2018-03-22 16:33:10

by Sinan Kaya

Subject: [PATCH v6] RDMA/qedr: eliminate duplicate barriers on weakly-ordered archs #2

Code includes wmb() followed by writel() in multiple places. writel()
already has a barrier on some architectures like arm64.

This ends up CPU observing two barriers back to back before executing the
register write.

Since code already has an explicit barrier call, changing writel() to
writel_relaxed().

Signed-off-by: Sinan Kaya <[email protected]>
---
drivers/infiniband/hw/qedr/verbs.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
index 53f00db..d1cf9a0 100644
--- a/drivers/infiniband/hw/qedr/verbs.c
+++ b/drivers/infiniband/hw/qedr/verbs.c
@@ -35,6 +35,7 @@
#include <net/ipv6.h>
#include <net/udp.h>
#include <linux/iommu.h>
+#include <linux/io-64-nonatomic-hi-lo.h>

#include <rdma/ib_verbs.h>
#include <rdma/ib_user_verbs.h>
@@ -860,7 +861,7 @@ static void doorbell_cq(struct qedr_cq *cq, u32 cons, u8 flags)
wmb();
cq->db.data.agg_flags = flags;
cq->db.data.value = cpu_to_le32(cons);
- writeq(cq->db.raw, cq->db_addr);
+ writeq_relaxed(cq->db.raw, cq->db_addr);

/* Make sure write would stick */
mmiowb();
@@ -3338,7 +3339,7 @@ int qedr_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,

qp->rq.db_data.data.value++;

- writel(qp->rq.db_data.raw, qp->rq.db);
+ writel_relaxed(qp->rq.db_data.raw, qp->rq.db);

/* Make sure write sticks */
mmiowb();
--
2.7.4

2018-03-23 19:14:59

by Sinan Kaya

Subject: Re: [PATCH v6] RDMA/qedr: eliminate duplicate barriers on weakly-ordered archs #2

On 3/22/2018 12:30 PM, Sinan Kaya wrote:
> Code includes wmb() followed by writel() in multiple places. writel()
> already has a barrier on some architectures like arm64.
>
> This ends up CPU observing two barriers back to back before executing the
> register write.
>
> Since code already has an explicit barrier call, changing writel() to
> writel_relaxed().
>
> Signed-off-by: Sinan Kaya <[email protected]>
> ---
> drivers/infiniband/hw/qedr/verbs.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
> index 53f00db..d1cf9a0 100644
> --- a/drivers/infiniband/hw/qedr/verbs.c
> +++ b/drivers/infiniband/hw/qedr/verbs.c
> @@ -35,6 +35,7 @@
> #include <net/ipv6.h>
> #include <net/udp.h>
> #include <linux/iommu.h>
> +#include <linux/io-64-nonatomic-hi-lo.h>

what should we do here? change include? drop writeq_relaxed?

>
> #include <rdma/ib_verbs.h>
> #include <rdma/ib_user_verbs.h>
> @@ -860,7 +861,7 @@ static void doorbell_cq(struct qedr_cq *cq, u32 cons, u8 flags)
> wmb();
> cq->db.data.agg_flags = flags;
> cq->db.data.value = cpu_to_le32(cons);
> - writeq(cq->db.raw, cq->db_addr);
> + writeq_relaxed(cq->db.raw, cq->db_addr);
>
> /* Make sure write would stick */
> mmiowb();
> @@ -3338,7 +3339,7 @@ int qedr_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
>
> qp->rq.db_data.data.value++;
>
> - writel(qp->rq.db_data.raw, qp->rq.db);
> + writel_relaxed(qp->rq.db_data.raw, qp->rq.db);
>
> /* Make sure write sticks */
> mmiowb();
>


--
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.