[PATCH 0/2] tracing: Fixes for filter

[PATCH 0/2] tracing: Fixes for filter

[Masami Hiramatsu (Google)]

Hi,

Here is a patch series to fix some issues on the trace event
and function filters.

The first patch fixes an issue that the event filter can not
handle the string pointer with BTF attribute tag. This happens
with CONFIG_DEBUG_INFO_BTF=y and PAHOLE_HAS_BTF_TAG=y.

The second patch fixes a selftest issue on the function glob
filter. Since mutex_trylock() can be an inline function, it
is not a good example for ftrace. This replaces it with
mutex_unlock().

Thank you,

---
Masami Hiramatsu (Google) (2):
      tracing: Handle "(const) char __attribute() *" as string ptr type
      selftests: tracing: Use mutex_unlock for testing glob filter

 kernel/trace/trace_events_filter.c                         | 5 +++++
 tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc | 2 +-
 2 files changed, 6 insertions(+), 1 deletion(-)

[PATCH 1/2] tracing: Handle "(const) char __attribute() *" as string ptr type

[Masami Hiramatsu (Google)]

From: Masami Hiramatsu (Google) <mhiramat@kernel.org>

With CONFIG_DEBUG_INFO_BTF=y and PAHOLE_HAS_BTF_TAG=y, `__user` is
converted to `__attribute((btf_type_tag("user")))`. In this case,
some syscall events have `const char __attribute(...) *` for __user
string, like below;

/sys/kernel/tracing # cat events/syscalls/sys_enter_openat/format
name: sys_enter_openat
ID: 720
format:
	field:unsigned short common_type;	offset:0;	size:2;	signed:0;
	field:unsigned char common_flags;	offset:2;	size:1;	signed:0;
	field:unsigned char common_preempt_count;	offset:3;	size:1;	signed:0;
	field:int common_pid;	offset:4;	size:4;	signed:1;

	field:int __syscall_nr;	offset:8;	size:4;	signed:1;
	field:int dfd;	offset:16;	size:8;	signed:0;
	field:const char __attribute__((btf_type_tag("user"))) * filename;	offset:24;	size:8;	signed:0;
	field:int flags;	offset:32;	size:8;	signed:0;
	field:umode_t mode;	offset:40;	size:8;	signed:0;


Then the trace event filter failes to set string acceptable flag
(FILTER_PTR_STRING) and rejects setting string filter;

# echo 'filename.ustring ~ "*ftracetest-dir.wbx24v*"' >> events/syscalls/sys_enter_openat/filter
sh: write error: Invalid argument
 # cat error_log
[  723.743637] event filter parse error: error: Expecting numeric field
  Command: filename.ustring ~ "*ftracetest-dir.wbx24v*"

Handle "(const) char __attribute() *" also as string ptr type.

Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
 kernel/trace/trace_events_filter.c |    5 +++++
 1 file changed, 5 insertions(+)

diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 3885aadc434d..5e27190a0377 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -1488,6 +1488,11 @@ int filter_assign_type(const char *type)
 	if (strcmp(type, "char *") == 0 || strcmp(type, "const char *") == 0)
 		return FILTER_PTR_STRING;
 
+	/* Ignore attributes */
+	if (glob_match("char __attribute(*) \\*", type) ||
+	    glob_match("const char __attribute(*) \\*", type))
+		return FILTER_PTR_STRING;
+
 	return FILTER_OTHER;
 }
 

Re: [PATCH 1/2] tracing: Handle "(const) char __attribute() *" as string ptr type

[Masami Hiramatsu]

On Thu,  3 Jul 2025 13:26:35 +0900
"Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:

> From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> 
> With CONFIG_DEBUG_INFO_BTF=y and PAHOLE_HAS_BTF_TAG=y, `__user` is
> converted to `__attribute((btf_type_tag("user")))`. In this case,
> some syscall events have `const char __attribute(...) *` for __user
> string, like below;
> 
> /sys/kernel/tracing # cat events/syscalls/sys_enter_openat/format
> name: sys_enter_openat
> ID: 720
> format:
> 	field:unsigned short common_type;	offset:0;	size:2;	signed:0;
> 	field:unsigned char common_flags;	offset:2;	size:1;	signed:0;
> 	field:unsigned char common_preempt_count;	offset:3;	size:1;	signed:0;
> 	field:int common_pid;	offset:4;	size:4;	signed:1;
> 
> 	field:int __syscall_nr;	offset:8;	size:4;	signed:1;
> 	field:int dfd;	offset:16;	size:8;	signed:0;
> 	field:const char __attribute__((btf_type_tag("user"))) * filename;	offset:24;

Ahh, sorry, it was "__attribute__()" not "__attribute()".

Anyway, I think we should sanitize the __attribute__ from
format.

Thank you,


	size:8;	signed:0;
> 	field:int flags;	offset:32;	size:8;	signed:0;
> 	field:umode_t mode;	offset:40;	size:8;	signed:0;
> 
> 
> Then the trace event filter failes to set string acceptable flag
> (FILTER_PTR_STRING) and rejects setting string filter;
> 
> # echo 'filename.ustring ~ "*ftracetest-dir.wbx24v*"' >> events/syscalls/sys_enter_openat/filter
> sh: write error: Invalid argument
>  # cat error_log
> [  723.743637] event filter parse error: error: Expecting numeric field
>   Command: filename.ustring ~ "*ftracetest-dir.wbx24v*"
> 
> Handle "(const) char __attribute() *" also as string ptr type.
> 
> Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> ---
>  kernel/trace/trace_events_filter.c |    5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
> index 3885aadc434d..5e27190a0377 100644
> --- a/kernel/trace/trace_events_filter.c
> +++ b/kernel/trace/trace_events_filter.c
> @@ -1488,6 +1488,11 @@ int filter_assign_type(const char *type)
>  	if (strcmp(type, "char *") == 0 || strcmp(type, "const char *") == 0)
>  		return FILTER_PTR_STRING;
>  
> +	/* Ignore attributes */
> +	if (glob_match("char __attribute(*) \\*", type) ||
> +	    glob_match("const char __attribute(*) \\*", type))
> +		return FILTER_PTR_STRING;
> +
>  	return FILTER_OTHER;
>  }
>  
> 


