LinuxLists.cc - [PATCH RESEND] rpmsg: Add driver_override device attribute for rpmsg

2018-01-10 13:17:57

Subject: [PATCH RESEND] rpmsg: Add driver_override device attribute for rpmsg_device

This patch adds "driver_override" device attribute for rpmsg_device which
will allow users to explicitly specify the rpmsg_driver to be used via
sysfs entry.

The "driver_override" device attribute implemented here is very similar
to "driver_override" implemented for platform, pci, and amba bus types.

One important use-case of "driver_override" device attribute is to force
use of rpmsg_chrdev driver for certain rpmsg_device instances.

Signed-off-by: Anup Patel <[email protected]>
---
Documentation/ABI/testing/sysfs-bus-rpmsg | 20 ++++++++++++++
drivers/rpmsg/rpmsg_core.c | 46 +++++++++++++++++++++++++++++--
2 files changed, 64 insertions(+), 2 deletions(-)

diff --git a/Documentation/ABI/testing/sysfs-bus-rpmsg b/Documentation/ABI/testing/sysfs-bus-rpmsg
index 189e419..1c8599e 100644
--- a/Documentation/ABI/testing/sysfs-bus-rpmsg
+++ b/Documentation/ABI/testing/sysfs-bus-rpmsg
@@ -73,3 +73,23 @@ Description:
This sysfs entry tells us whether the channel is a local
server channel that is announced (values are either
true or false).
+
+What: /sys/bus/rpmsg/devices/.../driver_override
+Date: January 2018
+KernelVersion: 4.16
+Contact: Ohad Ben-Cohen <[email protected]>
+Description:
+ Every rpmsg device is a communication channel with a remote
+ processor. Channels are identified by a textual name (see
+ /sys/bus/rpmsg/devices/.../name above) and have a local
+ ("source") rpmsg address, and remote ("destination") rpmsg
+ address.
+
+ The listening entity (or client) which communicates with a
+ remote processor is referred as rpmsg driver. The rpmsg device
+ and rpmsg driver are matched based on rpmsg device name and
+ rpmsg driver ID table.
+
+ This sysfs entry allows the rpmsg driver for a rpmsg device
+ to be specified which will override standard OF, ID table
+ and name matching.
diff --git a/drivers/rpmsg/rpmsg_core.c b/drivers/rpmsg/rpmsg_core.c
index dffa3aa..9a25e42 100644
--- a/drivers/rpmsg/rpmsg_core.c
+++ b/drivers/rpmsg/rpmsg_core.c
@@ -321,11 +321,11 @@ struct device *rpmsg_find_device(struct device *parent,
}
EXPORT_SYMBOL(rpmsg_find_device);

-/* sysfs show configuration fields */
+/* sysfs configuration fields */
#define rpmsg_show_attr(field, path, format_string) \
static ssize_t \
field##_show(struct device *dev, \
- struct device_attribute *attr, char *buf) \
+ struct device_attribute *attr, char *buf) \
{ \
struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
\
@@ -333,11 +333,52 @@ field##_show(struct device *dev, \
} \
static DEVICE_ATTR_RO(field);

+#define rpmsg_string_attr(field, path) \
+static ssize_t \
+field##_store(struct device *dev, \
+ struct device_attribute *attr, const char *buf, size_t sz)\
+{ \
+ struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
+ char *new, *old, *cp; \
+ \
+ new = kstrndup(buf, sz, GFP_KERNEL); \
+ if (!new) \
+ return -ENOMEM; \
+ \
+ cp = strchr(new, '\n'); \
+ if (cp) \
+ *cp = '\0'; \
+ \
+ device_lock(dev); \
+ old = rpdev->path; \
+ if (strlen(new)) { \
+ rpdev->path = new; \
+ } else { \
+ kfree(new); \
+ rpdev->path = NULL; \
+ } \
+ device_unlock(dev); \
+ \
+ kfree(old); \
+ \
+ return sz; \
+} \
+static ssize_t \
+field##_show(struct device *dev, \
+ struct device_attribute *attr, char *buf) \
+{ \
+ struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
+ \
+ return sprintf(buf, "%s\n", rpdev->path); \
+} \
+static DEVICE_ATTR_RW(field)
+
/* for more info, see Documentation/ABI/testing/sysfs-bus-rpmsg */
rpmsg_show_attr(name, id.name, "%s\n");
rpmsg_show_attr(src, src, "0x%x\n");
rpmsg_show_attr(dst, dst, "0x%x\n");
rpmsg_show_attr(announce, announce ? "true" : "false", "%s\n");
+rpmsg_string_attr(driver_override, driver_override);

