From: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
To: dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Mike Marciniszyn
<mike.marciniszyn-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Subject: [PATCH v2 05/20] IB/rdmavt, IB/hfi1, IB/qib: Make wc opcode translation driver dependent
Date: Mon, 20 Mar 2017 17:25:04 -0700 [thread overview]
Message-ID: <20170321002503.28538.44884.stgit@scvm10.sc.intel.com> (raw)
In-Reply-To: <20170321001900.28538.38175.stgit-9QXIwq+3FY+1XWohqUldA0EOCMrvLtNR@public.gmane.org>
From: Mike Marciniszyn <mike.marciniszyn-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
The work to create a completion helper moved the translation of send
wqe operations to completion opcodes to rdmvat.
This precludes having driver dependent operations. Make the translation
driver dependent by doing the translation in the driver prior to the
rvt_qp_swqe_complete() call using restored translation tables.
Fixes: Commit f2dc9cdce83c ("IB/rdmavt: Add a send completion helper")
Fixes: Commit 0771da5a6e9d ("IB/hfi1,IB/qib: Use new send completion helper")
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
drivers/infiniband/hw/hfi1/rc.c | 10 ++++++++--
drivers/infiniband/hw/hfi1/ruc.c | 5 ++++-
drivers/infiniband/hw/hfi1/verbs.c | 16 ++++++++++++++++
drivers/infiniband/hw/qib/qib_rc.c | 10 ++++++++--
drivers/infiniband/hw/qib/qib_ruc.c | 5 ++++-
drivers/infiniband/hw/qib/qib_verbs.c | 13 +++++++++++++
drivers/infiniband/sw/rdmavt/qp.c | 17 -----------------
include/rdma/rdmavt_qp.h | 3 ++-
8 files changed, 55 insertions(+), 24 deletions(-)
diff --git a/drivers/infiniband/hw/hfi1/rc.c b/drivers/infiniband/hw/hfi1/rc.c
index 7382be1..4649530 100644
--- a/drivers/infiniband/hw/hfi1/rc.c
+++ b/drivers/infiniband/hw/hfi1/rc.c
@@ -1034,7 +1034,10 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr)
/* see post_send() */
barrier();
rvt_put_swqe(wqe);
- rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS);
+ rvt_qp_swqe_complete(qp,
+ wqe,
+ ib_hfi1_wc_opcode[wqe->wr.opcode],
+ IB_WC_SUCCESS);
}
/*
* If we were waiting for sends to complete before re-sending,
@@ -1081,7 +1084,10 @@ static inline void update_last_psn(struct rvt_qp *qp, u32 psn)
qp->s_last = s_last;
/* see post_send() */
barrier();
- rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS);
+ rvt_qp_swqe_complete(qp,
+ wqe,
+ ib_hfi1_wc_opcode[wqe->wr.opcode],
+ IB_WC_SUCCESS);
} else {
struct hfi1_pportdata *ppd = ppd_from_ibp(ibp);
diff --git a/drivers/infiniband/hw/hfi1/ruc.c b/drivers/infiniband/hw/hfi1/ruc.c
index aa15bcb..d2eb793 100644
--- a/drivers/infiniband/hw/hfi1/ruc.c
+++ b/drivers/infiniband/hw/hfi1/ruc.c
@@ -920,7 +920,10 @@ void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe,
qp->ibqp.qp_type == IB_QPT_GSI)
atomic_dec(&ibah_to_rvtah(wqe->ud_wr.ah)->refcount);
- rvt_qp_swqe_complete(qp, wqe, status);
+ rvt_qp_swqe_complete(qp,
+ wqe,
+ ib_hfi1_wc_opcode[wqe->wr.opcode],
+ status);
if (qp->s_acked == old_last)
qp->s_acked = last;
diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c
index 222315f..815cb44 100644
--- a/drivers/infiniband/hw/hfi1/verbs.c
+++ b/drivers/infiniband/hw/hfi1/verbs.c
@@ -297,6 +297,22 @@ static inline bool wss_exceeds_threshold(void)
}
/*
+ * Translate ib_wr_opcode into ib_wc_opcode.
