edac: versalnet: Fix device_node leak in mc_probe()
Subsystems (2)
PATCH 1/1 edac: versalnet: Fix device_node leak in mc_probe()
of_parse_phandle() returns a device_node reference that must be released with of_node_put(). The original code never freed r5_core_node on any exit path, causing a memory leak.
Fix this by using the automatic cleanup attribute __free(device_node) which ensures of_node_put() is called when the variable goes out of scope.
--- drivers/edac/versalnet_edac.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/edac/versalnet_edac.c +2 -2
@@ -888,12 +888,12 @@ static void remove_versalnet(struct mc_priv *priv)static int mc_probe (struct platform_device *pdev) { -struct device_node *r5_core_node;struct mc_priv *priv;struct rproc *rp;int rc; - r5_core_node =of_parse_phandle (pdev->dev.of_node,"amd,rproc" ,0 ); + struct device_node *r5_core_node __free(device_node) = +of_parse_phandle (pdev->dev.of_node,"amd,rproc" ,0 );if (!r5_core_node) {dev_err (&pdev->dev,"amd,rproc: invalid phandle\n" );return -EINVAL ;
---
base-commit: 785f0eb2f85decbe7c1ef9ae922931f0194ffc2e
change-id: 20260322-versalnet-ee4feebc280e
Best regards,
Signature · 1 line
Felix Gu <[email protected]>
Borislav Petkov
maintainerQuoted 25 lines · Felix Gu
of_parse_phandle()returns adevice_nodereference that must be released withof_node_put(). The original code never freedr5_core_nodeon any exit path, causing a memory leak.Fix this by using the automatic cleanup attribute
__free(device_node)which ensuresof_node_put()is called when the variable goes out of scope.Fixes d5fe2fec6c40 ("EDAC: Add a driver for the AMD Versal NET DDR controller")Signed-off-by Felix Gu <[email protected]>--- drivers/edac/versalnet_edac.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)drivers/edac/versalnet_edac.c +0
-1@@ -888,12 +888,12 @@ static void remove_versalnet(struct mc_priv *priv)static int mc_probe (struct platform_device *pdev) { -struct device_node *r5_core_node;struct mc_priv *priv;struct rproc *rp;
drivers/edac/versalnet_edac.c static void remove_versalnet(struct mc_priv *priv)
Quoted 10 lines
drivers/edac/versalnet_edac.c +2
-2int rc; - r5_core_node =of_parse_phandle (pdev->dev.of_node,"amd,rproc" ,0 ); + struct device_node *r5_core_node __free(device_node) = +of_parse_phandle (pdev->dev.of_node,"amd,rproc" ,0 );if (!r5_core_node) {dev_err (&pdev->dev,"amd,rproc: invalid phandle\n" );return -EINVAL ; ---
Looks correct to me.
@Shubhrajyoti, can you pls test it?
Thx.
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette[AMD Official Use Only - AMD Internal Distribution Only]
Quoted 50 lines
-----Original Message-----
From: Borislav Petkov <[email protected]>
Sent: Sunday, March 22, 2026 10:01 PM
To: Felix Gu <[email protected]>
Cc: Datta, Shubhrajyoti <[email protected]>; Tony Luck <[email protected]>; [email protected]; linux- [email protected]
Subject: Re: [PATCH] edac: versalnet: Fixdevice_nodeleak inmc_probe()Caution: This message originated from an External Source. Use proper caution when opening attachments, clicking links, or responding.
Quoted 36 lines · Felix Gu
of_parse_phandle()returns adevice_nodereference that must be released withof_node_put(). The original code never freedr5_core_nodeon any exit path, causing a memory leak.Fix this by using the automatic cleanup attribute
__free(device_node)which ensuresof_node_put()is called when the variable goes out of scope.Fixes d5fe2fec6c40 ("EDAC: Add a driver for the AMD Versal NET DDRcontroller")
Signed-off-by Felix Gu <[email protected]>--- drivers/edac/versalnet_edac.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)(quoted diff) +0
-0
b/drivers/edac/versalnet_edac.cindex0b47ed7fed63..915bcd6166f7100644drivers/edac/versalnet_edac.c +2
-2@@ -888,12 +888,12 @@ static void remove_versalnet(struct mc_priv *priv)static int mc_probe (struct platform_device *pdev) { -struct device_node *r5_core_node;struct mc_priv *priv;struct rproc *rp;int rc; - r5_core_node =of_parse_phandle (pdev->dev.of_node,"amd,rproc" ,0 ); + struct device_node *r5_core_node __free(device_node) = +of_parse_phandle (pdev->dev.of_node,"amd,rproc" ,0 );if (!r5_core_node) {dev_err (&pdev->dev,"amd,rproc: invalid phandle\n" );return -EINVAL ;
Quoted 7 lines
Quoted 1 line
---
Looks correct to me.
@Shubhrajyoti, can you pls test it?
Thx.
I tested this change on hardware and it works as expected. Please feel free to add:
Quoted 6 lines
--
Regards/Gruss,
Boris.
Quoted 56 lines · Shubhrajyoti
[AMD Official Use Only - AMD Internal Distribution Only]
Quoted 54 lines
-----Original Message-----
From: Borislav Petkov <[email protected]>
Sent: Sunday, March 22, 2026 10:01 PM
To: Felix Gu <[email protected]>
Cc: Datta, Shubhrajyoti <[email protected]>; Tony Luck <[email protected]>; [email protected]; linux- [email protected]
Subject: Re: [PATCH] edac: versalnet: Fixdevice_nodeleak inmc_probe()Caution: This message originated from an External Source. Use proper caution when opening attachments, clicking links, or responding.
Quoted 27 lines · Felix Gu
of_parse_phandle()returns adevice_nodereference that must be released withof_node_put(). The original code never freedr5_core_nodeon any exit path, causing a memory leak.Fix this by using the automatic cleanup attribute
__free(device_node)which ensuresof_node_put()is called when the variable goes out of scope.Fixes d5fe2fec6c40 ("EDAC: Add a driver for the AMD Versal NET DDRcontroller")
Signed-off-by Felix Gu <[email protected]>--- drivers/edac/versalnet_edac.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)(quoted diff) +0
-0
b/drivers/edac/versalnet_edac.cindex0b47ed7fed63..915bcd6166f7100644drivers/edac/versalnet_edac.c +0
-1@@ -888,12 +888,12 @@ static void remove_versalnet(struct mc_priv *priv)static int mc_probe (struct platform_device *pdev) { -struct device_node *r5_core_node;struct mc_priv *priv;struct rproc *rp;drivers/edac/versalnet_edac.c static void remove_versalnet(struct mc_priv
Quoted 10 lines
drivers/edac/versalnet_edac.c +2
-2int rc; - r5_core_node =of_parse_phandle (pdev->dev.of_node,"amd,rproc" ,0 ); + struct device_node *r5_core_node __free(device_node) = +of_parse_phandle (pdev->dev.of_node,"amd,rproc" ,0 );if (!r5_core_node) {dev_err (&pdev->dev,"amd,rproc: invalid phandle\n" );return -EINVAL ; ---Looks correct to me.
Quoted 7 lines
Quoted 3 lines
@Shubhrajyoti, can you pls test it?
Thx.
I tested this change on hardware and it works as expected. Please feel free to add:
Reviewed-by Shubhrajyoti Datta <[email protected]>
Applied, thanks to both.
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette