Skip to contentSkip to conversation

mtd/maps: blackfin async flash maps: fix up memory leak

PATCH Applied [email protected] 4 participants 7 messages Feb 11, 2009 – Feb 19, 2009 Memory technology devices (MTD)

lore
Share & tools
lore
#0 lore raw

PATCH 4/4 mtd/maps: blackfin async flash maps: fix up memory leak

From: Mike Frysinger <[email protected]>

---

 drivers/mtd/maps/bfin-async-flash.c |    1 +
 1 file changed, 1 insertion(+)
drivers/mtd/maps/bfin-async-flash.c renamed from drivers/mtd/maps/bfin-async-flash.c~mtd-maps-blackfin-async-flash-maps-fix-up-memory-leak +1 -0
diff -puN drivers/mtd/maps/bfin-async-flash.c~mtd-maps-blackfin-async-flash-maps-fix-up-memory-leak drivers/mtd/maps/bfin-async-flash.c
@@ -170,6 +170,7 @@ static int __devinit bfin_flash_probe(st
 	if (ret > 0) {
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using commandline partition definition\n");
 		add_mtd_partitions(state->mtd, pdata->parts, ret);
+		kfree(pdata->parts);
 
 	} else if (pdata->nr_parts) {
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using board partition definition\n");

_

#1 lore raw

drivers/mtd/maps/bfin-async-flash.c static int __devinit bfin_flash_probe(st · line 176

Quoted 8 lines · On Wed, 11 Feb 2009 13:12:19 -0800, [email protected] wrote:
drivers/mtd/maps/bfin-async-flash.c renamed from drivers/mtd/maps/bfin-async-flash.c~mtd-maps-blackfin-async-flash-maps-fix-up-memory-leak +1 -0
@@ -170,6 +170,7 @@ static int __devinit bfin_flash_probe(st
 	if (ret > 0) {
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using commandline partition definition\n");
 		add_mtd_partitions(state->mtd, pdata->parts, ret);
+		kfree(pdata->parts);
 
 	} else if (pdata->nr_parts) {
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using board partition definition\n");

As Matthias Kaehlcke pointed out in mail titled "[BUG] physmap_flash_probe() frees memory still in use", this fix introduce another bug.

Revised patch for physmap driver is:
http://lists.infradead.org/pipermail/linux-mtd/2009-February/024590.html

Mike, please update this blackfin patch too.

---
Atsushi Nemoto

#2 lore raw
Quoted 14 lines · On Thu, 12 Feb 2009 09:35:31 +0900 (JST) Atsushi Nemoto <[email protected]>…

drivers/mtd/maps/bfin-async-flash.c static int __devinit bfin_flash_probe(st · line 176

Quoted 8 lines · On Wed, 11 Feb 2009 13:12:19 -0800, [email protected] wrote:
drivers/mtd/maps/bfin-async-flash.c renamed from drivers/mtd/maps/bfin-async-flash.c~mtd-maps-blackfin-async-flash-maps-fix-up-memory-leak +1 -0
@@ -170,6 +170,7 @@ static int __devinit bfin_flash_probe(st
 	if (ret > 0) {
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using commandline partition definition\n");
 		add_mtd_partitions(state->mtd, pdata->parts, ret);
+		kfree(pdata->parts);
 
 	} else if (pdata->nr_parts) {
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using board partition definition\n");

As Matthias Kaehlcke pointed out in mail titled "[BUG] physmap_flash_probe() frees memory still in use", this fix introduce another bug.

Quoted 5 lines

Revised patch for physmap driver is:
http://lists.infradead.org/pipermail/linux-mtd/2009-February/024590.html

Mike, please update this blackfin patch too.

I don't have a copy of that patch. You're on you own, guys..

#3 lore raw
Quoted 10 lines · Andrew Morton
Quoted 4 lines

Revised patch for physmap driver is:

http://lists.infradead.org/pipermail/linux-mtd/2009-February/024590.html

Quoted 3 lines

Mike, please update this blackfin patch too.

I don't have a copy of that patch. You're on you own, guys..

Er, wasn't your [1/4] actually Nemoto-san's latest physmap patch?

Mike, please can I have an updated version of the blackfin patch?

Nemoto-san, thank you for taking care of this.

Signature · 1 line
dwmw2
#4 lore raw
Quoted 8 lines · David Woodhouse
Quoted 7 lines · Andrew Morton
Quoted 2 lines

Revised patch for physmap driver is:

http://lists.infradead.org/pipermail/linux-mtd/2009-February/024590.html

Quoted 3 lines

Mike, please update this blackfin patch too.

Quoted 7 lines
Quoted 1 line

I don't have a copy of that patch. You're on you own, guys..

Er, wasn't your [1/4] actually Nemoto-san's latest physmap patch?

Mike, please can I have an updated version of the blackfin patch?

Nemoto-san, thank you for taking care of this.

i'm not entirely sure what the change is by reading that patch. what i've got is attached.
-mike

drivers/mtd/maps/bfin-async-flash.c +5 -0
@@ -41,4 +41,7 @@
 	uint32_t save_ambctl0, save_ambctl1;
 	unsigned long irq_flags;
+#ifdef CONFIG_MTD_PARTITIONS
+	struct mtd_partition *parts;
+#endif
 };
 
@@ -167,4 +174,5 @@
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using commandline partition definition\n");
 		add_mtd_partitions(state->mtd, pdata->parts, ret);
+		state->parts = pdata->parts;
 
 	} else if (pdata->nr_parts) {
@@ -190,4 +198,5 @@
 #ifdef CONFIG_MTD_PARTITIONS
 	del_mtd_partitions(state->mtd);
+	kfree(state->parts);
 #endif
 	map_destroy(state->mtd);
#5 lore raw
Quoted 2 lines · Mike Frysinger

i'm not entirely sure what the change is by reading that patch. what i've got is attached.

Looks fine for me. Thanks.
---
Atsushi Nemoto

#6 lore raw
Quoted 5 lines · Atsushi Nemoto
Quoted 2 lines · Mike Frysinger

i'm not entirely sure what the change is by reading that patch. what i've got is attached.

Looks fine for me. Thanks.

ok, i'll wait for my other patches against bfin-async-flash to make it into mainline and i'll post a follow up patch for this. -mike