Re: Linux 2.6.34-rc3 + CAN build problem

From: Oliver Hartkopp
Date: Sat Apr 10 2010 - 08:36:55 EST


NÃmeth MÃrton wrote:
> Eric Dumazet Ãrta:
>> Le samedi 10 avril 2010 Ã 10:13 +0200, NÃmeth MÃrton a Ãcrit :
>>> Hi,
>>>
>>> I have some problem building Liunux kernel 2.6.34-rc3 with the attached .config:
>>>
>>> $ make clean bzImage modules
>>> [...]
>>> CC net/socket.o
>>> LD net/802/built-in.o
>>> LD net/can/built-in.o
>>> CC [M] net/can/bcm.o
>>> CC [M] net/can/raw.o
>>> In file included from /mnt/store/nmarci/src/linux-2.6.34-rc3/arch/x86/include/asm/uaccess.h:571,
>>> from include/net/checksum.h:25,
>>> from include/linux/skbuff.h:28,
>>> from include/linux/if_ether.h:124,
>>> from include/linux/netdevice.h:29,
>>> from net/can/raw.c:48:
>>> In function âcopy_from_userâ,
>>> inlined from âraw_setsockoptâ at net/can/raw.c:447:
>>> /mnt/store/nmarci/src/linux-2.6.34-rc3/arch/x86/include/asm/uaccess_32.h:212: error: call to âcopy_from_user_overflowâ declared with attribute error:
>>> copy_from_user() buffer size is not provably correct
>>> make[2]: *** [net/can/raw.o] Error 1
>>> make[1]: *** [net/can] Error 2
>>> make: *** [net] Error 2
>>>
>>>
>> Could you give us your compiler version ?
>
> $ gcc --version
> gcc (Debian 4.4.2-9) 4.4.3 20100108 (prerelease)
> Copyright (C) 2009 Free Software Foundation, Inc.
> This is free software; see the source for copying conditions. There is NO
> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
>
>> Code is fine, but compiler a bit dumb :(
>>
>> [PATCH] can: avoids a false warning
>>
>> At this point optlen == sizeof(sfilter) but some compilers are dumb.
>>
>> Reported-by: NÃmeth MÃrton <nm127@xxxxxxxxxx
>> Signed-off-by: Eric Dumazet <eric.dumazet@xxxxxxxxx>

Acked-by: Oliver Hartkopp <oliver@xxxxxxxxxxxx>

Btw. i'm using the same compiler here (Debian Squeeze), and i did not have any
build problems with 2.6.24-rc3 and the following kernels like the currently
running 2.6.34-rc3-00288-gab195c5:

$ cat /proc/version
Linux version 2.6.34-rc3-00288-gab195c5 (hartko@vwagwolkf320) (gcc version
4.4.3 20100108 (prerelease) (Debian 4.4.2-9) ) #70 SMP Tue Apr 6 19:14:52 CEST
2010

So i wonder why Nemeth trapped into this problem ... probably an include file
mix-up?

Regards,
Oliver


>> ---
>> diff --git a/net/can/raw.c b/net/can/raw.c
>> index 3a7dffb..da99cf1 100644
>> --- a/net/can/raw.c
>> +++ b/net/can/raw.c
>> @@ -445,7 +445,7 @@ static int raw_setsockopt(struct socket *sock, int level, int optname,
>> return -EFAULT;
>> }
>> } else if (count == 1) {
>> - if (copy_from_user(&sfilter, optval, optlen))
>> + if (copy_from_user(&sfilter, optval, sizeof(sfilter)))
>> return -EFAULT;
>> }
>>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/