Skip to contentSkip to conversation

seqlock: <linux/seqlock.h>: fix multiple kernel-doc warnings

PATCH Applied Randy Dunlap 3 participants 4 messages Aug 17, 2020 – Aug 27, 2020 Locking primitives

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 seqlock: <linux/seqlock.h>: fix multiple kernel-doc warnings

Fix kernel-doc warnings in <linux/seqlock.h>.

../include/linux/seqlock.h:152: warning: Incorrect use of kernel-doc format: * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t ../include/linux/seqlock.h:164: warning: Incorrect use of kernel-doc format: * SEQCOUNT_LOCKTYPE() - Instantiate seqcount_LOCKNAME_t and helpers ../include/linux/seqlock.h:229: warning: Function parameter or member 'seq_name' not described in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Function parameter or member 'assoc_lock' not described in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Excess function parameter 'name' description in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Excess function parameter 'lock' description in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:695: warning: duplicate section name 'NOTE'

Demote kernel-doc notation for the macros "seqcount_LOCKNAME_init()" and "SEQCOUNT_LOCKTYPE()"; scripts/kernel-doc does not handle them correctly.

Rename function parameters in SEQCNT_LOCKNAME_ZERO() documentation to match the macro's argument names. Change the macro name in the documentation to SEQCOUNT_LOCKTYPE_ZERO() to match the macro's name.

For raw_write_seqcount_latch(), rename the second NOTE: to NOTE2:
to prevent a kernel-doc warning. However, the generated output is not quite as nice as it could be for this.

Fix a typo: s/LOCKTYPR/LOCKTYPE/

Fixes 0efc94c5d15c ("seqcount: Compress SEQCNT_LOCKNAME_ZERO()")
Fixes e4e9ab3f9f91 ("seqlock: Fold seqcount_LOCKNAME_init() definition")
Fixes a8772dccb2ec ("seqlock: Fold seqcount_LOCKNAME_t definition")
Cc Peter Zijlstra <[email protected]>
Cc Ingo Molnar <[email protected]>

---
v2: do not move the definition of seqcount_LOCKNAME_init(). Fix build errors reported by kernel test robot. Actually build a kernel with these changes.

 include/linux/seqlock.h |   13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)
include/linux/seqlock.h +6 -7
@@ -138,7 +138,7 @@ static inline void seqcount_lockdep_read
 #endif
 
 /**
- * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPR associated
+ * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPE associated
  * @seqcount:	The real sequence counter
  * @lock:	Pointer to the associated spinlock
  *
@@ -148,7 +148,7 @@ static inline void seqcount_lockdep_read
  * that the write side critical section is properly serialized.
  */
 
-/**
+/*
  * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t
  * @s:		Pointer to the seqcount_LOCKNAME_t instance
  * @lock:	Pointer to the associated LOCKTYPE
@@ -173,7 +173,6 @@ seqcount_##lockname##_init(seqcount_##lo
 	seqcount_init(&s->seqcount);					\
 	__SEQ_LOCK(s->lock = lock);					\
 }									\
-									\
 static __always_inline seqcount_t *					\
 __seqcount_##lockname##_ptr(seqcount_##lockname##_t *s)			\
 {									\
@@ -218,9 +217,9 @@ SEQCOUNT_LOCKTYPE(struct mutex,		mutex,
 SEQCOUNT_LOCKTYPE(struct ww_mutex,	ww_mutex,	true,	&s->lock->base)
 
 /**
- * SEQCNT_LOCKNAME_ZERO - static initializer for seqcount_LOCKNAME_t
- * @name:	Name of the seqcount_LOCKNAME_t instance
- * @lock:	Pointer to the associated LOCKTYPE
+ * SEQCOUNT_LOCKTYPE_ZERO - static initializer for seqcount_LOCKNAME_t
+ * @seq_name:	Name of the seqcount_LOCKNAME_t instance
+ * @assoc_lock:	Pointer to the associated LOCKTYPE
  */
 
