[PATCH 0/3] occ: Restore default behavior of polling OCC during init

[PATCH 0/3] occ: Restore default behavior of polling OCC during init

[Eddie James]

A recent change to the OCC hwmon driver modified the default behavior to
no longer poll the OCC during initialization. This does change the
interface, meaning that old applications will not work with the more
recent driver. To resolve this issue, introduce a new dts property to
control the behavior of the driver during initialization, similar to the
FSI master property "no-scan-on-init". Without the new
"ibm,inactive-on-init" boolean present, the driver will now do the
previous behavior of polling the OCC.

Eddie James (3):
  dt-bindings: hwmon: Add IBM OCC bindings
  fsi: occ: Support probing the hwmon child device from dts node
  hwmon: (occ) Check for device property for setting OCC active during
    probe

 .../bindings/hwmon/ibm,occ-hmwon.yaml         | 40 ++++++++++++++++++
 drivers/fsi/fsi-occ.c                         | 41 +++++++++++++++----
 drivers/hwmon/occ/common.c                    | 11 ++++-
 drivers/hwmon/occ/p9_sbe.c                    |  9 ++++
 4 files changed, 93 insertions(+), 8 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml

-- 
2.31.1

[PATCH 1/3] dt-bindings: hwmon: Add IBM OCC bindings

[Eddie James]

These bindings describe the POWER processor On Chip Controller accessed
from a service processor or baseboard management controller (BMC).

Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
 .../bindings/hwmon/ibm,occ-hmwon.yaml         | 40 +++++++++++++++++++
 1 file changed, 40 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml

diff --git a/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml b/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
new file mode 100644
index 000000000000..8f8c3b8d7129
--- /dev/null
+++ b/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
@@ -0,0 +1,40 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/hwmon/ibm,occ-hwmon.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: IBM On-Chip Controller (OCC) accessed from a service processor
+
+maintainers:
+  - Eddie James <eajames@linux.ibm.com>
+
+description: |
+  This binding describes a POWER processor On-Chip Controller (OCC)
+  accessed from a service processor or baseboard management controller
+  (BMC).
+
+properties:
+  compatible:
+    enum:
+      - ibm,p9-occ-hwmon
+      - ibm,p10-occ-hwmon
+
+  ibm,inactive-on-init:
+    description: This property describes whether or not the OCC should
+      be marked as active during device initialization. The alternative
+      is for user space to mark the device active based on higher level
+      communications between the BMC and the host processor.
+    type: boolean
+
+required:
+  - compatible
+
+additionalProperties: false
+
+examples:
+  - |
+    occ-hmwon {
+        compatible = "ibm,p9-occ-hwmon";
+        ibm,inactive-on-init;
+    };
-- 
2.31.1

[PATCH 2/3] fsi: occ: Support probing the hwmon child device from dts node

[Eddie James]

There is now a need for reading devicetree properties in the OCC
hwmon driver, which isn't current supported as the FSI driver just
instantiates a basic platform device. Add support for this use case
by checking for an "occ-hwmon" node and if present, creating an
OF device from it.

Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
 drivers/fsi/fsi-occ.c | 41 ++++++++++++++++++++++++++++++++++-------
 1 file changed, 34 insertions(+), 7 deletions(-)

diff --git a/drivers/fsi/fsi-occ.c b/drivers/fsi/fsi-occ.c
index 8f7f602b909d..abdd37d5507f 100644
--- a/drivers/fsi/fsi-occ.c
+++ b/drivers/fsi/fsi-occ.c
@@ -44,6 +44,7 @@ struct occ {
 	struct device *sbefifo;
 	char name[32];
 	int idx;
+	bool platform_hwmon;
 	u8 sequence_number;
 	void *buffer;
 	void *client_buffer;
@@ -598,7 +599,7 @@ int fsi_occ_submit(struct device *dev, const void *request, size_t req_len,
 }
 EXPORT_SYMBOL_GPL(fsi_occ_submit);
 
-static int occ_unregister_child(struct device *dev, void *data)
+static int occ_unregister_platform_child(struct device *dev, void *data)
 {
 	struct platform_device *hwmon_dev = to_platform_device(dev);
 