+ */
+const enum ib_wc_opcode ib_hfi1_wc_opcode[] = {
+ [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE,
+ [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE,
+ [IB_WR_SEND] = IB_WC_SEND,
+ [IB_WR_SEND_WITH_IMM] = IB_WC_SEND,
+ [IB_WR_RDMA_READ] = IB_WC_RDMA_READ,
+ [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP,
+ [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD,
+ [IB_WR_SEND_WITH_INV] = IB_WC_SEND,
+ [IB_WR_LOCAL_INV] = IB_WC_LOCAL_INV,
+ [IB_WR_REG_MR] = IB_WC_REG_MR
+};
+
+/*
* Length of header by opcode, 0 --> not supported
*/
const u8 hdr_len_by_opcode[256] = {
diff --git a/drivers/infiniband/hw/qib/qib_rc.c b/drivers/infiniband/hw/qib/qib_rc.c
index 12658e3..0234987 100644
--- a/drivers/infiniband/hw/qib/qib_rc.c
+++ b/drivers/infiniband/hw/qib/qib_rc.c
@@ -938,7 +938,10 @@ void qib_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr)
/* see post_send() */
barrier();
rvt_put_swqe(wqe);
- rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS);
+ rvt_qp_swqe_complete(qp,
+ wqe,
+ ib_qib_wc_opcode[wqe->wr.opcode],
+ IB_WC_SUCCESS);
}
/*
* If we were waiting for sends to complete before resending,
@@ -983,7 +986,10 @@ static inline void update_last_psn(struct rvt_qp *qp, u32 psn)
qp->s_last = s_last;
/* see post_send() */
barrier();
- rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS);
+ rvt_qp_swqe_complete(qp,
+ wqe,
+ ib_qib_wc_opcode[wqe->wr.opcode],
+ IB_WC_SUCCESS);
} else
this_cpu_inc(*ibp->rvp.rc_delayed_comp);
diff --git a/drivers/infiniband/hw/qib/qib_ruc.c b/drivers/infiniband/hw/qib/qib_ruc.c
index 17655cc..6e1adf7 100644
--- a/drivers/infiniband/hw/qib/qib_ruc.c
+++ b/drivers/infiniband/hw/qib/qib_ruc.c
@@ -769,7 +769,10 @@ void qib_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe,
qp->ibqp.qp_type == IB_QPT_GSI)
atomic_dec(&ibah_to_rvtah(wqe->ud_wr.ah)->refcount);
- rvt_qp_swqe_complete(qp, wqe, status);
+ rvt_qp_swqe_complete(qp,
+ wqe,
+ ib_qib_wc_opcode[wqe->wr.opcode],
+ status);
if (qp->s_acked == old_last)
qp->s_acked = last;
diff --git a/drivers/infiniband/hw/qib/qib_verbs.c b/drivers/infiniband/hw/qib/qib_verbs.c
index 83f8b5f..e120efe 100644
--- a/drivers/infiniband/hw/qib/qib_verbs.c
+++ b/drivers/infiniband/hw/qib/qib_verbs.c
@@ -114,6 +114,19 @@
MODULE_PARM_DESC(disable_sma, "Disable the SMA");
/*
+ * Translate ib_wr_opcode into ib_wc_opcode.
+ */
+const enum ib_wc_opcode ib_qib_wc_opcode[] = {
+ [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE,
+ [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE,
+ [IB_WR_SEND] = IB_WC_SEND,
+ [IB_WR_SEND_WITH_IMM] = IB_WC_SEND,
+ [IB_WR_RDMA_READ] = IB_WC_RDMA_READ,
+ [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP,
+ [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD
+};
+
+/*
* System image GUID.
*/
__be64 ib_qib_sys_image_guid;
diff --git a/drivers/infiniband/sw/rdmavt/qp.c b/drivers/infiniband/sw/rdmavt/qp.c
index f5ad8d4..28fb724 100644
--- a/drivers/infiniband/sw/rdmavt/qp.c
+++ b/drivers/infiniband/sw/rdmavt/qp.c
@@ -117,23 +117,6 @@
};
EXPORT_SYMBOL(ib_rvt_state_ops);
-/*
- * Translate ib_wr_opcode into ib_wc_opcode.