 #define SEQCOUNT_LOCKTYPE_ZERO(seq_name, assoc_lock) {			\
@@ -688,7 +687,7 @@ static inline int raw_read_seqcount_t_la
  *	to miss an entire modification sequence, once it resumes it might
  *	observe the new entry.
  *
- * NOTE:
+ * NOTE2:
  *
  *	When data is a dynamic data structure; one should use regular RCU
  *	patterns to manage the lifetimes of the objects within.
#1 lore raw

include/linux/seqlock.h seqcount_##lockname##_init(seqcount_##lo · line 178

Quoted 11 lines · Randy Dunlap
include/linux/seqlock.h +0 -1
@@ -173,7 +173,6 @@ seqcount_##lockname##_init(seqcount_##lo
 	seqcount_init(&s->seqcount);					\
 	__SEQ_LOCK(s->lock = lock);					\
 }									\
-									\
 static __always_inline seqcount_t *					\
 __seqcount_##lockname##_ptr(seqcount_##lockname##_t *s)			\
 {									\

I think I'd rather like that empty line there..

include/linux/seqlock.h SEQCOUNT_LOCKTYPE(struct mutex, mutex, · line 225

Quoted 13 lines
include/linux/seqlock.h +3 -3
@@ -218,9 +217,9 @@ SEQCOUNT_LOCKTYPE(struct mutex,		mutex,
 SEQCOUNT_LOCKTYPE(struct ww_mutex,	ww_mutex,	true,	&s->lock->base)
 
 /**
- * SEQCNT_LOCKNAME_ZERO - static initializer for seqcount_LOCKNAME_t
- * @name:	Name of the seqcount_LOCKNAME_t instance
- * @lock:	Pointer to the associated LOCKTYPE
+ * SEQCOUNT_LOCKTYPE_ZERO - static initializer for seqcount_LOCKNAME_t
+ * @seq_name:	Name of the seqcount_LOCKNAME_t instance
+ * @assoc_lock:	Pointer to the associated LOCKTYPE
  */
 
 #define SEQCOUNT_LOCKTYPE_ZERO(seq_name, assoc_lock) {			\

And this hunk seems wrong, SEQCOUNT_LOCKTYPE_ZERO() is not the intended API, SEQCNT_*_ZERO() are.

I've edited the patch like below, is that OK with you?

---
Subject: seqlock: Fix multiple kernel-doc warnings
From: Randy Dunlap <[email protected]>
Date: Sun, 16 Aug 2020 17:02:00 -0700

From: Randy Dunlap <[email protected]>

Fix kernel-doc warnings in <linux/seqlock.h>.

../include/linux/seqlock.h:152: warning: Incorrect use of kernel-doc format: * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t ../include/linux/seqlock.h:164: warning: Incorrect use of kernel-doc format: * SEQCOUNT_LOCKTYPE() - Instantiate seqcount_LOCKNAME_t and helpers ../include/linux/seqlock.h:229: warning: Function parameter or member 'seq_name' not described in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Function parameter or member 'assoc_lock' not described in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Excess function parameter 'name' description in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Excess function parameter 'lock' description in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:695: warning: duplicate section name 'NOTE'

Demote kernel-doc notation for the macros "seqcount_LOCKNAME_init()" and "SEQCOUNT_LOCKTYPE()"; scripts/kernel-doc does not handle them correctly.

Rename function parameters in SEQCNT_LOCKNAME_ZERO() documentation to match the macro's argument names. Change the macro name in the documentation to SEQCOUNT_LOCKTYPE_ZERO() to match the macro's name.

For raw_write_seqcount_latch(), rename the second NOTE: to NOTE2:
to prevent a kernel-doc warning. However, the generated output is not quite as nice as it could be for this.

