.type = CORESIGHT_DEV_TYPE_LINK;
desc.subtype.link_subtype = CORESIGHT_DEV_SUBTYPE_LINK_MERG;
desc.ops = &funnel_cs_ops;
diff --git a/drivers/hwtracing/coresight/coresight-replicator.c b/drivers/hwtracing/coresight/coresight-replicator.c
index f28bafd98995..64dfde7241c1 100644
--- a/drivers/hwtracing/coresight/coresight-replicator.c
+++ b/drivers/hwtracing/coresight/coresight-replicator.c
@@ -179,11 +179,6 @@ static int replicator_probe(struct device *dev, struct resource *res)
struct coresight_desc desc = { 0 };
void __iomem *base;
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- dev->platform_data = pdata;
-
if (is_of_node(dev_fwnode(dev)) &&
of_device_is_compatible(dev->of_node, "arm,coresight-replicator"))
pr_warn_once("Uses OBSOLETE CoreSight replicator binding\n");
@@ -215,6 +210,13 @@ static int replicator_probe(struct device *dev, struct resource *res)
dev_set_drvdata(dev, drvdata);
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto out_disable_clk;
+ }
+ dev->platform_data = pdata;
+
desc.type = CORESIGHT_DEV_TYPE_LINK;
desc.subtype.link_subtype = CORESIGHT_DEV_SUBTYPE_LINK_SPLIT;
desc.ops = &replicator_cs_ops;
diff --git a/drivers/hwtracing/coresight/coresight-stm.c b/drivers/hwtracing/coresight/coresight-stm.c
index 02031d93fb8b..03528f3fa9ff 100644
--- a/drivers/hwtracing/coresight/coresight-stm.c
+++ b/drivers/hwtracing/coresight/coresight-stm.c
@@ -810,10 +810,6 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id)
size_t bitmap_size;
struct coresight_desc desc = { 0 };
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- adev->dev.platform_data = pdata;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
@@ -866,6 +862,13 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id)
return -EPROBE_DEFER;
}
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto stm_unregister;
+ }
+ adev->dev.platform_data = pdata;
+
desc.type = CORESIGHT_DEV_TYPE_SOURCE;
desc.subtype.source_subtype = CORESIGHT_DEV_SUBTYPE_SOURCE_SOFTWARE;
desc.ops = &stm_cs_ops;
diff --git a/drivers/hwtracing/coresight/coresight-tmc.c b/drivers/hwtracing/coresight/coresight-tmc.c
index 44a571988219..212630e65cca 100644
--- a/drivers/hwtracing/coresight/coresight-tmc.c
+++ b/drivers/hwtracing/coresight/coresight-tmc.c
@@ -398,13 +398,6 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata)) {
- ret = PTR_ERR(pdata);
- goto out;
- }
- adev->dev.platform_data = pdata;
-
ret = -ENOMEM;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
@@ -434,7 +427,6 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
else
drvdata->size = readl_relaxed(drvdata->base + TMC_RSZ) * 4;
- desc.pdata = pdata;
desc.dev = dev;
desc.groups = coresight_tmc_groups;
desc.name = dev_name(dev);
@@ -467,6 +459,14 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
goto out;
}
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata)) {
+ ret = PTR_ERR(pdata);
+ goto out;
+ }
+ adev->dev.platform_data = pdata;
+ desc.pdata = pdata;
+
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
ret = PTR_ERR(drvdata->csdev);
diff --git a/drivers/hwtracing/coresight/coresight-tpiu.c b/drivers/hwtracing/coresight/coresight-tpiu.c
index d8a2e3991c7e..b699d613425d 100644
--- a/drivers/hwtracing/coresight/coresight-tpiu.c
+++ b/drivers/hwtracing/coresight/coresight-tpiu.c
@@ -125,11 +125,6 @@ static int tpiu_probe(struct amba_device *adev, const struct amba_id *id)
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
- pdata = coresight_get_platform_data(dev);
- if (IS_ERR(pdata))
- return PTR_ERR(pdata);
- dev->platform_data = pdata;
-
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
@@ -152,6 +147,11 @@ static int tpiu_probe(struct amba_device *adev, const struct amba_id *id)
/* Disable tpiu to support older devices */
tpiu_disable_hw(drvdata);
+ pdata = coresight_get_platform_data(dev);
+ if (IS_ERR(pdata))
+ return PTR_ERR(pdata);
+ dev->platform_data = pdata;
+
desc.type = CORESIGHT_DEV_TYPE_SINK;
desc.subtype.sink_subtype = CORESIGHT_DEV_SUBTYPE_SINK_PORT;
desc.ops = &tpiu_cs_ops;
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 11/28] coresight: Add support for releasing platform specific data
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (9 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 10/28] coresight: Rearrange platform data probing Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 12/28] coresight: platform: Use fwnode handle for device search Mathieu Poirier
` (17 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
Add a helper to clean up the platform specific data provided
by the firmware. This will be later used for dropping the necessary
references when we switch to the fwnode handles for tracking
connections.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-platform.c | 6 +++++-
drivers/hwtracing/coresight/coresight-priv.h | 4 ++++
drivers/hwtracing/coresight/coresight.c | 3 +++
3 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index f500de61e7f9..53d6eed44a20 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -17,6 +17,7 @@
#include <linux/cpumask.h>
#include <asm/smp_plat.h>
+#include "coresight-priv.h"
/*
* coresight_alloc_conns: Allocate connections record for each output
* port from the device.
@@ -311,7 +312,7 @@ struct coresight_platform_data *
coresight_get_platform_data(struct device *dev)
{
int ret = -ENOENT;
- struct coresight_platform_data *pdata;
+ struct coresight_platform_data *pdata = NULL;
struct fwnode_handle *fwnode = dev_fwnode(dev);
if (IS_ERR_OR_NULL(fwnode))
@@ -329,6 +330,9 @@ coresight_get_platform_data(struct device *dev)
if (!ret)
return pdata;
error:
+ if (!IS_ERR_OR_NULL(pdata))
+ /* Cleanup the connection information */
+ coresight_release_platform_data(pdata);
return ERR_PTR(ret);
}
EXPORT_SYMBOL_GPL(coresight_get_platform_data);
diff --git a/drivers/hwtracing/coresight/coresight-priv.h b/drivers/hwtracing/coresight/coresight-priv.h
index e0684d06e9ee..c21642114fc3 100644
--- a/drivers/hwtracing/coresight/coresight-priv.h
+++ b/drivers/hwtracing/coresight/coresight-priv.h
@@ -200,4 +200,8 @@ static inline void *coresight_get_uci_data(const struct amba_id *id)
return 0;
}
+static inline void
+coresight_release_platform_data(struct coresight_platform_data *pdata)
+{}
+
#endif
diff --git a/drivers/hwtracing/coresight/coresight.c b/drivers/hwtracing/coresight/coresight.c
index 96e15154a566..526141c2f876 100644
--- a/drivers/hwtracing/coresight/coresight.c
+++ b/drivers/hwtracing/coresight/coresight.c
@@ -1250,6 +1250,8 @@ struct coresight_device *coresight_register(struct coresight_desc *desc)
err_free_csdev:
kfree(csdev);
err_out:
+ /* Cleanup the connection information */
+ coresight_release_platform_data(desc->pdata);
return ERR_PTR(ret);
}
EXPORT_SYMBOL_GPL(coresight_register);
@@ -1259,6 +1261,7 @@ void coresight_unregister(struct coresight_device *csdev)
etm_perf_del_symlink_sink(csdev);
/* Remove references of that device in the topology */
coresight_remove_conns(csdev);
+ coresight_release_platform_data(csdev->pdata);
device_unregister(&csdev->dev);
}
EXPORT_SYMBOL_GPL(coresight_unregister);
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 12/28] coresight: platform: Use fwnode handle for device search
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (10 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 11/28] coresight: Add support for releasing platform specific data Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 13/28] coresight: Use fwnode handle instead of device names Mathieu Poirier
` (16 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
We match of_node while searching for a device. Make this
more generic in preparation for the ACPI support by using
fwnode_handle.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-platform.c | 16 +++++++++-------
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index 53d6eed44a20..4394095ad224 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -36,14 +36,13 @@ static int coresight_alloc_conns(struct device *dev,
return 0;
}
-#ifdef CONFIG_OF
-static int of_dev_node_match(struct device *dev, void *data)
+static int coresight_device_fwnode_match(struct device *dev, void *fwnode)
{
- return dev->of_node == data;
+ return dev_fwnode(dev) == fwnode;
}
static struct device *
-of_coresight_get_endpoint_device(struct device_node *endpoint)
+coresight_find_device_by_fwnode(struct fwnode_handle *fwnode)
{
struct device *dev = NULL;
@@ -52,7 +51,7 @@ of_coresight_get_endpoint_device(struct device_node *endpoint)
* platform bus.
*/
dev = bus_find_device(&platform_bus_type, NULL,
- endpoint, of_dev_node_match);
+ fwnode, coresight_device_fwnode_match);
if (dev)
return dev;
@@ -61,9 +60,10 @@ of_coresight_get_endpoint_device(struct device_node *endpoint)
* looking for the device that matches the endpoint node.
*/
return bus_find_device(&amba_bustype, NULL,
- endpoint, of_dev_node_match);
+ fwnode, coresight_device_fwnode_match);
}
+#ifdef CONFIG_OF
static inline bool of_coresight_legacy_ep_is_input(struct device_node *ep)
{
return of_property_read_bool(ep, "slave-mode");
@@ -191,6 +191,7 @@ static int of_coresight_parse_endpoint(struct device *dev,
struct device_node *rparent = NULL;
struct device_node *rep = NULL;
struct device *rdev = NULL;
+ struct fwnode_handle *rdev_fwnode;
do {
/* Parse the local port details */
@@ -209,8 +210,9 @@ static int of_coresight_parse_endpoint(struct device *dev,
if (of_graph_parse_endpoint(rep, &rendpoint))
break;
+ rdev_fwnode = of_fwnode_handle(rparent);
/* If the remote device is not available, defer probing */
- rdev = of_coresight_get_endpoint_device(rparent);
+ rdev = coresight_find_device_by_fwnode(rdev_fwnode);
if (!rdev) {
ret = -EPROBE_DEFER;
break;
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 13/28] coresight: Use fwnode handle instead of device names
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (11 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 12/28] coresight: platform: Use fwnode handle for device search Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 14/28] coresight: Use platform agnostic names Mathieu Poirier
` (15 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
We rely on the device names to find a CoreSight device on the
coresight bus. The device name however is obtained from the platform,
which is bound to the real platform/amba device. As we are about
to use different naming scheme for the coresight devices, we can't
rely on the platform device name to find the corresponding
coresight device. Instead we use the platform agnostic
"fwnode handle" of the parent device to find the devices.
We also reuse the same fwnode as the parent for the Coresight
device we create.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
.../hwtracing/coresight/coresight-platform.c | 14 +++++--
drivers/hwtracing/coresight/coresight-priv.h | 6 +--
drivers/hwtracing/coresight/coresight.c | 42 +++++++++++++++----
include/linux/coresight.h | 4 +-
4 files changed, 49 insertions(+), 17 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index 4394095ad224..49112a58478e 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -36,7 +36,7 @@ static int coresight_alloc_conns(struct device *dev,
return 0;
}
-static int coresight_device_fwnode_match(struct device *dev, void *fwnode)
+int coresight_device_fwnode_match(struct device *dev, void *fwnode)
{
return dev_fwnode(dev) == fwnode;
}
@@ -219,9 +219,15 @@ static int of_coresight_parse_endpoint(struct device *dev,
}
conn->outport = endpoint.port;
- conn->child_name = devm_kstrdup(dev,
- dev_name(rdev),
- GFP_KERNEL);
+ /*
+ * Hold the refcount to the target device. This could be
+ * released via:
+ * 1) coresight_release_platform_data() if the probe fails or
+ * this device is unregistered.
