ACPI, APEI, HEST Fix the unsuitable usage of platform_data.

platform_data in hest_parse_ghes() is used for saving the address of entry information of erst_tab. When the device is failed to be added, platform_data will be freed by platform_device_put(). But the value saved in platform_data should not be freed here. If it is done, it will make system panic.

So I think platform_data should save the address of allocated memory which saves entry information of erst_tab.

This patch fixed it and I confirmed it on x86_64 next-tree.

Signed-off-by Jin Dongming <[email protected]>

---
 drivers/acpi/apei/ghes.c |    2 +-
 drivers/acpi/apei/hest.c |   19 ++++++++++++++-----
 2 files changed, 15 insertions(+), 6 deletions(-)

drivers/acpi/apei/ghes.c +1 -1

index 385a605..b35c622 100644
@@ -302,7 +302,7 @@ static int __devinit ghes_probe(struct platform_device *ghes_dev)
 	struct ghes *ghes = NULL;
 	int rc = -EINVAL;
 
-	generic = ghes_dev->dev.platform_data;
+	generic = *(struct acpi_hest_generic **)(ghes_dev->dev.platform_data);
 	if (!generic->enabled)
 		return -ENODEV;
 

drivers/acpi/apei/hest.c +9 -4

index 343168d..5a65e2a 100644
@@ -137,20 +137,29 @@ static int hest_parse_ghes_count(struct acpi_hest_header *hest_hdr, void *data)
 
 static int hest_parse_ghes(struct acpi_hest_header *hest_hdr, void *data)
 {
-	struct acpi_hest_generic *generic;
+	void **platform_data;
 	struct platform_device *ghes_dev;
 	struct ghes_arr *ghes_arr = data;
 	int rc;
 
 	if (hest_hdr->type != ACPI_HEST_TYPE_GENERIC_ERROR)
 		return 0;
-	generic = (struct acpi_hest_generic *)hest_hdr;
-	if (!generic->enabled)
+
+	platform_data = kmalloc(sizeof(void *), GFP_KERNEL);
+	if (!platform_data)
+		return -ENOMEM;
+
+	*platform_data = hest_hdr;
+	if (!((struct acpi_hest_generic *)*platform_data)->enabled)
 		return 0;
+
 	ghes_dev = platform_device_alloc("GHES", hest_hdr->source_id);
-	if (!ghes_dev)

Quoted 4 lines · Jin Dongming

platform_data in hest_parse_ghes() is used for saving the address of entry information of erst_tab. When the device is failed to be added, platform_data will be freed by platform_device_put(). But the value saved in platform_data should not be freed here. If it is done, it will make system panic.

Good catch. Thanks.

Quoted 57 lines

So I think platform_data should save the address of allocated memory which saves entry information of erst_tab.

This patch fixed it and I confirmed it on x86_64 next-tree.

Signed-off-by Jin Dongming <[email protected]>

---
 drivers/acpi/apei/ghes.c |    2 +-
 drivers/acpi/apei/hest.c |   19 ++++++++++++++-----
 2 files changed, 15 insertions(+), 6 deletions(-)

drivers/acpi/apei/ghes.c +1 -1

index 385a605..b35c622 100644
@@ -302,7 +302,7 @@ static int __devinit ghes_probe(struct platform_device *ghes_dev)
 	struct ghes *ghes = NULL;
 	int rc = -EINVAL;
 
-	generic = ghes_dev->dev.platform_data;
+	generic = *(struct acpi_hest_generic **)(ghes_dev->dev.platform_data);
 	if (!generic->enabled)
 		return -ENODEV;
 

drivers/acpi/apei/hest.c +11 -5

index 343168d..5a65e2a 100644
@@ -137,20 +137,29 @@ static int hest_parse_ghes_count(struct acpi_hest_header *hest_hdr, void *data)
 
 static int hest_parse_ghes(struct acpi_hest_header *hest_hdr, void *data)
 {
-	struct acpi_hest_generic *generic;
+	void **platform_data;
 	struct platform_device *ghes_dev;
 	struct ghes_arr *ghes_arr = data;
 	int rc;
 
 	if (hest_hdr->type != ACPI_HEST_TYPE_GENERIC_ERROR)
 		return 0;
-	generic = (struct acpi_hest_generic *)hest_hdr;
-	if (!generic->enabled)
+
+	platform_data = kmalloc(sizeof(void *), GFP_KERNEL);
+	if (!platform_data)
+		return -ENOMEM;
+
+	*platform_data = hest_hdr;
+	if (!((struct acpi_hest_generic *)*platform_data)->enabled)
 		return 0;
+
 	ghes_dev = platform_device_alloc("GHES", hest_hdr->source_id);
-	if (!ghes_dev)
+	if (!ghes_dev) {
+		kfree(platform_data);
 		return -ENOMEM;
-	ghes_dev->dev.platform_data = generic;

drivers/acpi/apei/hest.c static int hest_parse_ghes_count(struct acpi_hest_header *hest_hdr, void *data) · line 165

Quoted 6 lines

drivers/acpi/apei/hest.c +3 -0

+	}
+
+	ghes_dev->dev.platform_data = platform_data;
 	rc = platform_device_add(ghes_dev);
 	if (rc)
 		goto err;

Why not use platform_device_add_data?

Best Regards,
Huang Ying

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Quoted 1 line

Why not use platform_device_add_data?

Here what should be done is to transport the pointer of hest_hdr to platform_data, so I think that it could be done successfully by allocating 8byte of memory. If the platform_device_add_data() is used, the more memory will be allocated.

So I think this method is better than using platform_device_add_data().

Quoted 5 lines · Jin Dongming

Quoted 1 line

Why not use platform_device_add_data?

Here what should be done is to transport the pointer of hest_hdr to platform_data, so I think that it could be done successfully by allocating 8byte of memory. If the platform_device_add_data() is used, the more memory will be allocated.

Quoted 1 line

So I think this method is better than using platform_device_add_data().

You can specify allocation size for platform_device_add_data(), so no more memory will be allocated.

Best Regards,
Huang Ying

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/