[PATCH v3 1/2] ARM: OMAP2+: Fix device node reference counts

[PATCH v3 1/2] ARM: OMAP2+: Fix device node reference counts

[Guenter Roeck]

After commit 0549bde0fcb1 ("of: fix of_node leak caused in
of_find_node_opts_by_path"), the following error may be
reported when running omap images.

OF: ERROR: Bad of_node_put() on /ocp@68000000
CPU: 0 PID: 0 Comm: swapper Not tainted 4.10.0-rc7-next-20170210 #1
Hardware name: Generic OMAP3-GP (Flattened Device Tree)
[<c0310604>] (unwind_backtrace) from [<c030bbf4>] (show_stack+0x10/0x14)
[<c030bbf4>] (show_stack) from [<c05add8c>] (dump_stack+0x98/0xac)
[<c05add8c>] (dump_stack) from [<c05af1b0>] (kobject_release+0x48/0x7c)
[<c05af1b0>] (kobject_release)
	from [<c0ad1aa4>] (of_find_node_by_name+0x74/0x94)
[<c0ad1aa4>] (of_find_node_by_name)
	from [<c1215bd4>] (omap3xxx_hwmod_is_hs_ip_block_usable+0x24/0x2c)
[<c1215bd4>] (omap3xxx_hwmod_is_hs_ip_block_usable) from
[<c1215d5c>] (omap3xxx_hwmod_init+0x180/0x274)
[<c1215d5c>] (omap3xxx_hwmod_init)
	from [<c120faa8>] (omap3_init_early+0xa0/0x11c)
[<c120faa8>] (omap3_init_early)
	from [<c120fb2c>] (omap3430_init_early+0x8/0x30)
[<c120fb2c>] (omap3430_init_early)
	from [<c1204710>] (setup_arch+0xc04/0xc34)
[<c1204710>] (setup_arch) from [<c1200948>] (start_kernel+0x68/0x38c)
[<c1200948>] (start_kernel) from [<8020807c>] (0x8020807c)

of_find_node_by_name() drops the reference to the passed device node.
The commit referenced above exposes this problem.

To fix the problem, use of_get_child_by_name() instead of
of_find_node_by_name(); of_get_child_by_name() does not drop
the reference count of passed device nodes. While semantically
different, we only look for immediate children of the passed
device node, so of_get_child_by_name() is a more appropriate
function to use anyway.

Release the reference to the device node obtained with
of_get_child_by_name() after it is no longer needed to avoid
another device node leak.

While at it, clean up the code and change the return type of
omap3xxx_hwmod_is_hs_ip_block_usable() to bool to match its use
and the return type of of_device_is_available().

Cc: Qi Hou <qi.hou@windriver.com>
Cc: Peter Rosin <peda@axentia.se>
Cc: Rob Herring <robh@kernel.org>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v3: Add commit SHA of patch exposing the problem
    Clarify that it does not cause the problem but exposes it
    Split into two patches
v2: Change subject ('Grab reference to device nodes where needed'
    didn't really cover all the changes made)
    Use of_get_child_by_name() instead of of_find_node_by_name()
    Drop references to device nodes as needed
    Change return type of omap3xxx_hwmod_is_hs_ip_block_usable()
    to bool

 arch/arm/mach-omap2/omap_hwmod_3xxx_data.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
index 56f917ec8621..44b5f9bbdde6 100644
--- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
@@ -3111,16 +3111,20 @@ static struct omap_hwmod_ocp_if *omap3xxx_dss_hwmod_ocp_ifs[] __initdata = {
  * Return: 0 if device named @dev_name is not likely to be accessible,
  * or 1 if it is likely to be accessible.
  */
-static int __init omap3xxx_hwmod_is_hs_ip_block_usable(struct device_node *bus,
-						       const char *dev_name)
+static bool __init omap3xxx_hwmod_is_hs_ip_block_usable(struct device_node *bus,
+							const char *dev_name)
 {
+	struct device_node *node;
+	bool available;
+
 	if (!bus)
-		return (omap_type() == OMAP2_DEVICE_TYPE_GP) ? 1 : 0;
+		return omap_type() == OMAP2_DEVICE_TYPE_GP;
 
-	if (of_device_is_available(of_find_node_by_name(bus, dev_name)))
-		return 1;
+	node = of_get_child_by_name(bus, dev_name);
+	available = of_device_is_available(node);
+	of_node_put(node);
 
-	return 0;
+	return available;
 }
 
 int __init omap3xxx_hwmod_init(void)
-- 
2.7.4

[PATCH v3 2/2] ARM: OMAP2+: Release device node after it is no longer needed.