@@ -607,12 +608,25 @@ static int occ_unregister_child(struct device *dev, void *data)
 	return 0;
 }
 
+static int occ_unregister_of_child(struct device *dev, void *data)
+{
+	struct platform_device *hwmon_dev = to_platform_device(dev);
+
+	of_device_unregister(hwmon_dev);
+	if (dev->of_node)
+		of_node_clear_flag(dev->of_node, OF_POPULATED);
+
+	return 0;
+}
+
 static int occ_probe(struct platform_device *pdev)
 {
 	int rc;
 	u32 reg;
+	char child_name[32];
 	struct occ *occ;
-	struct platform_device *hwmon_dev;
+	struct platform_device *hwmon_dev = NULL;
+	struct device_node *hwmon_node;
 	struct device *dev = &pdev->dev;
 	struct platform_device_info hwmon_dev_info = {
 		.parent = dev,
@@ -671,10 +685,20 @@ static int occ_probe(struct platform_device *pdev)
 		return rc;
 	}
 
-	hwmon_dev_info.id = occ->idx;
-	hwmon_dev = platform_device_register_full(&hwmon_dev_info);
-	if (IS_ERR(hwmon_dev))
-		dev_warn(dev, "failed to create hwmon device\n");
+	hwmon_node = of_get_child_by_name(dev->of_node, hwmon_dev_info.name);
+	if (hwmon_node) {
+		snprintf(child_name, sizeof(child_name), "%s.%d", hwmon_dev_info.name, occ->idx);
+		hwmon_dev = of_platform_device_create(hwmon_node, child_name, dev);
+		of_node_put(hwmon_node);
+	}
+
+	if (!hwmon_dev) {
+		occ->platform_hwmon = true;
+		hwmon_dev_info.id = occ->idx;
+		hwmon_dev = platform_device_register_full(&hwmon_dev_info);
+		if (IS_ERR(hwmon_dev))
+			dev_warn(dev, "failed to create hwmon device\n");
+	}
 
 	return 0;
 }
@@ -690,7 +714,10 @@ static int occ_remove(struct platform_device *pdev)
 	occ->buffer = NULL;
 	mutex_unlock(&occ->occ_lock);
 
-	device_for_each_child(&pdev->dev, NULL, occ_unregister_child);
+	if (occ->platform_hwmon)
+		device_for_each_child(&pdev->dev, NULL, occ_unregister_platform_child);
+	else
+		device_for_each_child(&pdev->dev, NULL, occ_unregister_of_child);
 
 	ida_simple_remove(&occ_ida, occ->idx);
 
-- 
2.31.1

[PATCH 3/3] hwmon: (occ) Check for device property for setting OCC active during probe

[Eddie James]

A previous commit changed the existing behavior of the driver to skip
attempting to communicate with the OCC during probe. Return to the
previous default behavior of automatically communicating with the OCC
and make it optional with a new device-tree property.

Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
 drivers/hwmon/occ/common.c | 11 ++++++++++-
 drivers/hwmon/occ/p9_sbe.c |  9 +++++++++
 2 files changed, 19 insertions(+), 1 deletion(-)

diff --git a/drivers/hwmon/occ/common.c b/drivers/hwmon/occ/common.c
index 45407b12db4b..95f16bb0e685 100644
--- a/drivers/hwmon/occ/common.c
+++ b/drivers/hwmon/occ/common.c
@@ -10,6 +10,7 @@
 #include <linux/math64.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
+#include <linux/property.h>
 #include <linux/sysfs.h>
 #include <asm/unaligned.h>
 