static ssize_t modalias_show(struct device *dev,
struct device_attribute *attr, char *buf)
@@ -359,6 +400,7 @@ static struct attribute *rpmsg_dev_attrs[] = {
&dev_attr_dst.attr,
&dev_attr_src.attr,
&dev_attr_announce.attr,
+ &dev_attr_driver_override.attr,
NULL,
};
ATTRIBUTE_GROUPS(rpmsg_dev);
--
2.7.4

2018-01-23 13:47:59

by Anup Patel

[permalink] [raw]

Subject: Re: [PATCH RESEND] rpmsg: Add driver_override device attribute for rpmsg_device

On Wed, Jan 10, 2018 at 6:47 PM, Anup Patel <[email protected]> wrote:
> This patch adds "driver_override" device attribute for rpmsg_device which
> will allow users to explicitly specify the rpmsg_driver to be used via
> sysfs entry.
>
> The "driver_override" device attribute implemented here is very similar
> to "driver_override" implemented for platform, pci, and amba bus types.
>
> One important use-case of "driver_override" device attribute is to force
> use of rpmsg_chrdev driver for certain rpmsg_device instances.
>
> Signed-off-by: Anup Patel <[email protected]>
> ---
> Documentation/ABI/testing/sysfs-bus-rpmsg | 20 ++++++++++++++
> drivers/rpmsg/rpmsg_core.c | 46 +++++++++++++++++++++++++++++--
> 2 files changed, 64 insertions(+), 2 deletions(-)
>
> diff --git a/Documentation/ABI/testing/sysfs-bus-rpmsg b/Documentation/ABI/testing/sysfs-bus-rpmsg
> index 189e419..1c8599e 100644
> --- a/Documentation/ABI/testing/sysfs-bus-rpmsg
> +++ b/Documentation/ABI/testing/sysfs-bus-rpmsg
> @@ -73,3 +73,23 @@ Description:
> This sysfs entry tells us whether the channel is a local
> server channel that is announced (values are either
> true or false).
> +
> +What: /sys/bus/rpmsg/devices/.../driver_override
> +Date: January 2018
> +KernelVersion: 4.16
> +Contact: Ohad Ben-Cohen <[email protected]>
> +Description:
> + Every rpmsg device is a communication channel with a remote
> + processor. Channels are identified by a textual name (see
> + /sys/bus/rpmsg/devices/.../name above) and have a local
> + ("source") rpmsg address, and remote ("destination") rpmsg
> + address.
> +
> + The listening entity (or client) which communicates with a
> + remote processor is referred as rpmsg driver. The rpmsg device
> + and rpmsg driver are matched based on rpmsg device name and
> + rpmsg driver ID table.
> +
> + This sysfs entry allows the rpmsg driver for a rpmsg device
> + to be specified which will override standard OF, ID table
> + and name matching.
> diff --git a/drivers/rpmsg/rpmsg_core.c b/drivers/rpmsg/rpmsg_core.c
> index dffa3aa..9a25e42 100644
> --- a/drivers/rpmsg/rpmsg_core.c
> +++ b/drivers/rpmsg/rpmsg_core.c
> @@ -321,11 +321,11 @@ struct device *rpmsg_find_device(struct device *parent,
> }
> EXPORT_SYMBOL(rpmsg_find_device);
>
> -/* sysfs show configuration fields */
> +/* sysfs configuration fields */
> #define rpmsg_show_attr(field, path, format_string) \
> static ssize_t \
> field##_show(struct device *dev, \
> - struct device_attribute *attr, char *buf) \
> + struct device_attribute *attr, char *buf) \
> { \
> struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
> \
> @@ -333,11 +333,52 @@ field##_show(struct device *dev, \
> } \
> static DEVICE_ATTR_RO(field);
>
> +#define rpmsg_string_attr(field, path) \
> +static ssize_t \
> +field##_store(struct device *dev, \
> + struct device_attribute *attr, const char *buf, size_t sz)\
> +{ \
> + struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
> + char *new, *old, *cp; \
> + \
> + new = kstrndup(buf, sz, GFP_KERNEL); \
> + if (!new) \
> + return -ENOMEM; \
> + \
> + cp = strchr(new, '\n'); \
> + if (cp) \
> + *cp = '\0'; \
> + \
> + device_lock(dev); \
> + old = rpdev->path; \
> + if (strlen(new)) { \
> + rpdev->path = new; \
> + } else { \
> + kfree(new); \
> + rpdev->path = NULL; \
> + } \
> + device_unlock(dev); \
> + \
> + kfree(old); \
> + \
> + return sz; \
> +} \
> +static ssize_t \
> +field##_show(struct device *dev, \
> + struct device_attribute *attr, char *buf) \
> +{ \
> + struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
> + \
> + return sprintf(buf, "%s\n", rpdev->path); \
> +} \
> +static DEVICE_ATTR_RW(field)
> +
> /* for more info, see Documentation/ABI/testing/sysfs-bus-rpmsg */
> rpmsg_show_attr(name, id.name, "%s\n");
> rpmsg_show_attr(src, src, "0x%x\n");
> rpmsg_show_attr(dst, dst, "0x%x\n");
> rpmsg_show_attr(announce, announce ? "true" : "false", "%s\n");
> +rpmsg_string_attr(driver_override, driver_override);
>
> static ssize_t modalias_show(struct device *dev,
> struct device_attribute *attr, char *buf)
> @@ -359,6 +400,7 @@ static struct attribute *rpmsg_dev_attrs[] = {
> &dev_attr_dst.attr,
> &dev_attr_src.attr,
> &dev_attr_announce.attr,
> + &dev_attr_driver_override.attr,
> NULL,
> };
> ATTRIBUTE_GROUPS(rpmsg_dev);
> --
> 2.7.4
>