[Guenter Roeck]

The device node returned by of_find_node_by_name() needs to be released
after it is no longer needed to avoid a device node leak.

Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v3: Split patch into two parts

 arch/arm/mach-omap2/omap_hwmod_3xxx_data.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
index 44b5f9bbdde6..042e460f4aa4 100644
--- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
@@ -3193,15 +3193,20 @@ int __init omap3xxx_hwmod_init(void)
 
 	if (h_sham && omap3xxx_hwmod_is_hs_ip_block_usable(bus, "sham")) {
 		r = omap_hwmod_register_links(h_sham);
-		if (r < 0)
+		if (r < 0) {
+			of_node_put(bus);
 			return r;
+		}
 	}
 
 	if (h_aes && omap3xxx_hwmod_is_hs_ip_block_usable(bus, "aes")) {
 		r = omap_hwmod_register_links(h_aes);
-		if (r < 0)
+		if (r < 0) {
+			of_node_put(bus);
 			return r;
+		}
 	}
+	of_node_put(bus);
 
 	/*
 	 * Register hwmod links specific to certain ES levels of a
-- 
2.7.4

Re: [PATCH v3 1/2] ARM: OMAP2+: Fix device node reference counts

[Tony Lindgren]

* Guenter Roeck <linux@roeck-us.net> [170304 07:04]:
> After commit 0549bde0fcb1 ("of: fix of_node leak caused in
> of_find_node_opts_by_path"), the following error may be
> reported when running omap images.
> 
> OF: ERROR: Bad of_node_put() on /ocp@68000000
> CPU: 0 PID: 0 Comm: swapper Not tainted 4.10.0-rc7-next-20170210 #1
> Hardware name: Generic OMAP3-GP (Flattened Device Tree)
> [<c0310604>] (unwind_backtrace) from [<c030bbf4>] (show_stack+0x10/0x14)
> [<c030bbf4>] (show_stack) from [<c05add8c>] (dump_stack+0x98/0xac)
> [<c05add8c>] (dump_stack) from [<c05af1b0>] (kobject_release+0x48/0x7c)
> [<c05af1b0>] (kobject_release)
> 	from [<c0ad1aa4>] (of_find_node_by_name+0x74/0x94)
> [<c0ad1aa4>] (of_find_node_by_name)
> 	from [<c1215bd4>] (omap3xxx_hwmod_is_hs_ip_block_usable+0x24/0x2c)
> [<c1215bd4>] (omap3xxx_hwmod_is_hs_ip_block_usable) from
> [<c1215d5c>] (omap3xxx_hwmod_init+0x180/0x274)
> [<c1215d5c>] (omap3xxx_hwmod_init)
> 	from [<c120faa8>] (omap3_init_early+0xa0/0x11c)
> [<c120faa8>] (omap3_init_early)
> 	from [<c120fb2c>] (omap3430_init_early+0x8/0x30)
> [<c120fb2c>] (omap3430_init_early)
> 	from [<c1204710>] (setup_arch+0xc04/0xc34)
> [<c1204710>] (setup_arch) from [<c1200948>] (start_kernel+0x68/0x38c)
> [<c1200948>] (start_kernel) from [<8020807c>] (0x8020807c)
> 
> of_find_node_by_name() drops the reference to the passed device node.
> The commit referenced above exposes this problem.
> 
> To fix the problem, use of_get_child_by_name() instead of
> of_find_node_by_name(); of_get_child_by_name() does not drop
> the reference count of passed device nodes. While semantically
> different, we only look for immediate children of the passed
> device node, so of_get_child_by_name() is a more appropriate
> function to use anyway.
> 
> Release the reference to the device node obtained with
> of_get_child_by_name() after it is no longer needed to avoid
> another device node leak.
> 
> While at it, clean up the code and change the return type of
> omap3xxx_hwmod_is_hs_ip_block_usable() to bool to match its use
> and the return type of of_device_is_available().
> 
> Cc: Qi Hou <qi.hou@windriver.com>
> Cc: Peter Rosin <peda@axentia.se>
> Cc: Rob Herring <robh@kernel.org>
> Signed-off-by: Guenter Roeck <linux@roeck-us.net>
> ---
> v3: Add commit SHA of patch exposing the problem
>     Clarify that it does not cause the problem but exposes it
>     Split into two patches
> v2: Change subject ('Grab reference to device nodes where needed'
>     didn't really cover all the changes made)
>     Use of_get_child_by_name() instead of of_find_node_by_name()
>     Drop references to device nodes as needed
>     Change return type of omap3xxx_hwmod_is_hs_ip_block_usable()
>     to bool

Thanks for the updates, applying both into omap-for-v4.11/fixes.

Regards,

Tony