From: Weihang Li <liweihang@hisilicon.com>
To: <jgg@ziepe.ca>, <leon@kernel.org>
Cc: <dledford@redhat.com>, <linux-rdma@vger.kernel.org>,
<linuxarm@huawei.com>
Subject: [PATCH rdma-core 5/7] libhns: Bugfix for updating qp params
Date: Thu, 21 Nov 2019 09:19:27 +0800 [thread overview]
Message-ID: <1574299169-31457-6-git-send-email-liweihang@hisilicon.com> (raw)
In-Reply-To: <1574299169-31457-1-git-send-email-liweihang@hisilicon.com>
From: Lijun Ou <oulijun@huawei.com>
In hns_roce_u_create_qp(), it's necessary to update rq params after
ibv_cmd_create_qp() running successfully, and then we will use it when
posting work request.
Fixes: b6cd213b276f ("libhns: Refactor for creating qp")
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Weihang Li <liweihang@hisilicon.com>
---
providers/hns/hns_roce_u_verbs.c | 16 +++++++++-------
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/providers/hns/hns_roce_u_verbs.c b/providers/hns/hns_roce_u_verbs.c
index 0acfd9a..a9c78f8 100644
--- a/providers/hns/hns_roce_u_verbs.c
+++ b/providers/hns/hns_roce_u_verbs.c
@@ -771,19 +771,14 @@ static void hns_roce_set_qp_params(struct ibv_pd *pd,
}
/* limit by the context queried during alloc context */
- qp->rq.max_post = min(ctx->max_qp_wr, qp->rq.wqe_cnt);
qp->sq.max_post = min(ctx->max_qp_wr, qp->sq.wqe_cnt);
qp->sq.max_gs = min(ctx->max_sge, qp->sq.max_gs);
- qp->rq.max_gs = min(ctx->max_sge, qp->rq.max_gs);
qp->sq_signal_bits = attr->sq_sig_all ? 0 : 1;
- qp->max_inline_data = HNS_ROCE_MAX_INLINE_DATA_LEN;
+ qp->max_inline_data = HNS_ROCE_MAX_INLINE_DATA_LEN;
/* update attr for creating qp */
attr->cap.max_send_wr = qp->sq.max_post;
- attr->cap.max_recv_wr = qp->rq.max_post;
- attr->cap.max_send_sge = qp->sq.max_gs;
- attr->cap.max_recv_sge = qp->rq.max_gs;
attr->cap.max_inline_data = qp->max_inline_data;
}
@@ -906,7 +901,14 @@ struct ibv_qp *hns_roce_u_create_qp(struct ibv_pd *pd,
}
pthread_mutex_unlock(&context->qp_table_mutex);
- qp->flags = resp.cap_flags;
+ /* adjust rq maxima to not exceed reported device maxima */
+ attr->cap.max_recv_wr = min(context->max_qp_wr, attr->cap.max_recv_wr);
+ attr->cap.max_recv_sge = min(context->max_sge, attr->cap.max_recv_sge);
+ qp->rq.wqe_cnt = attr->cap.max_recv_wr;
+ qp->rq.max_gs = attr->cap.max_recv_sge;
+ qp->rq.max_post = attr->cap.max_recv_wr;
+
+ qp->flags = resp.cap_flags;
return &qp->ibv_qp;
--
2.8.1
next prev parent reply other threads:[~2019-11-21 1:23 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-21 1:19 [PATCH rdma-core 0/7] libhns: Bugfix for hip08 Weihang Li
2019-11-21 1:19 ` [PATCH rdma-core 1/7] libhns: Fix calculation errors with ilog32() Weihang Li
2019-11-22 2:58 ` Zengtao (B)
2019-11-22 6:16 ` Weihang Li
2019-11-22 18:09 ` Jason Gunthorpe
2019-11-23 2:43 ` Weihang Li
2019-11-21 1:19 ` [PATCH rdma-core 2/7] libhns: Optimize bind_mw for fixing null pointer access Weihang Li
2019-11-22 3:02 ` Zengtao (B)
2019-11-22 6:40 ` Weihang Li
2019-11-21 1:19 ` [PATCH rdma-core 3/7] libhns: Bugfix for assigning sl Weihang Li
2019-11-21 1:19 ` [PATCH rdma-core 4/7] libhns: Bugfix for cleaning cq Weihang Li
2019-11-21 1:19 ` Weihang Li [this message]
2019-11-21 1:19 ` [PATCH rdma-core 6/7] libhns: Avoid null pointer operation Weihang Li
2019-11-21 1:19 ` [PATCH rdma-core 7/7] libhns: Return correct value of cqe num when flushing cqe failed Weihang Li
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1574299169-31457-6-git-send-email-liweihang@hisilicon.com \
--to=liweihang@hisilicon.com \
--cc=dledford@redhat.com \
--cc=jgg@ziepe.ca \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=linuxarm@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).