isert: convert to the new CQ API and a few random cleanups V2

isert: convert to the new CQ API and a few random cleanups V2

[Christoph Hellwig]

This series uses the new CQ API and cleans up a few random lose arounds
around the R/W path.  I have a few more changes for the RW path in the
works, but I think this is a good enough standalone series for now.


Changes since V1:

 - sort out pre-existing confusion about the login buffer sizes
 - add various little inline helpers for container_of constructs
 - assign the ->done handlers only once
 - updates commit messages


--
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

[PATCH 2/5] IB/isert: Split and properly type the login buffer

[Christoph Hellwig]

The login receive buffer is used as a iser_rx_desc, so type it as such
in struct isert_conn and allocate the exactly right space for it.  The
TX buffer is moved to a separate variable and properly sized as well.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 drivers/infiniband/ulp/isert/ib_isert.c | 56 ++++++++++++++++-----------------
 drivers/infiniband/ulp/isert/ib_isert.h |  3 +-
 2 files changed, 29 insertions(+), 30 deletions(-)

diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index 4dbb4a9..cd74195 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -597,10 +597,12 @@ isert_free_login_buf(struct isert_conn *isert_conn)
 
 	ib_dma_unmap_single(ib_dev, isert_conn->login_rsp_dma,
 			    ISER_RX_PAYLOAD_SIZE, DMA_TO_DEVICE);
+	kfree(isert_conn->login_rsp_buf);
+
 	ib_dma_unmap_single(ib_dev, isert_conn->login_req_dma,
-			    ISCSI_DEF_MAX_RECV_SEG_LEN,
+			    ISER_RX_PAYLOAD_SIZE,
 			    DMA_FROM_DEVICE);
-	kfree(isert_conn->login_buf);
+	kfree(isert_conn->login_req_buf);
 }
 
 static int
@@ -609,50 +611,48 @@ isert_alloc_login_buf(struct isert_conn *isert_conn,
 {
 	int ret;
 
-	isert_conn->login_buf = kzalloc(ISCSI_DEF_MAX_RECV_SEG_LEN +
-					ISER_RX_PAYLOAD_SIZE, GFP_KERNEL);
-	if (!isert_conn->login_buf) {
+	isert_conn->login_req_buf = kzalloc(sizeof(*isert_conn->login_req_buf),
+			GFP_KERNEL);
+	if (!isert_conn->login_req_buf) {
 		isert_err("Unable to allocate isert_conn->login_buf\n");
 		return -ENOMEM;
 	}
 
-	isert_conn->login_req_buf = isert_conn->login_buf;
-	isert_conn->login_rsp_buf = isert_conn->login_buf +
-				    ISCSI_DEF_MAX_RECV_SEG_LEN;
-
-	isert_dbg("Set login_buf: %p login_req_buf: %p login_rsp_buf: %p\n",
-		 isert_conn->login_buf, isert_conn->login_req_buf,
-		 isert_conn->login_rsp_buf);
-
 	isert_conn->login_req_dma = ib_dma_map_single(ib_dev,
-				(void *)isert_conn->login_req_buf,
-				ISCSI_DEF_MAX_RECV_SEG_LEN, DMA_FROM_DEVICE);
-
+				isert_conn->login_req_buf,
+				ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE);
 	ret = ib_dma_mapping_error(ib_dev, isert_conn->login_req_dma);
 	if (ret) {
 		isert_err("login_req_dma mapping error: %d\n", ret);
 		isert_conn->login_req_dma = 0;
-		goto out_login_buf;
+		goto out_free_login_req_buf;
+	}
+
+	isert_conn->login_rsp_buf = kzalloc(ISER_RX_PAYLOAD_SIZE, GFP_KERNEL);
+	if (!isert_conn->login_rsp_buf) {
+		isert_err("Unable to allocate isert_conn->login_rspbuf\n");
+		goto out_unmap_login_req_buf;
 	}
 
 	isert_conn->login_rsp_dma = ib_dma_map_single(ib_dev,
-					(void *)isert_conn->login_rsp_buf,
+					isert_conn->login_rsp_buf,
 					ISER_RX_PAYLOAD_SIZE, DMA_TO_DEVICE);
-
 	ret = ib_dma_mapping_error(ib_dev, isert_conn->login_rsp_dma);
 	if (ret) {
 		isert_err("login_rsp_dma mapping error: %d\n", ret);
 		isert_conn->login_rsp_dma = 0;
-		goto out_req_dma_map;
+		goto out_free_login_rsp_buf;
 	}
 
 	return 0;
 
-out_req_dma_map:
+out_free_login_rsp_buf:
+	kfree(isert_conn->login_rsp_buf);
+out_unmap_login_req_buf:
 	ib_dma_unmap_single(ib_dev, isert_conn->login_req_dma,
-			    ISCSI_DEF_MAX_RECV_SEG_LEN, DMA_FROM_DEVICE);
-out_login_buf:
-	kfree(isert_conn->login_buf);
+			    ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE);
+out_free_login_req_buf:
+	kfree(isert_conn->login_req_buf);
 	return ret;
 }
 
@@ -773,7 +773,7 @@ isert_connect_release(struct isert_conn *isert_conn)
 		ib_destroy_qp(isert_conn->qp);
 	}
 
-	if (isert_conn->login_buf)
+	if (isert_conn->login_req_buf)
 		isert_free_login_buf(isert_conn);
 
 	isert_device_put(device);
@@ -1218,7 +1218,7 @@ post_send:
 static void
 isert_rx_login_req(struct isert_conn *isert_conn)
 {
-	struct iser_rx_desc *rx_desc = (void *)isert_conn->login_req_buf;
+	struct iser_rx_desc *rx_desc = isert_conn->login_req_buf;
 	int rx_buflen = isert_conn->login_req_len;
 	struct iscsi_conn *conn = isert_conn->conn;
 	struct iscsi_login *login = conn->conn_login;
@@ -1596,7 +1596,7 @@ isert_rcv_completion(struct iser_rx_desc *desc,
 	u64 rx_dma;
 	int rx_buflen;
 
-	if ((char *)desc == isert_conn->login_req_buf) {
+	if (desc == isert_conn->login_req_buf) {
 		rx_dma = isert_conn->login_req_dma;
 		rx_buflen = ISER_RX_PAYLOAD_SIZE;
 		isert_dbg("login_buf: Using rx_dma: 0x%llx, rx_buflen: %d\n",
@@ -1615,7 +1615,7 @@ isert_rcv_completion(struct iser_rx_desc *desc,
 		 hdr->opcode, hdr->itt, hdr->flags,
 		 (int)(xfer_len - ISER_HEADERS_LEN));
 
-	if ((char *)desc == isert_conn->login_req_buf) {
+	if (desc == isert_conn->login_req_buf) {
 		isert_conn->login_req_len = xfer_len - ISER_HEADERS_LEN;
 		if (isert_conn->conn) {
 			struct iscsi_login *login = isert_conn->conn->conn_login;
diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h
index bd5e9b6..1240cf8 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.h
+++ b/drivers/infiniband/ulp/isert/ib_isert.h
@@ -182,8 +182,7 @@ struct isert_conn {
 	u32			initiator_depth;
 	bool			pi_support;
 	u32			max_sge;
-	char			*login_buf;
-	char			*login_req_buf;
+	struct iser_rx_desc	*login_req_buf;
 	char			*login_rsp_buf;
 	u64			login_req_dma;
 	int			login_req_len;
-- 
2.1.4

Re: [PATCH 2/5] IB/isert: Split and properly type the login buffer

[Sagi Grimberg]

> The login receive buffer is used as a iser_rx_desc, so type it as such
> in struct isert_conn and allocate the exactly right space for it.  The
> TX buffer is moved to a separate variable and properly sized as well.

As you said, we have room for cleanups in this area, but this looks
good.

Acked-by: Sagi Grimberg <sagig-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
--
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

Re: [PATCH 2/5] IB/isert: Split and properly type the login buffer

[Max Gurtovoy]

On 2/22/2016 3:49 PM, Christoph Hellwig wrote:
> The login receive buffer is used as a iser_rx_desc, so type it as such
> in struct isert_conn and allocate the exactly right space for it.  The
> TX buffer is moved to a separate variable and properly sized as well.

looks good.

Reviewed-by: Max Gurtovoy <maxg@mellanox.com>

[PATCH 1/5] IB/isert: Remove ISER_RECV_DATA_SEG_LEN

[Christoph Hellwig]

This is the same as ISCSI_DEF_MAX_RECV_SEG_LEN (and must be the same given
the structure layouts), so just use that constant instead.  This also
allows removing ISER_RX_LOGIN_SIZE in favor of ISER_RX_PAYLOAD_SIZE.

Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
 drivers/infiniband/ulp/isert/ib_isert.c | 10 +++++-----
 drivers/infiniband/ulp/isert/ib_isert.h |  8 +++-----
 2 files changed, 8 insertions(+), 10 deletions(-)

diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index f121e61..4dbb4a9 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -596,7 +596,7 @@ isert_free_login_buf(struct isert_conn *isert_conn)
 	struct ib_device *ib_dev = isert_conn->device->ib_device;
 
 	ib_dma_unmap_single(ib_dev, isert_conn->login_rsp_dma,
-			    ISER_RX_LOGIN_SIZE, DMA_TO_DEVICE);
+			    ISER_RX_PAYLOAD_SIZE, DMA_TO_DEVICE);
 	ib_dma_unmap_single(ib_dev, isert_conn->login_req_dma,
 			    ISCSI_DEF_MAX_RECV_SEG_LEN,
 			    DMA_FROM_DEVICE);
@@ -610,7 +610,7 @@ isert_alloc_login_buf(struct isert_conn *isert_conn,
 	int ret;
 
 	isert_conn->login_buf = kzalloc(ISCSI_DEF_MAX_RECV_SEG_LEN +
-					ISER_RX_LOGIN_SIZE, GFP_KERNEL);
+					ISER_RX_PAYLOAD_SIZE, GFP_KERNEL);
 	if (!isert_conn->login_buf) {
 		isert_err("Unable to allocate isert_conn->login_buf\n");
 		return -ENOMEM;
@@ -637,7 +637,7 @@ isert_alloc_login_buf(struct isert_conn *isert_conn,
 
 	isert_conn->login_rsp_dma = ib_dma_map_single(ib_dev,
 					(void *)isert_conn->login_rsp_buf,
-					ISER_RX_LOGIN_SIZE, DMA_TO_DEVICE);
+					ISER_RX_PAYLOAD_SIZE, DMA_TO_DEVICE);
 
 	ret = ib_dma_mapping_error(ib_dev, isert_conn->login_rsp_dma);
 	if (ret) {
@@ -1121,7 +1121,7 @@ isert_rdma_post_recvl(struct isert_conn *isert_conn)
 
 	memset(&sge, 0, sizeof(struct ib_sge));
 	sge.addr = isert_conn->login_req_dma;
-	sge.length = ISER_RX_LOGIN_SIZE;
+	sge.length = ISER_RX_PAYLOAD_SIZE;
 	sge.lkey = isert_conn->device->pd->local_dma_lkey;
 
 	isert_dbg("Setup sge: addr: %llx length: %d 0x%08x\n",
@@ -1598,7 +1598,7 @@ isert_rcv_completion(struct iser_rx_desc *desc,
 
 	if ((char *)desc == isert_conn->login_req_buf) {
 		rx_dma = isert_conn->login_req_dma;
-		rx_buflen = ISER_RX_LOGIN_SIZE;
+		rx_buflen = ISER_RX_PAYLOAD_SIZE;
 		isert_dbg("login_buf: Using rx_dma: 0x%llx, rx_buflen: %d\n",
 			 rx_dma, rx_buflen);
 	} else {
diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h
index 8d50453..bd5e9b6 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.h
+++ b/drivers/infiniband/ulp/isert/ib_isert.h
@@ -36,9 +36,7 @@
 /* Constant PDU lengths calculations */
 #define ISER_HEADERS_LEN	(sizeof(struct iser_ctrl) + \
 				 sizeof(struct iscsi_hdr))
-#define ISER_RECV_DATA_SEG_LEN	8192
-#define ISER_RX_PAYLOAD_SIZE	(ISER_HEADERS_LEN + ISER_RECV_DATA_SEG_LEN)
-#define ISER_RX_LOGIN_SIZE	(ISER_HEADERS_LEN + ISCSI_DEF_MAX_RECV_SEG_LEN)
+#define ISER_RX_PAYLOAD_SIZE	(ISER_HEADERS_LEN + ISCSI_DEF_MAX_RECV_SEG_LEN)
 
 /* QP settings */
 /* Maximal bounds on received asynchronous PDUs */
@@ -62,7 +60,7 @@
 				ISERT_MAX_TX_MISC_PDUS	+ \
 				ISERT_MAX_RX_MISC_PDUS)
 
-#define ISER_RX_PAD_SIZE	(ISER_RECV_DATA_SEG_LEN + 4096 - \
+#define ISER_RX_PAD_SIZE	(ISCSI_DEF_MAX_RECV_SEG_LEN + 4096 - \
 		(ISER_RX_PAYLOAD_SIZE + sizeof(u64) + sizeof(struct ib_sge)))
 
 #define ISCSI_ISER_SG_TABLESIZE		256
@@ -92,7 +90,7 @@ enum iser_conn_state {
 struct iser_rx_desc {
 	struct iser_ctrl iser_header;
 	struct iscsi_hdr iscsi_header;
-	char		data[ISER_RECV_DATA_SEG_LEN];
+	char		data[ISCSI_DEF_MAX_RECV_SEG_LEN];
 	u64		dma_addr;
 	struct ib_sge	rx_sg;
 	char		pad[ISER_RX_PAD_SIZE];
-- 
2.1.4

--
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

Re: [PATCH 1/5] IB/isert: Remove ISER_RECV_DATA_SEG_LEN

[Or Gerlitz]

On Mon, Feb 22, 2016 at 3:49 PM, Christoph Hellwig <hch@lst.de> wrote:

> @@ -36,9 +36,7 @@
> -#define ISER_RECV_DATA_SEG_LEN 8192

Sagi,

Can you please explain WW (When & Why) we changed this from 128 bytes
to 8192 bytes?!! do we use 8000++ bytes for every RX descriptor or
this is nowadays only for the login phase? what's the trick?

Or.

> @@ -92,7 +90,7 @@ enum iser_conn_state {
>  struct iser_rx_desc {
>         struct iser_ctrl iser_header;
>         struct iscsi_hdr iscsi_header;
> -       char            data[ISER_RECV_DATA_SEG_LEN];
> +       char            data[ISCSI_DEF_MAX_RECV_SEG_LEN];
>         u64             dma_addr;
>         struct ib_sge   rx_sg;
>         char            pad[ISER_RX_PAD_SIZE];
> --

Re: [PATCH 1/5] IB/isert: Remove ISER_RECV_DATA_SEG_LEN

[Or Gerlitz]

On Mon, Feb 22, 2016 at 4:55 PM, Or Gerlitz <gerlitz.or-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> On Mon, Feb 22, 2016 at 3:49 PM, Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org> wrote:
>
>> @@ -36,9 +36,7 @@
>> -#define ISER_RECV_DATA_SEG_LEN 8192
>
> Sagi,
>
> Can you please explain WW (When & Why) we changed this from 128 bytes
> to 8192 bytes?!! do we use 8000++ bytes for every RX descriptor or
> this is nowadays only for the login phase? what's the trick?

Oh, I see now that this is isert and not iser... if this descriptor is
used for every IO and set is such hard-coded way, that is terribly
inefficient, NM for now I guess, as this patchset doesn't introduce
this,

>> @@ -92,7 +90,7 @@ enum iser_conn_state {
>>  struct iser_rx_desc {
>>         struct iser_ctrl iser_header;
>>         struct iscsi_hdr iscsi_header;
>> -       char            data[ISER_RECV_DATA_SEG_LEN];
>> +       char            data[ISCSI_DEF_MAX_RECV_SEG_LEN];
>>         u64             dma_addr;
>>         struct ib_sge   rx_sg;
>>         char            pad[ISER_RX_PAD_SIZE];

BTW that pad is probably just c&p from the initiator, has nothing to do here
--
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

Re: [PATCH 1/5] IB/isert: Remove ISER_RECV_DATA_SEG_LEN

[Christoph Hellwig]

On Mon, Feb 22, 2016 at 04:57:50PM +0200, Or Gerlitz wrote:
> >> @@ -92,7 +90,7 @@ enum iser_conn_state {
> >>  struct iser_rx_desc {
> >>         struct iser_ctrl iser_header;
> >>         struct iscsi_hdr iscsi_header;
> >> -       char            data[ISER_RECV_DATA_SEG_LEN];
> >> +       char            data[ISCSI_DEF_MAX_RECV_SEG_LEN];
> >>         u64             dma_addr;
> >>         struct ib_sge   rx_sg;
> >>         char            pad[ISER_RX_PAD_SIZE];
> 
> BTW that pad is probably just c&p from the initiator, has nothing to do here

The way the RX descriptors work is rather odd in many ways. It's on
my todo list to investigate and rework them.

Re: [PATCH 1/5] IB/isert: Remove ISER_RECV_DATA_SEG_LEN

[Sagi Grimberg]

>>>> @@ -92,7 +90,7 @@ enum iser_conn_state {
>>>>   struct iser_rx_desc {
>>>>          struct iser_ctrl iser_header;
>>>>          struct iscsi_hdr iscsi_header;
>>>> -       char            data[ISER_RECV_DATA_SEG_LEN];
>>>> +       char            data[ISCSI_DEF_MAX_RECV_SEG_LEN];
>>>>          u64             dma_addr;
>>>>          struct ib_sge   rx_sg;
>>>>          char            pad[ISER_RX_PAD_SIZE];
>>
>> BTW that pad is probably just c&p from the initiator, has nothing to do here
>
> The way the RX descriptors work is rather odd in many ways. It's on
> my todo list to investigate and rework them.

Hey Christoph and Or,

Indeed the way RX descriptors are handled in isert has a lot of room
for cleanups (among other things...), its on my todo list as well. I
didn't have a lot of time to spend on isert lately but I do plan to
to resume the work there soon enough.

Christoph,
I took your patches for testing and evaluation (on top of some fixes
from Jenny I have in the pipe), basic functionality seems to work just
fine, I'll make sure to run some more tests and review the code.

Thanks!
--
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

Re: [PATCH 1/5] IB/isert: Remove ISER_RECV_DATA_SEG_LEN

[Sagi Grimberg]

Looks good,

Acked-by: Sagi Grimberg <sagig-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
--
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

Re: [PATCH 1/5] IB/isert: Remove ISER_RECV_DATA_SEG_LEN

[Max Gurtovoy]

I agree.

On 2/23/2016 1:03 PM, Sagi Grimberg wrote:
> Looks good,
>
> Acked-by: Sagi Grimberg <sagig-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
--
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

[PATCH 3/5] IB/isert: Convert to new CQ API

[Christoph Hellwig]

Use the workqueue based CQ type similar to what isert was using previously,
and properly split up the completion handlers.

Note that this also takes special care to handle the magic login WRs
separately, and also renames the submission functions so that it's clear
that they are only to be used for the login buffers.

Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
 drivers/infiniband/ulp/isert/ib_isert.c | 407 +++++++++++++-------------------
 drivers/infiniband/ulp/isert/ib_isert.h |  22 +-
 2 files changed, 182 insertions(+), 247 deletions(-)

diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index cd74195..1824692 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -59,12 +59,16 @@ isert_reg_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 static int
 isert_put_response(struct iscsi_conn *conn, struct iscsi_cmd *cmd);
 static int
-isert_rdma_post_recvl(struct isert_conn *isert_conn);
+isert_login_post_recv(struct isert_conn *isert_conn);
 static int
 isert_rdma_accept(struct isert_conn *isert_conn);
 struct rdma_cm_id *isert_setup_id(struct isert_np *isert_np);
 
 static void isert_release_work(struct work_struct *work);
+static void isert_recv_done(struct ib_cq *cq, struct ib_wc *wc);
+static void isert_send_done(struct ib_cq *cq, struct ib_wc *wc);
+static void isert_login_recv_done(struct ib_cq *cq, struct ib_wc *wc);
+static void isert_login_send_done(struct ib_cq *cq, struct ib_wc *wc);
 
 static inline bool
 isert_prot_cmd(struct isert_conn *conn, struct se_cmd *cmd)
@@ -177,12 +181,6 @@ err:
 	return ret;
 }
 
-static void
-isert_cq_event_callback(struct ib_event *e, void *context)
-{
-	isert_dbg("event: %d\n", e->event);
-}
-
 static int
 isert_alloc_rx_descriptors(struct isert_conn *isert_conn)
 {
@@ -212,6 +210,7 @@ isert_alloc_rx_descriptors(struct isert_conn *isert_conn)
 		rx_sg->addr = rx_desc->dma_addr;
 		rx_sg->length = ISER_RX_PAYLOAD_SIZE;
 		rx_sg->lkey = device->pd->local_dma_lkey;
+		rx_desc->rx_cqe.done = isert_recv_done;
 	}
 
 	return 0;
@@ -250,9 +249,6 @@ isert_free_rx_descriptors(struct isert_conn *isert_conn)
 	isert_conn->rx_descs = NULL;
 }
 
-static void isert_cq_work(struct work_struct *);
-static void isert_cq_callback(struct ib_cq *, void *);
-
 static void
 isert_free_comps(struct isert_device *device)
 {
@@ -261,10 +257,8 @@ isert_free_comps(struct isert_device *device)
 	for (i = 0; i < device->comps_used; i++) {
 		struct isert_comp *comp = &device->comps[i];
 
-		if (comp->cq) {
-			cancel_work_sync(&comp->work);
-			ib_destroy_cq(comp->cq);
-		}
+		if (comp->cq)
+			ib_free_cq(comp->cq);
 	}
 	kfree(device->comps);
 }
@@ -293,28 +287,17 @@ isert_alloc_comps(struct isert_device *device)
 	max_cqe = min(ISER_MAX_CQ_LEN, device->ib_device->attrs.max_cqe);
 
 	for (i = 0; i < device->comps_used; i++) {
-		struct ib_cq_init_attr cq_attr = {};
 		struct isert_comp *comp = &device->comps[i];
 
 		comp->device = device;
-		INIT_WORK(&comp->work, isert_cq_work);
-		cq_attr.cqe = max_cqe;
-		cq_attr.comp_vector = i;
-		comp->cq = ib_create_cq(device->ib_device,
-					isert_cq_callback,
-					isert_cq_event_callback,
-					(void *)comp,
-					&cq_attr);
+		comp->cq = ib_alloc_cq(device->ib_device, comp, max_cqe, i,
+				IB_POLL_WORKQUEUE);
 		if (IS_ERR(comp->cq)) {
 			isert_err("Unable to allocate cq\n");
 			ret = PTR_ERR(comp->cq);
 			comp->cq = NULL;
 			goto out_cq;
 		}
-
-		ret = ib_req_notify_cq(comp->cq, IB_CQ_NEXT_COMP);
-		if (ret)
-			goto out_cq;
 	}
 
 	return 0;
@@ -726,7 +709,7 @@ isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event)
 	if (ret)
 		goto out_conn_dev;
 
-	ret = isert_rdma_post_recvl(isert_conn);
+	ret = isert_login_post_recv(isert_conn);
 	if (ret)
 		goto out_conn_dev;
 
@@ -977,7 +960,8 @@ isert_post_recvm(struct isert_conn *isert_conn, u32 count)
 
 	for (rx_wr = isert_conn->rx_wr, i = 0; i < count; i++, rx_wr++) {
 		rx_desc = &isert_conn->rx_descs[i];
-		rx_wr->wr_id = (uintptr_t)rx_desc;
+
+		rx_wr->wr_cqe = &rx_desc->rx_cqe;
 		rx_wr->sg_list = &rx_desc->rx_sg;
 		rx_wr->num_sge = 1;
 		rx_wr->next = rx_wr + 1;
@@ -1002,7 +986,7 @@ isert_post_recv(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc)
 	struct ib_recv_wr *rx_wr_failed, rx_wr;
 	int ret;
 
-	rx_wr.wr_id = (uintptr_t)rx_desc;
+	rx_wr.wr_cqe = &rx_desc->rx_cqe;
 	rx_wr.sg_list = &rx_desc->rx_sg;
 	rx_wr.num_sge = 1;
 	rx_wr.next = NULL;
@@ -1018,7 +1002,7 @@ isert_post_recv(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc)
 }
 
 static int
-isert_post_send(struct isert_conn *isert_conn, struct iser_tx_desc *tx_desc)
+isert_login_post_send(struct isert_conn *isert_conn, struct iser_tx_desc *tx_desc)
 {
 	struct ib_device *ib_dev = isert_conn->cm_id->device;
 	struct ib_send_wr send_wr, *send_wr_failed;
@@ -1027,8 +1011,10 @@ isert_post_send(struct isert_conn *isert_conn, struct iser_tx_desc *tx_desc)
 	ib_dma_sync_single_for_device(ib_dev, tx_desc->dma_addr,
 				      ISER_HEADERS_LEN, DMA_TO_DEVICE);
 
+	tx_desc->tx_cqe.done = isert_login_send_done;
+
 	send_wr.next	= NULL;
-	send_wr.wr_id	= (uintptr_t)tx_desc;
+	send_wr.wr_cqe	= &tx_desc->tx_cqe;
 	send_wr.sg_list	= tx_desc->tx_sg;
 	send_wr.num_sge	= tx_desc->num_sge;
 	send_wr.opcode	= IB_WR_SEND;
@@ -1098,7 +1084,8 @@ isert_init_send_wr(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd,
 	struct iser_tx_desc *tx_desc = &isert_cmd->tx_desc;
 
 	isert_cmd->rdma_wr.iser_ib_op = ISER_IB_SEND;
-	send_wr->wr_id = (uintptr_t)&isert_cmd->tx_desc;
+	tx_desc->tx_cqe.done = isert_send_done;
+	send_wr->wr_cqe = &tx_desc->tx_cqe;
 
 	if (isert_conn->snd_w_inv && isert_cmd->inv_rkey) {
 		send_wr->opcode  = IB_WR_SEND_WITH_INV;
@@ -1113,7 +1100,7 @@ isert_init_send_wr(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd,
 }
 
 static int
-isert_rdma_post_recvl(struct isert_conn *isert_conn)
+isert_login_post_recv(struct isert_conn *isert_conn)
 {
 	struct ib_recv_wr rx_wr, *rx_wr_fail;
 	struct ib_sge sge;
@@ -1127,8 +1114,10 @@ isert_rdma_post_recvl(struct isert_conn *isert_conn)
 	isert_dbg("Setup sge: addr: %llx length: %d 0x%08x\n",
 		sge.addr, sge.length, sge.lkey);
 
+	isert_conn->login_req_buf->rx_cqe.done = isert_login_recv_done;
+
 	memset(&rx_wr, 0, sizeof(struct ib_recv_wr));
-	rx_wr.wr_id = (uintptr_t)isert_conn->login_req_buf;
+	rx_wr.wr_cqe = &isert_conn->login_req_buf->rx_cqe;
 	rx_wr.sg_list = &sge;
 	rx_wr.num_sge = 1;
 
@@ -1203,12 +1192,12 @@ isert_put_login_tx(struct iscsi_conn *conn, struct iscsi_login *login,
 			goto post_send;
 		}
 
-		ret = isert_rdma_post_recvl(isert_conn);
+		ret = isert_login_post_recv(isert_conn);
 		if (ret)
 			return ret;
 	}
 post_send:
-	ret = isert_post_send(isert_conn, tx_desc);
+	ret = isert_login_post_send(isert_conn, tx_desc);
 	if (ret)
 		return ret;
 
@@ -1551,12 +1540,44 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc,
 }
 
 static void
-isert_rx_do_work(struct iser_rx_desc *rx_desc, struct isert_conn *isert_conn)
+isert_print_wc(struct ib_wc *wc)
+{
+	if (wc->status != IB_WC_WR_FLUSH_ERR)
+		isert_err("%s (%d): cqe 0x%p vend_err %x\n",
+			  ib_wc_status_msg(wc->status), wc->status,
+			  wc->wr_cqe, wc->vendor_err);
+	else
+		isert_dbg("%s (%d): cqe 0x%p\n",
+			  ib_wc_status_msg(wc->status), wc->status,
+			  wc->wr_cqe);
+
+}
+
+static void
+isert_recv_done(struct ib_cq *cq, struct ib_wc *wc)
 {
+	struct isert_conn *isert_conn = wc->qp->qp_context;
+	struct ib_device *ib_dev = isert_conn->cm_id->device;
+	struct iser_rx_desc *rx_desc = cqe_to_rx_desc(wc->wr_cqe);
+	struct iscsi_hdr *hdr = &rx_desc->iscsi_header;
 	struct iser_ctrl *iser_ctrl = &rx_desc->iser_header;
 	uint64_t read_va = 0, write_va = 0;
 	uint32_t read_stag = 0, write_stag = 0;
 
