2014-07-31 20:26:07

by Steve Wise

Subject: [PATCH] svcrdma: remove rdma_create_qp() failure recovery logic

In svc_rdma_accept(), if rdma_create_qp() fails, there is useless
logic to try and call rdma_create_qp() again with reduced sge depths.
The assumption, I guess, was that perhaps the initial sge depths
chosen were too big. However they initial depths are selected based
on the rdma device attribute max_sge returned from ib_query_device().
If rdma_create_qp() fails, it would not be because the max_send_sge and
max_recv_sge values passed in exceed the device's max. So just remove
this code.

Signed-off-by: Steve Wise <[email protected]>
---

net/sunrpc/xprtrdma/svc_rdma_transport.c | 19 ++-----------------
1 files changed, 2 insertions(+), 17 deletions(-)

diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index e7323fb..282a43b 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -942,23 +942,8 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt)

ret = rdma_create_qp(newxprt->sc_cm_id, newxprt->sc_pd, &qp_attr);
if (ret) {
- /*
- * XXX: This is a hack. We need a xx_request_qp interface
- * that will adjust the qp_attr's with a best-effort
- * number
- */
- qp_attr.cap.max_send_sge -= 2;
- qp_attr.cap.max_recv_sge -= 2;
- ret = rdma_create_qp(newxprt->sc_cm_id, newxprt->sc_pd,
- &qp_attr);
- if (ret) {
- dprintk("svcrdma: failed to create QP, ret=%d\n", ret);
- goto errout;
- }
- newxprt->sc_max_sge = qp_attr.cap.max_send_sge;
- newxprt->sc_max_sge = qp_attr.cap.max_recv_sge;
- newxprt->sc_sq_depth = qp_attr.cap.max_send_wr;
- newxprt->sc_max_requests = qp_attr.cap.max_recv_wr;
+ dprintk("svcrdma: failed to create QP, ret=%d\n", ret);
+ goto errout;
}
newxprt->sc_qp = newxprt->sc_cm_id->qp;

2014-08-01 14:14:55

by Steve Wise

Subject: Re: [PATCH] svcrdma: remove rdma_create_qp() failure recovery logic

Tested over cxgb4 and mlx4.

Steve.

2014-08-05 19:10:19

by J. Bruce Fields

Subject: Re: [PATCH] svcrdma: remove rdma_create_qp() failure recovery logic

Thanks, applying for 3.17.--b.

On Thu, Jul 31, 2014 at 03:26:07PM -0500, Steve Wise wrote:
> In svc_rdma_accept(), if rdma_create_qp() fails, there is useless
> logic to try and call rdma_create_qp() again with reduced sge depths.
> The assumption, I guess, was that perhaps the initial sge depths
> chosen were too big. However they initial depths are selected based
> on the rdma device attribute max_sge returned from ib_query_device().
> If rdma_create_qp() fails, it would not be because the max_send_sge and
> max_recv_sge values passed in exceed the device's max. So just remove
> this code.
>
> Signed-off-by: Steve Wise <[email protected]>
> ---
>
> net/sunrpc/xprtrdma/svc_rdma_transport.c | 19 ++-----------------
> 1 files changed, 2 insertions(+), 17 deletions(-)
>
> diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
> index e7323fb..282a43b 100644
> --- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
> +++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
> @@ -942,23 +942,8 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt)
>
> ret = rdma_create_qp(newxprt->sc_cm_id, newxprt->sc_pd, &qp_attr);
> if (ret) {
> - /*
> - * XXX: This is a hack. We need a xx_request_qp interface
> - * that will adjust the qp_attr's with a best-effort
> - * number
> - */
> - qp_attr.cap.max_send_sge -= 2;
> - qp_attr.cap.max_recv_sge -= 2;
> - ret = rdma_create_qp(newxprt->sc_cm_id, newxprt->sc_pd,
> - &qp_attr);
> - if (ret) {
> - dprintk("svcrdma: failed to create QP, ret=%d\n", ret);
> - goto errout;
> - }
> - newxprt->sc_max_sge = qp_attr.cap.max_send_sge;
> - newxprt->sc_max_sge = qp_attr.cap.max_recv_sge;
> - newxprt->sc_sq_depth = qp_attr.cap.max_send_wr;
> - newxprt->sc_max_requests = qp_attr.cap.max_recv_wr;
> + dprintk("svcrdma: failed to create QP, ret=%d\n", ret);
> + goto errout;
> }
> newxprt->sc_qp = newxprt->sc_cm_id->qp;
>
>