* [PATCH v2 2/2] mm: mmap_lock: use DECLARE_EVENT_CLASS and DEFINE_EVENT_FN
@ 2021-10-09 7:12 Gang Li
2021-10-10 21:49 ` Steven Rostedt
2021-10-11 8:02 ` Vlastimil Babka
0 siblings, 2 replies; 3+ messages in thread
From: Gang Li @ 2021-10-09 7:12 UTC (permalink / raw)
To: rostedt
Cc: mingo, akpm, vbabka, axelrasmussen, linux-kernel, linux-mm, Gang Li
By using DECLARE_EVENT_CLASS and TRACE_EVENT_FN, we can save a lot
of space from duplicate code.
Signed-off-by: Gang Li <ligang.bdlg@bytedance.com>
---
include/trace/events/mmap_lock.h | 44 +++++++++-----------------------
1 file changed, 12 insertions(+), 32 deletions(-)
diff --git a/include/trace/events/mmap_lock.h b/include/trace/events/mmap_lock.h
index 5f980c92e3e9..14db8044c1ff 100644
--- a/include/trace/events/mmap_lock.h
+++ b/include/trace/events/mmap_lock.h
@@ -13,7 +13,7 @@ struct mm_struct;
extern int trace_mmap_lock_reg(void);
extern void trace_mmap_lock_unreg(void);
-TRACE_EVENT_FN(mmap_lock_start_locking,
+DECLARE_EVENT_CLASS(mmap_lock,
TP_PROTO(struct mm_struct *mm, const char *memcg_path, bool write),
@@ -36,11 +36,19 @@ TRACE_EVENT_FN(mmap_lock_start_locking,
__entry->mm,
__get_str(memcg_path),
__entry->write ? "true" : "false"
- ),
-
- trace_mmap_lock_reg, trace_mmap_lock_unreg
+ )
);
+#define DEFINE_MMAP_LOCK_EVENT(name) \
+ DEFINE_EVENT_FN(mmap_lock, name, \
+ TP_PROTO(struct mm_struct *mm, const char *memcg_path, \
+ bool write), \
+ TP_ARGS(mm, memcg_path, write), \
+ trace_mmap_lock_reg, trace_mmap_lock_unreg)
+
+DEFINE_MMAP_LOCK_EVENT(mmap_lock_start_locking);
+DEFINE_MMAP_LOCK_EVENT(mmap_lock_released);
+
TRACE_EVENT_FN(mmap_lock_acquire_returned,
TP_PROTO(struct mm_struct *mm, const char *memcg_path, bool write,
@@ -73,34 +81,6 @@ TRACE_EVENT_FN(mmap_lock_acquire_returned,
trace_mmap_lock_reg, trace_mmap_lock_unreg
);
-TRACE_EVENT_FN(mmap_lock_released,
-
- TP_PROTO(struct mm_struct *mm, const char *memcg_path, bool write),
-
- TP_ARGS(mm, memcg_path, write),
-
- TP_STRUCT__entry(
- __field(struct mm_struct *, mm)
- __string(memcg_path, memcg_path)
- __field(bool, write)
- ),
-
- TP_fast_assign(
- __entry->mm = mm;
- __assign_str(memcg_path, memcg_path);
- __entry->write = write;
- ),
-
- TP_printk(
- "mm=%p memcg_path=%s write=%s",
- __entry->mm,
- __get_str(memcg_path),
- __entry->write ? "true" : "false"
- ),
-
- trace_mmap_lock_reg, trace_mmap_lock_unreg
-);
-
#endif /* _TRACE_MMAP_LOCK_H */
/* This part must be outside protection */
--
2.20.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH v2 2/2] mm: mmap_lock: use DECLARE_EVENT_CLASS and DEFINE_EVENT_FN
2021-10-09 7:12 [PATCH v2 2/2] mm: mmap_lock: use DECLARE_EVENT_CLASS and DEFINE_EVENT_FN Gang Li
@ 2021-10-10 21:49 ` Steven Rostedt
2021-10-11 8:02 ` Vlastimil Babka
1 sibling, 0 replies; 3+ messages in thread
From: Steven Rostedt @ 2021-10-10 21:49 UTC (permalink / raw)
To: Gang Li; +Cc: mingo, akpm, vbabka, axelrasmussen, linux-kernel, linux-mm
On Sat, 9 Oct 2021 15:12:43 +0800
Gang Li <ligang.bdlg@bytedance.com> wrote:
> By using DECLARE_EVENT_CLASS and TRACE_EVENT_FN, we can save a lot
> of space from duplicate code.
Thanks for doing this. Looking at the events, they can certainly be
merged into one.
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
-- Steve
>
> Signed-off-by: Gang Li <ligang.bdlg@bytedance.com>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH v2 2/2] mm: mmap_lock: use DECLARE_EVENT_CLASS and DEFINE_EVENT_FN
2021-10-09 7:12 [PATCH v2 2/2] mm: mmap_lock: use DECLARE_EVENT_CLASS and DEFINE_EVENT_FN Gang Li
2021-10-10 21:49 ` Steven Rostedt
@ 2021-10-11 8:02 ` Vlastimil Babka
1 sibling, 0 replies; 3+ messages in thread
From: Vlastimil Babka @ 2021-10-11 8:02 UTC (permalink / raw)
To: Gang Li, rostedt; +Cc: mingo, akpm, axelrasmussen, linux-kernel, linux-mm
On 10/9/21 09:12, Gang Li wrote:
> By using DECLARE_EVENT_CLASS and TRACE_EVENT_FN, we can save a lot
> of space from duplicate code.
>
> Signed-off-by: Gang Li <ligang.bdlg@bytedance.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
> @@ -36,11 +36,19 @@ TRACE_EVENT_FN(mmap_lock_start_locking,
> __entry->mm,
> __get_str(memcg_path),
> __entry->write ? "true" : "false"
> - ),
> -
> - trace_mmap_lock_reg, trace_mmap_lock_unreg
> + )
> );
>
> +#define DEFINE_MMAP_LOCK_EVENT(name) \
> + DEFINE_EVENT_FN(mmap_lock, name, \
> + TP_PROTO(struct mm_struct *mm, const char *memcg_path, \
> + bool write), \
> + TP_ARGS(mm, memcg_path, write), \
> + trace_mmap_lock_reg, trace_mmap_lock_unreg)
A question (for Steven). I've several times wondered why DEFINE_EVENT has to
pass TP_PROTO/TP_ARGS even if they are often the same. Is it not possible to
have a variant that doesn't need that, and thus also doesn't need such extra
wrapping as above?
> +
> +DEFINE_MMAP_LOCK_EVENT(mmap_lock_start_locking);
> +DEFINE_MMAP_LOCK_EVENT(mmap_lock_released);
> +
> TRACE_EVENT_FN(mmap_lock_acquire_returned,
>
> TP_PROTO(struct mm_struct *mm, const char *memcg_path, bool write,
> @@ -73,34 +81,6 @@ TRACE_EVENT_FN(mmap_lock_acquire_returned,
> trace_mmap_lock_reg, trace_mmap_lock_unreg
> );
>
> -TRACE_EVENT_FN(mmap_lock_released,
> -
> - TP_PROTO(struct mm_struct *mm, const char *memcg_path, bool write),
> -
> - TP_ARGS(mm, memcg_path, write),
> -
> - TP_STRUCT__entry(
> - __field(struct mm_struct *, mm)
> - __string(memcg_path, memcg_path)
> - __field(bool, write)
> - ),
> -
> - TP_fast_assign(
> - __entry->mm = mm;
> - __assign_str(memcg_path, memcg_path);
> - __entry->write = write;
> - ),
> -
> - TP_printk(
> - "mm=%p memcg_path=%s write=%s",
> - __entry->mm,
> - __get_str(memcg_path),
> - __entry->write ? "true" : "false"
> - ),
> -
> - trace_mmap_lock_reg, trace_mmap_lock_unreg
> -);
> -
> #endif /* _TRACE_MMAP_LOCK_H */
>
> /* This part must be outside protection */
>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-10-11 8:02 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-09 7:12 [PATCH v2 2/2] mm: mmap_lock: use DECLARE_EVENT_CLASS and DEFINE_EVENT_FN Gang Li
2021-10-10 21:49 ` Steven Rostedt
2021-10-11 8:02 ` Vlastimil Babka
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).