Fix a typo: s/LOCKTYPR/LOCKTYPE/

Fixes 0efc94c5d15c ("seqcount: Compress SEQCNT_LOCKNAME_ZERO()")
Fixes e4e9ab3f9f91 ("seqlock: Fold seqcount_LOCKNAME_init() definition")
Fixes a8772dccb2ec ("seqlock: Fold seqcount_LOCKNAME_t definition")

---
v2: do not move the definition of seqcount_LOCKNAME_init(). Fix build errors reported by kernel test robot. Actually build a kernel with these changes.

 include/linux/seqlock.h |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
include/linux/seqlock.h +4 -4
@@ -138,7 +138,7 @@ static inline void seqcount_lockdep_read
 #endif
 
 /**
- * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPR associated
+ * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPE associated
  * @seqcount:	The real sequence counter
  * @lock:	Pointer to the associated spinlock
  *
@@ -148,7 +148,7 @@ static inline void seqcount_lockdep_read
  * that the write side critical section is properly serialized.
  */
 
-/**
+/*
  * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t
  * @s:		Pointer to the seqcount_LOCKNAME_t instance
  * @lock:	Pointer to the associated LOCKTYPE
@@ -217,7 +217,7 @@ SEQCOUNT_LOCKTYPE(rwlock_t,		rwlock,		fa
 SEQCOUNT_LOCKTYPE(struct mutex,		mutex,		true,	s->lock)
 SEQCOUNT_LOCKTYPE(struct ww_mutex,	ww_mutex,	true,	&s->lock->base)
 
-/**
+/*
  * SEQCNT_LOCKNAME_ZERO - static initializer for seqcount_LOCKNAME_t
  * @name:	Name of the seqcount_LOCKNAME_t instance
  * @lock:	Pointer to the associated LOCKTYPE
@@ -688,7 +688,7 @@ static inline int raw_read_seqcount_t_la
  *	to miss an entire modification sequence, once it resumes it might
  *	observe the new entry.
  *
- * NOTE:
+ * NOTE2:
  *
  *	When data is a dynamic data structure; one should use regular RCU
  *	patterns to manage the lifetimes of the objects within.
#2 lore raw
Quoted 28 lines · On 8/19/20 2:25 AM, [email protected] wrote:

include/linux/seqlock.h seqcount_##lockname##_init(seqcount_##lo · line 178

Quoted 11 lines · Randy Dunlap
include/linux/seqlock.h +0 -1
@@ -173,7 +173,6 @@ seqcount_##lockname##_init(seqcount_##lo
 	seqcount_init(&s->seqcount);					\
 	__SEQ_LOCK(s->lock = lock);					\
 }									\
-									\
 static __always_inline seqcount_t *					\
 __seqcount_##lockname##_ptr(seqcount_##lockname##_t *s)			\
 {									\

I think I'd rather like that empty line there..

Quoted 12 lines
include/linux/seqlock.h +3 -3
@@ -218,9 +217,9 @@ SEQCOUNT_LOCKTYPE(struct mutex,		mutex,
 SEQCOUNT_LOCKTYPE(struct ww_mutex,	ww_mutex,	true,	&s->lock->base)
 
 /**
- * SEQCNT_LOCKNAME_ZERO - static initializer for seqcount_LOCKNAME_t
- * @name:	Name of the seqcount_LOCKNAME_t instance
- * @lock:	Pointer to the associated LOCKTYPE
+ * SEQCOUNT_LOCKTYPE_ZERO - static initializer for seqcount_LOCKNAME_t
+ * @seq_name:	Name of the seqcount_LOCKNAME_t instance
+ * @assoc_lock:	Pointer to the associated LOCKTYPE
  */
 
Quoted 7 lines

include/linux/seqlock.h SEQCOUNT_LOCKTYPE(struct mutex, mutex, · line 225

Quoted 1 line
include/linux/seqlock.h +0 -0
 #define SEQCOUNT_LOCKTYPE_ZERO(seq_name, assoc_lock) {			\

And this hunk seems wrong, SEQCOUNT_LOCKTYPE_ZERO() is not the intended API, SEQCNT_*_ZERO() are.

