diff options
| author | Mike Marciniszyn <mike.marciniszyn@intel.com> | 2016-02-14 23:10:04 +0300 |
|---|---|---|
| committer | Doug Ledford <dledford@redhat.com> | 2016-03-11 04:38:07 +0300 |
| commit | 46a80d62e6e0ccfc9d8a05c5b773405b84a4afd7 (patch) | |
| tree | 077ea38ac2f7fd75c1334deadcc141ad6753a009 /drivers/infiniband/hw/qib/qib_ruc.c | |
| parent | 20f333b61300fa658952713ca9b8b4b72bbaed9f (diff) | |
| download | linux-46a80d62e6e0ccfc9d8a05c5b773405b84a4afd7.tar.xz | |
IB/qib, staging/rdma/hfi1: add s_hlock for use in post send
This patch adds an additional lock to reduce contention on the s_lock.
This lock is used in post_send() so that the post_send is not
serialized with the send engine and other send related processing.
To do this the s_next_psn is now maintained on post_send() while
post_send() related fields are moved to a new cache line. There is
an s_avail maintained for the post_send() to mitigate trading cache
lines with the send engine. The lock is released/acquired around
releasing the just built packet to the egress mechanism.
Reviewed-by: Jubin John <jubin.john@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Dean Luick <dean.luick@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Diffstat (limited to 'drivers/infiniband/hw/qib/qib_ruc.c')
| -rw-r--r-- | drivers/infiniband/hw/qib/qib_ruc.c | 11 |
1 files changed, 7 insertions, 4 deletions
diff --git a/drivers/infiniband/hw/qib/qib_ruc.c b/drivers/infiniband/hw/qib/qib_ruc.c index 2623684745f0..a5f07a64b228 100644 --- a/drivers/infiniband/hw/qib/qib_ruc.c +++ b/drivers/infiniband/hw/qib/qib_ruc.c @@ -391,7 +391,8 @@ static void qib_ruc_loopback(struct rvt_qp *sqp) sqp->s_flags |= RVT_S_BUSY; again: - if (sqp->s_last == sqp->s_head) + smp_read_barrier_depends(); /* see post_one_send() */ + if (sqp->s_last == ACCESS_ONCE(sqp->s_head)) goto clr_busy; wqe = rvt_get_swqe_ptr(sqp, sqp->s_last); @@ -765,22 +766,24 @@ void qib_do_send(struct rvt_qp *qp) qp->s_flags |= RVT_S_BUSY; - spin_unlock_irqrestore(&qp->s_lock, flags); - do { /* Check for a constructed packet to be sent. */ if (qp->s_hdrwords != 0) { + spin_unlock_irqrestore(&qp->s_lock, flags); /* * If the packet cannot be sent now, return and * the send tasklet will be woken up later. */ if (qib_verbs_send(qp, priv->s_hdr, qp->s_hdrwords, qp->s_cur_sge, qp->s_cur_size)) - break; + return; /* Record that s_hdr is empty. */ qp->s_hdrwords = 0; + spin_lock_irqsave(&qp->s_lock, flags); } } while (make_req(qp)); + + spin_unlock_irqrestore(&qp->s_lock, flags); } /* |