Hi All,

Any comments on this??

Regards,
Anup

2018-02-06 04:45:46

by Anup Patel

[permalink] [raw]

Subject: Re: [PATCH RESEND] rpmsg: Add driver_override device attribute for rpmsg_device

Hi Bjorn,

Can you please have a look at this patch?

Regards,
Anup

2018-02-21 11:41:04

by Anup Patel

[permalink] [raw]

Subject: Re: [PATCH RESEND] rpmsg: Add driver_override device attribute for rpmsg_device

2018-03-18 22:49:13

by Bjorn Andersson

[permalink] [raw]

Subject: Re: [PATCH RESEND] rpmsg: Add driver_override device attribute for rpmsg_device

On Wed 10 Jan 05:17 PST 2018, Anup Patel wrote:

> This patch adds "driver_override" device attribute for rpmsg_device which
> will allow users to explicitly specify the rpmsg_driver to be used via
> sysfs entry.
>
> The "driver_override" device attribute implemented here is very similar
> to "driver_override" implemented for platform, pci, and amba bus types.
>
> One important use-case of "driver_override" device attribute is to force
> use of rpmsg_chrdev driver for certain rpmsg_device instances.
>

I assume you mean specifically for the case where you want to prevent
some kernel driver to probe for some given channel?

The intention with rpmsg_char is that you through the rpmsg_ctrlX
interface create and destroy endpoints dynamically, so you wouldn't need
to use this mechanism to bind some specific channel to rpmsg_char.

That said, this does make sense for completeness sake.

[..]
> diff --git a/drivers/rpmsg/rpmsg_core.c b/drivers/rpmsg/rpmsg_core.c
> index dffa3aa..9a25e42 100644
> --- a/drivers/rpmsg/rpmsg_core.c
> +++ b/drivers/rpmsg/rpmsg_core.c
> @@ -321,11 +321,11 @@ struct device *rpmsg_find_device(struct device *parent,
> }
> EXPORT_SYMBOL(rpmsg_find_device);
>
> -/* sysfs show configuration fields */
> +/* sysfs configuration fields */
> #define rpmsg_show_attr(field, path, format_string) \
> static ssize_t \
> field##_show(struct device *dev, \
> - struct device_attribute *attr, char *buf) \
> + struct device_attribute *attr, char *buf) \

Seems unnecessary.