+	if (unlikely(wc->status != IB_WC_SUCCESS)) {
+		isert_print_wc(wc);
+		if (!--isert_conn->post_recv_buf_count)
+			iscsit_cause_connection_reinstatement(isert_conn->conn, 0);
+		return;
+	}
+
+	ib_dma_sync_single_for_cpu(ib_dev, rx_desc->dma_addr,
+			ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE);
+
+	isert_dbg("DMA: 0x%llx, iSCSI opcode: 0x%02x, ITT: 0x%08x, flags: 0x%02x dlen: %d\n",
+		 rx_desc->dma_addr, hdr->opcode, hdr->itt, hdr->flags,
+		 (int)(wc->byte_len - ISER_HEADERS_LEN));
+
 	switch (iser_ctrl->flags & 0xF0) {
 	case ISCSI_CTRL:
 		if (iser_ctrl->flags & ISER_RSV) {
@@ -1584,55 +1605,44 @@ isert_rx_do_work(struct iser_rx_desc *rx_desc, struct isert_conn *isert_conn)
 
 	isert_rx_opcode(isert_conn, rx_desc,
 			read_stag, read_va, write_stag, write_va);
+
+	ib_dma_sync_single_for_device(ib_dev, rx_desc->dma_addr,
+			ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE);
+
+	isert_conn->post_recv_buf_count--;
 }
 
 static void
-isert_rcv_completion(struct iser_rx_desc *desc,
-		     struct isert_conn *isert_conn,
-		     u32 xfer_len)
+isert_login_recv_done(struct ib_cq *cq, struct ib_wc *wc)
 {
+	struct isert_conn *isert_conn = wc->qp->qp_context;
 	struct ib_device *ib_dev = isert_conn->cm_id->device;
-	struct iscsi_hdr *hdr;
-	u64 rx_dma;
-	int rx_buflen;
-
-	if (desc == isert_conn->login_req_buf) {
-		rx_dma = isert_conn->login_req_dma;
-		rx_buflen = ISER_RX_PAYLOAD_SIZE;
-		isert_dbg("login_buf: Using rx_dma: 0x%llx, rx_buflen: %d\n",
-			 rx_dma, rx_buflen);
-	} else {
-		rx_dma = desc->dma_addr;
-		rx_buflen = ISER_RX_PAYLOAD_SIZE;
-		isert_dbg("req_buf: Using rx_dma: 0x%llx, rx_buflen: %d\n",
-			 rx_dma, rx_buflen);
+
+	if (unlikely(wc->status != IB_WC_SUCCESS)) {
+		isert_print_wc(wc);
+		if (!--isert_conn->post_recv_buf_count)
+			iscsit_cause_connection_reinstatement(isert_conn->conn, 0);
+		return;
 	}
 
-	ib_dma_sync_single_for_cpu(ib_dev, rx_dma, rx_buflen, DMA_FROM_DEVICE);
+	ib_dma_sync_single_for_cpu(ib_dev, isert_conn->login_req_dma,
+			ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE);
 
-	hdr = &desc->iscsi_header;
-	isert_dbg("iSCSI opcode: 0x%02x, ITT: 0x%08x, flags: 0x%02x dlen: %d\n",
-		 hdr->opcode, hdr->itt, hdr->flags,
-		 (int)(xfer_len - ISER_HEADERS_LEN));
+	isert_conn->login_req_len = wc->byte_len - ISER_HEADERS_LEN;
 
-	if (desc == isert_conn->login_req_buf) {
-		isert_conn->login_req_len = xfer_len - ISER_HEADERS_LEN;
-		if (isert_conn->conn) {
-			struct iscsi_login *login = isert_conn->conn->conn_login;
+	if (isert_conn->conn) {
+		struct iscsi_login *login = isert_conn->conn->conn_login;
 
-			if (login && !login->first_request)
-				isert_rx_login_req(isert_conn);
-		}
-		mutex_lock(&isert_conn->mutex);
-		complete(&isert_conn->login_req_comp);
-		mutex_unlock(&isert_conn->mutex);
-	} else {
-		isert_rx_do_work(desc, isert_conn);
+		if (login && !login->first_request)
+			isert_rx_login_req(isert_conn);
 	}
 
-	ib_dma_sync_single_for_device(ib_dev, rx_dma, rx_buflen,
-				      DMA_FROM_DEVICE);
+	mutex_lock(&isert_conn->mutex);
+	complete(&isert_conn->login_req_comp);
+	mutex_unlock(&isert_conn->mutex);
 
+	ib_dma_sync_single_for_device(ib_dev, isert_conn->login_req_dma,
+				ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE);
 	isert_conn->post_recv_buf_count--;
 }
 
@@ -1882,42 +1892,57 @@ fail_mr_status:
 }
 
 static void
-isert_completion_rdma_write(struct iser_tx_desc *tx_desc,
-			    struct isert_cmd *isert_cmd)
+isert_rdma_write_done(struct ib_cq *cq, struct ib_wc *wc)
 {
-	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
-	struct iscsi_cmd *cmd = isert_cmd->iscsi_cmd;
-	struct se_cmd *se_cmd = &cmd->se_cmd;
-	struct isert_conn *isert_conn = isert_cmd->conn;
+	struct isert_conn *isert_conn = wc->qp->qp_context;
 	struct isert_device *device = isert_conn->device;
+	struct iser_tx_desc *desc = cqe_to_tx_desc(wc->wr_cqe);
+	struct isert_cmd *isert_cmd = desc->isert_cmd;
+	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
+	struct se_cmd *cmd = &isert_cmd->iscsi_cmd->se_cmd;
 	int ret = 0;
 
+	if (unlikely(wc->status != IB_WC_SUCCESS)) {
+		isert_print_wc(wc);
+		isert_completion_put(desc, isert_cmd, device->ib_device, true);
+		return;
+	}
+
+	isert_dbg("Cmd %p\n", isert_cmd);
+
 	if (wr->fr_desc && wr->fr_desc->ind & ISERT_PROTECTED) {
-		ret = isert_check_pi_status(se_cmd,
-					    wr->fr_desc->pi_ctx->sig_mr);
+		ret = isert_check_pi_status(cmd, wr->fr_desc->pi_ctx->sig_mr);
 		wr->fr_desc->ind &= ~ISERT_PROTECTED;
 	}
 
 	device->unreg_rdma_mem(isert_cmd, isert_conn);
 	wr->rdma_wr_num = 0;
 	if (ret)
-		transport_send_check_condition_and_sense(se_cmd,
-							 se_cmd->pi_err, 0);
+		transport_send_check_condition_and_sense(cmd, cmd->pi_err, 0);
 	else
-		isert_put_response(isert_conn->conn, cmd);
+		isert_put_response(isert_conn->conn, isert_cmd->iscsi_cmd);
 }
 
 static void
-isert_completion_rdma_read(struct iser_tx_desc *tx_desc,
-			   struct isert_cmd *isert_cmd)
+isert_rdma_read_done(struct ib_cq *cq, struct ib_wc *wc)
 {
+	struct isert_conn *isert_conn = wc->qp->qp_context;
+	struct isert_device *device = isert_conn->device;
+	struct iser_tx_desc *desc = cqe_to_tx_desc(wc->wr_cqe);
+	struct isert_cmd *isert_cmd = desc->isert_cmd;
 	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
 	struct iscsi_cmd *cmd = isert_cmd->iscsi_cmd;
 	struct se_cmd *se_cmd = &cmd->se_cmd;
-	struct isert_conn *isert_conn = isert_cmd->conn;
-	struct isert_device *device = isert_conn->device;
 	int ret = 0;
 
+	if (unlikely(wc->status != IB_WC_SUCCESS)) {
+		isert_print_wc(wc);
+		isert_completion_put(desc, isert_cmd, device->ib_device, true);
+		return;
+	}
+
+	isert_dbg("Cmd %p\n", isert_cmd);
+
 	if (wr->fr_desc && wr->fr_desc->ind & ISERT_PROTECTED) {
 		ret = isert_check_pi_status(se_cmd,
 					    wr->fr_desc->pi_ctx->sig_mr);
@@ -1975,170 +2000,51 @@ isert_do_control_comp(struct work_struct *work)
 }
 
 static void
-isert_response_completion(struct iser_tx_desc *tx_desc,
-			  struct isert_cmd *isert_cmd,
-			  struct isert_conn *isert_conn,
-			  struct ib_device *ib_dev)
+isert_login_send_done(struct ib_cq *cq, struct ib_wc *wc)
 {
-	struct iscsi_cmd *cmd = isert_cmd->iscsi_cmd;
-
-	if (cmd->i_state == ISTATE_SEND_TASKMGTRSP ||
-	    cmd->i_state == ISTATE_SEND_LOGOUTRSP ||
-	    cmd->i_state == ISTATE_SEND_REJECT ||
-	    cmd->i_state == ISTATE_SEND_TEXTRSP) {
-		isert_unmap_tx_desc(tx_desc, ib_dev);
+	struct isert_conn *isert_conn = wc->qp->qp_context;
+	struct ib_device *ib_dev = isert_conn->cm_id->device;
+	struct iser_tx_desc *tx_desc = cqe_to_tx_desc(wc->wr_cqe);
 
-		INIT_WORK(&isert_cmd->comp_work, isert_do_control_comp);
-		queue_work(isert_comp_wq, &isert_cmd->comp_work);
-		return;
-	}
+	if (unlikely(wc->status != IB_WC_SUCCESS))
+		isert_print_wc(wc);
 
-	cmd->i_state = ISTATE_SENT_STATUS;
-	isert_completion_put(tx_desc, isert_cmd, ib_dev, false);
+	isert_unmap_tx_desc(tx_desc, ib_dev);
 }
 
 static void
-isert_snd_completion(struct iser_tx_desc *tx_desc,
-		      struct isert_conn *isert_conn)
+isert_send_done(struct ib_cq *cq, struct ib_wc *wc)
 {
+	struct isert_conn *isert_conn = wc->qp->qp_context;
 	struct ib_device *ib_dev = isert_conn->cm_id->device;
+	struct iser_tx_desc *tx_desc = cqe_to_tx_desc(wc->wr_cqe);
 	struct isert_cmd *isert_cmd = tx_desc->isert_cmd;
-	struct isert_rdma_wr *wr;
 
-	if (!isert_cmd) {
-		isert_unmap_tx_desc(tx_desc, ib_dev);
+	if (unlikely(wc->status != IB_WC_SUCCESS)) {
+		isert_print_wc(wc);
+		isert_completion_put(tx_desc, isert_cmd, ib_dev, true);
 		return;
 	}
-	wr = &isert_cmd->rdma_wr;
 
-	isert_dbg("Cmd %p iser_ib_op %d\n", isert_cmd, wr->iser_ib_op);
+	isert_dbg("Cmd %p\n", isert_cmd);
 