+ * 2) While removing the target device via
+ * coresight_remove_match()
+ */
+ conn->child_fwnode = fwnode_handle_get(rdev_fwnode);
conn->child_port = rendpoint.port;
/* Connection record updated */
ret = 1;
diff --git a/drivers/hwtracing/coresight/coresight-priv.h b/drivers/hwtracing/coresight/coresight-priv.h
index c21642114fc3..8b07fe55395a 100644
--- a/drivers/hwtracing/coresight/coresight-priv.h
+++ b/drivers/hwtracing/coresight/coresight-priv.h
@@ -200,8 +200,8 @@ static inline void *coresight_get_uci_data(const struct amba_id *id)
return 0;
}
-static inline void
-coresight_release_platform_data(struct coresight_platform_data *pdata)
-{}
+void coresight_release_platform_data(struct coresight_platform_data *pdata);
+
+int coresight_device_fwnode_match(struct device *dev, void *fwnode);
#endif
diff --git a/drivers/hwtracing/coresight/coresight.c b/drivers/hwtracing/coresight/coresight.c
index 526141c2f876..1287778c3be5 100644
--- a/drivers/hwtracing/coresight/coresight.c
+++ b/drivers/hwtracing/coresight/coresight.c
@@ -978,6 +978,7 @@ static void coresight_device_release(struct device *dev)
{
struct coresight_device *csdev = to_coresight_device(dev);
+ fwnode_handle_put(csdev->dev.fwnode);
kfree(csdev->refcnt);
kfree(csdev);
}
@@ -1009,13 +1010,11 @@ static int coresight_orphan_match(struct device *dev, void *data)
/* We have found at least one orphan connection */
if (conn->child_dev == NULL) {
/* Does it match this newly added device? */
- if (conn->child_name &&
- !strcmp(dev_name(&csdev->dev), conn->child_name)) {
+ if (conn->child_fwnode == csdev->dev.fwnode)
conn->child_dev = csdev;
- } else {
+ else
/* This component still has an orphan */
still_orphan = true;
- }
}
}
@@ -1047,9 +1046,9 @@ static void coresight_fixup_device_conns(struct coresight_device *csdev)
struct coresight_connection *conn = &csdev->pdata->conns[i];
struct device *dev = NULL;
- if (conn->child_name)
- dev = bus_find_device_by_name(&coresight_bustype, NULL,
- conn->child_name);
+ dev = bus_find_device(&coresight_bustype, NULL,
+ (void *)conn->child_fwnode,
+ coresight_device_fwnode_match);
if (dev) {
conn->child_dev = to_coresight_device(dev);
/* and put reference from 'bus_find_device()' */
@@ -1084,9 +1083,15 @@ static int coresight_remove_match(struct device *dev, void *data)
if (conn->child_dev == NULL)
continue;
- if (!strcmp(dev_name(&csdev->dev), conn->child_name)) {
+ if (csdev->dev.fwnode == conn->child_fwnode) {
iterator->orphan = true;
conn->child_dev = NULL;
+ /*
+ * Drop the reference to the handle for the remote
+ * device acquired in parsing the connections from
+ * platform data.
+ */
+ fwnode_handle_put(conn->child_fwnode);
/* No need to continue */
break;
}
@@ -1166,6 +1171,22 @@ static int __init coresight_init(void)
}
postcore_initcall(coresight_init);
+/*
+ * coresight_release_platform_data: Release references to the devices connected
+ * to the output port of this device.
+ */
+void coresight_release_platform_data(struct coresight_platform_data *pdata)
+{
+ int i;
+
+ for (i = 0; i < pdata->nr_outport; i++) {
+ if (pdata->conns[i].child_fwnode) {
+ fwnode_handle_put(pdata->conns[i].child_fwnode);
+ pdata->conns[i].child_fwnode = NULL;
+ }
+ }
+}
+
struct coresight_device *coresight_register(struct coresight_desc *desc)
{
int ret;
@@ -1210,6 +1231,11 @@ struct coresight_device *coresight_register(struct coresight_desc *desc)
csdev->dev.parent = desc->dev;
csdev->dev.release = coresight_device_release;
csdev->dev.bus = &coresight_bustype;
+ /*
+ * Hold the reference to our parent device. This will be
+ * dropped only in coresight_device_release().
+ */
+ csdev->dev.fwnode = fwnode_handle_get(dev_fwnode(desc->dev));
dev_set_name(&csdev->dev, "%s", desc->name);
ret = device_register(&csdev->dev);
diff --git a/include/linux/coresight.h b/include/linux/coresight.h
index b67d5074ece0..b40544bc06fe 100644
--- a/include/linux/coresight.h
+++ b/include/linux/coresight.h
@@ -126,15 +126,15 @@ struct coresight_desc {
/**
* struct coresight_connection - representation of a single connection
* @outport: a connection's output port number.
- * @chid_name: remote component's name.
* @child_port: remote component's port number @output is connected to.
+ * @chid_fwnode: remote component's fwnode handle.
* @child_dev: a @coresight_device representation of the component
connected to @outport.
*/
struct coresight_connection {
int outport;
- const char *child_name;
int child_port;
+ struct fwnode_handle *child_fwnode;
struct coresight_device *child_dev;
};
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 14/28] coresight: Use platform agnostic names
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (12 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 13/28] coresight: Use fwnode handle instead of device names Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 15/28] coresight: stm: ACPI support for parsing stimulus base Mathieu Poirier
` (14 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
So far we have reused the name of the "platform" device for
the CoreSight device. But this is not very intuitive when
we move to ACPI. Also, the ACPI device names have ":" in them
(e.g, ARMHC97C:01), which the perf tool doesn't like very much.
This patch introduces a generic naming scheme, givin more intuitive
names for the devices that appear on the CoreSight bus.
The names follow the pattern "prefix" followed by "index" (e.g, etm5).
We maintain a list of allocated devices per "prefix" to make sure
we don't allocate a new name when it is reprobed (e.g, due to
unsatisifed device dependencies). So, we maintain the list
of "fwnodes" of the parent devices to allocate a consistent name.
All devices except the ETMs get an index allocated in the order
of probing. ETMs get an index based on the CPU they are attached to.
TMC devices are named using "tmc_etf", "tmc_etb", and "tmc_etr"
prefixes depending on the configuration of the device.
The replicators and funnels are not classified as dynamic/static
anymore. One could easily figure that out by checking the presence
of "mgmt" registers under sysfs.
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-catu.c | 7 ++-
drivers/hwtracing/coresight/coresight-etb10.c | 7 ++-
drivers/hwtracing/coresight/coresight-etm3x.c | 4 +-
drivers/hwtracing/coresight/coresight-etm4x.c | 4 +-
.../hwtracing/coresight/coresight-funnel.c | 7 ++-
.../coresight/coresight-replicator.c | 7 ++-
drivers/hwtracing/coresight/coresight-stm.c | 12 +++-
drivers/hwtracing/coresight/coresight-tmc.c | 15 ++++-
drivers/hwtracing/coresight/coresight-tpiu.c | 7 ++-
drivers/hwtracing/coresight/coresight.c | 58 +++++++++++++++++++
include/linux/coresight.h | 25 +++++++-
11 files changed, 141 insertions(+), 12 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-catu.c b/drivers/hwtracing/coresight/coresight-catu.c
index 1c1ad1268b9d..16ebf38a9f66 100644
--- a/drivers/hwtracing/coresight/coresight-catu.c
+++ b/drivers/hwtracing/coresight/coresight-catu.c
@@ -28,6 +28,8 @@
#define catu_dbg(x, ...) do {} while (0)
#endif
+DEFINE_CORESIGHT_DEVLIST(catu_devs, "catu");
+
struct catu_etr_buf {
struct tmc_sg_table *catu_table;
dma_addr_t sladdr;
@@ -505,6 +507,10 @@ static int catu_probe(struct amba_device *adev, const struct amba_id *id)
struct device *dev = &adev->dev;
void __iomem *base;
+ catu_desc.name = coresight_alloc_device_name(&catu_devs, dev);
+ if (!catu_desc.name)
+ return -ENOMEM;
+
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata) {
ret = -ENOMEM;
@@ -551,7 +557,6 @@ static int catu_probe(struct amba_device *adev, const struct amba_id *id)
catu_desc.type = CORESIGHT_DEV_TYPE_HELPER;
catu_desc.subtype.helper_subtype = CORESIGHT_DEV_SUBTYPE_HELPER_CATU;
catu_desc.ops = &catu_ops;
- catu_desc.name = dev_name(dev);
drvdata->csdev = coresight_register(&catu_desc);
if (IS_ERR(drvdata->csdev))
diff --git a/drivers/hwtracing/coresight/coresight-etb10.c b/drivers/hwtracing/coresight/coresight-etb10.c
index fffaac3e3677..d5b9edecf76e 100644
--- a/drivers/hwtracing/coresight/coresight-etb10.c
+++ b/drivers/hwtracing/coresight/coresight-etb10.c
@@ -63,6 +63,8 @@
#define ETB_FFSR_BIT 1
#define ETB_FRAME_SIZE_WORDS 4
+DEFINE_CORESIGHT_DEVLIST(etb_devs, "etb");
+
/**
* struct etb_drvdata - specifics associated to an ETB component
* @base: memory mapped base address for this component.
@@ -733,6 +735,10 @@ static int etb_probe(struct amba_device *adev, const struct amba_id *id)
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
+ desc.name = coresight_alloc_device_name(&etb_devs, dev);
+ if (!desc.name)
+ return -ENOMEM;
+
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
@@ -777,7 +783,6 @@ static int etb_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &etb_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
- desc.name = dev_name(dev);
desc.groups = coresight_etb_groups;
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev))
diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index f2d461610a2f..bed729140718 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -815,6 +815,9 @@ static int etm_probe(struct amba_device *adev, const struct amba_id *id)
}
drvdata->cpu = coresight_get_cpu(dev);
+ desc.name = devm_kasprintf(dev, GFP_KERNEL, "etm%d", drvdata->cpu);
+ if (!desc.name)
+ return -ENOMEM;
cpus_read_lock();
etmdrvdata[drvdata->cpu] = drvdata;
@@ -856,7 +859,6 @@ static int etm_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &etm_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
- desc.name = dev_name(dev);
desc.groups = coresight_etm_groups;
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
diff --git a/drivers/hwtracing/coresight/coresight-etm4x.c b/drivers/hwtracing/coresight/coresight-etm4x.c
index 1609da1eaf83..7fe266194ab5 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x.c
+++ b/drivers/hwtracing/coresight/coresight-etm4x.c
@@ -1101,6 +1101,9 @@ static int etm4_probe(struct amba_device *adev, const struct amba_id *id)
spin_lock_init(&drvdata->spinlock);
drvdata->cpu = coresight_get_cpu(dev);
+ desc.name = devm_kasprintf(dev, GFP_KERNEL, "etm%d", drvdata->cpu);
+ if (!desc.name)
+ return -ENOMEM;
cpus_read_lock();
etmdrvdata[drvdata->cpu] = drvdata;
@@ -1144,7 +1147,6 @@ static int etm4_probe(struct amba_device *adev, const struct amba_id *id)
desc.pdata = pdata;
desc.dev = dev;
desc.groups = coresight_etmv4_groups;
- desc.name = dev_name(dev);
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
ret = PTR_ERR(drvdata->csdev);
diff --git a/drivers/hwtracing/coresight/coresight-funnel.c b/drivers/hwtracing/coresight/coresight-funnel.c
index 75fa2d3ad9b4..5867fcb4503b 100644
--- a/drivers/hwtracing/coresight/coresight-funnel.c
+++ b/drivers/hwtracing/coresight/coresight-funnel.c
@@ -29,6 +29,8 @@
#define FUNNEL_HOLDTIME (0x7 << FUNNEL_HOLDTIME_SHFT)
#define FUNNEL_ENSx_MASK 0xff
+DEFINE_CORESIGHT_DEVLIST(funnel_devs, "funnel");
+
/**
* struct funnel_drvdata - specifics associated to a funnel component
* @base: memory mapped base address for this component.
@@ -192,6 +194,10 @@ static int funnel_probe(struct device *dev, struct resource *res)
of_device_is_compatible(dev->of_node, "arm,coresight-funnel"))
pr_warn_once("Uses OBSOLETE CoreSight funnel binding\n");
+ desc.name = coresight_alloc_device_name(&funnel_devs, dev);
+ if (!desc.name)
+ return -ENOMEM;
+
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
@@ -231,7 +237,6 @@ static int funnel_probe(struct device *dev, struct resource *res)
desc.ops = &funnel_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
- desc.name = dev_name(dev);
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
ret = PTR_ERR(drvdata->csdev);
diff --git a/drivers/hwtracing/coresight/coresight-replicator.c b/drivers/hwtracing/coresight/coresight-replicator.c
index 64dfde7241c1..c0e42253dfe7 100644
--- a/drivers/hwtracing/coresight/coresight-replicator.c
+++ b/drivers/hwtracing/coresight/coresight-replicator.c
@@ -22,6 +22,8 @@
#define REPLICATOR_IDFILTER0 0x000
#define REPLICATOR_IDFILTER1 0x004
+DEFINE_CORESIGHT_DEVLIST(replicator_devs, "replicator");
+
/**
* struct replicator_drvdata - specifics associated to a replicator component
* @base: memory mapped base address for this component. Also indicates
@@ -183,6 +185,10 @@ static int replicator_probe(struct device *dev, struct resource *res)
of_device_is_compatible(dev->of_node, "arm,coresight-replicator"))
pr_warn_once("Uses OBSOLETE CoreSight replicator binding\n");
+ desc.name = coresight_alloc_device_name(&replicator_devs, dev);
+ if (!desc.name)
+ return -ENOMEM;
+
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
@@ -222,7 +228,6 @@ static int replicator_probe(struct device *dev, struct resource *res)
desc.ops = &replicator_cs_ops;
desc.pdata = dev->platform_data;
desc.dev = dev;
- desc.name = dev_name(dev);
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
diff --git a/drivers/hwtracing/coresight/coresight-stm.c b/drivers/hwtracing/coresight/coresight-stm.c
index 03528f3fa9ff..e3e2b000cfb7 100644
--- a/drivers/hwtracing/coresight/coresight-stm.c
+++ b/drivers/hwtracing/coresight/coresight-stm.c
@@ -107,6 +107,8 @@ struct channel_space {
unsigned long *guaranteed;
};
+DEFINE_CORESIGHT_DEVLIST(stm_devs, "stm");
+
/**
* struct stm_drvdata - specifics associated to an STM component
* @base: memory mapped base address for this component.
