Roland Dreier wrote:
Looks OK... probably not worth checking
ClassPortInfo:CapabilityMask.PortCountersXmitWaitSupported to make sure
this field is defined, although it is unfortunate that the IB spec says
that PortXmitWait is undefined rather than 0 when it isn't supported.
Anyway, one question:
> static PORT_PMA_ATTR(port_xmit_packets , 14, 32, 256);
> static PORT_PMA_ATTR(port_rcv_packets , 15, 32, 288);
> +/*
> + * There is no bit allocated for port_xmit_wait in the CounterSelect field
> + * (IB spec). However, since this bit is ignored when reading
> + * (show_pma_counter), the _counter field of port_xmit_wait can be set to
zero.
> + */
> +static PORT_PMA_ATTR(port_xmit_wait , 0, 32, 320);
I actually can't find any place where we look at the _counter field that
is passed into PORT_PMA_ATTR(), and this codfe was written so long ago
that I can't remember what reason (if any) we had for including it. Do
you know if there's any reason why not to just delete the whole _counter
thing entirely?
The _counter field is ignored by show_pma_counter, it will be relevant for
set_pma_counter, if we are going to add it.
Regards,
Vladimir
_______________________________________________
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general