cvs commit: src/sys/netinet tcp.h

[Sam Leffler]

Andre Oppermann wrote:
> Andrew Thompson wrote:
>> On Wed, May 23, 2007 at 07:16:21PM +0000, Andre Oppermann wrote:
>>> andre       2007-05-23 19:16:21 UTC
>>>
>>>   FreeBSD src repository
>>>
>>>   Modified files:
>>>     sys/netinet          tcp.h   Log:
>>>   Add CWR back into the PRINT_TH_FLAGS list as gcc42 doesn't complain
>>>   about \8 in a string anymore.
>>>     Revision  Changes    Path
>>>   1.39      +1 -1      src/sys/netinet/tcp.h
>>>
>>> - * $FreeBSD: /usr/local/www/cvsroot/FreeBSD/src/sys/netinet/tcp.h,v 
>>> 1.38 2007/05/18 19:58:37 andre Exp $
>>> + * $FreeBSD: /usr/local/www/cvsroot/FreeBSD/src/sys/netinet/tcp.h,v 
>>> 1.39 2007/05/23 19:16:21 andre Exp $
>>>   */
>>>  
>>>  #ifndef _NETINET_TCP_H_
>>> @@ -69,7 +69,7 @@ struct tcphdr {
>>>  #define    TH_ECE    0x40
>>>  #define    TH_CWR    0x80
>>>  #define    TH_FLAGS    
>>> (TH_FIN|TH_SYN|TH_RST|TH_PUSH|TH_ACK|TH_URG|TH_ECE|TH_CWR)
>>> -#define    PRINT_TH_FLAGS    "\20\1FIN\2SYN\3RST\4PUSH\5ACK\6URG\7ECE"
>>> +#define    PRINT_TH_FLAGS    
>>> "\20\1FIN\2SYN\3RST\4PUSH\5ACK\6URG\7ECE\8CWR"
>>
>> Shouldnt this be \10 instead of \8
> 
> man printf(9) doesn't say so.  However gcc42 doesn't accept it
> either, it just prints an error but continues.
> 
>  The arguments are made up of a sequence of bit identifiers.
>  Each bit identifier begins with an integer value which is
>  the number of the bit (starting from 1) this identifier
>  describes.  The rest of the identifier is a string of
>  characters containing the name of the bit.  The string is
>  terminated by either the bit number at the start of the next
>  bit identifier or NUL for the last bit identifier.
> 
> The whole situation is a bit confusing.
> 

The numbers are in octal.  Always have been.  You can probably use \x 
for hex instead if you want; I think that's c99.

	Sam