On 11/28/2018 2:02 PM, Ruhl, Michael J wrote:
b/drivers/infiniband/hw/qib/qib_ud.c
index 4d4c31e..8dab667 100644
--- a/drivers/infiniband/hw/qib/qib_ud.c
+++ b/drivers/infiniband/hw/qib/qib_ud.c
@@ -172,12 +172,8 @@ static void qib_ud_loopback(struct rvt_qp *sqp, struct
rvt_swqe *swqe)
ssge.num_sge = swqe->wr.num_sge;
sge = &ssge.sge;
while (length) {
- u32 len = sge->length;
+ u32 len = rvt_get_sge_length(sge, length);
- if (len > length)
- len = length;
- if (len > sge->sge_length)
- len = sge->sge_length;
BUG_ON(len == 0);
^^^^^^, above you are using WARN_ON_ONCE and below BUG_ON
I assume that everything can be safely removed.
I was focused mainly on patch footprint rather than other
issues and have not focused on the need for the BUG_ON().
I will add this to my list to follow up on this with a different patch.
We have a handful of BUG_ON calls in qib code. I'm ambivalent towards
them. Don't think it's worth anyone's time to mess with since they've
been there for 10+ years. On the other hand I would happily Ack a patch
that removes them.
-Denny
