Skip to contentSkip to conversation

ACPICA: Replace one-element array with flexible-array

PATCH In review Gustavo A. R. Silva 3 participants 4 messages Jun 2, 2020 – Jun 10, 2020

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 ACPICA: Replace one-element array with flexible-array

The current codebase makes use of one-element arrays in the following form:

struct something {
    int length;
    u8 data[1];
};

struct something *instance;

instance = kmalloc(sizeof(*instance) + size, GFP_KERNEL); instance->length = size;
memcpy(instance->data, source, size);

but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99:

struct foo {
        int stuff;
        struct boo array[];
};

By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on.

This issue was found with the help of Coccinelle and audited _manually_.

[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732932f ("cxgb3/l2t: Fix undefined behaviour")

---
Changes in v2:
- Don't use struct_size() for now.
- Update subject line and changelog text.

 drivers/acpi/acpica/utids.c | 2 +-
 include/acpi/actypes.h      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/acpi/acpica/utids.c +1 -1
index 3bb06935a2ad3..225f3c60203c7 100644
@@ -263,7 +263,7 @@ acpi_ut_execute_CID(struct acpi_namespace_node *device_node,
 	 * 3) Size of the actual CID strings
 	 */
 	cid_list_size = sizeof(struct acpi_pnp_device_id_list) +
-	    ((count - 1) * sizeof(struct acpi_pnp_device_id)) +
+	    count * sizeof(struct acpi_pnp_device_id) +
 	    string_area_size;
 
 	cid_list = ACPI_ALLOCATE_ZEROED(cid_list_size);
include/acpi/actypes.h +1 -1
index 4defed58ea338..c7bcda0ad366a 100644
@@ -1145,7 +1145,7 @@ struct acpi_pnp_device_id {
 struct acpi_pnp_device_id_list {
 	u32 count;		/* Number of IDs in Ids array */
 	u32 list_size;		/* Size of list, including ID strings */
-	struct acpi_pnp_device_id ids[1];	/* ID array */
+	struct acpi_pnp_device_id ids[];	/* ID array */
 };
 
 /*
Signature · 1 line
2.27.0
#1 lore raw
Quoted 28 lines · Gustavo A. R. Silva

The current codebase makes use of one-element arrays in the following form:

struct something {
    int length;
    u8 data[1];
};

struct something *instance;

instance = kmalloc(sizeof(*instance) + size, GFP_KERNEL); instance->length = size;
memcpy(instance->data, source, size);

but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99:

struct foo {
        int stuff;
        struct boo array[];
};

By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on.

Quoted 7 lines

This issue was found with the help of Coccinelle and audited _manually_.

[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732932f ("cxgb3/l2t: Fix undefined behaviour")

Signed-off-by Gustavo A. R. Silva <gustavoars(a)kernel.org>

Erik, can you take this to the upstream, please?

Quoted 38 lines

---
Changes in v2:
- Don't use struct_size() for now.
- Update subject line and changelog text.

 drivers/acpi/acpica/utids.c | 2 +-
 include/acpi/actypes.h      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/acpi/acpica/utids.c +1 -1
index 3bb06935a2ad3..225f3c60203c7 100644
@@ -263,7 +263,7 @@ acpi_ut_execute_CID(struct acpi_namespace_node *device_node,
         * 3) Size of the actual CID strings
         */
        cid_list_size = sizeof(struct acpi_pnp_device_id_list) +
-           ((count - 1) * sizeof(struct acpi_pnp_device_id)) +
+           count * sizeof(struct acpi_pnp_device_id) +
            string_area_size;
 
        cid_list = ACPI_ALLOCATE_ZEROED(cid_list_size);
include/acpi/actypes.h +1 -2
index 4defed58ea338..c7bcda0ad366a 100644
@@ -1145,7 +1145,7 @@ struct acpi_pnp_device_id {
 struct acpi_pnp_device_id_list {
        u32 count;              /* Number of IDs in Ids array */
        u32 list_size;          /* Size of list, including ID strings */
-       struct acpi_pnp_device_id ids[1];       /* ID array */
+       struct acpi_pnp_device_id ids[];        /* ID array */
 };
 