I've edited the patch like below, is that OK with you?

Hi Peter,
Yes, this looks good. Thanks for your help.

Quoted 84 lines

---
Subject: seqlock: Fix multiple kernel-doc warnings
From: Randy Dunlap <[email protected]>
Date: Sun, 16 Aug 2020 17:02:00 -0700

From: Randy Dunlap <[email protected]>

Fix kernel-doc warnings in <linux/seqlock.h>.

../include/linux/seqlock.h:152: warning: Incorrect use of kernel-doc format: * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t ../include/linux/seqlock.h:164: warning: Incorrect use of kernel-doc format: * SEQCOUNT_LOCKTYPE() - Instantiate seqcount_LOCKNAME_t and helpers ../include/linux/seqlock.h:229: warning: Function parameter or member 'seq_name' not described in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Function parameter or member 'assoc_lock' not described in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Excess function parameter 'name' description in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Excess function parameter 'lock' description in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:695: warning: duplicate section name 'NOTE'

Demote kernel-doc notation for the macros "seqcount_LOCKNAME_init()" and "SEQCOUNT_LOCKTYPE()"; scripts/kernel-doc does not handle them correctly.

Rename function parameters in SEQCNT_LOCKNAME_ZERO() documentation to match the macro's argument names. Change the macro name in the documentation to SEQCOUNT_LOCKTYPE_ZERO() to match the macro's name.

For raw_write_seqcount_latch(), rename the second NOTE: to NOTE2:
to prevent a kernel-doc warning. However, the generated output is not quite as nice as it could be for this.

Fix a typo: s/LOCKTYPR/LOCKTYPE/

Fixes 0efc94c5d15c ("seqcount: Compress SEQCNT_LOCKNAME_ZERO()")
Fixes e4e9ab3f9f91 ("seqlock: Fold seqcount_LOCKNAME_init() definition")
Fixes a8772dccb2ec ("seqlock: Fold seqcount_LOCKNAME_t definition")

---
v2: do not move the definition of seqcount_LOCKNAME_init(). Fix build errors reported by kernel test robot. Actually build a kernel with these changes.

 include/linux/seqlock.h |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
include/linux/seqlock.h +4 -4
@@ -138,7 +138,7 @@ static inline void seqcount_lockdep_read
 #endif
 
 /**
- * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPR associated
+ * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPE associated
  * @seqcount:	The real sequence counter
  * @lock:	Pointer to the associated spinlock
  *
@@ -148,7 +148,7 @@ static inline void seqcount_lockdep_read
  * that the write side critical section is properly serialized.
  */
 
-/**
+/*
  * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t
  * @s:		Pointer to the seqcount_LOCKNAME_t instance
  * @lock:	Pointer to the associated LOCKTYPE
@@ -217,7 +217,7 @@ SEQCOUNT_LOCKTYPE(rwlock_t,		rwlock,		fa
 SEQCOUNT_LOCKTYPE(struct mutex,		mutex,		true,	s->lock)
 SEQCOUNT_LOCKTYPE(struct ww_mutex,	ww_mutex,	true,	&s->lock->base)
 
-/**
+/*
  * SEQCNT_LOCKNAME_ZERO - static initializer for seqcount_LOCKNAME_t
  * @name:	Name of the seqcount_LOCKNAME_t instance
  * @lock:	Pointer to the associated LOCKTYPE
@@ -688,7 +688,7 @@ static inline int raw_read_seqcount_t_la
  *	to miss an entire modification sequence, once it resumes it might
  *	observe the new entry.
  *
- * NOTE:
+ * NOTE2:
  *
  *	When data is a dynamic data structure; one should use regular RCU
  *	patterns to manage the lifetimes of the objects within.
Signature · 2 lines
~Randy
Reported-by: Randy Dunlap <[email protected]>
merge bot Applied
#3 lore raw

[tip: locking/core] seqlock: Fix multiple kernel-doc warnings

In reply to Randy Dunlap
The following commit has been merged into the locking/core branch of tip:

The following commit has been merged into the locking/core branch of tip:

Commit-ID:     a28e884b966e713da29caefbb347efea77367d22
Gitweb:        https://git.kernel.org/tip/a28e884b966e713da29caefbb347efea77367d22
Author:        Randy Dunlap <[email protected]>
AuthorDate:    Sun, 16 Aug 2020 17:02:00 -07:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 26 Aug 2020 12:42:02 +02:00

seqlock: Fix multiple kernel-doc warnings

Fix kernel-doc warnings in <linux/seqlock.h>.

