| Message ID | 20200225012008.GA4309@embeddedor (mailing list archive) |
|---|---|
| State | Changes Requested |
| Delegated to: | Kalle Valo |
| Headers | show |
| Series | [next] prism54: Replace zero-length array with flexible-array member | expand |
On Mon, Feb 24, 2020 at 07:20:08PM -0600, 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] > > 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 <gustavo@embeddedor.com> I'd rather we just remove this driver completely, as it has a replacement upstream p54, and remained upstream just for a theoretical period of time someone was not able to use p54 anymore. I'll follow up with a removal of the driver. Luis
(changing the subject) Luis Chamberlain <mcgrof@kernel.org> writes: > On Mon, Feb 24, 2020 at 07:20:08PM -0600, 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] >> >> 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 <gustavo@embeddedor.com> > > I'd rather we just remove this driver completely, as it has a > replacement upstream p54, and remained upstream just for a theoretical > period of time someone was not able to use p54 anymore. I'll follow up > with a removal of the driver. Yeah, please do. I wonder if we should do other spring cleanup as well and remove drivers like ray_cs and wl3501, I have not seen any activity on those for years. Also rndis_wlan would be other candidate for removal. Anyone know if these drivers are used or if they even work anymore?
diff --git a/drivers/net/wireless/intersil/prism54/isl_oid.h b/drivers/net/wireless/intersil/prism54/isl_oid.h index 5441c1f9f2fc..1afc2ccf94ca 100644 --- a/drivers/net/wireless/intersil/prism54/isl_oid.h +++ b/drivers/net/wireless/intersil/prism54/isl_oid.h @@ -37,7 +37,7 @@ struct obj_mlmeex { u16 state; u16 code; u16 size; - u8 data[0]; + u8 data[]; } __packed; struct obj_buffer { @@ -68,12 +68,12 @@ struct obj_bss { struct obj_bsslist { u32 nr; - struct obj_bss bsslist[0]; + struct obj_bss bsslist[]; } __packed; struct obj_frequencies { u16 nr; - u16 mhz[0]; + u16 mhz[]; } __packed; struct obj_attachment { @@ -81,7 +81,7 @@ struct obj_attachment { char reserved; short id; short size; - char data[0]; + char data[]; } __packed; /* diff --git a/drivers/net/wireless/intersil/prism54/islpci_mgt.h b/drivers/net/wireless/intersil/prism54/islpci_mgt.h index d6bbbac46b4a..1f87d0aea60c 100644 --- a/drivers/net/wireless/intersil/prism54/islpci_mgt.h +++ b/drivers/net/wireless/intersil/prism54/islpci_mgt.h @@ -99,7 +99,7 @@ struct islpci_mgmtframe { pimfor_header_t *header; /* payload header, points into buf */ void *data; /* payload ex header, points into buf */ struct work_struct ws; /* argument for schedule_work() */ - char buf[0]; /* fragment buffer */ + char buf[]; /* fragment buffer */ }; int
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 <gustavo@embeddedor.com> --- drivers/net/wireless/intersil/prism54/isl_oid.h | 8 ++++---- drivers/net/wireless/intersil/prism54/islpci_mgt.h | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-)