@@ -1216,8 +1217,16 @@ int occ_setup(struct occ *occ)
 	occ->groups[0] = &occ->group;
 
 	rc = occ_setup_sysfs(occ);
-	if (rc)
+	if (rc) {
 		dev_err(occ->bus_dev, "failed to setup sysfs: %d\n", rc);
+		return rc;
+	}
+
+	if (!device_property_read_bool(occ->bus_dev, "ibm,inactive-on-init")) {
+		rc = occ_active(occ, true);
+		if (rc)
+			occ_shutdown_sysfs(occ);
+	}
 
 	return rc;
 }
diff --git a/drivers/hwmon/occ/p9_sbe.c b/drivers/hwmon/occ/p9_sbe.c
index 4a1fe4ee8e2c..bc0f190065aa 100644
--- a/drivers/hwmon/occ/p9_sbe.c
+++ b/drivers/hwmon/occ/p9_sbe.c
@@ -7,6 +7,7 @@
 #include <linux/fsi-occ.h>
 #include <linux/mm.h>
 #include <linux/module.h>
+#include <linux/mod_devicetable.h>
 #include <linux/mutex.h>
 #include <linux/platform_device.h>
 #include <linux/string.h>
@@ -179,9 +180,17 @@ static int p9_sbe_occ_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id p9_sbe_occ_of_match[] = {
+	{ .compatible = "ibm,p9-occ-hwmon" },
+	{ .compatible = "ibm,p10-occ-hwmon" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, p8_i2c_occ_of_match);
+
 static struct platform_driver p9_sbe_occ_driver = {
 	.driver = {
 		.name = "occ-hwmon",
+		.of_match_table = p9_sbe_occ_of_match,
 	},
 	.probe	= p9_sbe_occ_probe,
 	.remove = p9_sbe_occ_remove,
-- 
2.31.1

Re: [PATCH 1/3] dt-bindings: hwmon: Add IBM OCC bindings

[Rob Herring]

On Tue, 02 Aug 2022 14:46:54 -0500, Eddie James wrote:
> These bindings describe the POWER processor On Chip Controller accessed
> from a service processor or baseboard management controller (BMC).
> 
> Signed-off-by: Eddie James <eajames@linux.ibm.com>
> ---
>  .../bindings/hwmon/ibm,occ-hmwon.yaml         | 40 +++++++++++++++++++
>  1 file changed, 40 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
> 

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check'
on your patch (DT_CHECKER_FLAGS is new in v5.13):

yamllint warnings/errors:

dtschema/dtc warnings/errors:
./Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml: $id: relative path/filename doesn't match actual path or filename
	expected: http://devicetree.org/schemas/hwmon/ibm,occ-hmwon.yaml#

doc reference errors (make refcheckdocs):

See https://patchwork.ozlabs.org/patch/

This check can fail if there are any dependencies. The base for a patch
series is generally the most recent rc1.

If you already ran 'make dt_binding_check' and didn't see the above
error(s), then make sure 'yamllint' is installed and dt-schema is up to
date:

pip3 install dtschema --upgrade

Please check and re-submit.

Re: [PATCH 1/3] dt-bindings: hwmon: Add IBM OCC bindings

[Krzysztof Kozlowski]

On 02/08/2022 21:46, Eddie James wrote:
> These bindings describe the POWER processor On Chip Controller accessed
> from a service processor or baseboard management controller (BMC).
> 
> Signed-off-by: Eddie James <eajames@linux.ibm.com>
> ---
>  .../bindings/hwmon/ibm,occ-hmwon.yaml         | 40 +++++++++++++++++++
>  1 file changed, 40 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
> 
> diff --git a/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml b/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
> new file mode 100644
> index 000000000000..8f8c3b8d7129
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
> @@ -0,0 +1,40 @@
> +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/hwmon/ibm,occ-hwmon.yaml#

typo here

Does not look like you tested the bindings. Please run `make
dt_binding_check` (see
Documentation/devicetree/bindings/writing-schema.rst for instructions).

> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: IBM On-Chip Controller (OCC) accessed from a service processor
> +
> +maintainers:
> +  - Eddie James <eajames@linux.ibm.com>
> +
> +description: |
> +  This binding describes a POWER processor On-Chip Controller (OCC)

s/This binding describes a//
But instead describe the hardware. What is the OCC?

> +  accessed from a service processor or baseboard management controller
> +  (BMC).
> +
> +properties:
> +  compatible:
> +    enum:
> +      - ibm,p9-occ-hwmon
> +      - ibm,p10-occ-hwmon
> +
> +  ibm,inactive-on-init:
> +    description: This property describes whether or not the OCC should
> +      be marked as active during device initialization. The alternative
> +      is for user space to mark the device active based on higher level
> +      communications between the BMC and the host processor.

