On Mon, 2011-11-07 at 20:00 +0100, Bart Van Assche wrote: > On Mon, Nov 7, 2011 at 7:10 PM, Roland Dreier <[email protected]> wrote: > > On Sun, Nov 6, 2011 at 9:31 AM, Bart Van Assche <[email protected]> wrote: > >> Make sure that the last WQE event is handled properly if it arrives > >> before the QP has been transitioned into the error state. > > > > Do you mean if the event arrives before the modify QP operation > > returns? If so I agree this is a possible order of events... > > > > Otherwise I'm not sure I follow: the last WQE event can't be > > generated until the QP is in the error state (which of course > > can happen before ib_modify_qp() returns) > > The patch should be fine, but the description is not entirely correct: > it should have been "... if it arrives before ib_modify_qp() has been > invoked to transition the QP in the error state". Invoking > ib_modify_qp() is not the only way via which a QP can reach the error > state - transport errors can cause this too. >
Hi Bart, This patch fails to compile with the following: drivers/infiniband/ulp/srpt/ib_srpt.c: In function ‘srpt_perform_rdmas’: drivers/infiniband/ulp/srpt/ib_srpt.c:2936: error: ‘CH_RELEASING’ undeclared (first use in this function) drivers/infiniband/ulp/srpt/ib_srpt.c:2936: error: (Each undeclared identifier is reported only once drivers/infiniband/ulp/srpt/ib_srpt.c:2936: error: for each function it appears in.) I'm assuming this needs to be changed to CH_DRAINING instead, yes..? Thanks, --nab -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
