linux-trace-devel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Re: [PATCH] libtracefs: Document function_filter API
  2021-03-25 16:31 [PATCH] libtracefs: Document function_filter API Sameeruddin shaik
@ 2021-03-24 21:40 ` Steven Rostedt
  2021-03-26  0:24   ` sameeruddin shaik
  0 siblings, 1 reply; 4+ messages in thread
From: Steven Rostedt @ 2021-03-24 21:40 UTC (permalink / raw)
  To: Sameeruddin shaik; +Cc: linux-trace-devel, tz.stoyanov

Hi Sameer!

On Thu, 25 Mar 2021 22:01:30 +0530
Sameeruddin shaik <sameeruddin.shaik8@gmail.com> wrote:

> Added documentation for the below API:
> tracefs_function_filter()
> 
> Signed-off-by: Sameeruddin shaik <sameeruddin.shaik8@gmail.com>
> 
> diff --git a/Documentation/libtracefs-function-filter.txt b/Documentation/libtracefs-function-filter.txt
> new file mode 100644
> index 0000000..08eece7
> --- /dev/null
> +++ b/Documentation/libtracefs-function-filter.txt
> @@ -0,0 +1,116 @@
> +libtracefs(3)
> +=============
> +
> +NAME
> +----
> +tracefs_function_filter -  Function to write to set_ftrace_filter
> +file of specific instance or top tracing instance

Since we are abstracting out tracefs, I'd like to avoid the implementation
details, and keep it more about its functionality.

 - Function to limit kernel functions that are traced

> +
> +SYNOPSIS
> +--------
> +[verse]
> +--
> +*#include <tracefs.h>*
> +
> +int tracefs_function_filter(struct tracefs_instance *instance, const char **filters, const char *module, bool reset, const char ***errs);
> +--
> +
> +DESCRIPTION
> +-----------
> +This Function can be used to write to set_ftrace_file in the trace
> +filesystem, which will limit the trace to only those functions,
> +which were written in the file.

To avoid stating the implementation:

"This function can be used to limit the Linux kernel functions that is
traced by the function and function-graph tracers."


> +
> +It will take take _instance_argument, that can be NULL for the
> +top level tracing, _filters_, which is nothing but the array of the

No need to say "nothing"

"which is an array of strings that represent a list of filters that should
be applied to define what functions are to be traced. The array must end
with a NULL pointer."

> +strings to write into the set_ftrace_filter file ,_module_ , name of the
> +module to be traced, _reset_, to reset or append the filters to file

"_reset_ if set will clear the current set of filters and then apply the
filter list, otherwise the list of filters are added to the current set of
filters."

> + and _errs_ which is again an array of failed filters, which can be NULL

"and _errs_ is a pointer an array of strings, which will be allocated if
any of filters fail to any available function. If _errs_ is NULL, it will
be ignored."

> + if failed filters are not of concern, and  returns 0  on successful write,
> + 1 or -x (where x is an integer) on error.

I mentioned in a previous email that I think we need to change the return
value, but for now this is fine.

> +
> +RETURN VALUE
> +------------
> +tracefs_function_filter will return 0 on successful write to file, if there

Just need to say ".. return 0 on success."

> +is error, it will return 1 for general errors or negative number -x(x denotes
> +number of failed filters), if there are any failed filters.

Again, I think the error return may be changed, so we can leave this for
now.

> +
> +In case of negative return value, errs have to be checked and must be freed
> +using the free()
> +
> +EXAMPLE
> +-------
> +[source,c]
> +--
> +#include <tracefs.h>
> +
> +#define INST "dummy"
> +
> +const char *filters[] = { "run_init_process", "try_to_run_init_process", "dummy1", NULL };
> +
> +int main(int argc, char *argv[])
> +{
> +	struct tracefs_instance *inst = tracefs_instance_create(INST);
> +	const char **errs = NULL;
> +	bool reset = 1;
> +	int ret;
> +	int i = 0;
> +
> +	if (!inst) {
> +	   /* Error creating new trace instance*/
> +	   }
> +
> +	ret = tracefs_function_filter(inst, filters, NULL, reset, &errs);
> +
> +	if (ret < 0 && errs) {
> +	   	while (errs[i])
> +		      printf("%s\n", errs[i++]);
> +	      }
> +
> +	tracefs_instance_free(inst);
> +	tracefs_instance_destroy(inst);
> +	free(errs);
> +	return 0;

This looks fine.