 /*
--

2.27.0

#2 lore raw
Quoted 49 lines

-----Original Message-----
From: [email protected] <linux-acpi- [email protected]> On Behalf Of Rafael J. Wysocki
Sent: Wednesday, June 3, 2020 6:04 AM
To: Gustavo A. R. Silva <[email protected]>; Kaneda, Erik <[email protected]>
Cc: Moore, Robert <[email protected]>; Wysocki, Rafael J <[email protected]>; Len Brown <[email protected]>; ACPI Devel Maling List <[email protected]>; open list:ACPI COMPONENT ARCHITECTURE (ACPICA) <[email protected]>; Linux Kernel Mailing List <[email protected]>; Gustavo A. R. Silva <[email protected]>
Subject: Re: [PATCH v2] ACPICA: Replace one-element array with flexible- array

Quoted 30 lines · Gustavo A. R. Silva

The current codebase makes use of one-element arrays in the following form:

struct something {
    int length;
    u8 data[1];
};

struct something *instance;

instance = kmalloc(sizeof(*instance) + size, GFP_KERNEL); instance->length = size;
memcpy(instance->data, source, size);

but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99:

struct foo {
        int stuff;
        struct boo array[];
};

By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on.

This issue was found with the help of Coccinelle and audited _manually_.

Quoted 7 lines
Quoted 5 lines

[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732932f ("cxgb3/l2t: Fix undefined behaviour")

Erik, can you take this to the upstream, please?

Yes, We'll have to make additional changes to other structures with one-element arrays but we'll get this in time for the upcoming ACPICA release.

Thanks,
Erik

Quoted 40 lines

drivers/acpi/acpica/utids.c acpi_ut_execute_CID(struct acpi_namespace_node

Quoted 15 lines

---
Changes in v2:
- Don't use struct_size() for now.
- Update subject line and changelog text.

 drivers/acpi/acpica/utids.c | 2 +-
 include/acpi/actypes.h      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/acpi/acpica/utids.c +0 -0
index 3bb06935a2ad3..225f3c60203c7 100644
@@ -263,7 +263,7 @@ acpi_ut_execute_CID(struct acpi_namespace_node

*device_node,

Quoted 24 lines
drivers/acpi/acpica/utids.c +1 -1
         * 3) Size of the actual CID strings
         */
        cid_list_size = sizeof(struct acpi_pnp_device_id_list) +
-           ((count - 1) * sizeof(struct acpi_pnp_device_id)) +
+           count * sizeof(struct acpi_pnp_device_id) +
            string_area_size;
 
        cid_list = ACPI_ALLOCATE_ZEROED(cid_list_size);
include/acpi/actypes.h index renamed from include/acpi/actypes.h +0 -0

4defed58ea338..c7bcda0ad366a 100644

include/acpi/actypes.h +1 -2
@@ -1145,7 +1145,7 @@ struct acpi_pnp_device_id {  struct
 acpi_pnp_device_id_list {
        u32 count;              /* Number of IDs in Ids array */
        u32 list_size;          /* Size of list, including ID strings */
-       struct acpi_pnp_device_id ids[1];       /* ID array */
+       struct acpi_pnp_device_id ids[];        /* ID array */
 };
 
