On Tue, Apr 14, 2009 at 10:20 AM, Sasha Khapyorsky <[email protected]> wrote:
> On 10:12 Tue 14 Apr     , Hal Rosenstock wrote:
>> >
>> > Do you know any example where this will not work properly? I don't.
>>
>> Not currently but it's a spec compliance issue.
>
> I don't think we want to slow down fast path without a real needs.

Isn't compliance a real need ?

>> > The error status is printed originally, isn't it not sufficient?
>>
>> IMO it's not sufficient as redirection is not an error.
>
> It is handled as error in this patch, the only message is different.
>
>> > If not
>> > what about adding generic function which prints know error codes as
>> > string instead of handling some values separately?
>>
>> That's doable but a separate proposition IMO.
>
> This is better than doing duplicated status parsing flows.

Sure but still separate IMO.

-- Hal

> Sasha
_______________________________________________
general mailing list
[email protected]
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to