@@ -810,6 +812,10 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id)
size_t bitmap_size;
struct coresight_desc desc = { 0 };
+ desc.name = coresight_alloc_device_name(&stm_devs, dev);
+ if (!desc.name)
+ return -ENOMEM;
+
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
@@ -854,11 +860,12 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id)
spin_lock_init(&drvdata->spinlock);
stm_init_default_data(drvdata);
- stm_init_generic_data(drvdata, dev_name(dev));
+ stm_init_generic_data(drvdata, desc.name);
if (stm_register_device(dev, &drvdata->stm, THIS_MODULE)) {
dev_info(dev,
- "stm_register_device failed, probing deferred\n");
+ "%s : stm_register_device failed, probing deferred\n",
+ desc.name);
return -EPROBE_DEFER;
}
@@ -874,7 +881,6 @@ static int stm_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &stm_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
- desc.name = dev_name(dev);
desc.groups = coresight_stm_groups;
drvdata->csdev = coresight_register(&desc);
if (IS_ERR(drvdata->csdev)) {
diff --git a/drivers/hwtracing/coresight/coresight-tmc.c b/drivers/hwtracing/coresight/coresight-tmc.c
index 212630e65cca..be37aff573b4 100644
--- a/drivers/hwtracing/coresight/coresight-tmc.c
+++ b/drivers/hwtracing/coresight/coresight-tmc.c
@@ -27,6 +27,10 @@
#include "coresight-priv.h"
#include "coresight-tmc.h"
+DEFINE_CORESIGHT_DEVLIST(etb_devs, "tmc_etb");
+DEFINE_CORESIGHT_DEVLIST(etf_devs, "tmc_etf");
+DEFINE_CORESIGHT_DEVLIST(etr_devs, "tmc_etr");
+
void tmc_wait_for_tmcready(struct tmc_drvdata *drvdata)
{
/* Ensure formatter, unformatter and hardware fifo are empty */
@@ -397,6 +401,7 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
struct tmc_drvdata *drvdata;
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
+ struct coresight_dev_list *dev_list = NULL;
ret = -ENOMEM;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
@@ -429,13 +434,13 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
desc.dev = dev;
desc.groups = coresight_tmc_groups;
- desc.name = dev_name(dev);
switch (drvdata->config_type) {
case TMC_CONFIG_TYPE_ETB:
desc.type = CORESIGHT_DEV_TYPE_SINK;
desc.subtype.sink_subtype = CORESIGHT_DEV_SUBTYPE_SINK_BUFFER;
desc.ops = &tmc_etb_cs_ops;
+ dev_list = &etb_devs;
break;
case TMC_CONFIG_TYPE_ETR:
desc.type = CORESIGHT_DEV_TYPE_SINK;
@@ -447,11 +452,13 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
goto out;
idr_init(&drvdata->idr);
mutex_init(&drvdata->idr_mutex);
+ dev_list = &etr_devs;
break;
case TMC_CONFIG_TYPE_ETF:
desc.type = CORESIGHT_DEV_TYPE_LINKSINK;
desc.subtype.link_subtype = CORESIGHT_DEV_SUBTYPE_LINK_FIFO;
desc.ops = &tmc_etf_cs_ops;
+ dev_list = &etf_devs;
break;
default:
pr_err("%s: Unsupported TMC config\n", desc.name);
@@ -459,6 +466,12 @@ static int tmc_probe(struct amba_device *adev, const struct amba_id *id)
goto out;
}
+ desc.name = coresight_alloc_device_name(dev_list, dev);
+ if (!desc.name) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
pdata = coresight_get_platform_data(dev);
if (IS_ERR(pdata)) {
ret = PTR_ERR(pdata);
diff --git a/drivers/hwtracing/coresight/coresight-tpiu.c b/drivers/hwtracing/coresight/coresight-tpiu.c
index b699d613425d..f8583e4032a6 100644
--- a/drivers/hwtracing/coresight/coresight-tpiu.c
+++ b/drivers/hwtracing/coresight/coresight-tpiu.c
@@ -47,6 +47,8 @@
#define FFCR_FON_MAN BIT(6)
#define FFCR_STOP_FI BIT(12)
+DEFINE_CORESIGHT_DEVLIST(tpiu_devs, "tpiu");
+
/**
* @base: memory mapped base address for this component.
* @atclk: optional clock for the core parts of the TPIU.
@@ -125,6 +127,10 @@ static int tpiu_probe(struct amba_device *adev, const struct amba_id *id)
struct resource *res = &adev->res;
struct coresight_desc desc = { 0 };
+ desc.name = coresight_alloc_device_name(&tpiu_devs, dev);
+ if (!desc.name)
+ return -ENOMEM;
+
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
if (!drvdata)
return -ENOMEM;
@@ -157,7 +163,6 @@ static int tpiu_probe(struct amba_device *adev, const struct amba_id *id)
desc.ops = &tpiu_cs_ops;
desc.pdata = pdata;
desc.dev = dev;
- desc.name = dev_name(dev);
drvdata->csdev = coresight_register(&desc);
if (!IS_ERR(drvdata->csdev)) {
diff --git a/drivers/hwtracing/coresight/coresight.c b/drivers/hwtracing/coresight/coresight.c
index 1287778c3be5..86d1fc2c1bd4 100644
--- a/drivers/hwtracing/coresight/coresight.c
+++ b/drivers/hwtracing/coresight/coresight.c
@@ -1291,3 +1291,61 @@ void coresight_unregister(struct coresight_device *csdev)
device_unregister(&csdev->dev);
}
EXPORT_SYMBOL_GPL(coresight_unregister);
+
+
+/*
+ * coresight_search_device_idx - Search the fwnode handle of a device
+ * in the given dev_idx list. Must be called with the coresight_mutex held.
+ *
+ * Returns the index of the entry, when found. Otherwise, -ENOENT.
+ */
+static inline int coresight_search_device_idx(struct coresight_dev_list *dict,
+ struct fwnode_handle *fwnode)
+{
+ int i;
+
+ for (i = 0; i < dict->nr_idx; i++)
+ if (dict->fwnode_list[i] == fwnode)
+ return i;
+ return -ENOENT;
+}
+
+/*
+ * coresight_alloc_device_name - Get an index for a given device in the
+ * device index list specific to a driver. An index is allocated for a
+ * device and is tracked with the fwnode_handle to prevent allocating
+ * duplicate indices for the same device (e.g, if we defer probing of
+ * a device due to dependencies), in case the index is requested again.
+ */
+char *coresight_alloc_device_name(struct coresight_dev_list *dict,
+ struct device *dev)
+{
+ int idx;
+ char *name = NULL;
+ struct fwnode_handle **list;
+
+ mutex_lock(&coresight_mutex);
+
+ idx = coresight_search_device_idx(dict, dev_fwnode(dev));
+ if (idx < 0) {
+ /* Make space for the new entry */
+ idx = dict->nr_idx;
+ list = krealloc(dict->fwnode_list,
+ (idx + 1) * sizeof(*dict->fwnode_list),
+ GFP_KERNEL);
+ if (ZERO_OR_NULL_PTR(list)) {
+ idx = -ENOMEM;
+ goto done;
+ }
+
+ list[idx] = dev_fwnode(dev);
+ dict->fwnode_list = list;
+ dict->nr_idx = idx + 1;
+ }
+
+ name = devm_kasprintf(dev, GFP_KERNEL, "%s%d", dict->pfx, idx);
+done:
+ mutex_unlock(&coresight_mutex);
+ return name;
+}
+EXPORT_SYMBOL_GPL(coresight_alloc_device_name);
diff --git a/include/linux/coresight.h b/include/linux/coresight.h
index b40544bc06fe..a2b68823717b 100644
--- a/include/linux/coresight.h
+++ b/include/linux/coresight.h
@@ -168,6 +168,28 @@ struct coresight_device {
struct dev_ext_attribute *ea;
};
+/*
+ * coresight_dev_list - Mapping for devices to "name" index for device
+ * names.
+ *
+ * @nr_idx: Number of entries already allocated.
+ * @pfx: Prefix pattern for device name.
+ * @fwnode_list: Array of fwnode_handles associated with each allocated
+ * index, upto nr_idx entries.
+ */
+struct coresight_dev_list {
+ int nr_idx;
+ const char *pfx;
+ struct fwnode_handle **fwnode_list;
+};
+
+#define DEFINE_CORESIGHT_DEVLIST(var, dev_pfx) \
+static struct coresight_dev_list (var) = { \
+ .pfx = dev_pfx, \
+ .nr_idx = 0, \
+ .fwnode_list = NULL, \
+}
+
#define to_coresight_device(d) container_of(d, struct coresight_device, dev)
#define source_ops(csdev) csdev->ops->source_ops
@@ -261,7 +283,8 @@ extern int coresight_claim_device_unlocked(void __iomem *base);
extern void coresight_disclaim_device(void __iomem *base);
extern void coresight_disclaim_device_unlocked(void __iomem *base);
-
+extern char *coresight_alloc_device_name(struct coresight_dev_list *devs,
+ struct device *dev);
#else
static inline struct coresight_device *
coresight_register(struct coresight_desc *desc) { return NULL; }
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 15/28] coresight: stm: ACPI support for parsing stimulus base
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (13 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 14/28] coresight: Use platform agnostic names Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 16/28] coresight: Support for ACPI bindings Mathieu Poirier
` (13 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
The stimulus base for STM device must be listed as the second memory
resource, followed by the programming base address as described in
"Section 2.3 Resources" in ACPI for CoreSightTM 1.0 Platform Design
documen (DEN0067).
Add support for parsing the information for ACPI.
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-stm.c | 53 ++++++++++++++++++++-
1 file changed, 52 insertions(+), 1 deletion(-)
diff --git a/drivers/hwtracing/coresight/coresight-stm.c b/drivers/hwtracing/coresight/coresight-stm.c
index e3e2b000cfb7..b908ca104645 100644
--- a/drivers/hwtracing/coresight/coresight-stm.c
+++ b/drivers/hwtracing/coresight/coresight-stm.c
@@ -16,6 +16,7 @@
* (C) 2015-2016 Chunyan Zhang <zhang.chunyan@linaro.org>
*/
#include <asm/local.h>
+#include <linux/acpi.h>
#include <linux/amba/bus.h>
#include <linux/bitmap.h>
#include <linux/clk.h>
@@ -716,10 +717,60 @@ static inline int of_stm_get_stimulus_area(struct device *dev,
}
#endif
+#ifdef CONFIG_ACPI
+static int acpi_stm_get_stimulus_area(struct device *dev, struct resource *res)
+{
+ int rc;
+ bool found_base = false;
+ struct resource_entry *rent;
+ LIST_HEAD(res_list);
+
+ struct acpi_device *adev = ACPI_COMPANION(dev);
+
+ if (!adev)
+ return -ENODEV;
+ rc = acpi_dev_get_resources(adev, &res_list, NULL, NULL);
+ if (rc < 0)
+ return rc;
+
+ /*
+ * The stimulus base for STM device must be listed as the second memory
+ * resource, followed by the programming base address as described in
+ * "Section 2.3 Resources" in ACPI for CoreSightTM 1.0 Platform Design
+ * document (DEN0067).
+ */
+ rc = -ENOENT;
+ list_for_each_entry(rent, &res_list, node) {
+ if (resource_type(rent->res) != IORESOURCE_MEM)
+ continue;
+ if (found_base) {
+ *res = *rent->res;
+ rc = 0;
+ break;
+ }
+
+ found_base = true;
+ }
+
+ acpi_dev_free_resource_list(&res_list);
+ return rc;
+}
+#else
+static inline int acpi_stm_get_stimulus_area(struct device *dev,
+ struct resource *res)
+{
+ return -ENOENT;
+}
+#endif
+
static int stm_get_stimulus_area(struct device *dev, struct resource *res)
{
- if (is_of_node(dev_fwnode(dev)))
+ struct fwnode_handle *fwnode = dev_fwnode(dev);
+
+ if (is_of_node(fwnode))
return of_stm_get_stimulus_area(dev, res);
+ else if (is_acpi_node(fwnode))
+ return acpi_stm_get_stimulus_area(dev, res);
return -ENOENT;
}
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 16/28] coresight: Support for ACPI bindings
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (14 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 15/28] coresight: stm: ACPI support for parsing stimulus base Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 17/28] coresight: acpi: Support for AMBA components Mathieu Poirier
` (12 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
Add support for parsing the ACPI platform description
for CoreSight. The connections are encoded in a DSD graph
property with CoreSight specific variation of the property.
