[ofa-general] Re: [PATCH 2 of 3] IB/mthca: support multiple completion vectors

Roland Dreier rdreier at cisco.com
Thu May 3 11:37:43 PDT 2007


 > I don't know how many vectors make sense, so I decided to be
 > conservative here, since each EQ consumes a lot of memory by default.

I think #vectors == O(#CPUs) is what we should aim for.

Also another useful thing for NUMA systems might be to allocate the EQ
memory from the CPU where that interrupt will be targeted.  But I
don't know exactly the best way to do that, and I think we can leave
that for later.

 - R.



More information about the general mailing list