All of lore.kernel.org
 help / color / mirror / Atom feed
From: Suzuki K Poulose <suzuki.poulose@arm.com>
To: linux-arm-kernel@lists.infradead.org
Cc: coresight@lists.linaro.org, mathieu.poirier@linaro.org,
	anshuman.khandual@arm.com, mike.leach@linaro.org,
	leo.yan@linaro.org, linux-kernel@vger.kernel.org,
	jonathan.zhouwen@huawei.com, catalin.marinas@arm.com,
	Suzuki K Poulose <suzuki.poulose@arm.com>
Subject: [PATCH v7 12/28] coresight: etm4x: Hide sysfs attributes for unavailable registers
Date: Sun, 10 Jan 2021 22:48:34 +0000	[thread overview]
Message-ID: <20210110224850.1880240-13-suzuki.poulose@arm.com> (raw)
In-Reply-To: <20210110224850.1880240-1-suzuki.poulose@arm.com>

Some of the management registers in ETMv4.x are not accessible
via system register instructions. Thus we must hide the sysfs
files exposing them to the userspace, to prevent system crashes.

This patch adds an is_visible() routine to control the visibility
at runtime for the registers that may not be accessed.

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
New patch in v7
---
 .../coresight/coresight-etm4x-sysfs.c         | 51 +++++++++++++++++++
 drivers/hwtracing/coresight/coresight-etm4x.h |  6 +++
 2 files changed, 57 insertions(+)

diff --git a/drivers/hwtracing/coresight/coresight-etm4x-sysfs.c b/drivers/hwtracing/coresight/coresight-etm4x-sysfs.c
index ddbfeb24fc3f..e8fdda45ffca 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x-sysfs.c
+++ b/drivers/hwtracing/coresight/coresight-etm4x-sysfs.c
@@ -2370,6 +2370,56 @@ static ssize_t coresight_etm4x_reg_show(struct device *dev,
 	return scnprintf(buf, PAGE_SIZE, "0x%x\n", val);
 }
 
+static inline bool
+etm4x_register_implemented(struct etmv4_drvdata *drvdata, u32 offset)
+{
+	switch (offset) {
+	ETM4x_SYSREG_LIST_CASES
+		/*
+		 * Registers accessible via system instructions are always
+		 * implemented.
+		 */
+		return true;
+	ETM4x_MMAP_LIST_CASES
+		/*
+		 * Registers accessible only via memory-mapped registers
+		 * must not be accessed via system instructions.
+		 * We cannot access the drvdata->csdev here, as this
+		 * function is called during the device creation, via
+		 * coresight_register() and the csdev is not initialized
+		 * until that is done. So rely on the drvdata->base to
+		 * detect if we have a memory mapped access.
+		 */
+		return !!drvdata->base;
+	}
+
+	return false;
+}
+
+/*
+ * Hide the ETM4x registers that may not be available on the
+ * hardware.
+ * There are certain management registers unavailable via system
+ * instructions. Make those sysfs attributes hidden on such
+ * systems.
+ */
+static umode_t
+coresight_etm4x_attr_reg_implemented(struct kobject *kobj,
+				     struct attribute *attr, int unused)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct etmv4_drvdata *drvdata = dev_get_drvdata(dev->parent);
+	struct device_attribute *d_attr;
+	u32 offset;
+
+	d_attr = container_of(attr, struct device_attribute, attr);
+	offset = coresight_etm4x_attr_to_offset(d_attr);
+
+	if (etm4x_register_implemented(drvdata, offset))
+		return attr->mode;
+	return 0;
+}
+
 #define coresight_etm4x_reg(name, offset)				\
 	&((struct dev_ext_attribute[]) {				\
 	   {								\
@@ -2417,6 +2467,7 @@ static const struct attribute_group coresight_etmv4_group = {
 };
 
 static const struct attribute_group coresight_etmv4_mgmt_group = {
+	.is_visible = coresight_etm4x_attr_reg_implemented,
 	.attrs = coresight_etmv4_mgmt_attrs,
 	.name = "mgmt",
 };
diff --git a/drivers/hwtracing/coresight/coresight-etm4x.h b/drivers/hwtracing/coresight/coresight-etm4x.h
index 24ba0da5b096..193d2819afa7 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x.h
+++ b/drivers/hwtracing/coresight/coresight-etm4x.h
@@ -157,6 +157,9 @@
 #define CASE_WRITE(val, x)					\
 	case (x): { write_etm4x_sysreg_const_offset((val), (x)); break; }
 
+#define CASE_NOP(__unused, x)					\
+	case (x):	/* fall through */
+
 /* List of registers accessible via System instructions */
 #define ETM_SYSREG_LIST(op, val)		\
 	CASE_##op((val), TRCPRGCTLR)		\
@@ -369,6 +372,9 @@
 #define ETM4x_READ_SYSREG_CASES(res)	ETM_SYSREG_LIST(READ, (res))
 #define ETM4x_WRITE_SYSREG_CASES(val)	ETM_SYSREG_LIST(WRITE, (val))
 
