[openib-general] Re: User MAD support for cancel MAD

[Sean Hefty]

Michael S. Tsirkin wrote:

> Hello!
> Quoting r. shaharf (shaharf at voltaire.com) "RE: [openib-general] Re: User MAD support for cancel MAD":
> 
>>The interface issue is not so trivial. An IOCTL may do, but the problem
>>is the parameter of the IOCTL. The most straight forward way is to
>>specify a TID to cancel. This requires the usermode to avoid sending the
>>same mad until it is timeouts.
> 
> 
> And here I thought it is the kernel that selects the TID.
> No?

The MAD layer in the kernel assigns the only half of the TID.  This is 
done to ensure that TIDs are unique between clients.

Also, as a side note (and not a direct response), the current 
ib_cancel_mad API cancels MADs based on work request ID, not TID.

- Sean