The ETMs are listed as the children device of the respective
CPU.
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
.../hwtracing/coresight/coresight-platform.c | 464 ++++++++++++++++++
1 file changed, 464 insertions(+)
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index 49112a58478e..6fef8732eea5 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -3,6 +3,7 @@
* Copyright (c) 2012, The Linux Foundation. All rights reserved.
*/
+#include <linux/acpi.h>
#include <linux/types.h>
#include <linux/err.h>
#include <linux/slab.h>
@@ -308,10 +309,471 @@ of_get_coresight_platform_data(struct device *dev,
}
#endif
+#ifdef CONFIG_ACPI
+
+#include <acpi/actypes.h>
+#include <acpi/processor.h>
+
+/* ACPI Graph _DSD UUID : "ab02a46b-74c7-45a2-bd68-f7d344ef2153" */
+static const guid_t acpi_graph_uuid = GUID_INIT(0xab02a46b, 0x74c7, 0x45a2,
+ 0xbd, 0x68, 0xf7, 0xd3,
+ 0x44, 0xef, 0x21, 0x53);
+/* Coresight ACPI Graph UUID : "3ecbc8b6-1d0e-4fb3-8107-e627f805c6cd" */
+static const guid_t coresight_graph_uuid = GUID_INIT(0x3ecbc8b6, 0x1d0e, 0x4fb3,
+ 0x81, 0x07, 0xe6, 0x27,
+ 0xf8, 0x05, 0xc6, 0xcd);
+#define ACPI_CORESIGHT_LINK_SLAVE 0
+#define ACPI_CORESIGHT_LINK_MASTER 1
+
+static inline bool is_acpi_guid(const union acpi_object *obj)
+{
+ return (obj->type == ACPI_TYPE_BUFFER) && (obj->buffer.length == 16);
+}
+
+/*
+ * acpi_guid_matches - Checks if the given object is a GUID object and
+ * that it matches the supplied the GUID.
+ */
+static inline bool acpi_guid_matches(const union acpi_object *obj,
+ const guid_t *guid)
+{
+ return is_acpi_guid(obj) &&
+ guid_equal((guid_t *)obj->buffer.pointer, guid);
+}
+
+static inline bool is_acpi_dsd_graph_guid(const union acpi_object *obj)
+{
+ return acpi_guid_matches(obj, &acpi_graph_uuid);
+}
+
+static inline bool is_acpi_coresight_graph_guid(const union acpi_object *obj)
+{
+ return acpi_guid_matches(obj, &coresight_graph_uuid);
+}
+
+static inline bool is_acpi_coresight_graph(const union acpi_object *obj)
+{
+ const union acpi_object *graphid, *guid, *links;
+
+ if (obj->type != ACPI_TYPE_PACKAGE ||
+ obj->package.count < 3)
+ return false;
+
+ graphid = &obj->package.elements[0];
+ guid = &obj->package.elements[1];
+ links = &obj->package.elements[2];
+
+ if (graphid->type != ACPI_TYPE_INTEGER ||
+ links->type != ACPI_TYPE_INTEGER)
+ return false;
+
+ return is_acpi_coresight_graph_guid(guid);
+}
+
+/*
+ * acpi_validate_dsd_graph - Make sure the given _DSD graph conforms
+ * to the ACPI _DSD Graph specification.
+ *
+ * ACPI Devices Graph property has the following format:
+ * {
+ * Revision - Integer, must be 0
+ * NumberOfGraphs - Integer, N indicating the following list.
+ * Graph[1],
+ * ...
+ * Graph[N]
+ * }
+ *
+ * And each Graph entry has the following format:
+ * {
+ * GraphID - Integer, identifying a graph the device belongs to.
+ * UUID - UUID identifying the specification that governs
+ * this graph. (e.g, see is_acpi_coresight_graph())
+ * NumberOfLinks - Number "N" of connections on this node of the graph.
+ * Links[1]
+ * ...
+ * Links[N]
+ * }
+ *
+ * Where each "Links" entry has the following format:
+ *
+ * {
+ * SourcePortAddress - Integer
+ * DestinationPortAddress - Integer
+ * DestinationDeviceName - Reference to another device
+ * ( --- CoreSight specific extensions below ---)
+ * DirectionOfFlow - Integer 1 for output(master)
+ * 0 for input(slave)
+ * }
+ *
+ * e.g:
+ * For a Funnel device
+ *
+ * Device(MFUN) {
+ * ...
+ *
+ * Name (_DSD, Package() {
+ * // DSD Package contains tuples of { Proeprty_Type_UUID, Package() }
+ * ToUUID("daffd814-6eba-4d8c-8a91-bc9bbf4aa301"), //Std. Property UUID
+ * Package() {
+ * Package(2) { "property-name", <property-value> }
+ * },
+ *
+ * ToUUID("ab02a46b-74c7-45a2-bd68-f7d344ef2153"), // ACPI Graph UUID
+ * Package() {
+ * 0, // Revision
+ * 1, // NumberOfGraphs.
+ * Package() { // Graph[0] Package
+ * 1, // GraphID
+ * // Coresight Graph UUID
+ * ToUUID("3ecbc8b6-1d0e-4fb3-8107-e627f805c6cd"),
+ * 3, // NumberOfLinks aka ports
+ * // Link[0]: Output_0 -> Replicator:Input_0
+ * Package () { 0, 0, \_SB_.RPL0, 1 },
+ * // Link[1]: Input_0 <- Cluster0_Funnel0:Output_0
+ * Package () { 0, 0, \_SB_.CLU0.FUN0, 0 },
+ * // Link[2]: Input_1 <- Cluster1_Funnel0:Output_0
+ * Package () { 1, 0, \_SB_.CLU1.FUN0, 0 },
+ * } // End of Graph[0] Package
+ *
+ * }, // End of ACPI Graph Property
+ * })
+ */
+static inline bool acpi_validate_dsd_graph(const union acpi_object *graph)
+{
+ int i, n;
+ const union acpi_object *rev, *nr_graphs;
+
+ /* The graph must contain at least the Revision and Number of Graphs */
+ if (graph->package.count < 2)
+ return false;
+
+ rev = &graph->package.elements[0];
+ nr_graphs = &graph->package.elements[1];
+
+ if (rev->type != ACPI_TYPE_INTEGER ||
+ nr_graphs->type != ACPI_TYPE_INTEGER)
+ return false;
+
+ /* We only support revision 0 */
+ if (rev->integer.value != 0)
+ return false;
+
+ n = nr_graphs->integer.value;
+ /* CoreSight devices are only part of a single Graph */
+ if (n != 1)
+ return false;
+
+ /* Make sure the ACPI graph package has right number of elements */
+ if (graph->package.count != (n + 2))
+ return false;
+
+ /*
+ * Each entry must be a graph package with at least 3 members :
+ * { GraphID, UUID, NumberOfLinks(n), Links[.],... }
+ */
+ for (i = 2; i < n + 2; i++) {
+ const union acpi_object *obj = &graph->package.elements[i];
+
+ if (obj->type != ACPI_TYPE_PACKAGE ||
+ obj->package.count < 3)
+ return false;
+ }
+
+ return true;
+}
+
+/* acpi_get_dsd_graph - Find the _DSD Graph property for the given device. */
+const union acpi_object *
+acpi_get_dsd_graph(struct acpi_device *adev)
+{
+ int i;
+ struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER };
+ acpi_status status;
+ const union acpi_object *dsd;
+
+ status = acpi_evaluate_object_typed(adev->handle, "_DSD", NULL,
+ &buf, ACPI_TYPE_PACKAGE);
+ if (ACPI_FAILURE(status))
+ return NULL;
+
+ dsd = buf.pointer;
+
+ /*
+ * _DSD property consists tuples { Prop_UUID, Package() }
+ * Iterate through all the packages and find the Graph.
+ */
+ for (i = 0; i + 1 < dsd->package.count; i += 2) {
+ const union acpi_object *guid, *package;
+
+ guid = &dsd->package.elements[i];
+ package = &dsd->package.elements[i + 1];
+
+ /* All _DSD elements must have a UUID and a Package */
+ if (!is_acpi_guid(guid) || package->type != ACPI_TYPE_PACKAGE)
+ break;
+ /* Skip the non-Graph _DSD packages */
+ if (!is_acpi_dsd_graph_guid(guid))
+ continue;
+ if (acpi_validate_dsd_graph(package))
+ return package;
+ /* Invalid graph format, continue */
+ dev_warn(&adev->dev, "Invalid Graph _DSD property\n");
+ }
+
+ return NULL;
+}
+
+static inline bool
+acpi_validate_coresight_graph(const union acpi_object *cs_graph)
+{
+ int nlinks;
+
+ nlinks = cs_graph->package.elements[2].integer.value;
+ /*
+ * Graph must have the following fields :
+ * { GraphID, GraphUUID, NumberOfLinks, Links... }
+ */
+ if (cs_graph->package.count != (nlinks + 3))
+ return false;
+ /* The links are validated in acpi_coresight_parse_link() */
+ return true;
+}
+
+/*
+ * acpi_get_coresight_graph - Parse the device _DSD tables and find
+ * the Graph property matching the CoreSight Graphs.
+ *
+ * Returns the pointer to the CoreSight Graph Package when found. Otherwise
+ * returns NULL.
+ */
+const union acpi_object *
+acpi_get_coresight_graph(struct acpi_device *adev)
+{
+ const union acpi_object *graph_list, *graph;
+ int i, nr_graphs;
+
+ graph_list = acpi_get_dsd_graph(adev);
+ if (!graph_list)
+ return graph_list;
+
+ nr_graphs = graph_list->package.elements[1].integer.value;
+
+ for (i = 2; i < nr_graphs + 2; i++) {
+ graph = &graph_list->package.elements[i];
+ if (!is_acpi_coresight_graph(graph))
+ continue;
+ if (acpi_validate_coresight_graph(graph))
+ return graph;
+ /* Invalid graph format */
+ break;
+ }
+
+ return NULL;
+}
+
+/*
+ * acpi_coresight_parse_link - Parse the given Graph connection
+ * of the device and populate the coresight_connection for an output
+ * connection.
+ *
+ * CoreSight Graph specification mandates that the direction of the data
+ * flow must be specified in the link. i.e,
+ *
+ * SourcePortAddress, // Integer
+ * DestinationPortAddress, // Integer
+ * DestinationDeviceName, // Reference to another device
+ * DirectionOfFlow, // 1 for output(master), 0 for input(slave)
+ *
+ * Returns the direction of the data flow [ Input(slave) or Output(master) ]
+ * upon success.
+ * Returns an negative error number otherwise.
+ */
+static int acpi_coresight_parse_link(struct acpi_device *adev,
+ const union acpi_object *link,
+ struct coresight_connection *conn)
+{
+ int rc, dir;
+ const union acpi_object *fields;
+ struct acpi_device *r_adev;
+ struct device *rdev;
+
+ if (link->type != ACPI_TYPE_PACKAGE ||
+ link->package.count != 4)
+ return -EINVAL;
+
+ fields = link->package.elements;
+
+ if (fields[0].type != ACPI_TYPE_INTEGER ||
+ fields[1].type != ACPI_TYPE_INTEGER ||
+ fields[2].type != ACPI_TYPE_LOCAL_REFERENCE ||
+ fields[3].type != ACPI_TYPE_INTEGER)
+ return -EINVAL;
+
+ rc = acpi_bus_get_device(fields[2].reference.handle, &r_adev);
+ if (rc)
+ return rc;
+
+ dir = fields[3].integer.value;
+ if (dir == ACPI_CORESIGHT_LINK_MASTER) {
+ conn->outport = fields[0].integer.value;
+ conn->child_port = fields[1].integer.value;
+ rdev = coresight_find_device_by_fwnode(&r_adev->fwnode);
+ if (!rdev)
+ return -EPROBE_DEFER;
+ /*
+ * Hold the refcount to the target device. This could be
+ * released via:
+ * 1) coresight_release_platform_data() if the probe fails or
+ * this device is unregistered.
+ * 2) While removing the target device via
+ * coresight_remove_match().
+ */
+ conn->child_fwnode = fwnode_handle_get(&r_adev->fwnode);
+ }
+
+ return dir;
+}
+
+/*
+ * acpi_coresight_parse_graph - Parse the _DSD CoreSight graph
+ * connection information and populate the supplied coresight_platform_data
+ * instance.
+ */
+static int acpi_coresight_parse_graph(struct acpi_device *adev,
+ struct coresight_platform_data *pdata)
+{
+ int rc, i, nlinks;
+ const union acpi_object *graph;
+ struct coresight_connection *conns, *ptr;
+
+ pdata->nr_inport = pdata->nr_outport = 0;
+ graph = acpi_get_coresight_graph(adev);
+ if (!graph)
+ return -ENOENT;
+
+ nlinks = graph->package.elements[2].integer.value;
+ if (!nlinks)
+ return 0;
+
+ /*
+ * To avoid scanning the table twice (once for finding the number of
+ * output links and then later for parsing the output links),
+ * cache the links information in one go and then later copy
+ * it to the pdata.