Thanks!

-- Steve

> +}
> +--
> +
> +FILES
> +-----
> +[verse]
> +--
> +*tracefs.h*
> +	Header file to include in order to have access to the library APIs.
> +*-ltracefs*
> +	Linker switch to add when building a program that uses the library.
> +--
> +
> +SEE ALSO
> +--------
> +_libtracefs(3)_,
> +_libtraceevent(3)_,
> +_trace-cmd(1)_
> +
> +AUTHOR
> +------
> +[verse]
> +--
> +*Steven Rostedt* <rostedt@goodmis.org>
> +*Tzvetomir Stoyanov* <tz.stoyanov@gmail.com>
> +*sameeruddin shaik* <sameeruddin.shaik8@gmail.com>
> +--
> +REPORTING BUGS
> +--------------
> +Report bugs to  <linux-trace-devel@vger.kernel.org>
> +
> +LICENSE
> +-------
> +libtracefs is Free Software licensed under the GNU LGPL 2.1
> +
> +RESOURCES
> +---------
> +https://git.kernel.org/pub/scm/libs/libtrace/libtracefs.git/
> +
> +COPYING
> +-------
> +Copyright \(C) 2020 VMware, Inc. Free use of this software is granted under
> +the terms of the GNU Public License (GPL).


^ permalink raw reply	[flat|nested] 4+ messages in thread

* [PATCH] libtracefs: Document function_filter API
@ 2021-03-25 16:31 Sameeruddin shaik
  2021-03-24 21:40 ` Steven Rostedt
  0 siblings, 1 reply; 4+ messages in thread
From: Sameeruddin shaik @ 2021-03-25 16:31 UTC (permalink / raw)
  To: rostedt; +Cc: linux-trace-devel, tz.stoyanov, Sameeruddin shaik

Added documentation for the below API:
tracefs_function_filter()

Signed-off-by: Sameeruddin shaik <sameeruddin.shaik8@gmail.com>

diff --git a/Documentation/libtracefs-function-filter.txt b/Documentation/libtracefs-function-filter.txt
new file mode 100644
index 0000000..08eece7
--- /dev/null
+++ b/Documentation/libtracefs-function-filter.txt
@@ -0,0 +1,116 @@
+libtracefs(3)
+=============
+
+NAME
+----
+tracefs_function_filter -  Function to write to set_ftrace_filter
+file of specific instance or top tracing instance
+
+SYNOPSIS
+--------
+[verse]
+--
+*#include <tracefs.h>*
+
+int tracefs_function_filter(struct tracefs_instance *instance, const char **filters, const char *module, bool reset, const char ***errs);
+--
+
+DESCRIPTION
+-----------
+This Function can be used to write to set_ftrace_file in the trace
+filesystem, which will limit the trace to only those functions,
+which were written in the file.
+
+It will take take _instance_argument, that can be NULL for the
+top level tracing, _filters_, which is nothing but the array of the
+strings to write into the set_ftrace_filter file ,_module_ , name of the
+module to be traced, _reset_, to reset or append the filters to file
+ and _errs_ which is again an array of failed filters, which can be NULL
+ if failed filters are not of concern, and  returns 0  on successful write,
+ 1 or -x (where x is an integer) on error.
+
+RETURN VALUE
+------------
+tracefs_function_filter will return 0 on successful write to file, if there
+is error, it will return 1 for general errors or negative number -x(x denotes
+number of failed filters), if there are any failed filters.
+
+In case of negative return value, errs have to be checked and must be freed
+using the free()
+
+EXAMPLE
+-------
+[source,c]
+--
+#include <tracefs.h>
+
+#define INST "dummy"
+
+const char *filters[] = { "run_init_process", "try_to_run_init_process", "dummy1", NULL };
+
+int main(int argc, char *argv[])
+{
+	struct tracefs_instance *inst = tracefs_instance_create(INST);
+	const char **errs = NULL;
+	bool reset = 1;
+	int ret;
+	int i = 0;
+
+	if (!inst) {
+	   /* Error creating new trace instance*/
+	   }
+
+	ret = tracefs_function_filter(inst, filters, NULL, reset, &errs);
+
+	if (ret < 0 && errs) {
+	   	while (errs[i])
+		      printf("%s\n", errs[i++]);
+	      }
+
+	tracefs_instance_free(inst);
+	tracefs_instance_destroy(inst);
+	free(errs);
+	return 0;
+}
+--
+
+FILES
+-----
+[verse]
+--
+*tracefs.h*
+	Header file to include in order to have access to the library APIs.
+*-ltracefs*
+	Linker switch to add when building a program that uses the library.
+--
+
+SEE ALSO
+--------
+_libtracefs(3)_,
+_libtraceevent(3)_,
+_trace-cmd(1)_
+
+AUTHOR
+------
+[verse]
+--
+*Steven Rostedt* <rostedt@goodmis.org>
+*Tzvetomir Stoyanov* <tz.stoyanov@gmail.com>
+*sameeruddin shaik* <sameeruddin.shaik8@gmail.com>
+--
+REPORTING BUGS
+--------------
+Report bugs to  <linux-trace-devel@vger.kernel.org>
+
+LICENSE
+-------
+libtracefs is Free Software licensed under the GNU LGPL 2.1
+
+RESOURCES
+---------
+https://git.kernel.org/pub/scm/libs/libtrace/libtracefs.git/
+
+COPYING
+-------
+Copyright \(C) 2020 VMware, Inc. Free use of this software is granted under
+the terms of the GNU Public License (GPL).
-- 
2.7.4


