> This is the 3rd time in a row {(1) ipath (2) qib (3) OPA Gen1} for you guys to > implement IB transports in SW @ your low-level drivers. > > So... enough is enough, please put it in a (say) libibtrans kernel module > residing in the IB core and use it in this driver, to begin with. > I like this idea. Bear in mind ipath will go way from the tree entirely. The issue with this is the development effort would need to be incremental to preserve the functionality of both qib and hfi1. I would like to propose such an incremental approach with everyone's help, perhaps an IBTA object at a time as a follow-on. For namespace purposes, we pick and alternative to qib_/hfi1_, such as ibt or sib (soft ib). Mike ��.n��������+%������w��{.n�����{���fk��ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f