- */
-const enum ib_wc_opcode ib_rvt_wc_opcode[] = {
- [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE,
- [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE,
- [IB_WR_SEND] = IB_WC_SEND,
- [IB_WR_SEND_WITH_IMM] = IB_WC_SEND,
- [IB_WR_RDMA_READ] = IB_WC_RDMA_READ,
- [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP,
- [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD,
- [IB_WR_SEND_WITH_INV] = IB_WC_SEND,
- [IB_WR_LOCAL_INV] = IB_WC_LOCAL_INV,
- [IB_WR_REG_MR] = IB_WC_REG_MR
-};
-EXPORT_SYMBOL(ib_rvt_wc_opcode);
-
static void get_map_page(struct rvt_qpn_table *qpt,
struct rvt_qpn_map *map,
gfp_t gfp)
diff --git a/include/rdma/rdmavt_qp.h b/include/rdma/rdmavt_qp.h
index f381639..3cdd9e2 100644
--- a/include/rdma/rdmavt_qp.h
+++ b/include/rdma/rdmavt_qp.h
@@ -574,6 +574,7 @@ static inline void rvt_qp_wqe_unreserve(
static inline void rvt_qp_swqe_complete(
struct rvt_qp *qp,
struct rvt_swqe *wqe,
+ enum ib_wc_opcode opcode,
enum ib_wc_status status)
{
if (unlikely(wqe->wr.send_flags & RVT_SEND_RESERVE_USED))
@@ -586,7 +587,7 @@ static inline void rvt_qp_swqe_complete(
memset(&wc, 0, sizeof(wc));
wc.wr_id = wqe->wr.wr_id;
wc.status = status;
- wc.opcode = ib_rvt_wc_opcode[wqe->wr.opcode];
+ wc.opcode = opcode;
wc.qp = &qp->ibqp;
wc.byte_len = wqe->length;
rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc,
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2017-03-21 0:25 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-21 0:24 [PATCH v2 00/20] IB/hfi1, qib, rdmavt: Another round of patches for 4.11 Dennis Dalessandro
2017-03-21 0:24 ` Dennis Dalessandro
[not found] ` <20170321001900.28538.38175.stgit-9QXIwq+3FY+1XWohqUldA0EOCMrvLtNR@public.gmane.org>
2017-03-21 0:24 ` [PATCH v2 01/20] IB/hfi1: Force logical link down Dennis Dalessandro
2017-03-21 0:24 ` [PATCH v2 02/20] IB/hfi1: Race hazard avoidance in user SDMA driver Dennis Dalessandro
2017-03-21 0:24 ` [PATCH v2 03/20] IB/hfi1: Cache registers during state change Dennis Dalessandro
2017-03-21 0:24 ` [PATCH v2 04/20] IB/hfi1: NULL pointer dereference when freeing rhashtable Dennis Dalessandro
2017-03-21 0:25 ` Dennis Dalessandro [this message]
2017-03-21 0:25 ` [PATCH v2 06/20] IB/rdmavt: Add additional fields to post send trace Dennis Dalessandro
2017-03-21 0:25 ` [PATCH v2 07/20] IB/rdmavt: Add tracing for cq entry and poll Dennis Dalessandro
2017-03-21 0:25 ` [PATCH v2 08/20] IB/rdmavt: Add swqe completion trace Dennis Dalessandro
2017-03-21 0:25 ` [PATCH v2 09/20] IB/hfi1: Check device id early during init Dennis Dalessandro
2017-03-21 0:25 ` [PATCH v2 10/20] IB/hfi1: Protect the global dev_cntr_names and port_cntr_names Dennis Dalessandro
2017-03-21 0:25 ` [PATCH v2 11/20] IB/hfi1: Check for QSFP presence before attempting reads Dennis Dalessandro
2017-03-21 0:25 ` [PATCH v2 12/20] IB/hfi1: Add a patch value to the firmware version string Dennis Dalessandro
2017-03-21 0:25 ` [PATCH v2 13/20] IB/rdmavt, IB/hfi1: Fix timer migration regressions Dennis Dalessandro
2017-03-21 0:26 ` [PATCH v2 14/20] IB/rdmavt: Avoid reseting wqe send_flags in unreserve Dennis Dalessandro
2017-03-21 0:26 ` [PATCH v2 15/20] IB/hfi1: Ensure VL index is within bounds Dennis Dalessandro
2017-03-21 0:26 ` [PATCH v2 16/20] IB/hfi1: Add receive fault injection feature Dennis Dalessandro