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] libtracefs: Document function_filter API
  2021-03-24 21:40 ` Steven Rostedt
@ 2021-03-26  0:24   ` sameeruddin shaik
  0 siblings, 0 replies; 4+ messages in thread
From: sameeruddin shaik @ 2021-03-26  0:24 UTC (permalink / raw)
  To: Steven Rostedt; +Cc: linux-trace-devel, tz.stoyanov

hi steve,

On 25/03/21 3:10 am, Steven Rostedt wrote:
> Hi Sameer!
>
> On Thu, 25 Mar 2021 22:01:30 +0530
> Sameeruddin shaik <sameeruddin.shaik8@gmail.com> wrote:
>
>> Added documentation for the below API:
>> tracefs_function_filter()
>>
>> Signed-off-by: Sameeruddin shaik <sameeruddin.shaik8@gmail.com>
>>
>> diff --git a/Documentation/libtracefs-function-filter.txt b/Documentation/libtracefs-function-filter.txt
>> new file mode 100644
>> index 0000000..08eece7
>> --- /dev/null
>> +++ b/Documentation/libtracefs-function-filter.txt
>> @@ -0,0 +1,116 @@
>> +libtracefs(3)
>> +=============
>> +
>> +NAME
>> +----
>> +tracefs_function_filter -  Function to write to set_ftrace_filter
>> +file of specific instance or top tracing instance
> Since we are abstracting out tracefs, I'd like to avoid the implementation
> details, and keep it more about its functionality.
>
>   - Function to limit kernel functions that are traced
>
>> +
>> +SYNOPSIS
>> +--------
>> +[verse]
>> +--
>> +*#include <tracefs.h>*
>> +
>> +int tracefs_function_filter(struct tracefs_instance *instance, const char **filters, const char *module, bool reset, const char ***errs);
>> +--
>> +
>> +DESCRIPTION
>> +-----------
>> +This Function can be used to write to set_ftrace_file in the trace
>> +filesystem, which will limit the trace to only those functions,
>> +which were written in the file.
> To avoid stating the implementation:
>
> "This function can be used to limit the Linux kernel functions that is
> traced by the function and function-graph tracers."
>
>
>> +
>> +It will take take _instance_argument, that can be NULL for the
>> +top level tracing, _filters_, which is nothing but the array of the
> No need to say "nothing"
>
> "which is an array of strings that represent a list of filters that should
> be applied to define what functions are to be traced. The array must end
> with a NULL pointer."
>
>> +strings to write into the set_ftrace_filter file ,_module_ , name of the
>> +module to be traced, _reset_, to reset or append the filters to file
> "_reset_ if set will clear the current set of filters and then apply the
> filter list, otherwise the list of filters are added to the current set of
> filters."
>
>> + and _errs_ which is again an array of failed filters, which can be NULL
> "and _errs_ is a pointer an array of strings, which will be allocated if
