Message ID | 20230703123053.3117488-1-arnd@kernel.org (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | usb: functionfs: avoid memcpy() field overflow warning | expand |
On Mon, Jul 03, 2023 at 02:30:32PM +0200, Arnd Bergmann wrote: > From: Arnd Bergmann <arnd@arndb.de> > > __ffs_func_bind_do_os_desc() copies both the CompatibleID and SubCompatibleID > fields of the usb_ext_compat_desc structure into an array, which triggers > a warning in the fortified memcpy(): > > In file included from drivers/usb/gadget/function/f_fs.c:17: > In file included from include/linux/string.h:254: > include/linux/fortify-string.h:592:4: error: call to '__read_overflow2_field' declared with 'warning' attribute: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Werror,-Wattribute-warning] > __read_overflow2_field(q_size_field, size); > > Usually we can avoid this by using a struct_group() inside of the structure > definition, but this might cause problems in userspace since it is in a uapi > header. We use this in other uapi .h files, what is unique about these fields that makes it so that they can not be used? Because it's not the last field? thanks, greg k-h
On Mon, Jul 3, 2023, at 14:45, Greg Kroah-Hartman wrote: > On Mon, Jul 03, 2023 at 02:30:32PM +0200, Arnd Bergmann wrote: >> From: Arnd Bergmann <arnd@arndb.de> >> >> __ffs_func_bind_do_os_desc() copies both the CompatibleID and SubCompatibleID >> fields of the usb_ext_compat_desc structure into an array, which triggers >> a warning in the fortified memcpy(): >> >> In file included from drivers/usb/gadget/function/f_fs.c:17: >> In file included from include/linux/string.h:254: >> include/linux/fortify-string.h:592:4: error: call to '__read_overflow2_field' declared with 'warning' attribute: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Werror,-Wattribute-warning] >> __read_overflow2_field(q_size_field, size); >> >> Usually we can avoid this by using a struct_group() inside of the structure >> definition, but this might cause problems in userspace since it is in a uapi >> header. > > We use this in other uapi .h files, what is unique about these fields > that makes it so that they can not be used? Because it's not the last > field? It's probably ok, and I was overly cautious. I'll send a new version after some more testing. Arnd
diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index f41a385a5c421..b8f9e52e6db6b 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -2933,8 +2933,9 @@ static int __ffs_func_bind_do_os_desc(enum ffs_os_desc_type type, t = &func->function.os_desc_table[desc->bFirstInterfaceNumber]; t->if_id = func->interfaces_nums[desc->bFirstInterfaceNumber]; memcpy(t->os_desc->ext_compat_id, &desc->CompatibleID, - ARRAY_SIZE(desc->CompatibleID) + - ARRAY_SIZE(desc->SubCompatibleID)); + sizeof(desc->CompatibleID)); + memcpy(t->os_desc->ext_compat_id + sizeof(desc->CompatibleID), + &desc->SubCompatibleID, sizeof(desc->SubCompatibleID)); length = sizeof(*desc); } break;