Skip to contentSkip to conversation

drm/msm/adreno: Fix a reference leak in a6xx_gpu_init()

PATCH In review Felix Gu 4 participants 4 messages Jan 23 – Jan 29

Subsystems (4)
lore
Share & tools
lore

Felix Gu

author
#0 lore raw

PATCH v2 1/1 drm/msm/adreno: Fix a reference leak in a6xx_gpu_init()

In a6xx_gpu_init(), node is obtained via of_parse_phandle(). While there was a manual of_node_put() at the end of the common path, several early error returns would bypass this call, resulting in a reference leak.
Fix this by using the __free(device_node) cleanup handler to release the reference when the variable goes out of scope.

Fixes 5a903a44a984 ("drm/msm/a6xx: Introduce GMU wrapper support")

---
Changes in v2:
- Explicitly add header file.
- Link to v1: https://lore.kernel.org/r/[email protected]

---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/msm/adreno/a6xx_gpu.c +3 -3
index 2129d230a92b..604b0f861d27 100644
@@ -9,6 +9,7 @@
 #include "a6xx_gmu.xml.h"
 
 #include <linux/bitfield.h>
+#include <linux/cleanup.h>
 #include <linux/devfreq.h>
 #include <linux/firmware/qcom/qcom_scm.h>
 #include <linux/pm_domain.h>
@@ -2640,7 +2641,6 @@ static struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 	struct msm_drm_private *priv = dev->dev_private;
 	struct platform_device *pdev = priv->gpu_pdev;
 	struct adreno_platform_config *config = pdev->dev.platform_data;
-	struct device_node *node;
 	struct a6xx_gpu *a6xx_gpu;
 	struct adreno_gpu *adreno_gpu;
 	struct msm_gpu *gpu;
@@ -2660,7 +2660,8 @@ static struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 	adreno_gpu->registers = NULL;
 
 	/* Check if there is a GMU phandle and set it up */
-	node = of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0);
+	struct device_node *node __free(device_node) =
+		of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0);
 	/* FIXME: How do we gracefully handle this? */
 	BUG_ON(!node);
 
@@ -2702,7 +2703,6 @@ static struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 		ret = a6xx_gmu_wrapper_init(a6xx_gpu, node);
 	else
 		ret = a6xx_gmu_init(a6xx_gpu, node);
-	of_node_put(node);
 	if (ret) {
 		a6xx_destroy(&(a6xx_gpu->base.base));
 		return ERR_PTR(ret);

---
base-commit: a0c666c25aeefd16f4b088c6549a6fb6b65a8a1d
change-id: 20260123-a6xx_gpu-cbc095dbe423

Best regards,

Signature · 1 line
Felix Gu <[email protected]>
Reviewed-by on v2 1/1
#2 lore raw
In reply to Felix Gu
Quoted 9 lines · Felix Gu

In a6xx_gpu_init(), node is obtained via of_parse_phandle(). While there was a manual of_node_put() at the end of the common path, several early error returns would bypass this call, resulting in a reference leak.
Fix this by using the __free(device_node) cleanup handler to release the reference when the variable goes out of scope.

Fixes 5a903a44a984 ("drm/msm/a6xx: Introduce GMU wrapper support")
Quoted 8 lines

---
Changes in v2:
- Explicitly add header file.
- Link to v1: https://lore.kernel.org/r/[email protected]

---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
Signature · 2 lines
With best wishes
Dmitry
reviewer Reviewed-by on v2 1/1
#3 lore raw
In reply to Felix Gu
Quoted 9 lines · Felix Gu

In a6xx_gpu_init(), node is obtained via of_parse_phandle(). While there was a manual of_node_put() at the end of the common path, several early error returns would bypass this call, resulting in a reference leak.
Fix this by using the __free(device_node) cleanup handler to release the reference when the variable goes out of scope.

Fixes 5a903a44a984 ("drm/msm/a6xx: Introduce GMU wrapper support")

-Akhil

Quoted 52 lines

---
Changes in v2:
- Explicitly add header file.
- Link to v1: https://lore.kernel.org/r/[email protected]

---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/msm/adreno/a6xx_gpu.c +3 -4
index 2129d230a92b..604b0f861d27 100644
@@ -9,6 +9,7 @@
 #include "a6xx_gmu.xml.h"
 
 #include <linux/bitfield.h>
+#include <linux/cleanup.h>
 #include <linux/devfreq.h>
 #include <linux/firmware/qcom/qcom_scm.h>
 #include <linux/pm_domain.h>
@@ -2640,7 +2641,6 @@ static struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 	struct msm_drm_private *priv = dev->dev_private;
 	struct platform_device *pdev = priv->gpu_pdev;
 	struct adreno_platform_config *config = pdev->dev.platform_data;
-	struct device_node *node;
 	struct a6xx_gpu *a6xx_gpu;
 	struct adreno_gpu *adreno_gpu;
 	struct msm_gpu *gpu;
@@ -2660,7 +2660,8 @@ static struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 	adreno_gpu->registers = NULL;
 
 	/* Check if there is a GMU phandle and set it up */
-	node = of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0);
+	struct device_node *node __free(device_node) =
+		of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0);
 	/* FIXME: How do we gracefully handle this? */
 	BUG_ON(!node);
 
@@ -2702,7 +2703,6 @@ static struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 		ret = a6xx_gmu_wrapper_init(a6xx_gpu, node);
 	else
 		ret = a6xx_gmu_init(a6xx_gpu, node);
-	of_node_put(node);
 	if (ret) {
 		a6xx_destroy(&(a6xx_gpu->base.base));
 		return ERR_PTR(ret);
 
---

base-commit: a0c666c25aeefd16f4b088c6549a6fb6b65a8a1d
change-id: 20260123-a6xx_gpu-cbc095dbe423

Best regards,