Skip to contentSkip to conversation

block/rnbd-clt: improve find_or_create_sess() return check

PATCH In review [email protected] 3 participants 4 messages Jan 10, 2021 – Jan 11, 2021

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 block/rnbd-clt: improve find_or_create_sess() return check

From: Tom Rix <[email protected]>

clang static analysis reports this problem

rnbd-clt.c:1212:11: warning: Branch condition evaluates to a
garbage value
else if (!first)
^~~~~~

This is triggered in the find_and_get_or_create_sess() call because the variable first is not initialized and the earlier check is specifically for

	if (sess == ERR_PTR(-ENOMEM))

This is false positive.

But the if-check can be reduced by initializing first to false and then returning if the call to find_or_creat_sess() does not set it to true. When it remains false, either sess will be valid or not. The not case is caught by find_and_get_or_create_sess()'s caller rnbd_clt_map_device()

	sess = find_and_get_or_create_sess(...);
	if (IS_ERR(sess))
		return ERR_CAST(sess);

Since find_and_get_or_create_sess() initializes first to false setting it in find_or_create_sess() is not needed.

---
 drivers/block/rnbd/rnbd-clt.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
drivers/block/rnbd/rnbd-clt.c +4 -6
index 96e3f9fe8241..251f747cf10d 100644
@@ -919,6 +919,7 @@ static struct rnbd_clt_session *__find_and_get_sess(const char *sessname)
 	return NULL;
 }
 
+/* caller is responsible for initializing 'first' to false */
 static struct
 rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
 {
@@ -934,8 +935,7 @@ rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
 		}
 		list_add(&sess->list, &sess_list);
 		*first = true;
-	} else
-		*first = false;
+	}
 	mutex_unlock(&sess_lock);
 
 	return sess;
@@ -1203,13 +1203,11 @@ find_and_get_or_create_sess(const char *sessname,
 	struct rnbd_clt_session *sess;
 	struct rtrs_attrs attrs;
 	int err;
-	bool first;
+	bool first = false;
 	struct rtrs_clt_ops rtrs_ops;
 
 	sess = find_or_create_sess(sessname, &first);
-	if (sess == ERR_PTR(-ENOMEM))
-		return ERR_PTR(-ENOMEM);
-	else if (!first)
+	if (!first)
 		return sess;
 
 	if (!path_cnt) {
Signature · 1 line
2.27.0
Reviewed-by on v1 1/1
#1 lore raw
Quoted 8 lines · On Sun, Jan 10, 2021 at 01:57:26PM -0800, [email protected] wrote:

From: Tom Rix <[email protected]>

clang static analysis reports this problem

rnbd-clt.c:1212:11: warning: Branch condition evaluates to a
garbage value
else if (!first)
^~~~~~

Ah, is it complaining that the 'if (IS_ERR(sess)) {' section in find_or_create_sess() does not initialize first even though that will be caught by the 'if (sess == ERR_PTR(-ENOMEM))' in find_and_get_or_create_sess() because alloc_sess() only returns an -ENOMEM error pointer?

Quoted 14 lines

This is triggered in the find_and_get_or_create_sess() call because the variable first is not initialized and the earlier check is specifically for

	if (sess == ERR_PTR(-ENOMEM))

This is false positive.

But the if-check can be reduced by initializing first to false and then returning if the call to find_or_creat_sess() does not set it to true. When it remains false, either sess will be valid or not. The not case is caught by find_and_get_or_create_sess()'s caller rnbd_clt_map_device()

Quoted 8 lines
	sess = find_and_get_or_create_sess(...);
	if (IS_ERR(sess))
		return ERR_CAST(sess);

Since find_and_get_or_create_sess() initializes first to false setting it in find_or_create_sess() is not needed.

Every maintainer has their preference for where and how stuff gets initialized but this makes sense to me. I am not sure the commit above find_or_create_sess() is needed but again, personal preference.

Quoted 45 lines
---
 drivers/block/rnbd/rnbd-clt.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
drivers/block/rnbd/rnbd-clt.c +4 -6
index 96e3f9fe8241..251f747cf10d 100644
@@ -919,6 +919,7 @@ static struct rnbd_clt_session *__find_and_get_sess(const char *sessname)
 	return NULL;
 }
 
+/* caller is responsible for initializing 'first' to false */
 static struct
 rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
 {
@@ -934,8 +935,7 @@ rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
 		}
 		list_add(&sess->list, &sess_list);
 		*first = true;
-	} else
-		*first = false;
+	}
 	mutex_unlock(&sess_lock);
 
 	return sess;
@@ -1203,13 +1203,11 @@ find_and_get_or_create_sess(const char *sessname,
 	struct rnbd_clt_session *sess;
 	struct rtrs_attrs attrs;
 	int err;
-	bool first;
+	bool first = false;
 	struct rtrs_clt_ops rtrs_ops;
 
 	sess = find_or_create_sess(sessname, &first);
-	if (sess == ERR_PTR(-ENOMEM))
-		return ERR_PTR(-ENOMEM);
-	else if (!first)
+	if (!first)
 		return sess;
 
 	if (!path_cnt) {
Signature · 1 line
2.27.0
author
#2 lore raw
Quoted 9 lines · Nathan Chancellor
Quoted 8 lines · On Sun, Jan 10, 2021 at 01:57:26PM -0800, [email protected] wrote:

From: Tom Rix <[email protected]>

clang static analysis reports this problem

rnbd-clt.c:1212:11: warning: Branch condition evaluates to a
garbage value
else if (!first)
^~~~~~

Quoted 5 lines

Ah, is it complaining that the 'if (IS_ERR(sess)) {' section in find_or_create_sess() does not initialize first even though that will be caught by the 'if (sess == ERR_PTR(-ENOMEM))' in find_and_get_or_create_sess() because alloc_sess() only returns an -ENOMEM error pointer?