> { \
> struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
> \
> @@ -333,11 +333,52 @@ field##_show(struct device *dev, \
> } \
> static DEVICE_ATTR_RO(field);
>
> +#define rpmsg_string_attr(field, path) \

"path" is an odd name for these, I think it's a "member".

> +static ssize_t \
> +field##_store(struct device *dev, \
> + struct device_attribute *attr, const char *buf, size_t sz)\

field##_store(struct device *dev, struct device_attribute *attr, \
const char *buf, size_t sz) \

Is prettier

> +{ \
> + struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
> + char *new, *old, *cp; \
> + \
> + new = kstrndup(buf, sz, GFP_KERNEL); \
> + if (!new) \
> + return -ENOMEM; \
> + \
> + cp = strchr(new, '\n'); \
> + if (cp) \
> + *cp = '\0'; \

I prefer

new[strcspn(new, "\n")] = '\0';

Regards,
Bjorn

2018-03-21 08:01:12

by Anup Patel

[permalink] [raw]

Subject: Re: [PATCH RESEND] rpmsg: Add driver_override device attribute for rpmsg_device

On Mon, Mar 19, 2018 at 4:17 AM, Bjorn Andersson
<[email protected]> wrote:
> On Wed 10 Jan 05:17 PST 2018, Anup Patel wrote:
>
>> This patch adds "driver_override" device attribute for rpmsg_device which
>> will allow users to explicitly specify the rpmsg_driver to be used via
>> sysfs entry.
>>
>> The "driver_override" device attribute implemented here is very similar
>> to "driver_override" implemented for platform, pci, and amba bus types.
>>
>> One important use-case of "driver_override" device attribute is to force
>> use of rpmsg_chrdev driver for certain rpmsg_device instances.
>>
>
> I assume you mean specifically for the case where you want to prevent
> some kernel driver to probe for some given channel?

Yes, there are few use-cases where we want to prevent kernel
driver and rather access rpmsg device from user-space using
rpmsg_chrdev driver.

>
> The intention with rpmsg_char is that you through the rpmsg_ctrlX
> interface create and destroy endpoints dynamically, so you wouldn't need
> to use this mechanism to bind some specific channel to rpmsg_char.
>
>
> That said, this does make sense for completeness sake.
>
> [..]
>> diff --git a/drivers/rpmsg/rpmsg_core.c b/drivers/rpmsg/rpmsg_core.c
>> index dffa3aa..9a25e42 100644
>> --- a/drivers/rpmsg/rpmsg_core.c
>> +++ b/drivers/rpmsg/rpmsg_core.c
>> @@ -321,11 +321,11 @@ struct device *rpmsg_find_device(struct device *parent,
>> }
>> EXPORT_SYMBOL(rpmsg_find_device);
>>
>> -/* sysfs show configuration fields */
>> +/* sysfs configuration fields */
>> #define rpmsg_show_attr(field, path, format_string) \
>> static ssize_t \
>> field##_show(struct device *dev, \
>> - struct device_attribute *attr, char *buf) \
>> + struct device_attribute *attr, char *buf) \
>
> Seems unnecessary.

OK, I will drop these changes.

>
>> { \
>> struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
>> \
>> @@ -333,11 +333,52 @@ field##_show(struct device *dev, \
>> } \
>> static DEVICE_ATTR_RO(field);
>>
>> +#define rpmsg_string_attr(field, path) \
>
> "path" is an odd name for these, I think it's a "member".
>
>> +static ssize_t \
>> +field##_store(struct device *dev, \
>> + struct device_attribute *attr, const char *buf, size_t sz)\
>
> field##_store(struct device *dev, struct device_attribute *attr, \
> const char *buf, size_t sz) \
>
> Is prettier

OK, I will update this.

>
>> +{ \
>> + struct rpmsg_device *rpdev = to_rpmsg_device(dev); \
>> + char *new, *old, *cp; \
>> + \
>> + new = kstrndup(buf, sz, GFP_KERNEL); \
>> + if (!new) \
>> + return -ENOMEM; \
>> + \
>> + cp = strchr(new, '\n'); \
>> + if (cp) \
>> + *cp = '\0'; \
>
> I prefer
>
> new[strcspn(new, "\n")] = '\0';

Sure, I will update this.

Thanks,
Anup