+ */
+ conns = devm_kcalloc(&adev->dev, nlinks, sizeof(*conns), GFP_KERNEL);
+ if (!conns)
+ return -ENOMEM;
+ ptr = conns;
+ for (i = 0; i < nlinks; i++) {
+ const union acpi_object *link = &graph->package.elements[3 + i];
+ int dir;
+
+ dir = acpi_coresight_parse_link(adev, link, ptr);
+ if (dir < 0)
+ return dir;
+
+ if (dir == ACPI_CORESIGHT_LINK_MASTER) {
+ pdata->nr_outport++;
+ ptr++;
+ } else {
+ pdata->nr_inport++;
+ }
+ }
+
+ rc = coresight_alloc_conns(&adev->dev, pdata);
+ if (rc)
+ return rc;
+
+ /* Copy the connection information to the final location */
+ for (i = 0; i < pdata->nr_outport; i++)
+ pdata->conns[i] = conns[i];
+
+ devm_kfree(&adev->dev, conns);
+ return 0;
+}
+
+/*
+ * acpi_handle_to_logical_cpuid - Map a given acpi_handle to the
+ * logical CPU id of the corresponding CPU device.
+ *
+ * Returns the logical CPU id when found. Otherwise returns >= nr_cpus_id.
+ */
+static int
+acpi_handle_to_logical_cpuid(acpi_handle handle)
+{
+ int i;
+ struct acpi_processor *pr;
+
+ for_each_possible_cpu(i) {
+ pr = per_cpu(processors, i);
+ if (pr && pr->handle == handle)
+ break;
+ }
+
+ return i;
+}
+
+/*
+ * acpi_coresigh_get_cpu - Find the logical CPU id of the CPU associated
+ * with this coresight device. With ACPI bindings, the CoreSight components
+ * are listed as child device of the associated CPU.
+ *
+ * Returns the logical CPU id when found. Otherwise returns 0.
+ */
+static int acpi_coresight_get_cpu(struct device *dev)
+{
+ int cpu;
+ acpi_handle cpu_handle;
+ acpi_status status;
+ struct acpi_device *adev = ACPI_COMPANION(dev);
+
+ if (!adev)
+ return 0;
+ status = acpi_get_parent(adev->handle, &cpu_handle);
+ if (ACPI_FAILURE(status))
+ return 0;
+
+ cpu = acpi_handle_to_logical_cpuid(cpu_handle);
+ if (cpu >= nr_cpu_ids)
+ return 0;
+ return cpu;
+}
+
+static int
+acpi_get_coresight_platform_data(struct device *dev,
+ struct coresight_platform_data *pdata)
+{
+ struct acpi_device *adev;
+
+ adev = ACPI_COMPANION(dev);
+ if (!adev)
+ return -EINVAL;
+
+ return acpi_coresight_parse_graph(adev, pdata);
+}
+
+#else
+
+static inline int
+acpi_get_coresight_platform_data(struct device *dev,
+ struct coresight_platform_data *pdata)
+{
+ return -ENOENT;
+}
+
+static inline int acpi_coresight_get_cpu(struct device *dev)
+{
+ return 0;
+}
+#endif
+
int coresight_get_cpu(struct device *dev)
{
if (is_of_node(dev->fwnode))
return of_coresight_get_cpu(dev);
+ else if (is_acpi_device_node(dev->fwnode))
+ return acpi_coresight_get_cpu(dev);
return 0;
}
EXPORT_SYMBOL_GPL(coresight_get_cpu);
@@ -334,6 +796,8 @@ coresight_get_platform_data(struct device *dev)
if (is_of_node(fwnode))
ret = of_get_coresight_platform_data(dev, pdata);
+ else if (is_acpi_device_node(fwnode))
+ ret = acpi_get_coresight_platform_data(dev, pdata);
if (!ret)
return pdata;
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 17/28] coresight: acpi: Support for AMBA components
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (15 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 16/28] coresight: Support for ACPI bindings Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 18/28] coresight: acpi: Support for platform devices Mathieu Poirier
` (11 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
All AMBA devices are handled via ACPI AMBA scan notifier
infrastructure. The platform devices get the ACPI id
added to their driver.
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/acpi/acpi_amba.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/drivers/acpi/acpi_amba.c b/drivers/acpi/acpi_amba.c
index 7f77c071709a..e2142e3e0ceb 100644
--- a/drivers/acpi/acpi_amba.c
+++ b/drivers/acpi/acpi_amba.c
@@ -24,6 +24,15 @@
static const struct acpi_device_id amba_id_list[] = {
{"ARMH0061", 0}, /* PL061 GPIO Device */
+ {"ARMHC500", 0}, /* ARM CoreSight ETM4x */
+ {"ARMHC501", 0}, /* ARM CoreSight ETR */
+ {"ARMHC502", 0}, /* ARM CoreSight STM */
+ {"ARMHC503", 0}, /* ARM CoreSight Debug */
+ {"ARMHC979", 0}, /* ARM CoreSight TPIU */
+ {"ARMHC97C", 0}, /* ARM CoreSight SoC-400 TMC, SoC-600 ETF/ETB */
+ {"ARMHC98D", 0}, /* ARM CoreSight Dynamic Replicator */
+ {"ARMHC9CA", 0}, /* ARM CoreSight CATU */
+ {"ARMHC9FF", 0}, /* ARM CoreSight Dynamic Funnel */
{"", 0},
};
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 18/28] coresight: acpi: Support for platform devices
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (16 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 17/28] coresight: acpi: Support for AMBA components Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 19/28] coresight: Add dummy definition for of_coresight_get_cpu() Mathieu Poirier
` (10 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
Add support for platform devices which do not appear on the AMBA
bus.
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-replicator.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/drivers/hwtracing/coresight/coresight-replicator.c b/drivers/hwtracing/coresight/coresight-replicator.c
index c0e42253dfe7..542952759941 100644
--- a/drivers/hwtracing/coresight/coresight-replicator.c
+++ b/drivers/hwtracing/coresight/coresight-replicator.c
@@ -5,6 +5,7 @@
* Description: CoreSight Replicator driver
*/
+#include <linux/acpi.h>
#include <linux/amba/bus.h>
#include <linux/kernel.h>
#include <linux/device.h>
@@ -296,11 +297,18 @@ static const struct of_device_id static_replicator_match[] = {
{}
};
+#ifdef CONFIG_ACPI
+static const struct acpi_device_id static_replicator_acpi_ids[] = {
+ {"ARMHC985", 0}, /* ARM CoreSight Static Replicator */
+};
+#endif
+
static struct platform_driver static_replicator_driver = {
.probe = static_replicator_probe,
.driver = {
.name = "coresight-static-replicator",
- .of_match_table = static_replicator_match,
+ .of_match_table = of_match_ptr(static_replicator_match),
+ .acpi_match_table = ACPI_PTR(static_replicator_acpi_ids),
.pm = &replicator_dev_pm_ops,
.suppress_bind_attrs = true,
},
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 19/28] coresight: Add dummy definition for of_coresight_get_cpu()
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (17 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 18/28] coresight: acpi: Support for platform devices Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 20/28] coresight: tmc-etr: Do not call smp_processor_id() from preemptible Mathieu Poirier
` (9 subsequent siblings)
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
For !CONFIG_OF case, add a dummy definition for the of_coresight_get_cpu().
Reported-by: Mike Leach <mike.leach@linaro.org>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-platform.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
index 6fef8732eea5..3c5ceda8db24 100644
--- a/drivers/hwtracing/coresight/coresight-platform.c
+++ b/drivers/hwtracing/coresight/coresight-platform.c
@@ -307,6 +307,11 @@ of_get_coresight_platform_data(struct device *dev,
{
return -ENOENT;
}
+
+static inline int of_coresight_get_cpu(struct device *dev)
+{
+ return 0;
+}
#endif
#ifdef CONFIG_ACPI
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 20/28] coresight: tmc-etr: Do not call smp_processor_id() from preemptible
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (18 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 19/28] coresight: Add dummy definition for of_coresight_get_cpu() Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-20 6:02 ` Greg KH
2019-06-19 19:53 ` [PATCH 21/28] coresight: tmc-etr: alloc_perf_buf: Do not call smp_processor_id " Mathieu Poirier
` (8 subsequent siblings)
28 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
During a perf session we try to allocate buffers on the "node" associated
with the CPU the event is bound to. If it's not bound to a CPU, we use
the current CPU node, using smp_processor_id(). However this is unsafe
in a pre-emptible context and could generate the splats as below :
BUG: using smp_processor_id() in preemptible [00000000] code: perf/1743
caller is alloc_etr_buf.isra.6+0x80/0xa0
CPU: 1 PID: 1743 Comm: perf Not tainted 5.1.0-rc6-147786-g116841e #344
Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform, BIOS EDK II Feb 1 2019
Call trace:
dump_backtrace+0x0/0x150
show_stack+0x14/0x20
dump_stack+0x9c/0xc4
debug_smp_processor_id+0x10c/0x110
alloc_etr_buf.isra.6+0x80/0xa0
tmc_alloc_etr_buffer+0x12c/0x1f0
etm_setup_aux+0x1c4/0x230
rb_alloc_aux+0x1b8/0x2b8
perf_mmap+0x35c/0x478
mmap_region+0x34c/0x4f0
do_mmap+0x2d8/0x418
vm_mmap_pgoff+0xd0/0xf8
ksys_mmap_pgoff+0x88/0xf8
__arm64_sys_mmap+0x28/0x38
el0_svc_handler+0xd8/0x138
el0_svc+0x8/0xc
Use NUMA_NO_NODE hint instead of using the current node for events
not bound to CPUs.
Fixes: 855ab61c16bf70b646 ("coresight: tmc-etr: Refactor function tmc_etr_setup_perf_buf()")
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-tmc-etr.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
index ce0114a5435c..7c81f634ecb4 100644
--- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
+++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
@@ -1323,13 +1323,11 @@ static struct etr_perf_buffer *
tmc_etr_setup_perf_buf(struct tmc_drvdata *drvdata, struct perf_event *event,
int nr_pages, void **pages, bool snapshot)
{
- int node, cpu = event->cpu;
+ int node;
struct etr_buf *etr_buf;
struct etr_perf_buffer *etr_perf;
- if (cpu == -1)
- cpu = smp_processor_id();
- node = cpu_to_node(cpu);
+ node = (event->cpu == -1) ? NUMA_NO_NODE : cpu_to_node(event->cpu);
etr_perf = kzalloc_node(sizeof(*etr_perf), GFP_KERNEL, node);
if (!etr_perf)
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* Re: [PATCH 20/28] coresight: tmc-etr: Do not call smp_processor_id() from preemptible
2019-06-19 19:53 ` [PATCH 20/28] coresight: tmc-etr: Do not call smp_processor_id() from preemptible Mathieu Poirier
@ 2019-06-20 6:02 ` Greg KH
0 siblings, 0 replies; 41+ messages in thread
From: Greg KH @ 2019-06-20 6:02 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Wed, Jun 19, 2019 at 01:53:10PM -0600, Mathieu Poirier wrote:
> From: Suzuki K Poulose <suzuki.poulose@arm.com>
>
> During a perf session we try to allocate buffers on the "node" associated
> with the CPU the event is bound to. If it's not bound to a CPU, we use
> the current CPU node, using smp_processor_id(). However this is unsafe
> in a pre-emptible context and could generate the splats as below :
>
> BUG: using smp_processor_id() in preemptible [00000000] code: perf/1743
> caller is alloc_etr_buf.isra.6+0x80/0xa0
> CPU: 1 PID: 1743 Comm: perf Not tainted 5.1.0-rc6-147786-g116841e #344
> Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform, BIOS EDK II Feb 1 2019
> Call trace:
> dump_backtrace+0x0/0x150
> show_stack+0x14/0x20
> dump_stack+0x9c/0xc4
> debug_smp_processor_id+0x10c/0x110
> alloc_etr_buf.isra.6+0x80/0xa0
> tmc_alloc_etr_buffer+0x12c/0x1f0
> etm_setup_aux+0x1c4/0x230
> rb_alloc_aux+0x1b8/0x2b8
> perf_mmap+0x35c/0x478
> mmap_region+0x34c/0x4f0
> do_mmap+0x2d8/0x418
> vm_mmap_pgoff+0xd0/0xf8
> ksys_mmap_pgoff+0x88/0xf8
> __arm64_sys_mmap+0x28/0x38
> el0_svc_handler+0xd8/0x138
> el0_svc+0x8/0xc
>
> Use NUMA_NO_NODE hint instead of using the current node for events
> not bound to CPUs.
