[PATCH] treewide: Replace zero-length array with flexible-array

[PATCH] treewide: Replace zero-length array with flexible-array

[Gustavo A. R. Silva]

The current codebase makes use of the zero-length array language
extension to the C90 standard, 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.

Also, notice that, dynamic memory allocations won't be affected by
this change:

"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]

sizeof(flexible-array-member) triggers a warning because flexible array
members have incomplete type[1]. There are some instances of code in
which the sizeof operator is being incorrectly/erroneously applied to
zero-length arrays and the result is zero. Such instances may be hiding
some bugs. So, this work (flexible-array member conversions) will also
help to get completely rid of those sorts of issues.

This issue was found with the help of Coccinelle.

[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@kernel.org>
---
 samples/mei/mei-amt-version.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/samples/mei/mei-amt-version.c b/samples/mei/mei-amt-version.c
index 32234481ad7d..ad3e56042f96 100644
--- a/samples/mei/mei-amt-version.c
+++ b/samples/mei/mei-amt-version.c
@@ -267,7 +267,7 @@ struct amt_host_if_msg_header {
 struct amt_host_if_resp_header {
 	struct amt_host_if_msg_header header;
 	uint32_t status;
-	unsigned char data[0];
+	unsigned char data[];
 } __attribute__((packed));
 
 const uuid_le MEI_IAMTHIF = UUID_LE(0x12f80028, 0xb4b7, 0x4b2d,  \

RE: [PATCH] treewide: Replace zero-length array with flexible-array

[Winkler, Tomas]

> -----Original Message-----
> From: Gustavo A. R. Silva <gustavoars@kernel.org>
> Sent: Thursday, May 07, 2020 21:54
> To: Winkler, Tomas <tomas.winkler@intel.com>
> Cc: linux-kernel@vger.kernel.org
> Subject: [PATCH] treewide: Replace zero-length array with flexible-array


Ack. 

> 
> The current codebase makes use of the zero-length array language extension to
> the C90 standard, 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.
> 
> Also, notice that, dynamic memory allocations won't be affected by this change:
> 
> "Flexible array members have incomplete type, and so the sizeof operator may
> not be applied. As a quirk of the original implementation of zero-length arrays,
> sizeof evaluates to zero."[1]
> 
> sizeof(flexible-array-member) triggers a warning because flexible array
> members have incomplete type[1]. There are some instances of code in which
> the sizeof operator is being incorrectly/erroneously applied to zero-length
> arrays and the result is zero. Such instances may be hiding some bugs. So, this
> work (flexible-array member conversions) will also help to get completely rid of
> those sorts of issues.
> 
> This issue was found with the help of Coccinelle.
> 
> [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@kernel.org>
> ---
>  samples/mei/mei-amt-version.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/samples/mei/mei-amt-version.c b/samples/mei/mei-amt-version.c
> index 32234481ad7d..ad3e56042f96 100644
> --- a/samples/mei/mei-amt-version.c
> +++ b/samples/mei/mei-amt-version.c
> @@ -267,7 +267,7 @@ struct amt_host_if_msg_header {  struct
> amt_host_if_resp_header {
>  	struct amt_host_if_msg_header header;
>  	uint32_t status;
> -	unsigned char data[0];
> +	unsigned char data[];
>  } __attribute__((packed));
> 
>  const uuid_le MEI_IAMTHIF = UUID_LE(0x12f80028, 0xb4b7, 0x4b2d,  \

Re: [PATCH] treewide: Replace zero-length array with flexible-array

[Li Yang]

On Thu, May 7, 2020 at 1:49 PM Gustavo A. R. Silva
<gustavoars@kernel.org> wrote:
>
> The current codebase makes use of the zero-length array language
> extension to the C90 standard, 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.
>
> Also, notice that, dynamic memory allocations won't be affected by
> this change:
>
> "Flexible array members have incomplete type, and so the sizeof operator
> may not be applied. As a quirk of the original implementation of
> zero-length arrays, sizeof evaluates to zero."[1]
>
> sizeof(flexible-array-member) triggers a warning because flexible array
> members have incomplete type[1]. There are some instances of code in
> which the sizeof operator is being incorrectly/erroneously applied to
> zero-length arrays and the result is zero. Such instances may be hiding
> some bugs. So, this work (flexible-array member conversions) will also
> help to get completely rid of those sorts of issues.
>
> This issue was found with the help of Coccinelle.
>
> [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@kernel.org>
> ---
>  include/linux/fsl/bestcomm/bestcomm.h |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Applied for next.  Thanks.

Regards,
Leo

>
> diff --git a/include/linux/fsl/bestcomm/bestcomm.h b/include/linux/fsl/bestcomm/bestcomm.h
> index a0e2e6b19b57..154e541ce57e 100644
> --- a/include/linux/fsl/bestcomm/bestcomm.h
> +++ b/include/linux/fsl/bestcomm/bestcomm.h
> @@ -27,7 +27,7 @@
>   */
>  struct bcom_bd {
>         u32     status;
> -       u32     data[0];        /* variable payload size */
> +       u32     data[]; /* variable payload size */
>  };
>
>  /* ======================================================================== */
>

Re: [PATCH] treewide: Replace zero-length array with flexible-array

[Johan Hovold]

On Thu, May 07, 2020 at 01:53:13PM -0500, Gustavo A. R. Silva wrote:
> The current codebase makes use of the zero-length array language
> extension to the C90 standard, 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.
> 
> Also, notice that, dynamic memory allocations won't be affected by
> this change:
> 
> "Flexible array members have incomplete type, and so the sizeof operator
> may not be applied. As a quirk of the original implementation of
> zero-length arrays, sizeof evaluates to zero."[1]
> 
> sizeof(flexible-array-member) triggers a warning because flexible array
> members have incomplete type[1]. There are some instances of code in
> which the sizeof operator is being incorrectly/erroneously applied to
> zero-length arrays and the result is zero. Such instances may be hiding
> some bugs. So, this work (flexible-array member conversions) will also
> help to get completely rid of those sorts of issues.
> 
> This issue was found with the help of Coccinelle.
> 
> [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@kernel.org>

Applied for -next, thanks.