I find the combination property name with this description confusing. It
sounds like init of OCC and somehow it should be inactive? I assume if
you initialize device, it is active. Or maybe the "init" is of something
else? What is more, non-negation is easier to understand, so rather
"ibm,active-on-boot" (or something like that).

> +    type: boolean
> +
> +required:
> +  - compatible
> +
> +additionalProperties: false
> +
> +examples:
> +  - |
> +    occ-hmwon {

just "hwmon"

> +        compatible = "ibm,p9-occ-hwmon";
> +        ibm,inactive-on-init;
> +    };


Best regards,
Krzysztof

Re: [PATCH 3/3] hwmon: (occ) Check for device property for setting OCC active during probe

[kernel test robot]

Hi Eddie,

I love your patch! Yet something to improve:

[auto build test ERROR on next-20220728]
[also build test ERROR on v5.19]
[cannot apply to groeck-staging/hwmon-next linus/master v5.19 v5.19-rc8 v5.19-rc7]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Eddie-James/occ-Restore-default-behavior-of-polling-OCC-during-init/20220803-034854
base:    7c5e07b73ff3011c9b82d4a3286a3362b951ad2b
config: arm-randconfig-r021-20220803 (https://download.01.org/0day-ci/archive/20220803/202208032055.PEvfcqsc-lkp@intel.com/config)
compiler: clang version 16.0.0 (https://github.com/llvm/llvm-project 495519e5f8232d144ed26e9c18dbcbac6a5f25eb)
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # install arm cross compiling tool for clang build
        # apt-get install binutils-arm-linux-gnueabi
        # https://github.com/intel-lab-lkp/linux/commit/31dc5bad51ddf22f4e097c0c5862d14341708188
        git remote add linux-review https://github.com/intel-lab-lkp/linux
        git fetch --no-tags linux-review Eddie-James/occ-Restore-default-behavior-of-polling-OCC-during-init/20220803-034854
        git checkout 31dc5bad51ddf22f4e097c0c5862d14341708188
        # save the config file
        mkdir build_dir && cp config build_dir/.config
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=arm SHELL=/bin/bash drivers/hwmon/occ/

If you fix the issue, kindly add following tag where applicable
Reported-by: kernel test robot <lkp@intel.com>

All errors (new ones prefixed by >>):

>> drivers/hwmon/occ/p9_sbe.c:188:25: error: use of undeclared identifier 'p8_i2c_occ_of_match'; did you mean 'p9_sbe_occ_of_match'?
   MODULE_DEVICE_TABLE(of, p8_i2c_occ_of_match);
                           ^~~~~~~~~~~~~~~~~~~
                           p9_sbe_occ_of_match
   include/linux/module.h:244:15: note: expanded from macro 'MODULE_DEVICE_TABLE'
   extern typeof(name) __mod_##type##__##name##_device_table               \
                 ^
   drivers/hwmon/occ/p9_sbe.c:183:34: note: 'p9_sbe_occ_of_match' declared here
   static const struct of_device_id p9_sbe_occ_of_match[] = {
                                    ^
   1 error generated.


vim +188 drivers/hwmon/occ/p9_sbe.c

   182	
   183	static const struct of_device_id p9_sbe_occ_of_match[] = {
   184		{ .compatible = "ibm,p9-occ-hwmon" },
   185		{ .compatible = "ibm,p10-occ-hwmon" },
   186		{}
   187	};
 > 188	MODULE_DEVICE_TABLE(of, p8_i2c_occ_of_match);
   189	

-- 
0-DAY CI Kernel Test Service
https://01.org/lkp

Re: [PATCH 3/3] hwmon: (occ) Check for device property for setting OCC active during probe

[kernel test robot]

Hi Eddie,

I love your patch! Yet something to improve:

