Skip to contentSkip to conversation

soundwire: bus: Fix unbalanced pm_runtime_put() causing usage count underflow

PATCH Superseded Richard Fitzgerald 2 participants 2 messages Feb 27, 2023 SoundWire

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 soundwire: bus: Fix unbalanced pm_runtime_put() causing usage count underflow

This reverts commit
443a98e649b4 ("soundwire: bus: use pm_runtime_resume_and_get()")

Change calls to pm_runtime_resume_and_get() back to pm_runtime_get_sync(). This fixes a usage count underrun caused by doing a pm_runtime_put() even though pm_runtime_resume_and_get() returned an error.

The three affected functions ignore -EACCES error from trying to get pm_runtime, and carry on, including a put at the end of the function. But pm_runtime_resume_and_get() does not increment the usage count if it returns an error. So in the -EACCES case you must not call pm_runtime_put().

The documentation for pm_runtime_get_sync() says:
"Consider using pm_runtime_resume_and_get() ... as this is likely to result in cleaner code."

In this case I don't think it results in cleaner code because the pm_runtime_put() at the end of the function would have to be conditional on the return value from pm_runtime_resume_and_get() at the top of the function.

pm_runtime_get_sync() doesn't have this problem because it always increments the count, so always needs a put. The code can just flow through and do the pm_runtime_put() unconditionally.

---
 drivers/soundwire/bus.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)
drivers/soundwire/bus.c +10 -5
index 76515c33e639..4fd221d0cc81 100644
@@ -571,9 +571,11 @@ int sdw_nread(struct sdw_slave *slave, u32 addr, size_t count, u8 *val)
 {
 	int ret;
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
-	if (ret < 0 && ret != -EACCES)
+	ret = pm_runtime_get_sync(&slave->dev);
+	if (ret < 0 && ret != -EACCES) {
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
+	}
 
 	ret = sdw_nread_no_pm(slave, addr, count, val);
 
@@ -595,9 +597,11 @@ int sdw_nwrite(struct sdw_slave *slave, u32 addr, size_t count, const u8 *val)
 {
 	int ret;
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
-	if (ret < 0 && ret != -EACCES)
+	ret = pm_runtime_get_sync(&slave->dev);
+	if (ret < 0 && ret != -EACCES) {
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
+	}
 
 	ret = sdw_nwrite_no_pm(slave, addr, count, val);
 
@@ -1565,9 +1569,10 @@ static int sdw_handle_slave_alerts(struct sdw_slave *slave)
 
 	sdw_modify_slave_status(slave, SDW_SLAVE_ALERT);
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
+	ret = pm_runtime_get_sync(&slave->dev);
 	if (ret < 0 && ret != -EACCES) {
 		dev_err(&slave->dev, "Failed to resume device: %d\n", ret);
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
 	}
 
Signature · 1 line
2.30.2
Reviewed-by on v1 1/1
#1 lore raw
Quoted 22 lines · Richard Fitzgerald

This reverts commit
443a98e649b4 ("soundwire: bus: use pm_runtime_resume_and_get()")

Change calls to pm_runtime_resume_and_get() back to pm_runtime_get_sync(). This fixes a usage count underrun caused by doing a pm_runtime_put() even though pm_runtime_resume_and_get() returned an error.

The three affected functions ignore -EACCES error from trying to get pm_runtime, and carry on, including a put at the end of the function. But pm_runtime_resume_and_get() does not increment the usage count if it returns an error. So in the -EACCES case you must not call pm_runtime_put().

The documentation for pm_runtime_get_sync() says:
"Consider using pm_runtime_resume_and_get() ... as this is likely to result in cleaner code."

In this case I don't think it results in cleaner code because the pm_runtime_put() at the end of the function would have to be conditional on the return value from pm_runtime_resume_and_get() at the top of the function.

Quoted 5 lines

pm_runtime_get_sync() doesn't have this problem because it always increments the count, so always needs a put. The code can just flow through and do the pm_runtime_put() unconditionally.

Agreed, in hindsight the move to use pm_runtime_resume_and_get() mostly added issues left and right with limited benefits - just too hard to review and figure out what cases work and which ones don't.

Quoted 48 lines
---
 drivers/soundwire/bus.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)
drivers/soundwire/bus.c +10 -5
index 76515c33e639..4fd221d0cc81 100644
@@ -571,9 +571,11 @@ int sdw_nread(struct sdw_slave *slave, u32 addr, size_t count, u8 *val)
 {
 	int ret;
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
-	if (ret < 0 && ret != -EACCES)
+	ret = pm_runtime_get_sync(&slave->dev);
+	if (ret < 0 && ret != -EACCES) {
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
+	}
 
 	ret = sdw_nread_no_pm(slave, addr, count, val);
 
@@ -595,9 +597,11 @@ int sdw_nwrite(struct sdw_slave *slave, u32 addr, size_t count, const u8 *val)
 {
 	int ret;
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
-	if (ret < 0 && ret != -EACCES)
+	ret = pm_runtime_get_sync(&slave->dev);
+	if (ret < 0 && ret != -EACCES) {
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
+	}
 
 	ret = sdw_nwrite_no_pm(slave, addr, count, val);
 
@@ -1565,9 +1569,10 @@ static int sdw_handle_slave_alerts(struct sdw_slave *slave)
 
 	sdw_modify_slave_status(slave, SDW_SLAVE_ALERT);
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
+	ret = pm_runtime_get_sync(&slave->dev);
 	if (ret < 0 && ret != -EACCES) {
 		dev_err(&slave->dev, "Failed to resume device: %d\n", ret);
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
 	}