 /*
--

2.27.0

#3 lore raw
Quoted 72 lines

-----Original Message-----
From: [email protected] <linux-acpi- [email protected]> On Behalf Of Kaneda, Erik
Sent: Wednesday, June 3, 2020 4:13 PM
To: Rafael J. Wysocki <[email protected]>; Gustavo A. R. Silva <[email protected]>
Cc: Moore, Robert <[email protected]>; Wysocki, Rafael J <[email protected]>; Len Brown <[email protected]>; ACPI Devel Maling List <[email protected]>; open list:ACPI COMPONENT ARCHITECTURE (ACPICA) <[email protected]>; Linux Kernel Mailing List <[email protected]>; Gustavo A. R. Silva <[email protected]>
Subject: RE: [PATCH v2] ACPICA: Replace one-element array with flexible- array

Quoted 53 lines

-----Original Message-----
From: [email protected] <linux-acpi- [email protected]> On Behalf Of Rafael J. Wysocki
Sent: Wednesday, June 3, 2020 6:04 AM
To: Gustavo A. R. Silva <[email protected]>; Kaneda, Erik <[email protected]>
Cc: Moore, Robert <[email protected]>; Wysocki, Rafael J <[email protected]>; Len Brown <[email protected]>; ACPI Devel Maling List <[email protected]>; open list:ACPI COMPONENT ARCHITECTURE (ACPICA) <[email protected]>; Linux Kernel Mailing List <[email protected]>; Gustavo A. R. Silva <[email protected]>
Subject: Re: [PATCH v2] ACPICA: Replace one-element array with flexible- array

Quoted 35 lines · Gustavo A. R. Silva

The current codebase makes use of one-element arrays in the following
form:

struct something {
    int length;
    u8 data[1];
};

struct something *instance;

instance = kmalloc(sizeof(*instance) + size, GFP_KERNEL); instance->length = size;
memcpy(instance->data, source, size);

but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99:

struct foo {
        int stuff;
        struct boo array[];
};

By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on.

This issue was found with the help of Coccinelle and audited _manually_.

[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732932f ("cxgb3/l2t: Fix undefined behaviour")

Quoted 5 lines
Quoted 3 lines
Quoted 1 line

Erik, can you take this to the upstream, please?

Yes, We'll have to make additional changes to other structures with one- element arrays but we'll get this in time for the upcoming ACPICA release.

Pull request available here: https://github.com/acpica/acpica/pull/608

Erik

Quoted 45 lines

Thanks,
Erik

Quoted 10 lines
Quoted 10 lines

---
Changes in v2:
- Don't use struct_size() for now.
- Update subject line and changelog text.

 drivers/acpi/acpica/utids.c | 2 +-
 include/acpi/actypes.h      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
Quoted 6 lines

drivers/acpi/acpica/utids.c acpi_ut_execute_CID(struct

Quoted 6 lines
(quoted diff) +0 -0

b/drivers/acpi/acpica/utids.c index 3bb06935a2ad3..225f3c60203c7 100644

drivers/acpi/acpica/utids.c +0 -0
@@ -263,7 +263,7 @@ acpi_ut_execute_CID(struct

acpi_namespace_node

Quoted 25 lines

*device_node,

Quoted 24 lines
drivers/acpi/acpica/utids.c +1 -1
         * 3) Size of the actual CID strings
         */
        cid_list_size = sizeof(struct acpi_pnp_device_id_list) +
-           ((count - 1) * sizeof(struct acpi_pnp_device_id)) +
+           count * sizeof(struct acpi_pnp_device_id) +
            string_area_size;
 
        cid_list = ACPI_ALLOCATE_ZEROED(cid_list_size);
include/acpi/actypes.h index renamed from include/acpi/actypes.h +0 -0

4defed58ea338..c7bcda0ad366a 100644

include/acpi/actypes.h +1 -2
@@ -1145,7 +1145,7 @@ struct acpi_pnp_device_id {  struct
 acpi_pnp_device_id_list {
        u32 count;              /* Number of IDs in Ids array */
        u32 list_size;          /* Size of list, including ID strings */
-       struct acpi_pnp_device_id ids[1];       /* ID array */
+       struct acpi_pnp_device_id ids[];        /* ID array */
 };
 
 /*
--

2.27.0