[auto build test ERROR on next-20220728]
[also build test ERROR on v5.19]
[cannot apply to groeck-staging/hwmon-next linus/master v5.19 v5.19-rc8 v5.19-rc7]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Eddie-James/occ-Restore-default-behavior-of-polling-OCC-during-init/20220803-034854
base:    7c5e07b73ff3011c9b82d4a3286a3362b951ad2b
config: sh-allmodconfig (https://download.01.org/0day-ci/archive/20220803/202208032229.qba4UTzM-lkp@intel.com/config)
compiler: sh4-linux-gcc (GCC) 12.1.0
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # https://github.com/intel-lab-lkp/linux/commit/31dc5bad51ddf22f4e097c0c5862d14341708188
        git remote add linux-review https://github.com/intel-lab-lkp/linux
        git fetch --no-tags linux-review Eddie-James/occ-Restore-default-behavior-of-polling-OCC-during-init/20220803-034854
        git checkout 31dc5bad51ddf22f4e097c0c5862d14341708188
        # save the config file
        mkdir build_dir && cp config build_dir/.config
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=sh SHELL=/bin/bash drivers/hwmon/occ/

If you fix the issue, kindly add following tag where applicable
Reported-by: kernel test robot <lkp@intel.com>

All errors (new ones prefixed by >>):

   In file included from include/linux/device/driver.h:21,
                    from include/linux/device.h:32,
                    from drivers/hwmon/occ/p9_sbe.c:4:
>> drivers/hwmon/occ/p9_sbe.c:188:25: error: 'p8_i2c_occ_of_match' undeclared here (not in a function); did you mean 'p9_sbe_occ_of_match'?
     188 | MODULE_DEVICE_TABLE(of, p8_i2c_occ_of_match);
         |                         ^~~~~~~~~~~~~~~~~~~
   include/linux/module.h:244:15: note: in definition of macro 'MODULE_DEVICE_TABLE'
     244 | extern typeof(name) __mod_##type##__##name##_device_table               \
         |               ^~~~
>> include/linux/module.h:244:21: error: '__mod_of__p8_i2c_occ_of_match_device_table' aliased to undefined symbol 'p8_i2c_occ_of_match'
     244 | extern typeof(name) __mod_##type##__##name##_device_table               \
         |                     ^~~~~~
   drivers/hwmon/occ/p9_sbe.c:188:1: note: in expansion of macro 'MODULE_DEVICE_TABLE'
     188 | MODULE_DEVICE_TABLE(of, p8_i2c_occ_of_match);
         | ^~~~~~~~~~~~~~~~~~~


