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]
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 <[email protected]>
---
drivers/pci/controller/pci-hyperv.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
index 9977abff92fc..be957268f9d6 100644
--- a/drivers/pci/controller/pci-hyperv.c
+++ b/drivers/pci/controller/pci-hyperv.c
@@ -260,7 +260,7 @@ struct pci_packet {
int resp_packet_size);
void *compl_ctxt;
- struct pci_message message[0];
+ struct pci_message message[];
};
/*
@@ -296,7 +296,7 @@ struct pci_bus_d0_entry {
struct pci_bus_relations {
struct pci_incoming_message incoming;
u32 device_count;
- struct pci_function_description func[0];
+ struct pci_function_description func[];
} __packed;
struct pci_q_res_req_response {
@@ -508,7 +508,7 @@ struct hv_dr_work {
struct hv_dr_state {
struct list_head list_entry;
u32 device_count;
- struct pci_function_description func[0];
+ struct pci_function_description func[];
};
enum hv_pcichild_state {
--
2.23.0
> From: [email protected]
> <[email protected]> On Behalf Of Gustavo A. R. Silva
> Sent: Wednesday, February 12, 2020 4:51 PM
> ...
> 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]
>
> This issue was found with the help of Coccinelle.
Looks good to me. Thanks, Gustavo!
Reviewed-by: Dexuan Cui <[email protected]>
FWIW, it looks there are a lot of more to fix in the kernel tree: the below
commands return 1373 for me:
grep -nr '\[0\];$' * | grep '\.h:' | grep -v = | wc -l
Running the commands against the kernel/ directory returns 3.
Thanks,
-- Dexuan
On Thu, Feb 13, 2020 at 03:43:40AM +0000, Dexuan Cui wrote:
> > From: [email protected]
> > <[email protected]> On Behalf Of Gustavo A. R. Silva
> > Sent: Wednesday, February 12, 2020 4:51 PM
> > ...
> > 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]
> >
> > This issue was found with the help of Coccinelle.
>
> Looks good to me. Thanks, Gustavo!
>
> Reviewed-by: Dexuan Cui <[email protected]>
>
Lorenzo, will you be picking up this patch? It seems to me you've been
handling patches to pci-hyperv.c. This patch is not yet in pci/hv branch
in your repository.
Let me know what you think.
Wei.
On Wed, Mar 04, 2020 at 05:55:09PM +0000, Wei Liu wrote:
> On Thu, Feb 13, 2020 at 03:43:40AM +0000, Dexuan Cui wrote:
> > > From: [email protected]
> > > <[email protected]> On Behalf Of Gustavo A. R. Silva
> > > Sent: Wednesday, February 12, 2020 4:51 PM
> > > ...
> > > 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]
> > >
> > > This issue was found with the help of Coccinelle.
> >
> > Looks good to me. Thanks, Gustavo!
> >
> > Reviewed-by: Dexuan Cui <[email protected]>
> >
>
> Lorenzo, will you be picking up this patch? It seems to me you've been
> handling patches to pci-hyperv.c. This patch is not yet in pci/hv branch
> in your repository.
>
> Let me know what you think.
I shall pick it up, I checked patchwork and it was erroneously
assigned to Bjorn, that's why I have not taken it yet.
Fixed now, apologies, I will merge it shortly.
Lorenzo
On Wed, Mar 04, 2020 at 06:06:35PM +0000, Lorenzo Pieralisi wrote:
> On Wed, Mar 04, 2020 at 05:55:09PM +0000, Wei Liu wrote:
> > On Thu, Feb 13, 2020 at 03:43:40AM +0000, Dexuan Cui wrote:
> > > > From: [email protected]
> > > > <[email protected]> On Behalf Of Gustavo A. R. Silva
> > > > Sent: Wednesday, February 12, 2020 4:51 PM
> > > > ...
> > > > 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]
> > > >
> > > > This issue was found with the help of Coccinelle.
> > >
> > > Looks good to me. Thanks, Gustavo!
> > >
> > > Reviewed-by: Dexuan Cui <[email protected]>
> > >
> >
> > Lorenzo, will you be picking up this patch? It seems to me you've been
> > handling patches to pci-hyperv.c. This patch is not yet in pci/hv branch
> > in your repository.
> >
> > Let me know what you think.
>
> I shall pick it up, I checked patchwork and it was erroneously
> assigned to Bjorn, that's why I have not taken it yet.
>
> Fixed now, apologies, I will merge it shortly.
Thanks for picking it up.
Wei.
>
> Lorenzo
On 3/4/20 12:10, Wei Liu wrote:
>>>>
>>>> Looks good to me. Thanks, Gustavo!
>>>>
>>>> Reviewed-by: Dexuan Cui <[email protected]>
>>>>
>>>
>>> Lorenzo, will you be picking up this patch? It seems to me you've been
>>> handling patches to pci-hyperv.c. This patch is not yet in pci/hv branch
>>> in your repository.
>>>
>>> Let me know what you think.
>>
>> I shall pick it up, I checked patchwork and it was erroneously
>> assigned to Bjorn, that's why I have not taken it yet.
>>
>> Fixed now, apologies, I will merge it shortly.
>
> Thanks for picking it up.
>
Thank you all, guys. :)
--
Gustavo