> -----Original Message----- > From: Dennis Dalessandro [mailto:dennis.dalessandro@xxxxxxxxx] > Sent: Friday, January 26, 2018 9:59 AM > To: Jason Gunthorpe <jgg@xxxxxxxx> > Cc: dledford@xxxxxxxxxx; linux-rdma@xxxxxxxxxxxxxxx; Don Hiatt > <don.hiatt@xxxxxxxxx>; Sean Hefty <sean.hefty@xxxxxxxxx>; Ira Weiny > <ira.weiny@xxxxxxxxx>; Parav Pandit <parav@xxxxxxxxxxxx> > Subject: Re: [PATCH for-next 15/16] IB/core: Map iWarp AH type to undefined in > rdma_ah_find_type > > On 1/26/2018 10:45 AM, Jason Gunthorpe wrote: > > On Fri, Jan 26, 2018 at 06:33:18AM -0800, Dennis Dalessandro wrote: > >> From: Don Hiatt <don.hiatt@xxxxxxxxx> > >> > >> iWarp devices do not support the creation of address handles so > >> return AH_ATTR_TYPE_UNDEFINED for all iWarp devices. > >> > >> While we are here reduce the size of port_num to u8 and add a > >> comment. > >> > >> Fixes: 44c58487d51a ("IB/core: Define 'ib' and 'roce' rdma_ah_attr > >> types") > >> Reported-by: Parav Pandit <parav@xxxxxxxxxxxx> > >> CC: Sean Hefty <sean.hefty@xxxxxxxxx> > >> Reviewed-by: Ira Weiny <ira.weiny@xxxxxxxxx> > >> Signed-off-by: Don Hiatt <don.hiatt@xxxxxxxxx> > >> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@xxxxxxxxx> > >> include/rdma/ib_verbs.h | 15 +++++++++++---- > >> 1 files changed, 11 insertions(+), 4 deletions(-) > > > > What tree are you sending these patches again? This won't apply. > > Ugh, I thought I was up to date. Guess not. Let me rebase and respin. > Looks like you had a comment on one of the other patches so I'll wait for Alex to > address that and re-send. > Also can we have UNDEFINED AH enum as 0 rather than 3? This will allow to catch errors in the code that will decode 0 initialized field as IB? This also ensures that right AH type is initialized wherever needed. > -Denny ��.n��������+%������w��{.n�����{���fk��ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f