Reviewing the code, failure looks like it returns only -ENOMEM.

So the check is correct but brittle.

Quoted 19 lines
Quoted 19 lines

This is triggered in the find_and_get_or_create_sess() call because the variable first is not initialized and the earlier check is specifically for

	if (sess == ERR_PTR(-ENOMEM))

This is false positive.

But the if-check can be reduced by initializing first to false and then returning if the call to find_or_creat_sess() does not set it to true. When it remains false, either sess will be valid or not. The not case is caught by find_and_get_or_create_sess()'s caller rnbd_clt_map_device()

	sess = find_and_get_or_create_sess(...);
	if (IS_ERR(sess))
		return ERR_CAST(sess);
Quoted 7 lines
Quoted 4 lines

Since find_and_get_or_create_sess() initializes first to false setting it in find_or_create_sess() is not needed.

Every maintainer has their preference for where and how stuff gets initialized but this makes sense to me. I am not sure the commit above find_or_create_sess() is needed but again, personal preference.

Mostly this removes two unneeded branches at the cost of initializing a variable.

Secondary, the static analysis complaint is resolved.

Tom

Quoted 48 lines
Quoted 45 lines
---
 drivers/block/rnbd/rnbd-clt.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
drivers/block/rnbd/rnbd-clt.c +4 -6
index 96e3f9fe8241..251f747cf10d 100644
@@ -919,6 +919,7 @@ static struct rnbd_clt_session *__find_and_get_sess(const char *sessname)
 	return NULL;
 }
 
+/* caller is responsible for initializing 'first' to false */
 static struct
 rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
 {
@@ -934,8 +935,7 @@ rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
 		}
 		list_add(&sess->list, &sess_list);
 		*first = true;
-	} else
-		*first = false;
+	}
 	mutex_unlock(&sess_lock);
 
 	return sess;
@@ -1203,13 +1203,11 @@ find_and_get_or_create_sess(const char *sessname,
 	struct rnbd_clt_session *sess;
 	struct rtrs_attrs attrs;
 	int err;
-	bool first;
+	bool first = false;
 	struct rtrs_clt_ops rtrs_ops;
 
 	sess = find_or_create_sess(sessname, &first);
-	if (sess == ERR_PTR(-ENOMEM))
-		return ERR_PTR(-ENOMEM);
-	else if (!first)
+	if (!first)
 		return sess;
 
 	if (!path_cnt) {
Signature · 1 line
2.27.0
Acked-by on v1 1/1
#3 lore raw
In reply to [email protected]
Quoted 24 lines · On Sun, Jan 10, 2021 at 10:58 PM <[email protected]> wrote:

From: Tom Rix <[email protected]>

clang static analysis reports this problem

rnbd-clt.c:1212:11: warning: Branch condition evaluates to a
garbage value
else if (!first)
^~~~~~

This is triggered in the find_and_get_or_create_sess() call because the variable first is not initialized and the earlier check is specifically for

        if (sess == ERR_PTR(-ENOMEM))

This is false positive.

But the if-check can be reduced by initializing first to false and then returning if the call to find_or_creat_sess() does not set it to true. When it remains false, either sess will be valid or not. The not case is caught by find_and_get_or_create_sess()'s caller rnbd_clt_map_device()

Quoted 8 lines
        sess = find_and_get_or_create_sess(...);
        if (IS_ERR(sess))
                return ERR_CAST(sess);

Since find_and_get_or_create_sess() initializes first to false setting it in find_or_create_sess() is not needed.

Less LOC is better :)

Thanks Tom and Nathan!

Quoted 45 lines
---
 drivers/block/rnbd/rnbd-clt.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
drivers/block/rnbd/rnbd-clt.c +4 -7
index 96e3f9fe8241..251f747cf10d 100644
@@ -919,6 +919,7 @@ static struct rnbd_clt_session *__find_and_get_sess(const char *sessname)
        return NULL;
 }
 
+/* caller is responsible for initializing 'first' to false */
 static struct
 rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
 {
@@ -934,8 +935,7 @@ rnbd_clt_session *find_or_create_sess(const char *sessname, bool *first)
                }
                list_add(&sess->list, &sess_list);
                *first = true;
-       } else
-               *first = false;
+       }
        mutex_unlock(&sess_lock);
 
        return sess;
@@ -1203,13 +1203,11 @@ find_and_get_or_create_sess(const char *sessname,
        struct rnbd_clt_session *sess;
        struct rtrs_attrs attrs;
        int err;
-       bool first;
+       bool first = false;
        struct rtrs_clt_ops rtrs_ops;
 
        sess = find_or_create_sess(sessname, &first);
-       if (sess == ERR_PTR(-ENOMEM))
-               return ERR_PTR(-ENOMEM);
-       else if (!first)
+       if (!first)
                return sess;
 
        if (!path_cnt) {
--

2.27.0