Skip to contentSkip to conversation

clocksource/drivers: Fix memory leaks in ttc_setup_clockevent and ttc_setup_clocksource

PATCH In review Navid Emamdoost 4 participants 4 messages Dec 20, 2019 – Dec 23, 2019

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

PATCH 1/1 clocksource/drivers: Fix memory leaks in ttc_setup_clockevent and ttc_setup_clocksource

In the implementation of ttc_setup_clockevent() and ttc_setup_clocksource(), the allocated memory for ttccs is leaked when clk_notifier_register() fails. Use goto to direct the execution into error handling path.

Fixes 70504f311d4b ("clocksource/drivers/cadence_ttc: Convert init function to return error")
---
 drivers/clocksource/timer-cadence-ttc.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)
drivers/clocksource/timer-cadence-ttc.c +13 -9
index c6d11a1cb238..46d69982ad33 100644
@@ -328,10 +328,8 @@ static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base,
 	ttccs->ttc.clk = clk;
 
 	err = clk_prepare_enable(ttccs->ttc.clk);
-	if (err) {
-		kfree(ttccs);
-		return err;
-	}
+	if (err)
+		goto release_ttcce;
 
 	ttccs->ttc.freq = clk_get_rate(ttccs->ttc.clk);
 
@@ -341,8 +339,10 @@ static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base,
 
 	err = clk_notifier_register(ttccs->ttc.clk,
 				    &ttccs->ttc.clk_rate_change_nb);
-	if (err)
+	if (err) {
 		pr_warn("Unable to register clock notifier.\n");
+		goto release_ttcce;
+	}
 
 	ttccs->ttc.base_addr = base;
 	ttccs->cs.name = "ttc_clocksource";
@@ -363,16 +363,20 @@ static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base,
 		     ttccs->ttc.base_addr + TTC_CNT_CNTRL_OFFSET);
 
 	err = clocksource_register_hz(&ttccs->cs, ttccs->ttc.freq / PRESCALE);
-	if (err) {
-		kfree(ttccs);
-		return err;
-	}
+	if (err)
+		goto release_ttcce;
 
 	ttc_sched_clock_val_reg = base + TTC_COUNT_VAL_OFFSET;
 	sched_clock_register(ttc_sched_clock_read, timer_width,
 			     ttccs->ttc.freq / PRESCALE);
 
 	return 0;
+
+release_ttcce:
+
+	kfree(ttcce);
+	return err;
+
 }
 
 static int ttc_rate_change_clockevent_cb(struct notifier_block *nb,
Signature · 1 line
2.17.1
#1 lore raw
Quoted 4 lines · Navid Emamdoost

In the implementation of ttc_setup_clockevent() and ttc_setup_clocksource(), the allocated memory for ttccs is leaked when clk_notifier_register() fails. Use goto to direct the execution into error handling path.

No, that memory was never leaked since that function did not return on registration errors before your patch.

Quoted 1 line
Fixes 70504f311d4b ("clocksource/drivers/cadence_ttc: Convert init function to return error")

Perhaps you meant to fix the actual leak that was added by this commit in a different function, ttc_setup_clockevent(), when returning on notifier-registration errors?

Also should the clock be left enabled on errors?

Quoted 23 lines
---
 drivers/clocksource/timer-cadence-ttc.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)
drivers/clocksource/timer-cadence-ttc.c +2 -4
index c6d11a1cb238..46d69982ad33 100644
@@ -328,10 +328,8 @@ static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base,
 	ttccs->ttc.clk = clk;
 
 	err = clk_prepare_enable(ttccs->ttc.clk);
-	if (err) {
-		kfree(ttccs);
-		return err;
-	}
+	if (err)
+		goto release_ttcce;
 
 	ttccs->ttc.freq = clk_get_rate(ttccs->ttc.clk);
 

drivers/clocksource/timer-cadence-ttc.c static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base, · line 345

Quoted 8 lines
drivers/clocksource/timer-cadence-ttc.c +3 -1
@@ -341,8 +339,10 @@ static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base,
 
 	err = clk_notifier_register(ttccs->ttc.clk,
 				    &ttccs->ttc.clk_rate_change_nb);
-	if (err)
+	if (err) {
 		pr_warn("Unable to register clock notifier.\n");
+		goto release_ttcce;
+	}

Johan

#2 lore raw
Quoted 41 lines · Johan Hovold
Quoted 4 lines · Navid Emamdoost

In the implementation of ttc_setup_clockevent() and ttc_setup_clocksource(), the allocated memory for ttccs is leaked when clk_notifier_register() fails. Use goto to direct the execution into error handling path.

No, that memory was never leaked since that function did not return on registration errors before your patch.

Quoted 1 line
Fixes 70504f311d4b ("clocksource/drivers/cadence_ttc: Convert init function to return error")

Perhaps you meant to fix the actual leak that was added by this commit in a different function, ttc_setup_clockevent(), when returning on notifier-registration errors?

Also should the clock be left enabled on errors?

Quoted 24 lines
---
 drivers/clocksource/timer-cadence-ttc.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)
drivers/clocksource/timer-cadence-ttc.c +2 -4
index c6d11a1cb238..46d69982ad33 100644
@@ -328,10 +328,8 @@ static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base,
 	ttccs->ttc.clk = clk;
 
 	err = clk_prepare_enable(ttccs->ttc.clk);
-	if (err) {
-		kfree(ttccs);
-		return err;
-	}
+	if (err)
+		goto release_ttcce;
 
 	ttccs->ttc.freq = clk_get_rate(ttccs->ttc.clk);
 
@@ -341,8 +339,10 @@ static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base,
 
Quoted 8 lines

drivers/clocksource/timer-cadence-ttc.c static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base, · line 345

Quoted 7 lines
drivers/clocksource/timer-cadence-ttc.c +3 -1
 	err = clk_notifier_register(ttccs->ttc.clk,
 				    &ttccs->ttc.clk_rate_change_nb);
-	if (err)
+	if (err) {
 		pr_warn("Unable to register clock notifier.\n");
+		goto release_ttcce;
+	}

+ this is IMHO v3 version. It means just label it properly and also keep changes log.

Thanks,
Michal

#3 lore raw

Re: [PATCH] clocksource/drivers: Fix something in ttc_setup_clockevent and ttc_setup_clocksource

In reply to Navid Emamdoost
Quoted 3 lines

In the implementation of ttc_setup_clockevent() and ttc_setup_clocksource(), the allocated memory for ttccs is leaked when clk_notifier_register() fails.

I suggest to correct your commit message.

Quoted 1 line

Use goto to direct the execution into error handling path.

* Should the desired completion of the exception handling be described by an other wording?

* Will it be better to improve the affected functions by separate update steps?

* How do you think about to add the tag “Reported-by” for Michal Simek?
https://lore.kernel.org/linux-arm-kernel/[email protected]/

…

Quoted 1 line

+++ b/drivers/clocksource/timer-cadence-ttc.c

…

Quoted 1 line
(quoted diff) +0 -0
@@ -363,16 +363,20 @@  static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base,

…

drivers/clocksource/timer-cadence-ttc.c static int __init ttc_setup_clocksource(struct clk *clk, void __iomem *base, · line 378

Quoted 6 lines
drivers/clocksource/timer-cadence-ttc.c +5 -0
 	return 0;
+
+release_ttcce:
+
+	kfree(ttcce);
+	return err;

…

* Please omit a blank line after the label.

* Is there a need to call the function “clk_disable_unprepare” by another jump target?

Regards,
Markus

_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel