> -----Original Message-----
> From: ext Maxim Uvarov [mailto:[email protected]]
> Sent: Monday, September 08, 2014 2:13 PM
> To: Savolainen, Petri (NSN - FI/Espoo); Anders Roxell
> Cc: [email protected]
> Subject: Re: [lng-odp] [PATCH 2/2] odph_icmp: add ODPH_ prefix
> 
> On 09/08/2014 12:10 PM, Savolainen, Petri (NSN - FI/Espoo) wrote:
> >
> >> -----Original Message-----
> >> From: [email protected] [mailto:lng-odp-
> >> [email protected]] On Behalf Of ext Maxim Uvarov
> >> Sent: Saturday, September 06, 2014 11:56 PM
> >> To: Anders Roxell
> >> Cc: [email protected]
> >> Subject: Re: [lng-odp] [PATCH 2/2] odph_icmp: add ODPH_ prefix
> >>
> >> On 09/04/2014 11:48 PM, Anders Roxell wrote:
> >>> Feels dangerous, that means that one application with two code
> branches
> >>> can use the same define they think... however, that can be defined
> with
> >>> different values right?
> >> Not. If they will define to different values then you will see
> >> compilation error.
> >> Gcc will say that you are trying to redefine value.
> >>
> >>> If not, then we can drop this patch, but then we need to remove ODPH_
> >>> from all the defines that we have in ODP right?
> >> Need to walk over them. If it's standard system values, then we don't
> >> need to redefine them.
> >> We should be very careful with changes due to changing this things
> >> require changes in all odp
> >> applications.
> >>
> > All ODP helper definitions should have ODPH_ or odph_ prefix. Those are
> only helpers - there's no guarantees (== limited SW definitions and
> maintenance) to support a full featured protocol implementation. Such an
> implementation would be another project then.
> >
> > Also what are standard system values (definitions)? Defined in
> POSIX/Linux/Unix/RTOS X/Vendor SDK Y/IETF RFC Z?
> >
> > -Petri
> 
> It does not matter where they defined. Important thing that they do not
> change, like:
> 
> -#define ICMP_TIME_EXCEEDED   11      /**< Time Exceeded              */
> -#define ICMP_PARAMETERPROB   12      /**< Parameter Problem          */
> -#define ICMP_TIMESTAMP               13      /**< Timestamp Request          
> */
> -#define ICMP_TIMESTAMPREPLY  14      /**< Timestamp Reply            */
> -#define ICMP_INFO_REQUEST    15      /**< Information Request        */
> 
> Do you now where this values are different?
> 

How user can be sure that e.g. ICMP_TIMESTAMP is defined only for ICMP 
timestamp request operation code in system XYZ (which may not be Linux) and not 
for something else? With ODPH_ we can ensure that. Application is not portable 
if it does not build (due to name space clash).

-Petri


> Thanks,
> Maxim.
> 
> 
> >
> >> Maxim.
> >>
> >>> Cheers,
> >>> Anders
> >>
> >> _______________________________________________
> >> lng-odp mailing list
> >> [email protected]
> >> http://lists.linaro.org/mailman/listinfo/lng-odp


_______________________________________________
lng-odp mailing list
[email protected]
http://lists.linaro.org/mailman/listinfo/lng-odp

Reply via email to