> any of filters fail to any available function. If _errs_ is NULL, it will
> be ignored."
>
>> + if failed filters are not of concern, and  returns 0  on successful write,
>> + 1 or -x (where x is an integer) on error.
> I mentioned in a previous email that I think we need to change the return
> value, but for now this is fine.
In future, when the changes were made, i will change this man page 
accordingly.
>> +
>> +RETURN VALUE
>> +------------
>> +tracefs_function_filter will return 0 on successful write to file, if there
> Just need to say ".. return 0 on success."
>
>> +is error, it will return 1 for general errors or negative number -x(x denotes
>> +number of failed filters), if there are any failed filters.
> Again, I think the error return may be changed, so we can leave this for
> now.
>
>> +
>> +In case of negative return value, errs have to be checked and must be freed
>> +using the free()
>> +
>> +EXAMPLE
>> +-------
>> +[source,c]
>> +--
>> +#include <tracefs.h>
>> +
>> +#define INST "dummy"
>> +
>> +const char *filters[] = { "run_init_process", "try_to_run_init_process", "dummy1", NULL };
>> +
>> +int main(int argc, char *argv[])
>> +{
>> +	struct tracefs_instance *inst = tracefs_instance_create(INST);
>> +	const char **errs = NULL;
>> +	bool reset = 1;
>> +	int ret;
>> +	int i = 0;
>> +
>> +	if (!inst) {
>> +	   /* Error creating new trace instance*/
>> +	   }
>> +
>> +	ret = tracefs_function_filter(inst, filters, NULL, reset, &errs);
>> +
>> +	if (ret < 0 && errs) {
>> +	   	while (errs[i])
>> +		      printf("%s\n", errs[i++]);
>> +	      }
>> +
>> +	tracefs_instance_free(inst);
>> +	tracefs_instance_destroy(inst);
>> +	free(errs);
>> +	return 0;
> This looks fine.
>
> Thanks!
>
> -- Steve
>
>> +}
>> +--
>> +
>> +FILES
>> +-----
>> +[verse]
>> +--
>> +*tracefs.h*
>> +	Header file to include in order to have access to the library APIs.
>> +*-ltracefs*
>> +	Linker switch to add when building a program that uses the library.
>> +--
>> +
>> +SEE ALSO
>> +--------
>> +_libtracefs(3)_,
>> +_libtraceevent(3)_,
>> +_trace-cmd(1)_
>> +
>> +AUTHOR
>> +------
>> +[verse]
>> +--
>> +*Steven Rostedt* <rostedt@goodmis.org>
>> +*Tzvetomir Stoyanov* <tz.stoyanov@gmail.com>
>> +*sameeruddin shaik* <sameeruddin.shaik8@gmail.com>
>> +--
>> +REPORTING BUGS
>> +--------------
>> +Report bugs to  <linux-trace-devel@vger.kernel.org>
>> +
>> +LICENSE
>> +-------
>> +libtracefs is Free Software licensed under the GNU LGPL 2.1
>> +
>> +RESOURCES
>> +---------
>> +https://git.kernel.org/pub/scm/libs/libtrace/libtracefs.git/
>> +
>> +COPYING
>> +-------
>> +Copyright \(C) 2020 VMware, Inc. Free use of this software is granted under
>> +the terms of the GNU Public License (GPL).