+#define ETM4x_SYSREG_LIST_CASES		ETM_SYSREG_LIST(NOP, __unused)
+#define ETM4x_MMAP_LIST_CASES		ETM_MMAP_LIST(NOP, __unused)
+
 #define read_etm4x_sysreg_offset(offset, _64bit)				\
 	({									\
 		u64 __val;							\
-- 
2.24.1


WARNING: multiple messages have this Message-ID (diff)
From: Suzuki K Poulose <suzuki.poulose@arm.com>
To: linux-arm-kernel@lists.infradead.org
Cc: mathieu.poirier@linaro.org, anshuman.khandual@arm.com,
	catalin.marinas@arm.com, coresight@lists.linaro.org,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	linux-kernel@vger.kernel.org, leo.yan@linaro.org,
	jonathan.zhouwen@huawei.com, mike.leach@linaro.org
Subject: [PATCH v7 12/28] coresight: etm4x: Hide sysfs attributes for unavailable registers
Date: Sun, 10 Jan 2021 22:48:34 +0000	[thread overview]
Message-ID: <20210110224850.1880240-13-suzuki.poulose@arm.com> (raw)
In-Reply-To: <20210110224850.1880240-1-suzuki.poulose@arm.com>

Some of the management registers in ETMv4.x are not accessible
via system register instructions. Thus we must hide the sysfs
files exposing them to the userspace, to prevent system crashes.

This patch adds an is_visible() routine to control the visibility
at runtime for the registers that may not be accessed.

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
New patch in v7
---
 .../coresight/coresight-etm4x-sysfs.c         | 51 +++++++++++++++++++
 drivers/hwtracing/coresight/coresight-etm4x.h |  6 +++
 2 files changed, 57 insertions(+)

diff --git a/drivers/hwtracing/coresight/coresight-etm4x-sysfs.c b/drivers/hwtracing/coresight/coresight-etm4x-sysfs.c
index ddbfeb24fc3f..e8fdda45ffca 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x-sysfs.c
+++ b/drivers/hwtracing/coresight/coresight-etm4x-sysfs.c
@@ -2370,6 +2370,56 @@ static ssize_t coresight_etm4x_reg_show(struct device *dev,
 	return scnprintf(buf, PAGE_SIZE, "0x%x\n", val);
 }
 
+static inline bool
+etm4x_register_implemented(struct etmv4_drvdata *drvdata, u32 offset)
+{
+	switch (offset) {
+	ETM4x_SYSREG_LIST_CASES
+		/*
+		 * Registers accessible via system instructions are always
+		 * implemented.
+		 */
+		return true;
+	ETM4x_MMAP_LIST_CASES
+		/*
+		 * Registers accessible only via memory-mapped registers
+		 * must not be accessed via system instructions.
+		 * We cannot access the drvdata->csdev here, as this
+		 * function is called during the device creation, via
+		 * coresight_register() and the csdev is not initialized
+		 * until that is done. So rely on the drvdata->base to
+		 * detect if we have a memory mapped access.
+		 */
+		return !!drvdata->base;
+	}
+
+	return false;
+}
+
+/*
+ * Hide the ETM4x registers that may not be available on the
+ * hardware.
+ * There are certain management registers unavailable via system
+ * instructions. Make those sysfs attributes hidden on such
+ * systems.
+ */
+static umode_t
+coresight_etm4x_attr_reg_implemented(struct kobject *kobj,
+				     struct attribute *attr, int unused)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct etmv4_drvdata *drvdata = dev_get_drvdata(dev->parent);
+	struct device_attribute *d_attr;
+	u32 offset;
+
+	d_attr = container_of(attr, struct device_attribute, attr);
+	offset = coresight_etm4x_attr_to_offset(d_attr);
+
+	if (etm4x_register_implemented(drvdata, offset))
+		return attr->mode;
+	return 0;
+}
+
 #define coresight_etm4x_reg(name, offset)				\
 	&((struct dev_ext_attribute[]) {				\
 	   {								\
@@ -2417,6 +2467,7 @@ static const struct attribute_group coresight_etmv4_group = {
 };
 
 static const struct attribute_group coresight_etmv4_mgmt_group = {
+	.is_visible = coresight_etm4x_attr_reg_implemented,
 	.attrs = coresight_etmv4_mgmt_attrs,
 	.name = "mgmt",
 };
diff --git a/drivers/hwtracing/coresight/coresight-etm4x.h b/drivers/hwtracing/coresight/coresight-etm4x.h
index 24ba0da5b096..193d2819afa7 100644
--- a/drivers/hwtracing/coresight/coresight-etm4x.h
+++ b/drivers/hwtracing/coresight/coresight-etm4x.h
@@ -157,6 +157,9 @@
 #define CASE_WRITE(val, x)					\
 	case (x): { write_etm4x_sysreg_const_offset((val), (x)); break; }
 