vim +188 drivers/hwmon/occ/p9_sbe.c

     3	
   > 4	#include <linux/device.h>
     5	#include <linux/errno.h>
     6	#include <linux/slab.h>
     7	#include <linux/fsi-occ.h>
     8	#include <linux/mm.h>
     9	#include <linux/module.h>
    10	#include <linux/mod_devicetable.h>
    11	#include <linux/mutex.h>
    12	#include <linux/platform_device.h>
    13	#include <linux/string.h>
    14	#include <linux/sysfs.h>
    15	
    16	#include "common.h"
    17	
    18	#define OCC_CHECKSUM_RETRIES	3
    19	
    20	struct p9_sbe_occ {
    21		struct occ occ;
    22		bool sbe_error;
    23		void *ffdc;
    24		size_t ffdc_len;
    25		size_t ffdc_size;
    26		struct mutex sbe_error_lock;	/* lock access to ffdc data */
    27		struct device *sbe;
    28	};
    29	
    30	#define to_p9_sbe_occ(x)	container_of((x), struct p9_sbe_occ, occ)
    31	
    32	static ssize_t ffdc_read(struct file *filp, struct kobject *kobj,
    33				 struct bin_attribute *battr, char *buf, loff_t pos,
    34				 size_t count)
    35	{
    36		ssize_t rc = 0;
    37		struct occ *occ = dev_get_drvdata(kobj_to_dev(kobj));
    38		struct p9_sbe_occ *ctx = to_p9_sbe_occ(occ);
    39	
    40		mutex_lock(&ctx->sbe_error_lock);
    41		if (ctx->sbe_error) {
    42			rc = memory_read_from_buffer(buf, count, &pos, ctx->ffdc,
    43						     ctx->ffdc_len);
    44			if (pos >= ctx->ffdc_len)
    45				ctx->sbe_error = false;
    46		}
    47		mutex_unlock(&ctx->sbe_error_lock);
    48	
    49		return rc;
    50	}
    51	static BIN_ATTR_RO(ffdc, OCC_MAX_RESP_WORDS * 4);
    52	
    53	static bool p9_sbe_occ_save_ffdc(struct p9_sbe_occ *ctx, const void *resp,
    54					 size_t resp_len)
    55	{
    56		bool notify = false;
    57	
    58		mutex_lock(&ctx->sbe_error_lock);
    59		if (!ctx->sbe_error) {
    60			if (resp_len > ctx->ffdc_size) {
    61				kvfree(ctx->ffdc);
    62				ctx->ffdc = kvmalloc(resp_len, GFP_KERNEL);
    63				if (!ctx->ffdc) {
    64					ctx->ffdc_len = 0;
    65					ctx->ffdc_size = 0;
    66					goto done;
    67				}
    68	
    69				ctx->ffdc_size = resp_len;
    70			}
    71	
    72			notify = true;
    73			ctx->sbe_error = true;
    74			ctx->ffdc_len = resp_len;
    75			memcpy(ctx->ffdc, resp, resp_len);
    76		}
    77	
    78	done:
    79		mutex_unlock(&ctx->sbe_error_lock);
    80		return notify;
    81	}
    82	
    83	static int p9_sbe_occ_send_cmd(struct occ *occ, u8 *cmd, size_t len,
    84				       void *resp, size_t resp_len)
    85	{
    86		struct p9_sbe_occ *ctx = to_p9_sbe_occ(occ);
    87		int rc, i;
    88	
    89		for (i = 0; i < OCC_CHECKSUM_RETRIES; ++i) {
    90			rc = fsi_occ_submit(ctx->sbe, cmd, len, resp, &resp_len);
    91			if (rc >= 0)
    92				break;
    93			if (resp_len) {
    94				if (p9_sbe_occ_save_ffdc(ctx, resp, resp_len))
    95					sysfs_notify(&occ->bus_dev->kobj, NULL,
    96						     bin_attr_ffdc.attr.name);
    97				return rc;
    98			}
    99			if (rc != -EBADE)
   100				return rc;
   101		}
   102	
   103		switch (((struct occ_response *)resp)->return_status) {
   104		case OCC_RESP_CMD_IN_PRG:
   105			rc = -ETIMEDOUT;
   106			break;
   107		case OCC_RESP_SUCCESS:
   108			rc = 0;
   109			break;
   110		case OCC_RESP_CMD_INVAL:
   111		case OCC_RESP_CMD_LEN_INVAL:
   112		case OCC_RESP_DATA_INVAL:
   113		case OCC_RESP_CHKSUM_ERR:
   114			rc = -EINVAL;
   115			break;
   116		case OCC_RESP_INT_ERR:
   117		case OCC_RESP_BAD_STATE:
   118		case OCC_RESP_CRIT_EXCEPT:
   119		case OCC_RESP_CRIT_INIT:
   120		case OCC_RESP_CRIT_WATCHDOG:
   121		case OCC_RESP_CRIT_OCB:
   122		case OCC_RESP_CRIT_HW:
   123			rc = -EREMOTEIO;
   124			break;
   125		default:
   126			rc = -EPROTO;
   127		}
   128	
   129		return rc;
   130	}
   131	
   132	static int p9_sbe_occ_probe(struct platform_device *pdev)
   133	{
   134		int rc;
   135		struct occ *occ;
   136		struct p9_sbe_occ *ctx = devm_kzalloc(&pdev->dev, sizeof(*ctx),
   137						      GFP_KERNEL);
   138		if (!ctx)
   139			return -ENOMEM;
   140	
   141		mutex_init(&ctx->sbe_error_lock);
   142	
   143		ctx->sbe = pdev->dev.parent;
   144		occ = &ctx->occ;
   145		occ->bus_dev = &pdev->dev;
   146		platform_set_drvdata(pdev, occ);
   147	
   148		occ->powr_sample_time_us = 500;
   149		occ->poll_cmd_data = 0x20;		/* P9 OCC poll data */
   150		occ->send_cmd = p9_sbe_occ_send_cmd;
   151	
   152		rc = occ_setup(occ);
   153		if (rc == -ESHUTDOWN)
   154			rc = -ENODEV;	/* Host is shutdown, don't spew errors */
   155	
   156		if (!rc) {
   157			rc = device_create_bin_file(occ->bus_dev, &bin_attr_ffdc);
   158			if (rc) {
   159				dev_warn(occ->bus_dev,
   160					 "failed to create SBE error ffdc file\n");
   161				rc = 0;
   162			}
   163		}
   164	
   165		return rc;
   166	}
   167	
   168	static int p9_sbe_occ_remove(struct platform_device *pdev)
   169	{
   170		struct occ *occ = platform_get_drvdata(pdev);
   171		struct p9_sbe_occ *ctx = to_p9_sbe_occ(occ);
   172	
   173		device_remove_bin_file(occ->bus_dev, &bin_attr_ffdc);
   174	
   175		ctx->sbe = NULL;
   176		occ_shutdown(occ);
   177	
   178		kvfree(ctx->ffdc);
   179	
   180		return 0;
   181	}
   182	
   183	static const struct of_device_id p9_sbe_occ_of_match[] = {
   184		{ .compatible = "ibm,p9-occ-hwmon" },
   185		{ .compatible = "ibm,p10-occ-hwmon" },
   186		{}
   187	};
 > 188	MODULE_DEVICE_TABLE(of, p8_i2c_occ_of_match);
   189	