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [PATCH] libtracefs: Document function_filter API
@ 2021-03-26  0:43 Sameeruddin shaik
  0 siblings, 0 replies; 4+ messages in thread
From: Sameeruddin shaik @ 2021-03-26  0:43 UTC (permalink / raw)
  To: rostedt; +Cc: linux-trace-devel, tz.stoyanov, Sameeruddin shaik

Added documentation for the below API:
tracefs_function_filter()

Signed-off-by: Sameeruddin shaik <sameeruddin.shaik8@gmail.com>

diff --git a/Documentation/libtracefs-function-filter.txt b/Documentation/libtracefs-function-filter.txt
new file mode 100644
index 0000000..34d8979
--- /dev/null
+++ b/Documentation/libtracefs-function-filter.txt
@@ -0,0 +1,120 @@
+libtracefs(3)
+=============
+
+NAME
+----
+tracefs_function_filter - Function to limit kernel functions that are traced
+
+SYNOPSIS
+--------
+[verse]
+--
+*#include <tracefs.h>*
+
+int tracefs_function_filter(struct tracefs_instance *instance, const char **filters, const char *module, bool reset, const char ***errs);
+--
+
+DESCRIPTION
+-----------
+This function can be used to limit the Linux kernel functions that were
+traced by the function and function-graph tracers
+
+It will take
+_instance_ , that can be NULL for the top level tracing.
+_filters_, which is an array of the strings that represent a list of filters that should
+be applied to define what functions are to be traced and The array must end
+with a NULL pointer.
+_module_ , name of the module to be traced.
+_reset_ if set will clear the current set of filters and then apply the
+filter list, otherwise the list of filters are added to the current set of
+filters,
+_errs_, is a pointer to an array of strings, which will be allocated if
+any of filters fail to match any available function, If _errs_ is NULL, it will
+be ignored.
+
+returns 0  on success, 1 or -x (where x is an integer) on error.
+
+RETURN VALUE
+------------
+return 0 on success, if there is error, it will return 1 for general errors or
+negative number -x(x denotes number of failed filters), if there are any failed filters.
+
+In case of negative return value, errs have to be checked and must be freed
+using the free()
+
+EXAMPLE
+-------
+[source,c]
+--
+#include <tracefs.h>
+
+#define INST "dummy"
+
+const char *filters[] = { "run_init_process", "try_to_run_init_process", "dummy1", NULL };
+
+int main(int argc, char *argv[])
+{
+	struct tracefs_instance *inst = tracefs_instance_create(INST);
+	const char **errs = NULL;
+	bool reset = 1;
+	int ret;
+	int i = 0;
+
+	if (!inst) {
+	   /* Error creating new trace instance*/
+	   }
+
+	ret = tracefs_function_filter(inst, filters, NULL, reset, &errs);
+
+	if (ret < 0 && errs) {
+	   	while (errs[i])
+		      printf("%s\n", errs[i++]);
+	      }
+
+	tracefs_instance_free(inst);
+	tracefs_instance_destroy(inst);
+	free(errs);
+	return 0;
+}
+--
+
+FILES
+-----
+[verse]
+--
+*tracefs.h*
+	Header file to include in order to have access to the library APIs.
+*-ltracefs*
+	Linker switch to add when building a program that uses the library.
+--
+
+SEE ALSO
+--------
+_libtracefs(3)_,
+_libtraceevent(3)_,
+_trace-cmd(1)_
+
+AUTHOR
+------
+[verse]
+--
+*Steven Rostedt* <rostedt@goodmis.org>
+*Tzvetomir Stoyanov* <tz.stoyanov@gmail.com>
+*sameeruddin shaik* <sameeruddin.shaik8@gmail.com>
+--
+REPORTING BUGS
+--------------
+Report bugs to  <linux-trace-devel@vger.kernel.org>
+
+LICENSE
+-------
+libtracefs is Free Software licensed under the GNU LGPL 2.1
+
+RESOURCES
+---------
+https://git.kernel.org/pub/scm/libs/libtrace/libtracefs.git/
+
+COPYING
+-------
+Copyright \(C) 2020 VMware, Inc. Free use of this software is granted under
+the terms of the GNU Public License (GPL).
-- 
2.7.4


^ permalink raw reply related	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2021-03-25  0:44 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-03-25 16:31 [PATCH] libtracefs: Document function_filter API Sameeruddin shaik
2021-03-24 21:40 ` Steven Rostedt
2021-03-26  0:24   ` sameeruddin shaik
2021-03-26  0:43 Sameeruddin shaik

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).