../include/linux/seqlock.h:152: warning: Incorrect use of kernel-doc format: * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t ../include/linux/seqlock.h:164: warning: Incorrect use of kernel-doc format: * SEQCOUNT_LOCKTYPE() - Instantiate seqcount_LOCKNAME_t and helpers ../include/linux/seqlock.h:229: warning: Function parameter or member 'seq_name' not described in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Function parameter or member 'assoc_lock' not described in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Excess function parameter 'name' description in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:229: warning: Excess function parameter 'lock' description in 'SEQCOUNT_LOCKTYPE_ZERO' ../include/linux/seqlock.h:695: warning: duplicate section name 'NOTE'

Demote kernel-doc notation for the macros "seqcount_LOCKNAME_init()" and "SEQCOUNT_LOCKTYPE()"; scripts/kernel-doc does not handle them correctly.

Rename function parameters in SEQCNT_LOCKNAME_ZERO() documentation to match the macro's argument names. Change the macro name in the documentation to SEQCOUNT_LOCKTYPE_ZERO() to match the macro's name.

For raw_write_seqcount_latch(), rename the second NOTE: to NOTE2:
to prevent a kernel-doc warning. However, the generated output is not quite as nice as it could be for this.

Fix a typo: s/LOCKTYPR/LOCKTYPE/

Fixes 0efc94c5d15c ("seqcount: Compress SEQCNT_LOCKNAME_ZERO()")
Fixes e4e9ab3f9f91 ("seqlock: Fold seqcount_LOCKNAME_init() definition")
Fixes a8772dccb2ec ("seqlock: Fold seqcount_LOCKNAME_t definition")
---
 include/linux/seqlock.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
include/linux/seqlock.h +4 -4
index 962d976..300cbf3 100644
@@ -138,7 +138,7 @@ static inline void seqcount_lockdep_reader_access(const seqcount_t *s)
 #endif
 
 /**
- * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPR associated
+ * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPE associated
  * @seqcount:	The real sequence counter
  * @lock:	Pointer to the associated spinlock
  *
@@ -148,7 +148,7 @@ static inline void seqcount_lockdep_reader_access(const seqcount_t *s)
  * that the write side critical section is properly serialized.
  */
 
-/**
+/*
  * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t
  * @s:		Pointer to the seqcount_LOCKNAME_t instance
  * @lock:	Pointer to the associated LOCKTYPE
@@ -217,7 +217,7 @@ SEQCOUNT_LOCKTYPE(rwlock_t,		rwlock,		false,	s->lock)
 SEQCOUNT_LOCKTYPE(struct mutex,		mutex,		true,	s->lock)
 SEQCOUNT_LOCKTYPE(struct ww_mutex,	ww_mutex,	true,	&s->lock->base)
 
-/**
+/*
  * SEQCNT_LOCKNAME_ZERO - static initializer for seqcount_LOCKNAME_t
  * @name:	Name of the seqcount_LOCKNAME_t instance
  * @lock:	Pointer to the associated LOCKTYPE
@@ -688,7 +688,7 @@ static inline int raw_read_seqcount_t_latch(seqcount_t *s)
  *	to miss an entire modification sequence, once it resumes it might
  *	observe the new entry.
  *
- * NOTE:
+ * NOTE2:
  *
  *	When data is a dynamic data structure; one should use regular RCU
  *	patterns to manage the lifetimes of the objects within.