-- 
0-DAY CI Kernel Test Service
https://01.org/lkp

Re: [PATCH 1/3] dt-bindings: hwmon: Add IBM OCC bindings

[Eddie James]

On 8/3/22 01:55, Krzysztof Kozlowski wrote:
> On 02/08/2022 21:46, Eddie James wrote:
>> These bindings describe the POWER processor On Chip Controller accessed
>> from a service processor or baseboard management controller (BMC).
>>
>> Signed-off-by: Eddie James <eajames@linux.ibm.com>
>> ---
>>   .../bindings/hwmon/ibm,occ-hmwon.yaml         | 40 +++++++++++++++++++
>>   1 file changed, 40 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml b/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
>> new file mode 100644
>> index 000000000000..8f8c3b8d7129
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/hwmon/ibm,occ-hmwon.yaml
>> @@ -0,0 +1,40 @@
>> +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/hwmon/ibm,occ-hwmon.yaml#
> typo here
>
> Does not look like you tested the bindings. Please run `make
> dt_binding_check` (see
> Documentation/devicetree/bindings/writing-schema.rst for instructions).


I actually did but it didn't catch that somehow. I had to use a somewhat 
hacked together python/pip on my system so perhaps that's to blame.


>
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: IBM On-Chip Controller (OCC) accessed from a service processor
>> +
>> +maintainers:
>> +  - Eddie James <eajames@linux.ibm.com>
>> +
>> +description: |
>> +  This binding describes a POWER processor On-Chip Controller (OCC)
> s/This binding describes a//
> But instead describe the hardware. What is the OCC?


OK I'll fix that. It's a management engine for system power and thermals.


>
>> +  accessed from a service processor or baseboard management controller
>> +  (BMC).
>> +
>> +properties:
>> +  compatible:
>> +    enum:
>> +      - ibm,p9-occ-hwmon
>> +      - ibm,p10-occ-hwmon
>> +
>> +  ibm,inactive-on-init:
>> +    description: This property describes whether or not the OCC should
>> +      be marked as active during device initialization. The alternative
>> +      is for user space to mark the device active based on higher level
>> +      communications between the BMC and the host processor.
> I find the combination property name with this description confusing. It
> sounds like init of OCC and somehow it should be inactive? I assume if
> you initialize device, it is active. Or maybe the "init" is of something
> else? What is more, non-negation is easier to understand, so rather
> "ibm,active-on-boot" (or something like that).