-	switch (wr->iser_ib_op) {
-	case ISER_IB_SEND:
-		isert_response_completion(tx_desc, isert_cmd,
-					  isert_conn, ib_dev);
-		break;
-	case ISER_IB_RDMA_WRITE:
-		isert_completion_rdma_write(tx_desc, isert_cmd);
-		break;
-	case ISER_IB_RDMA_READ:
-		isert_completion_rdma_read(tx_desc, isert_cmd);
-		break;
+	switch (isert_cmd->iscsi_cmd->i_state) {
+	case ISTATE_SEND_TASKMGTRSP:
+	case ISTATE_SEND_LOGOUTRSP:
+	case ISTATE_SEND_REJECT:
+	case ISTATE_SEND_TEXTRSP:
+		isert_unmap_tx_desc(tx_desc, ib_dev);
+
+		INIT_WORK(&isert_cmd->comp_work, isert_do_control_comp);
+		queue_work(isert_comp_wq, &isert_cmd->comp_work);
+		return;
 	default:
-		isert_err("Unknown wr->iser_ib_op: 0x%x\n", wr->iser_ib_op);
-		dump_stack();
+		isert_cmd->iscsi_cmd->i_state = ISTATE_SENT_STATUS;
+		isert_completion_put(tx_desc, isert_cmd, ib_dev, false);
 		break;
 	}
 }
 
-/**
- * is_isert_tx_desc() - Indicate if the completion wr_id
- *     is a TX descriptor or not.
- * @isert_conn: iser connection
- * @wr_id: completion WR identifier
- *
- * Since we cannot rely on wc opcode in FLUSH errors
- * we must work around it by checking if the wr_id address
- * falls in the iser connection rx_descs buffer. If so
- * it is an RX descriptor, otherwize it is a TX.
- */
-static inline bool
-is_isert_tx_desc(struct isert_conn *isert_conn, void *wr_id)
-{
-	void *start = isert_conn->rx_descs;
-	int len = ISERT_QP_MAX_RECV_DTOS * sizeof(*isert_conn->rx_descs);
-
-	if (wr_id >= start && wr_id < start + len)
-		return false;
-
-	return true;
-}
-
-static void
-isert_cq_comp_err(struct isert_conn *isert_conn, struct ib_wc *wc)
-{
-	if (wc->wr_id == ISER_BEACON_WRID) {
-		isert_info("conn %p completing wait_comp_err\n",
-			   isert_conn);
-		complete(&isert_conn->wait_comp_err);
-	} else if (is_isert_tx_desc(isert_conn, (void *)(uintptr_t)wc->wr_id)) {
-		struct ib_device *ib_dev = isert_conn->cm_id->device;
-		struct isert_cmd *isert_cmd;
-		struct iser_tx_desc *desc;
-
-		desc = (struct iser_tx_desc *)(uintptr_t)wc->wr_id;
-		isert_cmd = desc->isert_cmd;
-		if (!isert_cmd)
-			isert_unmap_tx_desc(desc, ib_dev);
-		else
-			isert_completion_put(desc, isert_cmd, ib_dev, true);
-	} else {
-		isert_conn->post_recv_buf_count--;
-		if (!isert_conn->post_recv_buf_count)
-			iscsit_cause_connection_reinstatement(isert_conn->conn, 0);
-	}
-}
-
-static void
-isert_handle_wc(struct ib_wc *wc)
-{
-	struct isert_conn *isert_conn;
-	struct iser_tx_desc *tx_desc;
-	struct iser_rx_desc *rx_desc;
-
-	isert_conn = wc->qp->qp_context;
-	if (likely(wc->status == IB_WC_SUCCESS)) {
-		if (wc->opcode == IB_WC_RECV) {
-			rx_desc = (struct iser_rx_desc *)(uintptr_t)wc->wr_id;
-			isert_rcv_completion(rx_desc, isert_conn, wc->byte_len);
-		} else {
-			tx_desc = (struct iser_tx_desc *)(uintptr_t)wc->wr_id;
-			isert_snd_completion(tx_desc, isert_conn);
-		}
-	} else {
-		if (wc->status != IB_WC_WR_FLUSH_ERR)
-			isert_err("%s (%d): wr id %llx vend_err %x\n",
-				  ib_wc_status_msg(wc->status), wc->status,
-				  wc->wr_id, wc->vendor_err);
-		else
-			isert_dbg("%s (%d): wr id %llx\n",
-				  ib_wc_status_msg(wc->status), wc->status,
-				  wc->wr_id);
-
-		if (wc->wr_id != ISER_FASTREG_LI_WRID)
-			isert_cq_comp_err(isert_conn, wc);
-	}
-}
-
-static void
-isert_cq_work(struct work_struct *work)
-{
-	enum { isert_poll_budget = 65536 };
-	struct isert_comp *comp = container_of(work, struct isert_comp,
-					       work);
-	struct ib_wc *const wcs = comp->wcs;
-	int i, n, completed = 0;
-
-	while ((n = ib_poll_cq(comp->cq, ARRAY_SIZE(comp->wcs), wcs)) > 0) {
-		for (i = 0; i < n; i++)
-			isert_handle_wc(&wcs[i]);
-
-		completed += n;
-		if (completed >= isert_poll_budget)
-			break;
-	}
-
-	ib_req_notify_cq(comp->cq, IB_CQ_NEXT_COMP);
-}
-
-static void
-isert_cq_callback(struct ib_cq *cq, void *context)
-{
-	struct isert_comp *comp = context;
-
-	queue_work(isert_comp_wq, &comp->work);
-}
-
 static int
 isert_post_response(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd)
 {
@@ -2395,7 +2301,8 @@ isert_build_rdma_wr(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd,
 	page_off = offset % PAGE_SIZE;
 
 	rdma_wr->wr.sg_list = ib_sge;
-	rdma_wr->wr.wr_id = (uintptr_t)&isert_cmd->tx_desc;
+	rdma_wr->wr.wr_cqe = &isert_cmd->tx_desc.tx_cqe;
+
 	/*
 	 * Perform mapping of TCM scatterlist memory ib_sge dma_addr.
 	 */
@@ -2478,6 +2385,8 @@ isert_map_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 
 		rdma_wr->wr.send_flags = 0;
 		if (wr->iser_ib_op == ISER_IB_RDMA_WRITE) {
+			isert_cmd->tx_desc.tx_cqe.done = isert_rdma_write_done;
+
 			rdma_wr->wr.opcode = IB_WR_RDMA_WRITE;
 			rdma_wr->remote_addr = isert_cmd->read_va + offset;
 			rdma_wr->rkey = isert_cmd->read_stag;
@@ -2486,6 +2395,8 @@ isert_map_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 			else
 				rdma_wr->wr.next = &wr->rdma_wr[i + 1].wr;
 		} else {
+			isert_cmd->tx_desc.tx_cqe.done = isert_rdma_read_done;
+
 			rdma_wr->wr.opcode = IB_WR_RDMA_READ;
 			rdma_wr->remote_addr = isert_cmd->write_va + va_offset;
 			rdma_wr->rkey = isert_cmd->write_stag;
@@ -2517,7 +2428,7 @@ isert_inv_rkey(struct ib_send_wr *inv_wr, struct ib_mr *mr)
 	u32 rkey;
 
 	memset(inv_wr, 0, sizeof(*inv_wr));
-	inv_wr->wr_id = ISER_FASTREG_LI_WRID;
+	inv_wr->wr_cqe = NULL;
 	inv_wr->opcode = IB_WR_LOCAL_INV;
 	inv_wr->ex.invalidate_rkey = mr->rkey;
 
@@ -2573,7 +2484,7 @@ isert_fast_reg_mr(struct isert_conn *isert_conn,
 
 	reg_wr.wr.next = NULL;
 	reg_wr.wr.opcode = IB_WR_REG_MR;
-	reg_wr.wr.wr_id = ISER_FASTREG_LI_WRID;
+	reg_wr.wr.wr_cqe = NULL;
 	reg_wr.wr.send_flags = 0;
 	reg_wr.wr.num_sge = 0;
 	reg_wr.mr = mr;
@@ -2684,7 +2595,7 @@ isert_reg_sig_mr(struct isert_conn *isert_conn,
 
 	memset(&sig_wr, 0, sizeof(sig_wr));
 	sig_wr.wr.opcode = IB_WR_REG_SIG_MR;
-	sig_wr.wr.wr_id = ISER_FASTREG_LI_WRID;
+	sig_wr.wr.wr_cqe = NULL;
 	sig_wr.wr.sg_list = &rdma_wr->ib_sg[DATA];
 	sig_wr.wr.num_sge = 1;
 	sig_wr.access_flags = IB_ACCESS_LOCAL_WRITE;
@@ -2839,14 +2750,18 @@ isert_reg_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 	rdma_wr = &isert_cmd->rdma_wr.s_rdma_wr;
 	rdma_wr->wr.sg_list = &wr->s_ib_sge;
 	rdma_wr->wr.num_sge = 1;
-	rdma_wr->wr.wr_id = (uintptr_t)&isert_cmd->tx_desc;
+	rdma_wr->wr.wr_cqe = &isert_cmd->tx_desc.tx_cqe;
 	if (wr->iser_ib_op == ISER_IB_RDMA_WRITE) {
+		isert_cmd->tx_desc.tx_cqe.done = isert_rdma_write_done;
+
 		rdma_wr->wr.opcode = IB_WR_RDMA_WRITE;
 		rdma_wr->remote_addr = isert_cmd->read_va;
 		rdma_wr->rkey = isert_cmd->read_stag;
 		rdma_wr->wr.send_flags = !isert_prot_cmd(isert_conn, se_cmd) ?
 				      0 : IB_SEND_SIGNALED;
 	} else {
+		isert_cmd->tx_desc.tx_cqe.done = isert_rdma_read_done;
+
 		rdma_wr->wr.opcode = IB_WR_RDMA_READ;
 		rdma_wr->remote_addr = isert_cmd->write_va;
 		rdma_wr->rkey = isert_cmd->write_stag;
@@ -3310,14 +3225,26 @@ isert_wait4cmds(struct iscsi_conn *conn)
 }
 
 static void
+isert_beacon_done(struct ib_cq *cq, struct ib_wc *wc)
+{
+	struct isert_conn *isert_conn = wc->qp->qp_context;
+
+	isert_print_wc(wc);
+
+	isert_info("conn %p completing wait_comp_err\n", isert_conn);
+	complete(&isert_conn->wait_comp_err);
+}
+
+static void
 isert_wait4flush(struct isert_conn *isert_conn)
 {
 	struct ib_recv_wr *bad_wr;
+	static struct ib_cqe cqe = { .done = isert_beacon_done };
 
 	isert_info("conn %p\n", isert_conn);
 
 	init_completion(&isert_conn->wait_comp_err);
-	isert_conn->beacon.wr_id = ISER_BEACON_WRID;
+	isert_conn->beacon.wr_cqe = &cqe;
 	/* post an indication that all flush errors were consumed */
 	if (ib_post_recv(isert_conn->qp, &isert_conn->beacon, &bad_wr)) {
 		isert_err("conn %p failed to post beacon", isert_conn);
diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h
index 1240cf8..df494c7 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.h
+++ b/drivers/infiniband/ulp/isert/ib_isert.h
@@ -61,11 +61,10 @@
 				ISERT_MAX_RX_MISC_PDUS)
 