2017-03-21 0:26 ` [PATCH v2 17/20] IB/hfi1: Add transmit " Dennis Dalessandro
[not found] ` <20170321002619.28538.31428.stgit-9QXIwq+3FY+1XWohqUldA0EOCMrvLtNR@public.gmane.org>
2017-04-05 18:34 ` Doug Ledford
[not found] ` <1491417255.2923.5.camel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2017-04-05 18:49 ` Dennis Dalessandro
2017-03-21 0:26 ` [PATCH v2 18/20] IB/hfi1: Eliminate synchronize_rcu() in mr delete Dennis Dalessandro
2017-03-21 0:26 ` [PATCH v2 19/20] IB/rdmavt, IB/qib, IB/hfi1: Make percpu refcount optional for user MRs Dennis Dalessandro
[not found] ` <20170321002631.28538.2121.stgit-9QXIwq+3FY+1XWohqUldA0EOCMrvLtNR@public.gmane.org>
2017-04-05 18:38 ` Doug Ledford
[not found] ` <1491417489.2923.6.camel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2017-04-05 19:46 ` Dennis Dalessandro
[not found] ` <f008c532-340e-01f2-80e6-4bea74175e3e-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2017-04-05 19:51 ` Leon Romanovsky
[not found] ` <CALq1K=JsjSCiSBeZVe4kHQmjw7tznL36JcsamZTVGZ5RhBvZPw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-04-05 20:09 ` Marciniszyn, Mike
[not found] ` <32E1700B9017364D9B60AED9960492BC342EA858-RjuIdWtd+YbTXloPLtfHfbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2017-04-06 7:49 ` Leon Romanovsky
[not found] ` <20170406074955.GG2269-U/DQcQFIOTAAJjI8aNfphQ@public.gmane.org>
2017-04-06 11:45 ` Dennis Dalessandro
[not found] ` <8cdf2fbb-f2a9-0b4b-b144-397ee73d1569-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2017-04-06 12:37 ` Leon Romanovsky
[not found] ` <20170406123726.GH2269-U/DQcQFIOTAAJjI8aNfphQ@public.gmane.org>
2017-04-06 13:00 ` Dennis Dalessandro
[not found] ` <f1703866-9c5c-a30a-0d95-9f6a33cc4f75-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2017-04-06 13:33 ` Leon Romanovsky
2017-04-06 14:47 ` Marciniszyn, Mike
[not found] ` <32E1700B9017364D9B60AED9960492BC342EABD0-RjuIdWtd+YbTXloPLtfHfbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2017-04-06 17:13 ` Jason Gunthorpe
[not found] ` <20170406171354.GA19854-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-04-06 17:16 ` Marciniszyn, Mike
[not found] ` <32E1700B9017364D9B60AED9960492BC342EADEE-RjuIdWtd+YbTXloPLtfHfbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2017-04-06 17:44 ` Jason Gunthorpe
[not found] ` <20170406174438.GA20020-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-04-07 21:12 ` Marciniszyn, Mike
[not found] ` <32E1700B9017364D9B60AED9960492BC342EBA18-RjuIdWtd+YbTXloPLtfHfbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2017-04-07 22:06 ` Jason Gunthorpe
[not found] ` <20170407220618.GA29138-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-04-09 6:26 ` Leon Romanovsky
2017-04-05 19:47 ` Leon Romanovsky
2017-03-21 0:26 ` [PATCH v2 20/20] IB/core: If the MGID/MLID pair is not on the list return an error Dennis Dalessandro
2017-04-05 18:50 ` [PATCH v2 00/20] IB/hfi1, qib, rdmavt: Another round of patches for 4.11 Doug Ledford
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=20170321002503.28538.44884.stgit@scvm10.sc.intel.com \
--to=dennis.dalessandro-ral2jqcrhueavxtiumwx3w@public.gmane.org \
--cc=dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=mike.marciniszyn-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.