Well, the host processor initializes the OCC during it's boot, but this 
document is describing a binding to be used by a service processor 
talking to the OCC. So the OCC may be in any state. The init meant 
driver init, but I will simply the description and change the property 
to be more explicit: ibm,no-poll-on-init since that is what is actually 
happening. Similar to the FSI binding for no-scan-on-init.


>
>> +    type: boolean
>> +
>> +required:
>> +  - compatible
>> +
>> +additionalProperties: false
>> +
>> +examples:
>> +  - |
>> +    occ-hmwon {
> just "hwmon"
>
>> +        compatible = "ibm,p9-occ-hwmon";
>> +        ibm,inactive-on-init;
>> +    };


Thanks for the review!

Eddie


>
> Best regards,
> Krzysztof

Re: [PATCH 1/3] dt-bindings: hwmon: Add IBM OCC bindings

[Krzysztof Kozlowski]

On 09/08/2022 22:42, Eddie James wrote:
>>> +  ibm,inactive-on-init:
>>> +    description: This property describes whether or not the OCC should
>>> +      be marked as active during device initialization. The alternative
>>> +      is for user space to mark the device active based on higher level
>>> +      communications between the BMC and the host processor.
>> I find the combination property name with this description confusing. It
>> sounds like init of OCC and somehow it should be inactive? I assume if
>> you initialize device, it is active. Or maybe the "init" is of something
>> else? What is more, non-negation is easier to understand, so rather
>> "ibm,active-on-boot" (or something like that).
> 
> 
> Well, the host processor initializes the OCC during it's boot, but this 
> document is describing a binding to be used by a service processor 
> talking to the OCC. So the OCC may be in any state. The init meant 
> driver init, but I will simply the description and change the property 
> to be more explicit: ibm,no-poll-on-init since that is what is actually 
> happening. Similar to the FSI binding for no-scan-on-init.

no-scan-on-init is not a good example. It wasn't even reviewed by Rob
(looking at commit). In both cases you describe driver behavior which is
not appropriate for bindings. Instead you should describe the hardware
characteristics/feature/bug/state which require skipping the initialization.


Best regards,
Krzysztof

Re: [PATCH 1/3] dt-bindings: hwmon: Add IBM OCC bindings

[Eddie James]

On 8/10/22 02:43, Krzysztof Kozlowski wrote:
> On 09/08/2022 22:42, Eddie James wrote:
>>>> +  ibm,inactive-on-init:
>>>> +    description: This property describes whether or not the OCC should
>>>> +      be marked as active during device initialization. The alternative
>>>> +      is for user space to mark the device active based on higher level
>>>> +      communications between the BMC and the host processor.
>>> I find the combination property name with this description confusing. It
>>> sounds like init of OCC and somehow it should be inactive? I assume if
>>> you initialize device, it is active. Or maybe the "init" is of something
>>> else? What is more, non-negation is easier to understand, so rather
>>> "ibm,active-on-boot" (or something like that).
>>
>> Well, the host processor initializes the OCC during it's boot, but this
>> document is describing a binding to be used by a service processor
>> talking to the OCC. So the OCC may be in any state. The init meant
>> driver init, but I will simply the description and change the property
>> to be more explicit: ibm,no-poll-on-init since that is what is actually
>> happening. Similar to the FSI binding for no-scan-on-init.
> no-scan-on-init is not a good example. It wasn't even reviewed by Rob
> (looking at commit). In both cases you describe driver behavior which is
> not appropriate for bindings. Instead you should describe the hardware
> characteristics/feature/bug/state which require skipping the initialization.


Yep... there is no such hardware thing. The driver should never poll the 
OCC during driver initialization (since host state isn't known), but it 
did for the first couple of years of the drivers existence because we 
didn't catch that it could cause problems. I submitted a patch to change 
the driver behavior, but it does change the user space interface, so the 
argument is that the new behavior should be optional.

I suppose one correct way to control that would be a module parameter, 
but that really doesn't work well on embedded-like systems like ours.

Thanks very much for your feedback Krzysztof. Joel, any suggestions here?

Eddie


>
>
> Best regards,
> Krzysztof