>
> Fixes: 855ab61c16bf70b646 ("coresight: tmc-etr: Refactor function tmc_etr_setup_perf_buf()")
> Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> ---
> drivers/hwtracing/coresight/coresight-tmc-etr.c | 6 ++----
> 1 file changed, 2 insertions(+), 4 deletions(-)
Why isn't this being submitted for 5.2-final? Why wait until 5.3 to fix
this?
thanks,
greg k-h
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread
* [PATCH 21/28] coresight: tmc-etr: alloc_perf_buf: Do not call smp_processor_id from preemptible
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (19 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 20/28] coresight: tmc-etr: Do not call smp_processor_id() from preemptible Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-20 6:03 ` Greg KH
2019-06-19 19:53 ` [PATCH 22/28] coresight: tmc-etf: " Mathieu Poirier
` (7 subsequent siblings)
28 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
During a perf session we try to allocate buffers on the "node" associated
with the CPU the event is bound to. If it is not bound to a CPU, we
use the current CPU node, using smp_processor_id(). However this is unsafe
in a pre-emptible context and could generate the splats as below :
BUG: using smp_processor_id() in preemptible [00000000] code: perf/1743
caller is tmc_alloc_etr_buffer+0x1bc/0x1f0
CPU: 1 PID: 1743 Comm: perf Not tainted 5.1.0-rc6-147786-g116841e #344
Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform, BIOS EDK II Feb 1 2019
Call trace:
dump_backtrace+0x0/0x150
show_stack+0x14/0x20
dump_stack+0x9c/0xc4
debug_smp_processor_id+0x10c/0x110
tmc_alloc_etr_buffer+0x1bc/0x1f0
etm_setup_aux+0x1c4/0x230
rb_alloc_aux+0x1b8/0x2b8
perf_mmap+0x35c/0x478
mmap_region+0x34c/0x4f0
do_mmap+0x2d8/0x418
vm_mmap_pgoff+0xd0/0xf8
ksys_mmap_pgoff+0x88/0xf8
__arm64_sys_mmap+0x28/0x38
el0_svc_handler+0xd8/0x138
el0_svc+0x8/0xc
Use NUMA_NO_NODE hint instead of using the current node for events
not bound to CPUs.
Fixes: 22f429f19c4135d51e9 ("coresight: etm-perf: Add support for ETR backend")
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-tmc-etr.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
index 7c81f634ecb4..5d2bf6d18961 100644
--- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
+++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
@@ -1184,14 +1184,11 @@ static struct etr_buf *
alloc_etr_buf(struct tmc_drvdata *drvdata, struct perf_event *event,
int nr_pages, void **pages, bool snapshot)
{
- int node, cpu = event->cpu;
+ int node;
struct etr_buf *etr_buf;
unsigned long size;
- if (cpu == -1)
- cpu = smp_processor_id();
- node = cpu_to_node(cpu);
-
+ node = (event->cpu == -1) ? NUMA_NO_NODE : cpu_to_node(event->cpu);
/*
* Try to match the perf ring buffer size if it is larger
* than the size requested via sysfs.
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* Re: [PATCH 21/28] coresight: tmc-etr: alloc_perf_buf: Do not call smp_processor_id from preemptible
2019-06-19 19:53 ` [PATCH 21/28] coresight: tmc-etr: alloc_perf_buf: Do not call smp_processor_id " Mathieu Poirier
@ 2019-06-20 6:03 ` Greg KH
0 siblings, 0 replies; 41+ messages in thread
From: Greg KH @ 2019-06-20 6:03 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Wed, Jun 19, 2019 at 01:53:11PM -0600, Mathieu Poirier wrote:
> From: Suzuki K Poulose <suzuki.poulose@arm.com>
>
> During a perf session we try to allocate buffers on the "node" associated
> with the CPU the event is bound to. If it is not bound to a CPU, we
> use the current CPU node, using smp_processor_id(). However this is unsafe
> in a pre-emptible context and could generate the splats as below :
>
> BUG: using smp_processor_id() in preemptible [00000000] code: perf/1743
> caller is tmc_alloc_etr_buffer+0x1bc/0x1f0
> CPU: 1 PID: 1743 Comm: perf Not tainted 5.1.0-rc6-147786-g116841e #344
> Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform, BIOS EDK II Feb 1 2019
> Call trace:
> dump_backtrace+0x0/0x150
> show_stack+0x14/0x20
> dump_stack+0x9c/0xc4
> debug_smp_processor_id+0x10c/0x110
> tmc_alloc_etr_buffer+0x1bc/0x1f0
> etm_setup_aux+0x1c4/0x230
> rb_alloc_aux+0x1b8/0x2b8
> perf_mmap+0x35c/0x478
> mmap_region+0x34c/0x4f0
> do_mmap+0x2d8/0x418
> vm_mmap_pgoff+0xd0/0xf8
> ksys_mmap_pgoff+0x88/0xf8
> __arm64_sys_mmap+0x28/0x38
> el0_svc_handler+0xd8/0x138
> el0_svc+0x8/0xc
>
> Use NUMA_NO_NODE hint instead of using the current node for events
> not bound to CPUs.
>
> Fixes: 22f429f19c4135d51e9 ("coresight: etm-perf: Add support for ETR backend")
> Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Same here, why isn't this 5.2-final and stable@vger.kernel.org
material?
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread
* [PATCH 22/28] coresight: tmc-etf: Do not call smp_processor_id from preemptible
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (20 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 21/28] coresight: tmc-etr: alloc_perf_buf: Do not call smp_processor_id " Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-20 6:03 ` Greg KH
2019-06-19 19:53 ` [PATCH 23/28] coresight: etb10: " Mathieu Poirier
` (6 subsequent siblings)
28 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
During a perf session we try to allocate buffers on the "node" associated
with the CPU the event is bound to. If it is not bound to a CPU, we
use the current CPU node, using smp_processor_id(). However this is unsafe
in a pre-emptible context and could generate the splats as below :
BUG: using smp_processor_id() in preemptible [00000000] code: perf/2544
caller is tmc_alloc_etf_buffer+0x5c/0x60
CPU: 2 PID: 2544 Comm: perf Not tainted 5.1.0-rc6-147786-g116841e #344
Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform, BIOS EDK II Feb 1 2019
Call trace:
dump_backtrace+0x0/0x150
show_stack+0x14/0x20
dump_stack+0x9c/0xc4
debug_smp_processor_id+0x10c/0x110
tmc_alloc_etf_buffer+0x5c/0x60
etm_setup_aux+0x1c4/0x230
rb_alloc_aux+0x1b8/0x2b8
perf_mmap+0x35c/0x478
mmap_region+0x34c/0x4f0
do_mmap+0x2d8/0x418
vm_mmap_pgoff+0xd0/0xf8
ksys_mmap_pgoff+0x88/0xf8
__arm64_sys_mmap+0x28/0x38
el0_svc_handler+0xd8/0x138
el0_svc+0x8/0xc
Use NUMA_NO_NODE hint instead of using the current node for events
not bound to CPUs.
Fixes: 2e499bbc1a929ac ("coresight: tmc: implementing TMC-ETF AUX space API")
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-tmc-etf.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-tmc-etf.c b/drivers/hwtracing/coresight/coresight-tmc-etf.c
index b89e29c5b39d..23b7ff00af5c 100644
--- a/drivers/hwtracing/coresight/coresight-tmc-etf.c
+++ b/drivers/hwtracing/coresight/coresight-tmc-etf.c
@@ -377,12 +377,10 @@ static void *tmc_alloc_etf_buffer(struct coresight_device *csdev,
struct perf_event *event, void **pages,
int nr_pages, bool overwrite)
{
- int node, cpu = event->cpu;
+ int node;
struct cs_buffers *buf;
- if (cpu == -1)
- cpu = smp_processor_id();
- node = cpu_to_node(cpu);
+ node = (event->cpu == -1) ? NUMA_NO_NODE : cpu_to_node(event->cpu);
/* Allocate memory structure for interaction with Perf */
buf = kzalloc_node(sizeof(struct cs_buffers), GFP_KERNEL, node);
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* Re: [PATCH 22/28] coresight: tmc-etf: Do not call smp_processor_id from preemptible
2019-06-19 19:53 ` [PATCH 22/28] coresight: tmc-etf: " Mathieu Poirier
@ 2019-06-20 6:03 ` Greg KH
0 siblings, 0 replies; 41+ messages in thread
From: Greg KH @ 2019-06-20 6:03 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Wed, Jun 19, 2019 at 01:53:12PM -0600, Mathieu Poirier wrote:
> From: Suzuki K Poulose <suzuki.poulose@arm.com>
>
> During a perf session we try to allocate buffers on the "node" associated
> with the CPU the event is bound to. If it is not bound to a CPU, we
> use the current CPU node, using smp_processor_id(). However this is unsafe
> in a pre-emptible context and could generate the splats as below :
>
> BUG: using smp_processor_id() in preemptible [00000000] code: perf/2544
> caller is tmc_alloc_etf_buffer+0x5c/0x60
> CPU: 2 PID: 2544 Comm: perf Not tainted 5.1.0-rc6-147786-g116841e #344
> Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform, BIOS EDK II Feb 1 2019
> Call trace:
> dump_backtrace+0x0/0x150
> show_stack+0x14/0x20
> dump_stack+0x9c/0xc4
> debug_smp_processor_id+0x10c/0x110
> tmc_alloc_etf_buffer+0x5c/0x60
> etm_setup_aux+0x1c4/0x230
> rb_alloc_aux+0x1b8/0x2b8
> perf_mmap+0x35c/0x478
> mmap_region+0x34c/0x4f0
> do_mmap+0x2d8/0x418
> vm_mmap_pgoff+0xd0/0xf8
> ksys_mmap_pgoff+0x88/0xf8
> __arm64_sys_mmap+0x28/0x38
> el0_svc_handler+0xd8/0x138
> el0_svc+0x8/0xc
>
> Use NUMA_NO_NODE hint instead of using the current node for events
> not bound to CPUs.
>
> Fixes: 2e499bbc1a929ac ("coresight: tmc: implementing TMC-ETF AUX space API")
> Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> ---
> drivers/hwtracing/coresight/coresight-tmc-etf.c | 6 ++----
> 1 file changed, 2 insertions(+), 4 deletions(-)
And again, should go into 5.2-final and stable.
thanks,
greg k-h
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread
* [PATCH 23/28] coresight: etb10: Do not call smp_processor_id from preemptible
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (21 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 22/28] coresight: tmc-etf: " Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-20 6:04 ` Greg KH
2019-06-19 19:53 ` [PATCH 24/28] coresight: Potential uninitialized variable in probe() Mathieu Poirier
` (5 subsequent siblings)
28 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
During a perf session we try to allocate buffers on the "node" associated
with the CPU the event is bound to. If it is not bound to a CPU, we
use the current CPU node, using smp_processor_id(). However this is unsafe
in a pre-emptible context and could generate the splats as below :
BUG: using smp_processor_id() in preemptible [00000000] code: perf/2544
Use NUMA_NO_NODE hint instead of using the current node for events
not bound to CPUs.
Fixes: 2997aa4063d97fdb39 ("coresight: etb10: implementing AUX API")
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-etb10.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-etb10.c b/drivers/hwtracing/coresight/coresight-etb10.c
index d5b9edecf76e..3810290e6d07 100644
--- a/drivers/hwtracing/coresight/coresight-etb10.c
+++ b/drivers/hwtracing/coresight/coresight-etb10.c
@@ -374,12 +374,10 @@ static void *etb_alloc_buffer(struct coresight_device *csdev,
struct perf_event *event, void **pages,
int nr_pages, bool overwrite)
{
- int node, cpu = event->cpu;
+ int node;
struct cs_buffers *buf;
- if (cpu == -1)
- cpu = smp_processor_id();
- node = cpu_to_node(cpu);
+ node = (event->cpu == -1) ? NUMA_NO_NODE : cpu_to_node(event->cpu);
buf = kzalloc_node(sizeof(struct cs_buffers), GFP_KERNEL, node);
if (!buf)
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* Re: [PATCH 23/28] coresight: etb10: Do not call smp_processor_id from preemptible
2019-06-19 19:53 ` [PATCH 23/28] coresight: etb10: " Mathieu Poirier
@ 2019-06-20 6:04 ` Greg KH
0 siblings, 0 replies; 41+ messages in thread
From: Greg KH @ 2019-06-20 6:04 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Wed, Jun 19, 2019 at 01:53:13PM -0600, Mathieu Poirier wrote:
> From: Suzuki K Poulose <suzuki.poulose@arm.com>
>
> During a perf session we try to allocate buffers on the "node" associated
> with the CPU the event is bound to. If it is not bound to a CPU, we
> use the current CPU node, using smp_processor_id(). However this is unsafe
> in a pre-emptible context and could generate the splats as below :
>
> BUG: using smp_processor_id() in preemptible [00000000] code: perf/2544
>
> Use NUMA_NO_NODE hint instead of using the current node for events
> not bound to CPUs.
