[PATCH v1 05/12] tracefs: add TP_printk_no_nl - RFC

Łukasz Bartosik posted 12 patches 2 years, 1 month ago
There is a newer version of this series
[PATCH v1 05/12] tracefs: add TP_printk_no_nl - RFC
Posted by Łukasz Bartosik 2 years, 1 month ago
From: Jim Cromie <jim.cromie@gmail.com>

This variant of TP_printk() does *not* add the trailing newline.  It
is for use by printk/debug-ish events which already have a trailing
newline.  Its here to support:

https://lore.kernel.org/lkml/
20200825153338.17061-1-vincent.whitchurch@axis.com/
which taught dyndbg to send pr_debug() msgs to tracefs, via -x/T flag.

It "reused" the include/trace/events/printk.h console event,
which does the following:

	TP_fast_assign(
		/*
		 * Each trace entry is printed in a new line.
		 * If the msg finishes with '\n', cut it off
		 * to avoid blank lines in the trace.
		 */
		if ((len > 0) && (text[len-1] == '\n'))
			len -= 1;

		memcpy(__get_str(msg), text, len);
		__get_str(msg)[len] = 0;
	),

That trim work could be avoided, *iff* all pr_debug() callers are
known to have no '\n' to strip.  While thats not true for *all*
callsites, it is 99+% true for DRM.debug callsites, and can be made
true for some subsets of prdbg/dyndbg callsites.

WANTED: macros to validate that a literal format-str has or doesn't
have a trailing newline, or to provide or trim trailing newline(s?).
Should be usable in TP_printk* defns, for use in new event defns.

Cc: <rostedt@goodmis.org>
Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: <daniel@ffwll.ch>
Cc: <pmladek@suse.com>
Cc: <sergey.senozhatsky@gmail.com>
Cc: <john.ogness@linutronix.de>
Cc: Simon Ser <contact@emersion.fr>
Cc: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
---
 include/trace/stages/stage3_trace_output.h | 3 +++
 include/trace/stages/stage7_class_define.h | 3 +++
 2 files changed, 6 insertions(+)

diff --git a/include/trace/stages/stage3_trace_output.h b/include/trace/stages/stage3_trace_output.h
index c1fb1355d309..5f5c1374fa10 100644
--- a/include/trace/stages/stage3_trace_output.h
+++ b/include/trace/stages/stage3_trace_output.h
@@ -8,6 +8,9 @@
 #undef TP_printk
 #define TP_printk(fmt, args...) fmt "\n", args
 
+#undef TP_printk_no_nl
+#define TP_printk_no_nl(fmt, args...) fmt, args
+
 #undef __get_dynamic_array
 #define __get_dynamic_array(field)	\
 		((void *)__entry + (__entry->__data_loc_##field & 0xffff))
diff --git a/include/trace/stages/stage7_class_define.h b/include/trace/stages/stage7_class_define.h
index bcb960d16fc0..8247e4478f19 100644
--- a/include/trace/stages/stage7_class_define.h
+++ b/include/trace/stages/stage7_class_define.h
@@ -37,3 +37,6 @@
 
 #undef TP_printk
 #define TP_printk(fmt, args...) "\"" fmt "\", "  __stringify(args)
+
+#undef TP_printk_no_nl
+#define TP_printk_no_nl(fmt, args...) "\"" fmt "\", "  __stringify(args)
-- 
2.42.0.869.gea05f2083d-goog
Re: [PATCH v1 05/12] tracefs: add TP_printk_no_nl - RFC
Posted by jim.cromie@gmail.com 2 years, 1 month ago
On Fri, Nov 3, 2023 at 7:10 AM Łukasz Bartosik <lb@semihalf.com> wrote:
>
> From: Jim Cromie <jim.cromie@gmail.com>
>
> This variant of TP_printk() does *not* add the trailing newline.  It
> is for use by printk/debug-ish events which already have a trailing
> newline.  Its here to support:
>
> https://lore.kernel.org/lkml/
> 20200825153338.17061-1-vincent.whitchurch@axis.com/
> which taught dyndbg to send pr_debug() msgs to tracefs, via -x/T flag.
>
> It "reused" the include/trace/events/printk.h console event,
> which does the following:
>
>         TP_fast_assign(
>                 /*
>                  * Each trace entry is printed in a new line.
>                  * If the msg finishes with '\n', cut it off
>                  * to avoid blank lines in the trace.
>                  */
>                 if ((len > 0) && (text[len-1] == '\n'))
>                         len -= 1;
>
>                 memcpy(__get_str(msg), text, len);
>                 __get_str(msg)[len] = 0;
>         ),
>
> That trim work could be avoided, *iff* all pr_debug() callers are
> known to have no '\n' to strip.  While thats not true for *all*
> callsites, it is 99+% true for DRM.debug callsites, and can be made
> true for some subsets of prdbg/dyndbg callsites.