+#define CASE_NOP(__unused, x)					\
+	case (x):	/* fall through */
+
 /* List of registers accessible via System instructions */
 #define ETM_SYSREG_LIST(op, val)		\
 	CASE_##op((val), TRCPRGCTLR)		\
@@ -369,6 +372,9 @@
 #define ETM4x_READ_SYSREG_CASES(res)	ETM_SYSREG_LIST(READ, (res))
 #define ETM4x_WRITE_SYSREG_CASES(val)	ETM_SYSREG_LIST(WRITE, (val))
 
+#define ETM4x_SYSREG_LIST_CASES		ETM_SYSREG_LIST(NOP, __unused)
+#define ETM4x_MMAP_LIST_CASES		ETM_MMAP_LIST(NOP, __unused)
+
 #define read_etm4x_sysreg_offset(offset, _64bit)				\
 	({									\
 		u64 __val;							\
-- 
2.24.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  parent reply	other threads:[~2021-01-10 22:50 UTC|newest]

Thread overview: 74+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-10 22:48 [PATCH v7 00/28] coresight: etm4x: Support for system instructions Suzuki K Poulose
2021-01-10 22:48 ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 01/28] coresight: etm4x: Handle access to TRCSSPCICRn Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 02/28] coresight: etm4x: Skip accessing TRCPDCR in save/restore Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-25 18:39   ` Mathieu Poirier
2021-01-25 18:39     ` Mathieu Poirier
2021-01-10 22:48 ` [PATCH v7 03/28] coresight: Introduce device access abstraction Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-25 18:42   ` Mathieu Poirier
2021-01-25 18:42     ` Mathieu Poirier
2021-01-10 22:48 ` [PATCH v7 04/28] coresight: tpiu: Prepare for using coresight " Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 05/28] coresight: Convert coresight_timeout to use " Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 06/28] coresight: Convert claim/disclaim operations to use access wrappers Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 07/28] coresight: etm4x: Always read the registers on the host CPU Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 08/28] coresight: etm4x: Convert all register accesses Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 09/28] coresight: etm4x: Make offset available for sysfs attributes Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 10/28] coresight: etm4x: Add commentary on the registers Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 11/28] coresight: etm4x: Add sysreg access helpers Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` Suzuki K Poulose [this message]
2021-01-10 22:48   ` [PATCH v7 12/28] coresight: etm4x: Hide sysfs attributes for unavailable registers Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 13/28] coresight: etm4x: Define DEVARCH register fields Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 14/28] coresight: etm4x: Check for Software Lock Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 15/28] coresight: etm4x: Cleanup secure exception level masks Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 16/28] coresight: etm4x: Clean up " Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 17/28] coresight: etm4x: Handle ETM architecture version Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 18/28] coresight: etm4x: Detect access early on the target CPU Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 19/28] coresight: etm4x: Use TRCDEVARCH for component discovery Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 20/28] coresight: etm4x: Expose trcdevarch via sysfs Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 21/28] coresight: etm4x: Add necessary synchronization for sysreg access Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 22/28] coresight: etm4x: Detect system instructions support Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 23/28] coresight: etm4x: Refactor probing routine Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 24/28] coresight: etm4x: Run arch feature detection on the CPU Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 25/28] coresight: etm4x: Add support for sysreg only devices Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 26/28] dts: bindings: coresight: ETM system register access only units Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 27/28] arm64: Add TRFCR_ELx definitions Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-01-10 22:48 ` [PATCH v7 28/28] coresight: Add support for v8.4 SelfHosted tracing Suzuki K Poulose
2021-01-10 22:48   ` Suzuki K Poulose
2021-02-12 10:34   ` Mike Leach
2021-02-12 10:34     ` Mike Leach
2021-02-12 15:36     ` Suzuki K Poulose
2021-02-12 15:36       ` Suzuki K Poulose
2021-02-12 17:30       ` Mike Leach
2021-02-12 17:30         ` Mike Leach
2021-02-18 14:51         ` Suzuki K Poulose
2021-02-18 14:51           ` Suzuki K Poulose
2021-01-25 18:49 ` [PATCH v7 00/28] coresight: etm4x: Support for system instructions Mathieu Poirier
2021-01-25 18:49   ` Mathieu Poirier
2021-01-25 22:05   ` Suzuki K Poulose
2021-01-25 22:05     ` Suzuki K Poulose

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210110224850.1880240-13-suzuki.poulose@arm.com \
    --to=suzuki.poulose@arm.com \
    --cc=anshuman.khandual@arm.com \
    --cc=catalin.marinas@arm.com \
    --cc=coresight@lists.linaro.org \
    --cc=jonathan.zhouwen@huawei.com \
    --cc=leo.yan@linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.poirier@linaro.org \
    --cc=mike.leach@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.