 #define ISER_RX_PAD_SIZE	(ISCSI_DEF_MAX_RECV_SEG_LEN + 4096 - \
-		(ISER_RX_PAYLOAD_SIZE + sizeof(u64) + sizeof(struct ib_sge)))
+		(ISER_RX_PAYLOAD_SIZE + sizeof(u64) + sizeof(struct ib_sge) + \
+		 sizeof(struct ib_cqe)))
 
 #define ISCSI_ISER_SG_TABLESIZE		256
-#define ISER_FASTREG_LI_WRID		0xffffffffffffffffULL
-#define ISER_BEACON_WRID               0xfffffffffffffffeULL
 
 enum isert_desc_type {
 	ISCSI_TX_CONTROL,
@@ -93,20 +92,33 @@ struct iser_rx_desc {
 	char		data[ISCSI_DEF_MAX_RECV_SEG_LEN];
 	u64		dma_addr;
 	struct ib_sge	rx_sg;
+	struct ib_cqe	rx_cqe;
 	char		pad[ISER_RX_PAD_SIZE];
 } __packed;
 
+static inline struct iser_rx_desc *cqe_to_rx_desc(struct ib_cqe *cqe)
+{
+	return container_of(cqe, struct iser_rx_desc, rx_cqe);
+}
+
 struct iser_tx_desc {
 	struct iser_ctrl iser_header;
 	struct iscsi_hdr iscsi_header;
 	enum isert_desc_type type;
 	u64		dma_addr;
 	struct ib_sge	tx_sg[2];
+	struct ib_cqe	tx_cqe;
 	int		num_sge;
 	struct isert_cmd *isert_cmd;
 	struct ib_send_wr send_wr;
 } __packed;
 
+static inline struct iser_tx_desc *cqe_to_tx_desc(struct ib_cqe *cqe)
+{
+	return container_of(cqe, struct iser_tx_desc, tx_cqe);
+}
+
+
 enum isert_indicator {
 	ISERT_PROTECTED		= 1 << 0,
 	ISERT_DATA_KEY_VALID	= 1 << 1,
@@ -218,17 +230,13 @@ struct isert_conn {
  *
  * @device:     pointer to device handle
  * @cq:         completion queue
- * @wcs:        work completion array
  * @active_qps: Number of active QPs attached
  *              to completion context
- * @work:       completion work handle
  */
 struct isert_comp {
 	struct isert_device     *device;
 	struct ib_cq		*cq;
-	struct ib_wc		 wcs[16];
 	int                      active_qps;
-	struct work_struct	 work;
 };
 
 struct isert_device {
-- 
2.1.4

--
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

Re: [PATCH 3/5] IB/isert: Convert to new CQ API

[Sagi Grimberg]

> Use the workqueue based CQ type similar to what isert was using previously,
> and properly split up the completion handlers.

Ideally, we'd have a separate done handler for each time of response
(text, logout, tmr, normal task etc..) but it requires more work so that
would be possible, we can do it incrementally.

This patch had a minor conflict on top of a fix patch from Jenny, I
fixed the fuzz, would you mind if I resend a combined set once the
code passes our regression tests?

> Note that this also takes special care to handle the magic login WRs
> separately, and also renames the submission functions so that it's clear
> that they are only to be used for the login buffers.

That's very helpful.

>   static void
> +isert_beacon_done(struct ib_cq *cq, struct ib_wc *wc)
> +{
> +	struct isert_conn *isert_conn = wc->qp->qp_context;
> +
> +	isert_print_wc(wc);
> +
> +	isert_info("conn %p completing wait_comp_err\n", isert_conn);
> +	complete(&isert_conn->wait_comp_err);
> +}
> +

I have a patch that replaces this with the new ib_drain_qp() from
Steve, I'll add it to the set.

Looks good,

Acked-by: Sagi Grimberg <sagig@mellanox.com>

Re: [PATCH 3/5] IB/isert: Convert to new CQ API

[Christoph Hellwig]

On Tue, Feb 23, 2016 at 01:15:05PM +0200, Sagi Grimberg wrote:
>
>> Use the workqueue based CQ type similar to what isert was using previously,
>> and properly split up the completion handlers.
>
> Ideally, we'd have a separate done handler for each time of response
> (text, logout, tmr, normal task etc..) but it requires more work so that
> would be possible, we can do it incrementally.

Except for TMR and logout they are treated very similar in isert,
so I'm not sure such a fine grained split is worth it.  But there's
some opportunity for cleaning up the completions hanlders for sure.

> This patch had a minor conflict on top of a fix patch from Jenny, I
> fixed the fuzz, would you mind if I resend a combined set once the
> code passes our regression tests?

Sure, please go ahead!

>>   static void
>> +isert_beacon_done(struct ib_cq *cq, struct ib_wc *wc)
>> +{
>> +	struct isert_conn *isert_conn = wc->qp->qp_context;
>> +
>> +	isert_print_wc(wc);
>> +
>> +	isert_info("conn %p completing wait_comp_err\n", isert_conn);
>> +	complete(&isert_conn->wait_comp_err);
>> +}
>> +
>
> I have a patch that replaces this with the new ib_drain_qp() from
> Steve, I'll add it to the set.

Great!
--
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

Re: [PATCH 3/5] IB/isert: Convert to new CQ API

[Sagi Grimberg]

> -isert_rx_do_work(struct iser_rx_desc *rx_desc, struct isert_conn *isert_conn)
> +isert_print_wc(struct ib_wc *wc)
> +{
> +	if (wc->status != IB_WC_WR_FLUSH_ERR)
> +		isert_err("%s (%d): cqe 0x%p vend_err %x\n",
> +			  ib_wc_status_msg(wc->status), wc->status,
> +			  wc->wr_cqe, wc->vendor_err);
> +	else
> +		isert_dbg("%s (%d): cqe 0x%p\n",
> +			  ib_wc_status_msg(wc->status), wc->status,
> +			  wc->wr_cqe);
> +
> +}

I don't find the wr_cqe address all that useful. I modified
isert_print_wc to get a error type instead ("send", "recv", "rdma
write"...) sounds fine?
--
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

Re: [PATCH 3/5] IB/isert: Convert to new CQ API

[Christoph Hellwig]

On Wed, Feb 24, 2016 at 06:01:27PM +0200, Sagi Grimberg wrote:
> I don't find the wr_cqe address all that useful. I modified
> isert_print_wc to get a error type instead ("send", "recv", "rdma
> write"...) sounds fine?

Fine with me.

Re: [PATCH 3/5] IB/isert: Convert to new CQ API

[Max Gurtovoy]

On 2/24/2016 8:43 PM, Christoph Hellwig wrote:
> On Wed, Feb 24, 2016 at 06:01:27PM +0200, Sagi Grimberg wrote:
>> I don't find the wr_cqe address all that useful. I modified
>> isert_print_wc to get a error type instead ("send", "recv", "rdma
>> write"...) sounds fine?
>
> Fine with me.
>

sounds good.

[PATCH 5/5] IB/isert: Kill the ->isert_cmd back pointer in struct iser_tx_desc

[Christoph Hellwig]

We only use the pointer when processing regular iSER commands, and it then
always points to the struct iser_cmd that contains the TX descriptor.

Remove it and rely on container_of to save a little space and avoid a
pointer that is updated multiple times per processed command.

Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
---
 drivers/infiniband/ulp/isert/ib_isert.c | 11 +++--------
 drivers/infiniband/ulp/isert/ib_isert.h |  6 +++++-
 2 files changed, 8 insertions(+), 9 deletions(-)

diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index 02fb3e6..5e95ddd 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -1040,7 +1040,6 @@ isert_create_send_desc(struct isert_conn *isert_conn,
 	tx_desc->iser_header.flags = ISCSI_CTRL;
 
 	tx_desc->num_sge = 1;
-	tx_desc->isert_cmd = isert_cmd;
 
 	if (tx_desc->tx_sg[0].lkey != device->pd->local_dma_lkey) {
 		tx_desc->tx_sg[0].lkey = device->pd->local_dma_lkey;
@@ -1891,7 +1890,7 @@ isert_rdma_write_done(struct ib_cq *cq, struct ib_wc *wc)
 	struct isert_conn *isert_conn = wc->qp->qp_context;
 	struct isert_device *device = isert_conn->device;
 	struct iser_tx_desc *desc = cqe_to_tx_desc(wc->wr_cqe);
-	struct isert_cmd *isert_cmd = desc->isert_cmd;
+	struct isert_cmd *isert_cmd = tx_desc_to_cmd(desc);
 	struct se_cmd *cmd = &isert_cmd->iscsi_cmd->se_cmd;
 	int ret = 0;
 