some or all of DRM.debug messages (that I audited / caught)
were merged by Maxime recently, I;ll go back (later) to see if I missed any.

>
> WANTED: macros to validate that a literal format-str has or doesn't
> have a trailing newline, or to provide or trim trailing newline(s?).
> Should be usable in TP_printk* defns, for use in new event defns.

that might be over-optimizing

Steve,
IIRC you considered adding \n where needed.
is there anything gained (conceivably) by not just adding the trailing
\n when "needed" ?
statistically, macros could get us to 99.99+ "compliance"
IIRC - the "needed" seems correct.


>
> Cc: <rostedt@goodmis.org>
> Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
> Cc: <daniel@ffwll.ch>
> Cc: <pmladek@suse.com>
> Cc: <sergey.senozhatsky@gmail.com>
> Cc: <john.ogness@linutronix.de>
> Cc: Simon Ser <contact@emersion.fr>
> Cc: Sean Paul <seanpaul@chromium.org>
> Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
> ---
>  include/trace/stages/stage3_trace_output.h | 3 +++
>  include/trace/stages/stage7_class_define.h | 3 +++
>  2 files changed, 6 insertions(+)
>
> diff --git a/include/trace/stages/stage3_trace_output.h b/include/trace/stages/stage3_trace_output.h
> index c1fb1355d309..5f5c1374fa10 100644
> --- a/include/trace/stages/stage3_trace_output.h
> +++ b/include/trace/stages/stage3_trace_output.h
> @@ -8,6 +8,9 @@
>  #undef TP_printk
>  #define TP_printk(fmt, args...) fmt "\n", args
>
> +#undef TP_printk_no_nl
> +#define TP_printk_no_nl(fmt, args...) fmt, args
> +
>  #undef __get_dynamic_array
>  #define __get_dynamic_array(field)     \
>                 ((void *)__entry + (__entry->__data_loc_##field & 0xffff))
> diff --git a/include/trace/stages/stage7_class_define.h b/include/trace/stages/stage7_class_define.h
> index bcb960d16fc0..8247e4478f19 100644
> --- a/include/trace/stages/stage7_class_define.h
> +++ b/include/trace/stages/stage7_class_define.h
> @@ -37,3 +37,6 @@
>
>  #undef TP_printk
>  #define TP_printk(fmt, args...) "\"" fmt "\", "  __stringify(args)
> +
> +#undef TP_printk_no_nl
> +#define TP_printk_no_nl(fmt, args...) "\"" fmt "\", "  __stringify(args)
> --
> 2.42.0.869.gea05f2083d-goog
>
Re: [PATCH v1 05/12] tracefs: add TP_printk_no_nl - RFC
Posted by Steven Rostedt 2 years, 1 month ago
On Fri, 3 Nov 2023 21:40:13 -0600
jim.cromie@gmail.com wrote:

> Steve,
> IIRC you considered adding \n where needed.
> is there anything gained (conceivably) by not just adding the trailing
> \n when "needed" ?
> statistically, macros could get us to 99.99+ "compliance"
> IIRC - the "needed" seems correct.
> 

I replied to the next patch suggesting to do this on the read side.

-- Steve