[ofa-general] max inline data query ?

[Dotan Barak]

Hi.

If one wishes to use maximum inline data, he should open a QP with 
maximum number of s/g entries in the QP->SQ
(ibv_qp_init_attr.cap.max_send_sge). The actual number of bytes which 
can be send as inline will be returned in the
attribute ibv_qp_init_attr.cap.max_inline_data after ibv_create_qp will 
be finished.

The inline size effects the size of memory which is being used (and 
pinned) for the QP buffers/queues
(per WR in the QP->SQ).

I'm sorry, but i don't really know what is exactly the size which is 
being consumed for a QP
(I remember that in the past, Roland mentioned the exact calculation in 
one of the emails that he sent
to this mailing list).

thanks
Dotan

Tang, Changqing wrote:
> Does the inline size causes QP to use more pinned memory ? Is there any document to tell
> how a QP consumes physical memory, and what the relationship with inline data size is?
>
> Thanks.
>
> --CQ
>
>   
>> -----Original Message-----
>> From: general-bounces at lists.openfabrics.org
>> [mailto:general-bounces at lists.openfabrics.org] On Behalf Of
>> Arlin Davis
>> Sent: Monday, March 10, 2008 2:55 PM
>> To: OpenFabrics General; Roland Dreier; Sean Hefty
>> Subject: [ofa-general] max inline data query ?
>>
>> Sorry, I know that we had this discussion before but could
>> someone refresh my memory. Are there plans to add a query for
>> max inline data size?
>>
>> Right now I have no other mechanism but to call
>> ibv_create_qp, increasing inline size for a given sge count,
>> until it fails.
>>
>> Thanks,
>>
>> -arlin
>> _______________________________________________
>> general mailing list
>> general at lists.openfabrics.org
>> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
>>
>> To unsubscribe, please visit
>> http://openib.org/mailman/listinfo/openib-general
>>
>>     
> _______________________________________________
> general mailing list
> general at lists.openfabrics.org
> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
>
> To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
>
>