@@ -1923,7 +1922,7 @@ isert_rdma_read_done(struct ib_cq *cq, struct ib_wc *wc)
 	struct isert_conn *isert_conn = wc->qp->qp_context;
 	struct isert_device *device = isert_conn->device;
 	struct iser_tx_desc *desc = cqe_to_tx_desc(wc->wr_cqe);
-	struct isert_cmd *isert_cmd = desc->isert_cmd;
+	struct isert_cmd *isert_cmd = tx_desc_to_cmd(desc);
 	struct iscsi_cmd *cmd = isert_cmd->iscsi_cmd;
 	struct se_cmd *se_cmd = &cmd->se_cmd;
 	int ret = 0;
@@ -2011,7 +2010,7 @@ isert_send_done(struct ib_cq *cq, struct ib_wc *wc)
 	struct isert_conn *isert_conn = wc->qp->qp_context;
 	struct ib_device *ib_dev = isert_conn->cm_id->device;
 	struct iser_tx_desc *tx_desc = cqe_to_tx_desc(wc->wr_cqe);
-	struct isert_cmd *isert_cmd = tx_desc->isert_cmd;
+	struct isert_cmd *isert_cmd = tx_desc_to_cmd(tx_desc);
 
 	if (unlikely(wc->status != IB_WC_SUCCESS)) {
 		isert_print_wc(wc);
@@ -2339,8 +2338,6 @@ isert_map_rdma(struct isert_cmd *isert_cmd, struct iscsi_conn *conn)
 	u32 offset, data_len, data_left, rdma_write_max, va_offset = 0;
 	int ret = 0, i, ib_sge_cnt;
 
-	isert_cmd->tx_desc.isert_cmd = isert_cmd;
-
 	offset = isert_cmd->iser_ib_op == ISER_IB_RDMA_READ ?
 			cmd->write_data_done : 0;
 	ret = isert_map_data_buf(isert_conn, isert_cmd, se_cmd->t_data_sg,
@@ -2702,8 +2699,6 @@ isert_reg_rdma(struct isert_cmd *isert_cmd, struct iscsi_conn *conn)
 	int ret = 0;
 	unsigned long flags;
 
-	isert_cmd->tx_desc.isert_cmd = isert_cmd;
-
 	offset = isert_cmd->iser_ib_op == ISER_IB_RDMA_READ ?
 			cmd->write_data_done : 0;
 	ret = isert_map_data_buf(isert_conn, isert_cmd, se_cmd->t_data_sg,
diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h
index 6103e68..e04f1804 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.h
+++ b/drivers/infiniband/ulp/isert/ib_isert.h
@@ -109,7 +109,6 @@ struct iser_tx_desc {
 	struct ib_sge	tx_sg[2];
 	struct ib_cqe	tx_cqe;
 	int		num_sge;
-	struct isert_cmd *isert_cmd;
 	struct ib_send_wr send_wr;
 } __packed;
 
@@ -180,6 +179,11 @@ struct isert_cmd {
 	struct scatterlist	sg;
 };
 
+static inline struct isert_cmd *tx_desc_to_cmd(struct iser_tx_desc *desc)
+{
+	return container_of(desc, struct isert_cmd, tx_desc);
+}
+
 struct isert_device;
 
 struct isert_conn {
-- 
2.1.4

--
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

Re: [PATCH 5/5] IB/isert: Kill the ->isert_cmd back pointer in struct iser_tx_desc

[Sagi Grimberg]

> We only use the pointer when processing regular iSER commands, and it then
> always points to the struct iser_cmd that contains the TX descriptor.

That was I hack I knew we will be able to get rid of once we move to
the new CQ API. The iscsi login response is not allocate a command
structure and this back-pointer was used to differentiate that. Now
that the login response has it's own done handler we can get rid of
it.

Thanks!

Reviewed-by: Sagi Grimberg <sagig@mellanox.com>

Re: isert: convert to the new CQ API and a few random cleanups V2

[Sagi Grimberg]

Thanks Christoph,

This is very good progress overall, and we can move on with
other needed cleanups and simplifications.

As I said, I have a couple of fixes from Jenny that I want
this to be on top of. I already took this set and added it
to our regression and so far things look fine, no breakage.

I'll submit the combined series soon enough!

Cheers,
Sagi.
--
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

[PATCH 4/5] IB/isert: Kill struct isert_rdma_wr

[Christoph Hellwig]

There is exactly one instance per struct isert_cmd, so merge the two to
simplify everyones life.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 drivers/infiniband/ulp/isert/ib_isert.c | 219 ++++++++++++++++----------------
 drivers/infiniband/ulp/isert/ib_isert.h |  30 ++---
 2 files changed, 119 insertions(+), 130 deletions(-)

diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index 1824692..02fb3e6 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -49,13 +49,11 @@ static struct workqueue_struct *isert_release_wq;
 static void
 isert_unmap_cmd(struct isert_cmd *isert_cmd, struct isert_conn *isert_conn);
 static int
-isert_map_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
-	       struct isert_rdma_wr *wr);
+isert_map_rdma(struct isert_cmd *isert_cmd, struct iscsi_conn *conn);
 static void
 isert_unreg_rdma(struct isert_cmd *isert_cmd, struct isert_conn *isert_conn);
 static int
-isert_reg_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
-	       struct isert_rdma_wr *wr);
+isert_reg_rdma(struct isert_cmd *isert_cmd, struct iscsi_conn *conn);
 static int
 isert_put_response(struct iscsi_conn *conn, struct iscsi_cmd *cmd);
 static int
@@ -1083,7 +1081,7 @@ isert_init_send_wr(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd,
 {
 	struct iser_tx_desc *tx_desc = &isert_cmd->tx_desc;
 
-	isert_cmd->rdma_wr.iser_ib_op = ISER_IB_SEND;
+	isert_cmd->iser_ib_op = ISER_IB_SEND;
 	tx_desc->tx_cqe.done = isert_send_done;
 	send_wr->wr_cqe = &tx_desc->tx_cqe;
 
@@ -1693,54 +1691,50 @@ isert_unmap_data_buf(struct isert_conn *isert_conn, struct isert_data_buf *data)
 static void
 isert_unmap_cmd(struct isert_cmd *isert_cmd, struct isert_conn *isert_conn)
 {
-	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
-
 	isert_dbg("Cmd %p\n", isert_cmd);
 
-	if (wr->data.sg) {
+	if (isert_cmd->data.sg) {
 		isert_dbg("Cmd %p unmap_sg op\n", isert_cmd);
-		isert_unmap_data_buf(isert_conn, &wr->data);
+		isert_unmap_data_buf(isert_conn, &isert_cmd->data);
 	}
 
-	if (wr->rdma_wr) {
+	if (isert_cmd->rdma_wr) {
 		isert_dbg("Cmd %p free send_wr\n", isert_cmd);
-		kfree(wr->rdma_wr);
-		wr->rdma_wr = NULL;
+		kfree(isert_cmd->rdma_wr);
+		isert_cmd->rdma_wr = NULL;
 	}
 
-	if (wr->ib_sge) {
+	if (isert_cmd->ib_sge) {
 		isert_dbg("Cmd %p free ib_sge\n", isert_cmd);
-		kfree(wr->ib_sge);
-		wr->ib_sge = NULL;
+		kfree(isert_cmd->ib_sge);
+		isert_cmd->ib_sge = NULL;
 	}
 }
 
 static void
 isert_unreg_rdma(struct isert_cmd *isert_cmd, struct isert_conn *isert_conn)
 {
-	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
-
 	isert_dbg("Cmd %p\n", isert_cmd);
 
-	if (wr->fr_desc) {
-		isert_dbg("Cmd %p free fr_desc %p\n", isert_cmd, wr->fr_desc);
-		if (wr->fr_desc->ind & ISERT_PROTECTED) {
-			isert_unmap_data_buf(isert_conn, &wr->prot);
-			wr->fr_desc->ind &= ~ISERT_PROTECTED;
+	if (isert_cmd->fr_desc) {
+		isert_dbg("Cmd %p free fr_desc %p\n", isert_cmd, isert_cmd->fr_desc);
+		if (isert_cmd->fr_desc->ind & ISERT_PROTECTED) {
+			isert_unmap_data_buf(isert_conn, &isert_cmd->prot);
+			isert_cmd->fr_desc->ind &= ~ISERT_PROTECTED;
 		}
 		spin_lock_bh(&isert_conn->pool_lock);
-		list_add_tail(&wr->fr_desc->list, &isert_conn->fr_pool);
+		list_add_tail(&isert_cmd->fr_desc->list, &isert_conn->fr_pool);
 		spin_unlock_bh(&isert_conn->pool_lock);
-		wr->fr_desc = NULL;
+		isert_cmd->fr_desc = NULL;
 	}
 
-	if (wr->data.sg) {
+	if (isert_cmd->data.sg) {
 		isert_dbg("Cmd %p unmap_sg op\n", isert_cmd);
-		isert_unmap_data_buf(isert_conn, &wr->data);
+		isert_unmap_data_buf(isert_conn, &isert_cmd->data);
 	}
 
-	wr->ib_sge = NULL;
-	wr->rdma_wr = NULL;
+	isert_cmd->ib_sge = NULL;
+	isert_cmd->rdma_wr = NULL;
 }
 
 static void
@@ -1898,7 +1892,6 @@ isert_rdma_write_done(struct ib_cq *cq, struct ib_wc *wc)
 	struct isert_device *device = isert_conn->device;
 	struct iser_tx_desc *desc = cqe_to_tx_desc(wc->wr_cqe);
 	struct isert_cmd *isert_cmd = desc->isert_cmd;
-	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
 	struct se_cmd *cmd = &isert_cmd->iscsi_cmd->se_cmd;
 	int ret = 0;
 
@@ -1910,13 +1903,14 @@ isert_rdma_write_done(struct ib_cq *cq, struct ib_wc *wc)
 
 	isert_dbg("Cmd %p\n", isert_cmd);
 
-	if (wr->fr_desc && wr->fr_desc->ind & ISERT_PROTECTED) {
-		ret = isert_check_pi_status(cmd, wr->fr_desc->pi_ctx->sig_mr);
-		wr->fr_desc->ind &= ~ISERT_PROTECTED;
+	if (isert_cmd->fr_desc && isert_cmd->fr_desc->ind & ISERT_PROTECTED) {
+		ret = isert_check_pi_status(cmd,
+				isert_cmd->fr_desc->pi_ctx->sig_mr);
+		isert_cmd->fr_desc->ind &= ~ISERT_PROTECTED;
 	}
 
 	device->unreg_rdma_mem(isert_cmd, isert_conn);
-	wr->rdma_wr_num = 0;
+	isert_cmd->rdma_wr_num = 0;
 	if (ret)
 		transport_send_check_condition_and_sense(cmd, cmd->pi_err, 0);
 	else
@@ -1930,7 +1924,6 @@ isert_rdma_read_done(struct ib_cq *cq, struct ib_wc *wc)
 	struct isert_device *device = isert_conn->device;
 	struct iser_tx_desc *desc = cqe_to_tx_desc(wc->wr_cqe);
 	struct isert_cmd *isert_cmd = desc->isert_cmd;
-	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
 	struct iscsi_cmd *cmd = isert_cmd->iscsi_cmd;
 	struct se_cmd *se_cmd = &cmd->se_cmd;
 	int ret = 0;
@@ -1943,16 +1936,16 @@ isert_rdma_read_done(struct ib_cq *cq, struct ib_wc *wc)
 
 	isert_dbg("Cmd %p\n", isert_cmd);
 
