| Message ID | 1713459125-14914-4-git-send-email-kotaranov@linux.microsoft.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | RDMA/mana_ib: Implement RNIC CQs | expand |
> Subject: [PATCH rdma-next 3/6] RDMA/mana_ib: replace duplicate cqe with > buf_size I don't understand this commit message on "duplicate" cqe. I couldn't find a duplicate of it in the existing code.
> From: Long Li <longli@microsoft.com> > Sent: Wednesday, 24 April 2024 01:35 > To: Konstantin Taranov <kotaranov@linux.microsoft.com>; Konstantin > Taranov <kotaranov@microsoft.com>; sharmaajay@microsoft.com; > jgg@ziepe.ca; leon@kernel.org > Cc: linux-rdma@vger.kernel.org; linux-kernel@vger.kernel.org > Subject: RE: [PATCH rdma-next 3/6] RDMA/mana_ib: replace duplicate cqe > with buf_size > > > Subject: [PATCH rdma-next 3/6] RDMA/mana_ib: replace duplicate cqe > > with buf_size > > I don't understand this commit message on "duplicate" cqe. I couldn't find a > duplicate of it in the existing code. If we need cqe, we could use it at cq->ibcq.cqe. The patch does not assign it as it is not used, but if you want I can add "cq->ibcq.cqe = attr->cqe;" in v2. - Konstantin
> Subject: RE: [PATCH rdma-next 3/6] RDMA/mana_ib: replace duplicate cqe with > buf_size > > > From: Long Li <longli@microsoft.com> > > Sent: Wednesday, 24 April 2024 01:35 > > To: Konstantin Taranov <kotaranov@linux.microsoft.com>; Konstantin > > Taranov <kotaranov@microsoft.com>; sharmaajay@microsoft.com; > > jgg@ziepe.ca; leon@kernel.org > > Cc: linux-rdma@vger.kernel.org; linux-kernel@vger.kernel.org > > Subject: RE: [PATCH rdma-next 3/6] RDMA/mana_ib: replace duplicate cqe > > with buf_size > > > > > Subject: [PATCH rdma-next 3/6] RDMA/mana_ib: replace duplicate cqe > > > with buf_size > > > > I don't understand this commit message on "duplicate" cqe. I couldn't > > find a duplicate of it in the existing code. > > If we need cqe, we could use it at cq->ibcq.cqe. The patch does not assign it as it > is not used, but if you want I can add "cq->ibcq.cqe = attr->cqe;" in v2. > > - Konstantin I see. We don't need buf_size because it can be computed from cq->ibcq.cqe? The commit message is confusing enough to make people think cqe is a duplicate of buf_size. Long
diff --git a/drivers/infiniband/hw/mana/cq.c b/drivers/infiniband/hw/mana/cq.c index dc931b9..0467ee8 100644 --- a/drivers/infiniband/hw/mana/cq.c +++ b/drivers/infiniband/hw/mana/cq.c @@ -33,8 +33,8 @@ int mana_ib_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr, return -EINVAL; } - cq->cqe = attr->cqe; - err = mana_ib_create_queue(mdev, ucmd.buf_addr, cq->cqe * COMP_ENTRY_SIZE, &cq->queue); + cq->buf_size = attr->cqe * COMP_ENTRY_SIZE; + err = mana_ib_create_queue(mdev, ucmd.buf_addr, cq->buf_size, &cq->queue); if (err) { ibdev_dbg(ibdev, "Failed to create queue for create cq, %d\n", err); return err; diff --git a/drivers/infiniband/hw/mana/mana_ib.h b/drivers/infiniband/hw/mana/mana_ib.h index 9162f29..9c07021 100644 --- a/drivers/infiniband/hw/mana/mana_ib.h +++ b/drivers/infiniband/hw/mana/mana_ib.h @@ -90,7 +90,7 @@ struct mana_ib_mr { struct mana_ib_cq { struct ib_cq ibcq; struct mana_ib_queue queue; - int cqe; + u32 buf_size; u32 comp_vector; mana_handle_t cq_handle; }; diff --git a/drivers/infiniband/hw/mana/qp.c b/drivers/infiniband/hw/mana/qp.c index 280e85a..c4fb8b4 100644 --- a/drivers/infiniband/hw/mana/qp.c +++ b/drivers/infiniband/hw/mana/qp.c @@ -196,7 +196,7 @@ static int mana_ib_create_qp_rss(struct ib_qp *ibqp, struct ib_pd *pd, wq_spec.queue_size = wq->wq_buf_size; cq_spec.gdma_region = cq->queue.gdma_region; - cq_spec.queue_size = cq->cqe * COMP_ENTRY_SIZE; + cq_spec.queue_size = cq->buf_size; cq_spec.modr_ctx_id = 0; eq = &mpc->ac->eqs[cq->comp_vector]; cq_spec.attached_eq = eq->eq->id; @@ -355,7 +355,7 @@ static int mana_ib_create_qp_raw(struct ib_qp *ibqp, struct ib_pd *ibpd, wq_spec.queue_size = ucmd.sq_buf_size; cq_spec.gdma_region = send_cq->queue.gdma_region; - cq_spec.queue_size = send_cq->cqe * COMP_ENTRY_SIZE; + cq_spec.queue_size = send_cq->buf_size; cq_spec.modr_ctx_id = 0; eq_vec = send_cq->comp_vector; eq = &mpc->ac->eqs[eq_vec];