>
> Fixes: 2997aa4063d97fdb39 ("coresight: etb10: implementing AUX API")
> Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> ---
> drivers/hwtracing/coresight/coresight-etb10.c | 6 ++----
> 1 file changed, 2 insertions(+), 4 deletions(-)
Again, 5.2 and stable.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread
* [PATCH 24/28] coresight: Potential uninitialized variable in probe()
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (22 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 23/28] coresight: etb10: " Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-20 6:04 ` Greg KH
2019-06-19 19:53 ` [PATCH 25/28] coresight: etm3x: Smatch: Fix potential NULL pointer dereference Mathieu Poirier
` (4 subsequent siblings)
28 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Dan Carpenter <dan.carpenter@oracle.com>
The "drvdata->atclk" clock is optional, but if it gets set to an error
pointer then we're accidentally return an uninitialized variable instead
of success.
Fixes: 78e6427b4e7b ("coresight: funnel: Support static funnel")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-funnel.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/hwtracing/coresight/coresight-funnel.c b/drivers/hwtracing/coresight/coresight-funnel.c
index 5867fcb4503b..fa97cb9ab4f9 100644
--- a/drivers/hwtracing/coresight/coresight-funnel.c
+++ b/drivers/hwtracing/coresight/coresight-funnel.c
@@ -244,6 +244,7 @@ static int funnel_probe(struct device *dev, struct resource *res)
}
pm_runtime_put(dev);
+ ret = 0;
out_disable_clk:
if (ret && !IS_ERR_OR_NULL(drvdata->atclk))
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 25/28] coresight: etm3x: Smatch: Fix potential NULL pointer dereference
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (23 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 24/28] coresight: Potential uninitialized variable in probe() Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-20 6:04 ` Greg KH
2019-06-19 19:53 ` [PATCH 26/28] coresight: tmc: " Mathieu Poirier
` (3 subsequent siblings)
28 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
Based on the following report from Smatch tool, make sure we have a
valid drvdata before we dereference it to find the real dev.
The patch 21d26b905c05: "coresight: etm: Clean up device specific
data" from May 22, 2019, leads to the following Smatch complaint:
./drivers/hwtracing/coresight/coresight-etm3x.c:460 etm_get_trace_id()
warn: variable dereferenced before check 'drvdata' (see line 458)
./drivers/hwtracing/coresight/coresight-etm3x.c
457 int trace_id = -1;
458 struct device *etm_dev = drvdata->csdev->dev.parent;
^^^^^^^^^
New dereference
459
460 if (!drvdata)
^^^^^^^^
Checked too late. Delete the check?
461 goto out;
462
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-etm3x.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
index bed729140718..225c2982e4fe 100644
--- a/drivers/hwtracing/coresight/coresight-etm3x.c
+++ b/drivers/hwtracing/coresight/coresight-etm3x.c
@@ -455,11 +455,12 @@ int etm_get_trace_id(struct etm_drvdata *drvdata)
{
unsigned long flags;
int trace_id = -1;
- struct device *etm_dev = drvdata->csdev->dev.parent;
+ struct device *etm_dev;
if (!drvdata)
goto out;
+ etm_dev = drvdata->csdev->dev.parent;
if (!local_read(&drvdata->mode))
return drvdata->traceid;
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* Re: [PATCH 25/28] coresight: etm3x: Smatch: Fix potential NULL pointer dereference
2019-06-19 19:53 ` [PATCH 25/28] coresight: etm3x: Smatch: Fix potential NULL pointer dereference Mathieu Poirier
@ 2019-06-20 6:04 ` Greg KH
2019-06-20 21:38 ` Mathieu Poirier
0 siblings, 1 reply; 41+ messages in thread
From: Greg KH @ 2019-06-20 6:04 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Wed, Jun 19, 2019 at 01:53:15PM -0600, Mathieu Poirier wrote:
> From: Suzuki K Poulose <suzuki.poulose@arm.com>
>
> Based on the following report from Smatch tool, make sure we have a
> valid drvdata before we dereference it to find the real dev.
>
> The patch 21d26b905c05: "coresight: etm: Clean up device specific
> data" from May 22, 2019, leads to the following Smatch complaint:
>
> ./drivers/hwtracing/coresight/coresight-etm3x.c:460 etm_get_trace_id()
> warn: variable dereferenced before check 'drvdata' (see line 458)
>
> ./drivers/hwtracing/coresight/coresight-etm3x.c
> 457 int trace_id = -1;
> 458 struct device *etm_dev = drvdata->csdev->dev.parent;
> ^^^^^^^^^
> New dereference
>
> 459
> 460 if (!drvdata)
> ^^^^^^^^
> Checked too late. Delete the check?
>
> 461 goto out;
> 462
>
> Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> Cc: Dan Carpenter <dan.carpenter@oracle.com>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> ---
> drivers/hwtracing/coresight/coresight-etm3x.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
> index bed729140718..225c2982e4fe 100644
> --- a/drivers/hwtracing/coresight/coresight-etm3x.c
> +++ b/drivers/hwtracing/coresight/coresight-etm3x.c
> @@ -455,11 +455,12 @@ int etm_get_trace_id(struct etm_drvdata *drvdata)
> {
> unsigned long flags;
> int trace_id = -1;
> - struct device *etm_dev = drvdata->csdev->dev.parent;
> + struct device *etm_dev;
>
> if (!drvdata)
> goto out;
>
> + etm_dev = drvdata->csdev->dev.parent;
> if (!local_read(&drvdata->mode))
> return drvdata->traceid;
>
> --
> 2.17.1
>
Looks like 5.2 and possibly stable material, right?
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread* Re: [PATCH 25/28] coresight: etm3x: Smatch: Fix potential NULL pointer dereference
2019-06-20 6:04 ` Greg KH
@ 2019-06-20 21:38 ` Mathieu Poirier
2019-06-21 6:59 ` Greg KH
0 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-20 21:38 UTC (permalink / raw)
To: Greg KH; +Cc: linux-arm-kernel
On Thu, 20 Jun 2019 at 00:05, Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Wed, Jun 19, 2019 at 01:53:15PM -0600, Mathieu Poirier wrote:
> > From: Suzuki K Poulose <suzuki.poulose@arm.com>
> >
> > Based on the following report from Smatch tool, make sure we have a
> > valid drvdata before we dereference it to find the real dev.
> >
> > The patch 21d26b905c05: "coresight: etm: Clean up device specific
> > data" from May 22, 2019, leads to the following Smatch complaint:
> >
> > ./drivers/hwtracing/coresight/coresight-etm3x.c:460 etm_get_trace_id()
> > warn: variable dereferenced before check 'drvdata' (see line 458)
> >
> > ./drivers/hwtracing/coresight/coresight-etm3x.c
> > 457 int trace_id = -1;
> > 458 struct device *etm_dev = drvdata->csdev->dev.parent;
> > ^^^^^^^^^
> > New dereference
> >
> > 459
> > 460 if (!drvdata)
> > ^^^^^^^^
> > Checked too late. Delete the check?
> >
> > 461 goto out;
> > 462
> >
> > Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> > Cc: Dan Carpenter <dan.carpenter@oracle.com>
> > Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> > Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> > ---
> > drivers/hwtracing/coresight/coresight-etm3x.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
> > index bed729140718..225c2982e4fe 100644
> > --- a/drivers/hwtracing/coresight/coresight-etm3x.c
> > +++ b/drivers/hwtracing/coresight/coresight-etm3x.c
> > @@ -455,11 +455,12 @@ int etm_get_trace_id(struct etm_drvdata *drvdata)
> > {
> > unsigned long flags;
> > int trace_id = -1;
> > - struct device *etm_dev = drvdata->csdev->dev.parent;
> > + struct device *etm_dev;
> >
> > if (!drvdata)
> > goto out;
> >
> > + etm_dev = drvdata->csdev->dev.parent;
> > if (!local_read(&drvdata->mode))
> > return drvdata->traceid;
> >
> > --
> > 2.17.1
> >
>
> Looks like 5.2 and possibly stable material, right?
This patch is fixing code that was introduced by patch 12/45 [1] of
the original set - it should really be applied on top you
char-misc-next for the 5.3 merge window.
[1]. https://www.spinics.net/lists/arm-kernel/msg736150.html
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread* Re: [PATCH 25/28] coresight: etm3x: Smatch: Fix potential NULL pointer dereference
2019-06-20 21:38 ` Mathieu Poirier
@ 2019-06-21 6:59 ` Greg KH
0 siblings, 0 replies; 41+ messages in thread
From: Greg KH @ 2019-06-21 6:59 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Thu, Jun 20, 2019 at 03:38:58PM -0600, Mathieu Poirier wrote:
> On Thu, 20 Jun 2019 at 00:05, Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Wed, Jun 19, 2019 at 01:53:15PM -0600, Mathieu Poirier wrote:
> > > From: Suzuki K Poulose <suzuki.poulose@arm.com>
> > >
> > > Based on the following report from Smatch tool, make sure we have a
> > > valid drvdata before we dereference it to find the real dev.
> > >
> > > The patch 21d26b905c05: "coresight: etm: Clean up device specific
> > > data" from May 22, 2019, leads to the following Smatch complaint:
> > >
> > > ./drivers/hwtracing/coresight/coresight-etm3x.c:460 etm_get_trace_id()
> > > warn: variable dereferenced before check 'drvdata' (see line 458)
> > >
> > > ./drivers/hwtracing/coresight/coresight-etm3x.c
> > > 457 int trace_id = -1;
> > > 458 struct device *etm_dev = drvdata->csdev->dev.parent;
> > > ^^^^^^^^^
> > > New dereference
> > >
> > > 459
> > > 460 if (!drvdata)
> > > ^^^^^^^^
> > > Checked too late. Delete the check?
> > >
> > > 461 goto out;
> > > 462
> > >
> > > Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> > > Cc: Dan Carpenter <dan.carpenter@oracle.com>
> > > Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> > > Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> > > ---
> > > drivers/hwtracing/coresight/coresight-etm3x.c | 3 ++-
> > > 1 file changed, 2 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
> > > index bed729140718..225c2982e4fe 100644
> > > --- a/drivers/hwtracing/coresight/coresight-etm3x.c
> > > +++ b/drivers/hwtracing/coresight/coresight-etm3x.c
> > > @@ -455,11 +455,12 @@ int etm_get_trace_id(struct etm_drvdata *drvdata)
> > > {
> > > unsigned long flags;
> > > int trace_id = -1;
> > > - struct device *etm_dev = drvdata->csdev->dev.parent;
> > > + struct device *etm_dev;
> > >
> > > if (!drvdata)
> > > goto out;
> > >
> > > + etm_dev = drvdata->csdev->dev.parent;
> > > if (!local_read(&drvdata->mode))
> > > return drvdata->traceid;
> > >
> > > --
> > > 2.17.1
> > >
> >
> > Looks like 5.2 and possibly stable material, right?
>
> This patch is fixing code that was introduced by patch 12/45 [1] of
> the original set - it should really be applied on top you
> char-misc-next for the 5.3 merge window.
>
> [1]. https://www.spinics.net/lists/arm-kernel/msg736150.html
Ok, can you resend it then?
thanks,
greg k-h
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread
* [PATCH 26/28] coresight: tmc: Smatch: Fix potential NULL pointer dereference
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (24 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 25/28] coresight: etm3x: Smatch: Fix potential NULL pointer dereference Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-20 6:05 ` Greg KH
2019-06-19 19:53 ` [PATCH 27/28] coresight: platform: add OF/APCI dependency Mathieu Poirier
` (2 subsequent siblings)
28 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Suzuki K Poulose <suzuki.poulose@arm.com>
Based on the following report from Smatch, fix the potential
NULL pointer dereference check.
The patch 743256e214e8: "coresight: tmc: Clean up device specific
data" from May 22, 2019, leads to the following Smatch complaint:
drivers/hwtracing/coresight/coresight-tmc-etr.c:625 tmc_etr_free_flat_buf()
warn: variable dereferenced before check 'flat_buf' (see line 623)
drivers/hwtracing/coresight/coresight-tmc-etr.c
622 struct etr_flat_buf *flat_buf = etr_buf->private;
623 struct device *real_dev = flat_buf->dev->parent;
^^^^^^^^^^
The patch introduces a new NULL check
624
625 if (flat_buf && flat_buf->daddr)
^^^^^^^^
but the existing code assumed it can be NULL.
626 dma_free_coherent(real_dev, flat_buf->size,
627 flat_buf->vaddr, flat_buf->daddr);
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-tmc-etr.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
index 5d2bf6d18961..17006705287a 100644
--- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
+++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
@@ -620,11 +620,13 @@ static int tmc_etr_alloc_flat_buf(struct tmc_drvdata *drvdata,
static void tmc_etr_free_flat_buf(struct etr_buf *etr_buf)
{
struct etr_flat_buf *flat_buf = etr_buf->private;
- struct device *real_dev = flat_buf->dev->parent;
- if (flat_buf && flat_buf->daddr)
+ if (flat_buf && flat_buf->daddr) {
+ struct device *real_dev = flat_buf->dev->parent;
+
dma_free_coherent(real_dev, flat_buf->size,
flat_buf->vaddr, flat_buf->daddr);
+ }
kfree(flat_buf);
}
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* Re: [PATCH 26/28] coresight: tmc: Smatch: Fix potential NULL pointer dereference
2019-06-19 19:53 ` [PATCH 26/28] coresight: tmc: " Mathieu Poirier
@ 2019-06-20 6:05 ` Greg KH
2019-06-20 21:42 ` Mathieu Poirier
0 siblings, 1 reply; 41+ messages in thread
From: Greg KH @ 2019-06-20 6:05 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Wed, Jun 19, 2019 at 01:53:16PM -0600, Mathieu Poirier wrote:
> From: Suzuki K Poulose <suzuki.poulose@arm.com>
>
> Based on the following report from Smatch, fix the potential
> NULL pointer dereference check.