-- 
Masami Hiramatsu (Google) <mhiramat@kernel.org>

[PATCH 2/2] selftests: tracing: Use mutex_unlock for testing glob filter

[Masami Hiramatsu (Google)]

From: Masami Hiramatsu (Google) <mhiramat@kernel.org>

Since commit c5b6ababd21a ("locking/mutex: implement
mutex_trylock_nested") makes mutex_trylock() as an inlined
function if CONFIG_DEBUG_LOCK_ALLOC=y, we can not use
mutex_trylock() for testing the glob filter of ftrace.

Use mutex_unlock instead.

Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
 .../ftrace/test.d/ftrace/func-filter-glob.tc       |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc b/tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc
index 4b994b6df5ac..ed81eaf2afd6 100644
--- a/tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc
+++ b/tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc
@@ -29,7 +29,7 @@ ftrace_filter_check 'schedule*' '^schedule.*$'
 ftrace_filter_check '*pin*lock' '.*pin.*lock$'
 
 # filter by start*mid*
-ftrace_filter_check 'mutex*try*' '^mutex.*try.*'
+ftrace_filter_check 'mutex*unl*' '^mutex.*unl.*'
 
 # Advanced full-glob matching feature is recently supported.
 # Skip the tests if we are sure the kernel does not support it.

Re: [PATCH 2/2] selftests: tracing: Use mutex_unlock for testing glob filter

[Steven Rostedt]

On Thu,  3 Jul 2025 13:26:43 +0900
"Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:

> From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> 
> Since commit c5b6ababd21a ("locking/mutex: implement
> mutex_trylock_nested") makes mutex_trylock() as an inlined
> function if CONFIG_DEBUG_LOCK_ALLOC=y, we can not use
> mutex_trylock() for testing the glob filter of ftrace.
> 
> Use mutex_unlock instead.
> 
> Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>

Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>

-- Steve

Re: [PATCH 2/2] selftests: tracing: Use mutex_unlock for testing glob filter

[Masami Hiramatsu]

Shuah,

Can you take this patch (fix) to your tree since this is
a fix for the selftest?

Thank you,

On Thu, 3 Jul 2025 10:30:39 -0400
Steven Rostedt <rostedt@goodmis.org> wrote:

> On Thu,  3 Jul 2025 13:26:43 +0900
> "Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:
> 
> > From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> > 
> > Since commit c5b6ababd21a ("locking/mutex: implement
> > mutex_trylock_nested") makes mutex_trylock() as an inlined
> > function if CONFIG_DEBUG_LOCK_ALLOC=y, we can not use
> > mutex_trylock() for testing the glob filter of ftrace.
> > 
> > Use mutex_unlock instead.
> > 
> > Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> 
> Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
> 
> -- Steve


-- 
Masami Hiramatsu (Google) <mhiramat@kernel.org>

Re: [PATCH 2/2] selftests: tracing: Use mutex_unlock for testing glob filter

[Shuah Khan]

On 7/4/25 00:04, Masami Hiramatsu (Google) wrote:
> Shuah,
> 
> Can you take this patch (fix) to your tree since this is
> a fix for the selftest?
> 
> Thank you,
> 
> On Thu, 3 Jul 2025 10:30:39 -0400
> Steven Rostedt <rostedt@goodmis.org> wrote:
> 
>> On Thu,  3 Jul 2025 13:26:43 +0900
>> "Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:
>>
>>> From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
>>>
>>> Since commit c5b6ababd21a ("locking/mutex: implement
>>> mutex_trylock_nested") makes mutex_trylock() as an inlined
>>> function if CONFIG_DEBUG_LOCK_ALLOC=y, we can not use
>>> mutex_trylock() for testing the glob filter of ftrace.
>>>
>>> Use mutex_unlock instead.
>>>
>>> Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
>>
>> Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
>>
>> -- Steve
> 
> 

Applied to linux-kselftest next

thanks,
-- Shuah

Re: [PATCH 0/2] tracing: Fixes for filter

[Steven Rostedt]

On Thu,  3 Jul 2025 13:26:26 +0900
"Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:

> Hi,
> 
> Here is a patch series to fix some issues on the trace event
> and function filters.
> 
> The first patch fixes an issue that the event filter can not
> handle the string pointer with BTF attribute tag. This happens
> with CONFIG_DEBUG_INFO_BTF=y and PAHOLE_HAS_BTF_TAG=y.
> 
> The second patch fixes a selftest issue on the function glob
> filter. Since mutex_trylock() can be an inline function, it
> is not a good example for ftrace. This replaces it with
> mutex_unlock().
> 
> Thank you,
> 
> ---
> Masami Hiramatsu (Google) (2):
>       tracing: Handle "(const) char __attribute() *" as string ptr type

I can take the first patch.

>       selftests: tracing: Use mutex_unlock for testing glob filter

This patch should go via Shuah's tree.

-- Steve

> 
>  kernel/trace/trace_events_filter.c                         | 5 +++++
>  tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc | 2 +-
>  2 files changed, 6 insertions(+), 1 deletion(-)