Johan

Re: [PATCH] treewide: Replace zero-length array with flexible-array

[Luck, Tony]

On Thu, May 07, 2020 at 01:53:34PM -0500, Gustavo A. R. Silva wrote:
> The current codebase makes use of the zero-length array language
> extension to the C90 standard, 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.
> 
> Also, notice that, dynamic memory allocations won't be affected by
> this change:
> 
> "Flexible array members have incomplete type, and so the sizeof operator
> may not be applied. As a quirk of the original implementation of
> zero-length arrays, sizeof evaluates to zero."[1]
> 
> sizeof(flexible-array-member) triggers a warning because flexible array
> members have incomplete type[1]. There are some instances of code in
> which the sizeof operator is being incorrectly/erroneously applied to
> zero-length arrays and the result is zero. Such instances may be hiding
> some bugs. So, this work (flexible-array member conversions) will also
> help to get completely rid of those sorts of issues.
> 
> This issue was found with the help of Coccinelle.
> 
> [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@kernel.org>

I'm fine with this specific change ... so

Acked-by: Tony Luck <tony.luck@intel.com>

for anyone picking up all of these treewide: patches
(if they don't, then I can apply this to the ia64 tree)


But a question ... is sizeof still ok on a structure that
contains a flexible-array-member?

E.g. I'd expect:

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

	printk("size of foo = %d\n", sizeof(struct foo));

To not give any complilation warnings and to tell me that
the size of the structure is 4 bytes.

Is that still true and OK?

-Tony

[PATCH] treewide: Replace zero-length array with flexible-array

[Gustavo A. R. Silva]

The current codebase makes use of the zero-length array language
extension to the C90 standard, 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.

Also, notice that, dynamic memory allocations won't be affected by
this change:

"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]

sizeof(flexible-array-member) triggers a warning because flexible array
members have incomplete type[1]. There are some instances of code in
which the sizeof operator is being incorrectly/erroneously applied to
zero-length arrays and the result is zero. Such instances may be hiding
some bugs. So, this work (flexible-array member conversions) will also
help to get completely rid of those sorts of issues.

This issue was found with the help of Coccinelle.

[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@kernel.org>
---
 arch/ia64/kernel/unwind_i.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/ia64/kernel/unwind_i.h b/arch/ia64/kernel/unwind_i.h
index 67994a7e5816..1dd57ba44327 100644
--- a/arch/ia64/kernel/unwind_i.h
+++ b/arch/ia64/kernel/unwind_i.h
@@ -42,7 +42,7 @@ enum unw_register_index {
 
 struct unw_info_block {
 	u64 header;
-	u64 desc[0];		/* unwind descriptors */
+	u64 desc[];		/* unwind descriptors */
 	/* personality routine and language-specific data follow behind descriptors */
 };
 

[PATCH] treewide: Replace zero-length array with flexible-array

[Gustavo A. R. Silva]

The current codebase makes use of the zero-length array language
extension to the C90 standard, 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.

Also, notice that, dynamic memory allocations won't be affected by
this change:

"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]

sizeof(flexible-array-member) triggers a warning because flexible array
members have incomplete type[1]. There are some instances of code in
which the sizeof operator is being incorrectly/erroneously applied to
zero-length arrays and the result is zero. Such instances may be hiding
some bugs. So, this work (flexible-array member conversions) will also
help to get completely rid of those sorts of issues.

This issue was found with the help of Coccinelle.

[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@kernel.org>
---
 drivers/gnss/serial.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gnss/serial.h b/drivers/gnss/serial.h
index 980ffdc86c2a..621953f7821d 100644
--- a/drivers/gnss/serial.h
+++ b/drivers/gnss/serial.h
@@ -16,7 +16,7 @@ struct gnss_serial {
 	struct gnss_device *gdev;
 	speed_t	speed;
 	const struct gnss_serial_ops *ops;
-	unsigned long drvdata[0];
+	unsigned long drvdata[];
 };
 
 enum gnss_serial_pm_state {

[PATCH] treewide: Replace zero-length array with flexible-array

[Gustavo A. R. Silva]

The current codebase makes use of the zero-length array language
extension to the C90 standard, 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.

Also, notice that, dynamic memory allocations won't be affected by
this change:

"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]

sizeof(flexible-array-member) triggers a warning because flexible array
members have incomplete type[1]. There are some instances of code in
which the sizeof operator is being incorrectly/erroneously applied to
zero-length arrays and the result is zero. Such instances may be hiding
some bugs. So, this work (flexible-array member conversions) will also
help to get completely rid of those sorts of issues.

This issue was found with the help of Coccinelle.

[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@kernel.org>
---
 include/linux/fsl/bestcomm/bestcomm.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/fsl/bestcomm/bestcomm.h b/include/linux/fsl/bestcomm/bestcomm.h
index a0e2e6b19b57..154e541ce57e 100644
--- a/include/linux/fsl/bestcomm/bestcomm.h
+++ b/include/linux/fsl/bestcomm/bestcomm.h
@@ -27,7 +27,7 @@
  */
 struct bcom_bd {
 	u32	status;
-	u32	data[0];	/* variable payload size */
+	u32	data[];	/* variable payload size */
 };
 
 /* ======================================================================== */