>
> The patch 743256e214e8: "coresight: tmc: Clean up device specific
> data" from May 22, 2019, leads to the following Smatch complaint:
>
> drivers/hwtracing/coresight/coresight-tmc-etr.c:625 tmc_etr_free_flat_buf()
> warn: variable dereferenced before check 'flat_buf' (see line 623)
>
> drivers/hwtracing/coresight/coresight-tmc-etr.c
> 622 struct etr_flat_buf *flat_buf = etr_buf->private;
> 623 struct device *real_dev = flat_buf->dev->parent;
> ^^^^^^^^^^
> The patch introduces a new NULL check
>
> 624
> 625 if (flat_buf && flat_buf->daddr)
> ^^^^^^^^
> but the existing code assumed it can be NULL.
>
> 626 dma_free_coherent(real_dev, flat_buf->size,
> 627 flat_buf->vaddr, flat_buf->daddr);
>
> Cc: Dan Carpenter <dan.carpenter@oracle.com>
> Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> ---
> drivers/hwtracing/coresight/coresight-tmc-etr.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
> index 5d2bf6d18961..17006705287a 100644
> --- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
> +++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
> @@ -620,11 +620,13 @@ static int tmc_etr_alloc_flat_buf(struct tmc_drvdata *drvdata,
> static void tmc_etr_free_flat_buf(struct etr_buf *etr_buf)
> {
> struct etr_flat_buf *flat_buf = etr_buf->private;
> - struct device *real_dev = flat_buf->dev->parent;
>
> - if (flat_buf && flat_buf->daddr)
> + if (flat_buf && flat_buf->daddr) {
> + struct device *real_dev = flat_buf->dev->parent;
> +
> dma_free_coherent(real_dev, flat_buf->size,
> flat_buf->vaddr, flat_buf->daddr);
> + }
> kfree(flat_buf);
> }
>
> --
> 2.17.1
>
Again, 5.2-final and stable...
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread* Re: [PATCH 26/28] coresight: tmc: Smatch: Fix potential NULL pointer dereference
2019-06-20 6:05 ` Greg KH
@ 2019-06-20 21:42 ` Mathieu Poirier
2019-06-21 6:59 ` Greg KH
0 siblings, 1 reply; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-20 21:42 UTC (permalink / raw)
To: Greg KH; +Cc: linux-arm-kernel
On Thu, 20 Jun 2019 at 00:05, Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Wed, Jun 19, 2019 at 01:53:16PM -0600, Mathieu Poirier wrote:
> > From: Suzuki K Poulose <suzuki.poulose@arm.com>
> >
> > Based on the following report from Smatch, fix the potential
> > NULL pointer dereference check.
> >
> > The patch 743256e214e8: "coresight: tmc: Clean up device specific
> > data" from May 22, 2019, leads to the following Smatch complaint:
> >
> > drivers/hwtracing/coresight/coresight-tmc-etr.c:625 tmc_etr_free_flat_buf()
> > warn: variable dereferenced before check 'flat_buf' (see line 623)
> >
> > drivers/hwtracing/coresight/coresight-tmc-etr.c
> > 622 struct etr_flat_buf *flat_buf = etr_buf->private;
> > 623 struct device *real_dev = flat_buf->dev->parent;
> > ^^^^^^^^^^
> > The patch introduces a new NULL check
> >
> > 624
> > 625 if (flat_buf && flat_buf->daddr)
> > ^^^^^^^^
> > but the existing code assumed it can be NULL.
> >
> > 626 dma_free_coherent(real_dev, flat_buf->size,
> > 627 flat_buf->vaddr, flat_buf->daddr);
> >
> > Cc: Dan Carpenter <dan.carpenter@oracle.com>
> > Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> > Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> > Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> > ---
> > drivers/hwtracing/coresight/coresight-tmc-etr.c | 6 ++++--
> > 1 file changed, 4 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
> > index 5d2bf6d18961..17006705287a 100644
> > --- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
> > +++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
> > @@ -620,11 +620,13 @@ static int tmc_etr_alloc_flat_buf(struct tmc_drvdata *drvdata,
> > static void tmc_etr_free_flat_buf(struct etr_buf *etr_buf)
> > {
> > struct etr_flat_buf *flat_buf = etr_buf->private;
> > - struct device *real_dev = flat_buf->dev->parent;
> >
> > - if (flat_buf && flat_buf->daddr)
> > + if (flat_buf && flat_buf->daddr) {
> > + struct device *real_dev = flat_buf->dev->parent;
> > +
> > dma_free_coherent(real_dev, flat_buf->size,
> > flat_buf->vaddr, flat_buf->daddr);
> > + }
> > kfree(flat_buf);
> > }
> >
> > --
> > 2.17.1
> >
>
> Again, 5.2-final and stable...
So is this one, if addresses a deficiency introduced in patch 8/45
[1]. I have a new set ready for the other ones you flagged.
Thanks,
Mathieu
[1]. https://www.spinics.net/lists/arm-kernel/msg736144.html
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread* Re: [PATCH 26/28] coresight: tmc: Smatch: Fix potential NULL pointer dereference
2019-06-20 21:42 ` Mathieu Poirier
@ 2019-06-21 6:59 ` Greg KH
0 siblings, 0 replies; 41+ messages in thread
From: Greg KH @ 2019-06-21 6:59 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Thu, Jun 20, 2019 at 03:42:18PM -0600, Mathieu Poirier wrote:
> On Thu, 20 Jun 2019 at 00:05, Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Wed, Jun 19, 2019 at 01:53:16PM -0600, Mathieu Poirier wrote:
> > > From: Suzuki K Poulose <suzuki.poulose@arm.com>
> > >
> > > Based on the following report from Smatch, fix the potential
> > > NULL pointer dereference check.
> > >
> > > The patch 743256e214e8: "coresight: tmc: Clean up device specific
> > > data" from May 22, 2019, leads to the following Smatch complaint:
> > >
> > > drivers/hwtracing/coresight/coresight-tmc-etr.c:625 tmc_etr_free_flat_buf()
> > > warn: variable dereferenced before check 'flat_buf' (see line 623)
> > >
> > > drivers/hwtracing/coresight/coresight-tmc-etr.c
> > > 622 struct etr_flat_buf *flat_buf = etr_buf->private;
> > > 623 struct device *real_dev = flat_buf->dev->parent;
> > > ^^^^^^^^^^
> > > The patch introduces a new NULL check
> > >
> > > 624
> > > 625 if (flat_buf && flat_buf->daddr)
> > > ^^^^^^^^
> > > but the existing code assumed it can be NULL.
> > >
> > > 626 dma_free_coherent(real_dev, flat_buf->size,
> > > 627 flat_buf->vaddr, flat_buf->daddr);
> > >
> > > Cc: Dan Carpenter <dan.carpenter@oracle.com>
> > > Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
> > > Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> > > Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
> > > ---
> > > drivers/hwtracing/coresight/coresight-tmc-etr.c | 6 ++++--
> > > 1 file changed, 4 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c
> > > index 5d2bf6d18961..17006705287a 100644
> > > --- a/drivers/hwtracing/coresight/coresight-tmc-etr.c
> > > +++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c
> > > @@ -620,11 +620,13 @@ static int tmc_etr_alloc_flat_buf(struct tmc_drvdata *drvdata,
> > > static void tmc_etr_free_flat_buf(struct etr_buf *etr_buf)
> > > {
> > > struct etr_flat_buf *flat_buf = etr_buf->private;
> > > - struct device *real_dev = flat_buf->dev->parent;
> > >
> > > - if (flat_buf && flat_buf->daddr)
> > > + if (flat_buf && flat_buf->daddr) {
> > > + struct device *real_dev = flat_buf->dev->parent;
> > > +
> > > dma_free_coherent(real_dev, flat_buf->size,
> > > flat_buf->vaddr, flat_buf->daddr);
> > > + }
> > > kfree(flat_buf);
> > > }
> > >
> > > --
> > > 2.17.1
> > >
> >
> > Again, 5.2-final and stable...
>
> So is this one, if addresses a deficiency introduced in patch 8/45
> [1]. I have a new set ready for the other ones you flagged.
Ok, can you please resend?
thanks,
greg k-h
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread
* [PATCH 27/28] coresight: platform: add OF/APCI dependency
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (25 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 26/28] coresight: tmc: " Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-19 19:53 ` [PATCH 28/28] coresight: replicator: Add terminate entry for acpi_device_id tables Mathieu Poirier
2019-06-20 6:08 ` [PATCH 00/28] coresight: next v5.2-rc5 (V2) Greg KH
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Arnd Bergmann <arnd@arndb.de>
When neither CONFIG_OF nor CONFIG_ACPI are set, we get a harmless
build warning:
drivers/hwtracing/coresight/coresight-platform.c:26:12: error: unused function 'coresight_alloc_conns'
[-Werror,-Wunused-function]
static int coresight_alloc_conns(struct device *dev,
^
drivers/hwtracing/coresight/coresight-platform.c:46:1: error: unused function 'coresight_find_device_by_fwnode'
[-Werror,-Wunused-function]
coresight_find_device_by_fwnode(struct fwnode_handle *fwnode)
As the code is useless in that configuration anyway, just add
a Kconfig dependency that only allows building when at least
one of the two is set.
This should not hinder compile-testing, as CONFIG_OF can be
enabled on any architecture.
Fixes: ac0e232c12f0 ("coresight: platform: Use fwnode handle for device search")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/hwtracing/coresight/Kconfig b/drivers/hwtracing/coresight/Kconfig
index 5487d4a1abc2..14638db4991d 100644
--- a/drivers/hwtracing/coresight/Kconfig
+++ b/drivers/hwtracing/coresight/Kconfig
@@ -4,6 +4,7 @@
#
menuconfig CORESIGHT
bool "CoreSight Tracing Support"
+ depends on OF || ACPI
select ARM_AMBA
select PERF_EVENTS
help
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* [PATCH 28/28] coresight: replicator: Add terminate entry for acpi_device_id tables
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (26 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 27/28] coresight: platform: add OF/APCI dependency Mathieu Poirier
@ 2019-06-19 19:53 ` Mathieu Poirier
2019-06-20 6:08 ` [PATCH 00/28] coresight: next v5.2-rc5 (V2) Greg KH
28 siblings, 0 replies; 41+ messages in thread
From: Mathieu Poirier @ 2019-06-19 19:53 UTC (permalink / raw)
To: gregkh; +Cc: linux-arm-kernel
From: Wei Yongjun <weiyongjun1@huawei.com>
Make sure acpi_device_id tables have terminate entry.
Fixes: fe446287ec9f ("coresight: acpi: Support for platform devices")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
---
drivers/hwtracing/coresight/coresight-replicator.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/hwtracing/coresight/coresight-replicator.c b/drivers/hwtracing/coresight/coresight-replicator.c
index 542952759941..b7d6d59d56db 100644
--- a/drivers/hwtracing/coresight/coresight-replicator.c
+++ b/drivers/hwtracing/coresight/coresight-replicator.c
@@ -300,6 +300,7 @@ static const struct of_device_id static_replicator_match[] = {
#ifdef CONFIG_ACPI
static const struct acpi_device_id static_replicator_acpi_ids[] = {
{"ARMHC985", 0}, /* ARM CoreSight Static Replicator */
+ {}
};
#endif
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 41+ messages in thread* Re: [PATCH 00/28] coresight: next v5.2-rc5 (V2)
2019-06-19 19:52 [PATCH 00/28] coresight: next v5.2-rc5 (V2) Mathieu Poirier
` (27 preceding siblings ...)
2019-06-19 19:53 ` [PATCH 28/28] coresight: replicator: Add terminate entry for acpi_device_id tables Mathieu Poirier
@ 2019-06-20 6:08 ` Greg KH
28 siblings, 0 replies; 41+ messages in thread
From: Greg KH @ 2019-06-20 6:08 UTC (permalink / raw)
To: Mathieu Poirier; +Cc: linux-arm-kernel
On Wed, Jun 19, 2019 at 01:52:50PM -0600, Mathieu Poirier wrote:
> Here is the rest of the collection - it applies cleanly on your
> char-misc-testing branch (538eca25ff8f).
I did not apply patches 20-26 in this series based on the comments I
wrote on them. Please fix those up and send as a separate patch series
for 5.2-final.
thanks,
greg k-h
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 41+ messages in thread