* [PATCH] mtd: ubi: nvmem: Fix device_node leak in ubi_nvmem_add()
@ 2026-10-27 9:15 Vishal Satle
2026-10-27 9:18 ` Daniel Golle
2026-10-27 11:18 ` Zhihao Cheng
0 siblings, 2 replies; 3+ messages in thread
From: Vishal Satle @ 2026-10-27 9:15 UTC (permalink / raw)
To: Richard Weinberger, Daniel Golle, Miquel Raynal, Vignesh Raghavendra
Cc: Zhihao Cheng, linux-mtd, linux-kernel, stable, Vishal Satle
ubi_nvmem_add() calls of_get_child_by_name() to verify that the UBI
volume device node contains an "nvmem-layout" subnode.
of_get_child_by_name() returns a pointer to the child device_node
with its reference count incremented.
Because the returned pointer was only evaluated as a boolean condition
and not retained or put, the reference count of the "nvmem-layout" node
was leaked whenever an NVMEM-enabled UBI volume was registered.
Assign the child node to a local variable and release it with
of_node_put() once existence has been validated.
Fixes: 34a5d8525b6a ("mtd: ubi: add NVMEM support")
Cc: stable@vger.kernel.org
Signed-off-by: Vishal Satle <vishalsatleit24@gmail.com>
---
drivers/mtd/ubi/nvmem.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/mtd/ubi/nvmem.c b/drivers/mtd/ubi/nvmem.c
index 2dade9727..15737d511 100644
--- a/drivers/mtd/ubi/nvmem.c
+++ b/drivers/mtd/ubi/nvmem.c
@@ -61,6 +61,7 @@ static int ubi_nvmem_reg_read(void *priv, unsigned int from,
static int ubi_nvmem_add(struct ubi_volume_info *vi)
{
struct device_node *np = dev_of_node(vi->dev);
+ struct device_node *layout;
struct nvmem_config config = {};
struct ubi_nvmem *unv;
int ret;
@@ -68,9 +69,12 @@ static int ubi_nvmem_add(struct ubi_volume_info *vi)
if (!np)
return 0;
- if (!of_get_child_by_name(np, "nvmem-layout"))
+ layout = of_get_child_by_name(np, "nvmem-layout");
+ if (!layout)
return 0;
+ of_node_put(layout);
+
if (WARN_ON_ONCE(vi->usable_leb_size <= 0) ||
WARN_ON_ONCE(vi->size <= 0))
return -EINVAL;
--
2.51.0
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] mtd: ubi: nvmem: Fix device_node leak in ubi_nvmem_add()
2026-10-27 9:15 [PATCH] mtd: ubi: nvmem: Fix device_node leak in ubi_nvmem_add() Vishal Satle
@ 2026-10-27 9:18 ` Daniel Golle
2026-10-27 11:18 ` Zhihao Cheng
1 sibling, 0 replies; 3+ messages in thread
From: Daniel Golle @ 2026-10-27 9:18 UTC (permalink / raw)
To: Vishal Satle
Cc: Richard Weinberger, Miquel Raynal, Vignesh Raghavendra,
Zhihao Cheng, linux-mtd, linux-kernel, stable
On Tue, Oct 27, 2026 at 02:45:07PM +0530, Vishal Satle wrote:
> ubi_nvmem_add() calls of_get_child_by_name() to verify that the UBI
> volume device node contains an "nvmem-layout" subnode.
> of_get_child_by_name() returns a pointer to the child device_node
> with its reference count incremented.
>
> Because the returned pointer was only evaluated as a boolean condition
> and not retained or put, the reference count of the "nvmem-layout" node
> was leaked whenever an NVMEM-enabled UBI volume was registered.
>
> Assign the child node to a local variable and release it with
> of_node_put() once existence has been validated.
>
> Fixes: 34a5d8525b6a ("mtd: ubi: add NVMEM support")
> Cc: stable@vger.kernel.org
> Signed-off-by: Vishal Satle <vishalsatleit24@gmail.com>
Thanks for fixing this.
Reviewed-by: Daniel Golle <daniel@makrotopia.org>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] mtd: ubi: nvmem: Fix device_node leak in ubi_nvmem_add()
2026-10-27 9:15 [PATCH] mtd: ubi: nvmem: Fix device_node leak in ubi_nvmem_add() Vishal Satle
2026-10-27 9:18 ` Daniel Golle
@ 2026-10-27 11:18 ` Zhihao Cheng
1 sibling, 0 replies; 3+ messages in thread
From: Zhihao Cheng @ 2026-10-27 11:18 UTC (permalink / raw)
To: Vishal Satle, Richard Weinberger, Daniel Golle, Miquel Raynal,
Vignesh Raghavendra
Cc: linux-mtd, linux-kernel, stable
在 2026/10/27 17:15, Vishal Satle 写道:
> ubi_nvmem_add() calls of_get_child_by_name() to verify that the UBI
> volume device node contains an "nvmem-layout" subnode.
> of_get_child_by_name() returns a pointer to the child device_node
> with its reference count incremented.
>
> Because the returned pointer was only evaluated as a boolean condition
> and not retained or put, the reference count of the "nvmem-layout" node
> was leaked whenever an NVMEM-enabled UBI volume was registered.
>
> Assign the child node to a local variable and release it with
> of_node_put() once existence has been validated.
>
> Fixes: 34a5d8525b6a ("mtd: ubi: add NVMEM support")
> Cc: stable@vger.kernel.org
> Signed-off-by: Vishal Satle <vishalsatleit24@gmail.com>
> ---
> drivers/mtd/ubi/nvmem.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
Reviewed-by: Zhihao Cheng <chengzhihao1@huawei.com>
>
> diff --git a/drivers/mtd/ubi/nvmem.c b/drivers/mtd/ubi/nvmem.c
> index 2dade9727..15737d511 100644
> --- a/drivers/mtd/ubi/nvmem.c
> +++ b/drivers/mtd/ubi/nvmem.c
> @@ -61,6 +61,7 @@ static int ubi_nvmem_reg_read(void *priv, unsigned int from,
> static int ubi_nvmem_add(struct ubi_volume_info *vi)
> {
> struct device_node *np = dev_of_node(vi->dev);
> + struct device_node *layout;
> struct nvmem_config config = {};
> struct ubi_nvmem *unv;
> int ret;
> @@ -68,9 +69,12 @@ static int ubi_nvmem_add(struct ubi_volume_info *vi)
> if (!np)
> return 0;
>
> - if (!of_get_child_by_name(np, "nvmem-layout"))
> + layout = of_get_child_by_name(np, "nvmem-layout");
> + if (!layout)
> return 0;
>
> + of_node_put(layout);
> +
> if (WARN_ON_ONCE(vi->usable_leb_size <= 0) ||
> WARN_ON_ONCE(vi->size <= 0))
> return -EINVAL;
>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2026-10-27 11:18 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-10-27 9:15 [PATCH] mtd: ubi: nvmem: Fix device_node leak in ubi_nvmem_add() Vishal Satle
2026-10-27 9:18 ` Daniel Golle
2026-10-27 11:18 ` Zhihao Cheng
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®