Re: [PATCH 1/4] drivers/serial/ucc_uart.c: Add of_node_put to avoid memory leak

Previous thread: [PATCH 4/4] arch/powerpc/platforms/chrp/nvram.c: Add of_node_put to avoid memory leak by Julia Lawall on Tuesday, August 31, 2010 - 8:48 am. (2 messages)

Next thread: [git pull] PCI fixes by Jesse Barnes on Tuesday, August 31, 2010 - 8:49 am. (1 message)
From: Julia Lawall
Date: Tuesday, August 31, 2010 - 8:48 am

These patches introduce calls to of_node_put after various functions that
call of_node_get.

--

From: Julia Lawall
Date: Tuesday, August 31, 2010 - 8:48 am

Add a call to of_node_put in the error handling code following a call to
of_find_compatible_node or of_find_node_by_type.

This patch also substantially reorganizes the error handling code in the
function, to that it is possible first to jump to code that frees qe_port
and then to jump to code that also puts np.

The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@r exists@
local idexpression x;
expression E,E1,E2;
statement S;
@@

*x = 
(of_find_node_by_path
|of_find_node_by_name
|of_find_node_by_phandle
|of_get_parent
|of_get_next_parent
|of_get_next_child
|of_find_compatible_node
|of_match_node
|of_find_node_by_type
|of_find_node_with_property
|of_find_matching_node
|of_parse_phandle
)(...);
...
if (x == NULL) S
<... when != x = E
*if (...) {
  ... when != of_node_put(x)
      when != if (...) { ... of_node_put(x); ... }
(
  return <+...x...+>;
|
*  return ...;
)
}
...>
(
E2 = x;
|
of_node_put(x);
)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>

---
 drivers/serial/ucc_uart.c |   67 ++++++++++++++++++++++++----------------------
 1 file changed, 35 insertions(+), 32 deletions(-)

diff --git a/drivers/serial/ucc_uart.c b/drivers/serial/ucc_uart.c
index 3f4848e..38a5ef0 100644
--- a/drivers/serial/ucc_uart.c
+++ b/drivers/serial/ucc_uart.c
@@ -1270,13 +1270,12 @@ static int ucc_uart_probe(struct platform_device *ofdev,
 	ret = of_address_to_resource(np, 0, &res);
 	if (ret) {
 		dev_err(&ofdev->dev, "missing 'reg' property in device tree\n");
-		kfree(qe_port);
-		return ret;
+		goto out_free;
 	}
 	if (!res.start) {
 		dev_err(&ofdev->dev, "invalid 'reg' property in device tree\n");
-		kfree(qe_port);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto out_free;
 	}
 	qe_port->port.mapbase = res.start;
 
@@ -1286,17 +1285,17 @@ static int ucc_uart_probe(struct platform_device *ofdev,
 	if (!iprop) {
 		iprop = of_get_property(np, "device-id", NULL);
 		if (!iprop) ...
From: Grant Likely
Date: Wednesday, September 8, 2010 - 12:59 pm

Looks good to me.

--

From: Wolfram Sang
Date: Tuesday, August 31, 2010 - 9:51 am

Hmm, the function you change returns -ENODEV if a clock cannot be found, but
the return type is u32. This should better be 0. Can you take care of this as
well? You could then add my


-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
From: Julia Lawall
Date: Tuesday, August 31, 2010 - 10:44 am

Add a call to of_node_put in the error handling code following a call to
of_find_matching_node.

This patch also moves the existing call to of_node_put after the call to
iounmap in the error handling code, to make it possible to jump to
of_node_put without doing iounmap.  These appear to be disjoint operations,
so the ordering doesn't matter.

This patch furthermore changes the -ENODEV result in the error handling
code for of_find_matching_node to a return of 0, as found in the error
handling code for of_iomap, because the return type of the function is
unsigned.

The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@r exists@
local idexpression x;
expression E,E1,E2;
statement S;
@@

*x = 
(of_find_node_by_path
|of_find_node_by_name
|of_find_node_by_phandle
|of_get_parent
|of_get_next_parent
|of_get_next_child
|of_find_compatible_node
|of_match_node
|of_find_node_by_type
|of_find_node_with_property
|of_find_matching_node
|of_parse_phandle
)(...);
...
if (x == NULL) S
<... when != x = E
*if (...) {
  ... when != of_node_put(x)
      when != if (...) { ... of_node_put(x); ... }
(
  return <+...x...+>;
|
*  return ...;
)
}
...>
(
E2 = x;
|
of_node_put(x);
)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>

---
 drivers/net/can/mscan/mpc5xxx_can.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/can/mscan/mpc5xxx_can.c b/drivers/net/can/mscan/mpc5xxx_can.c
--- a/drivers/net/can/mscan/mpc5xxx_can.c
+++ b/drivers/net/can/mscan/mpc5xxx_can.c
@@ -143,12 +143,12 @@ static u32 __devinit mpc512x_can_get_clock(struct platform_device *ofdev,
 	np_clock = of_find_matching_node(NULL, mpc512x_clock_ids);
 	if (!np_clock) {
 		dev_err(&ofdev->dev, "couldn't find clock node\n");
-		return -ENODEV;
+		return 0;
 	}
 	clockctl = of_iomap(np_clock, 0);
 	if (!clockctl) {
 		dev_err(&ofdev->dev, "couldn't map clock registers\n");
-		return 0;
+		goto exit_put;
 ...
From: Wolfgang Grandegger
Date: Tuesday, August 31, 2010 - 12:22 pm

Acked-by: Wolfgang Grandegger <wg@grandegger.com>
--

Previous thread: [PATCH 4/4] arch/powerpc/platforms/chrp/nvram.c: Add of_node_put to avoid memory leak by Julia Lawall on Tuesday, August 31, 2010 - 8:48 am. (2 messages)

Next thread: [git pull] PCI fixes by Jesse Barnes on Tuesday, August 31, 2010 - 8:49 am. (1 message)