-	if (wr->fr_desc && wr->fr_desc->ind & ISERT_PROTECTED) {
+	if (isert_cmd->fr_desc && isert_cmd->fr_desc->ind & ISERT_PROTECTED) {
 		ret = isert_check_pi_status(se_cmd,
-					    wr->fr_desc->pi_ctx->sig_mr);
-		wr->fr_desc->ind &= ~ISERT_PROTECTED;
+					    isert_cmd->fr_desc->pi_ctx->sig_mr);
+		isert_cmd->fr_desc->ind &= ~ISERT_PROTECTED;
 	}
 
 	iscsit_stop_dataout_timer(cmd);
 	device->unreg_rdma_mem(isert_cmd, isert_conn);
-	cmd->write_data_done = wr->data.len;
-	wr->rdma_wr_num = 0;
+	cmd->write_data_done = isert_cmd->data.len;
+	isert_cmd->rdma_wr_num = 0;
 
 	isert_dbg("Cmd: %p RDMA_READ comp calling execute_cmd\n", isert_cmd);
 	spin_lock_bh(&cmd->istate_lock);
@@ -2335,13 +2328,12 @@ isert_build_rdma_wr(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd,
 }
 
 static int
-isert_map_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
-	       struct isert_rdma_wr *wr)
+isert_map_rdma(struct isert_cmd *isert_cmd, struct iscsi_conn *conn)
 {
+	struct iscsi_cmd *cmd = isert_cmd->iscsi_cmd;
 	struct se_cmd *se_cmd = &cmd->se_cmd;
-	struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd);
 	struct isert_conn *isert_conn = conn->context;
-	struct isert_data_buf *data = &wr->data;
+	struct isert_data_buf *data = &isert_cmd->data;
 	struct ib_rdma_wr *rdma_wr;
 	struct ib_sge *ib_sge;
 	u32 offset, data_len, data_left, rdma_write_max, va_offset = 0;
@@ -2349,10 +2341,12 @@ isert_map_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 
 	isert_cmd->tx_desc.isert_cmd = isert_cmd;
 
-	offset = wr->iser_ib_op == ISER_IB_RDMA_READ ? cmd->write_data_done : 0;
+	offset = isert_cmd->iser_ib_op == ISER_IB_RDMA_READ ?
+			cmd->write_data_done : 0;
 	ret = isert_map_data_buf(isert_conn, isert_cmd, se_cmd->t_data_sg,
 				 se_cmd->t_data_nents, se_cmd->data_length,
-				 offset, wr->iser_ib_op, &wr->data);
+				 offset, isert_cmd->iser_ib_op,
+				 &isert_cmd->data);
 	if (ret)
 		return ret;
 
@@ -2365,45 +2359,44 @@ isert_map_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 		ret = -ENOMEM;
 		goto unmap_cmd;
 	}
-	wr->ib_sge = ib_sge;
+	isert_cmd->ib_sge = ib_sge;
 
-	wr->rdma_wr_num = DIV_ROUND_UP(data->nents, isert_conn->max_sge);
-	wr->rdma_wr = kzalloc(sizeof(struct ib_rdma_wr) * wr->rdma_wr_num,
-				GFP_KERNEL);
-	if (!wr->rdma_wr) {
-		isert_dbg("Unable to allocate wr->rdma_wr\n");
+	isert_cmd->rdma_wr_num = DIV_ROUND_UP(data->nents, isert_conn->max_sge);
+	isert_cmd->rdma_wr = kzalloc(sizeof(struct ib_rdma_wr) *
+			isert_cmd->rdma_wr_num, GFP_KERNEL);
+	if (!isert_cmd->rdma_wr) {
+		isert_dbg("Unable to allocate isert_cmd->rdma_wr\n");
 		ret = -ENOMEM;
 		goto unmap_cmd;
 	}
 
-	wr->isert_cmd = isert_cmd;
 	rdma_write_max = isert_conn->max_sge * PAGE_SIZE;
 
-	for (i = 0; i < wr->rdma_wr_num; i++) {
-		rdma_wr = &isert_cmd->rdma_wr.rdma_wr[i];
+	for (i = 0; i < isert_cmd->rdma_wr_num; i++) {
+		rdma_wr = &isert_cmd->rdma_wr[i];
 		data_len = min(data_left, rdma_write_max);
 
 		rdma_wr->wr.send_flags = 0;
-		if (wr->iser_ib_op == ISER_IB_RDMA_WRITE) {
+		if (isert_cmd->iser_ib_op == ISER_IB_RDMA_WRITE) {
 			isert_cmd->tx_desc.tx_cqe.done = isert_rdma_write_done;
 
 			rdma_wr->wr.opcode = IB_WR_RDMA_WRITE;
 			rdma_wr->remote_addr = isert_cmd->read_va + offset;
 			rdma_wr->rkey = isert_cmd->read_stag;
-			if (i + 1 == wr->rdma_wr_num)
+			if (i + 1 == isert_cmd->rdma_wr_num)
 				rdma_wr->wr.next = &isert_cmd->tx_desc.send_wr;
 			else
-				rdma_wr->wr.next = &wr->rdma_wr[i + 1].wr;
+				rdma_wr->wr.next = &isert_cmd->rdma_wr[i + 1].wr;
 		} else {
 			isert_cmd->tx_desc.tx_cqe.done = isert_rdma_read_done;
 
 			rdma_wr->wr.opcode = IB_WR_RDMA_READ;
 			rdma_wr->remote_addr = isert_cmd->write_va + va_offset;
 			rdma_wr->rkey = isert_cmd->write_stag;
-			if (i + 1 == wr->rdma_wr_num)
+			if (i + 1 == isert_cmd->rdma_wr_num)
 				rdma_wr->wr.send_flags = IB_SEND_SIGNALED;
 			else
-				rdma_wr->wr.next = &wr->rdma_wr[i + 1].wr;
+				rdma_wr->wr.next = &isert_cmd->rdma_wr[i + 1].wr;
 		}
 
 		ib_sge_cnt = isert_build_rdma_wr(isert_conn, isert_cmd, ib_sge,
@@ -2571,10 +2564,10 @@ isert_set_prot_checks(u8 prot_checks)
 
 static int
 isert_reg_sig_mr(struct isert_conn *isert_conn,
-		 struct se_cmd *se_cmd,
-		 struct isert_rdma_wr *rdma_wr,
+		 struct isert_cmd *isert_cmd,
 		 struct fast_reg_descriptor *fr_desc)
 {
+	struct se_cmd *se_cmd = &isert_cmd->iscsi_cmd->se_cmd;
 	struct ib_sig_handover_wr sig_wr;
 	struct ib_send_wr inv_wr, *bad_wr, *wr = NULL;
 	struct pi_context *pi_ctx = fr_desc->pi_ctx;
@@ -2596,13 +2589,13 @@ isert_reg_sig_mr(struct isert_conn *isert_conn,
 	memset(&sig_wr, 0, sizeof(sig_wr));
 	sig_wr.wr.opcode = IB_WR_REG_SIG_MR;
 	sig_wr.wr.wr_cqe = NULL;
-	sig_wr.wr.sg_list = &rdma_wr->ib_sg[DATA];
+	sig_wr.wr.sg_list = &isert_cmd->ib_sg[DATA];
 	sig_wr.wr.num_sge = 1;
 	sig_wr.access_flags = IB_ACCESS_LOCAL_WRITE;
 	sig_wr.sig_attrs = &sig_attrs;
 	sig_wr.sig_mr = pi_ctx->sig_mr;
 	if (se_cmd->t_prot_sg)
-		sig_wr.prot = &rdma_wr->ib_sg[PROT];
+		sig_wr.prot = &isert_cmd->ib_sg[PROT];
 
 	if (!wr)
 		wr = &sig_wr.wr;
@@ -2616,35 +2609,34 @@ isert_reg_sig_mr(struct isert_conn *isert_conn,
 	}
 	fr_desc->ind &= ~ISERT_SIG_KEY_VALID;
 
-	rdma_wr->ib_sg[SIG].lkey = pi_ctx->sig_mr->lkey;
-	rdma_wr->ib_sg[SIG].addr = 0;
-	rdma_wr->ib_sg[SIG].length = se_cmd->data_length;
+	isert_cmd->ib_sg[SIG].lkey = pi_ctx->sig_mr->lkey;
+	isert_cmd->ib_sg[SIG].addr = 0;
+	isert_cmd->ib_sg[SIG].length = se_cmd->data_length;
 	if (se_cmd->prot_op != TARGET_PROT_DIN_STRIP &&
 	    se_cmd->prot_op != TARGET_PROT_DOUT_INSERT)
 		/*
 		 * We have protection guards on the wire
 		 * so we need to set a larget transfer
 		 */
-		rdma_wr->ib_sg[SIG].length += se_cmd->prot_length;
+		isert_cmd->ib_sg[SIG].length += se_cmd->prot_length;
 
 	isert_dbg("sig_sge: addr: 0x%llx  length: %u lkey: %x\n",
-		  rdma_wr->ib_sg[SIG].addr, rdma_wr->ib_sg[SIG].length,
-		  rdma_wr->ib_sg[SIG].lkey);
+		  isert_cmd->ib_sg[SIG].addr, isert_cmd->ib_sg[SIG].length,
+		  isert_cmd->ib_sg[SIG].lkey);
 err:
 	return ret;
 }
 
 static int
 isert_handle_prot_cmd(struct isert_conn *isert_conn,
-		      struct isert_cmd *isert_cmd,
-		      struct isert_rdma_wr *wr)
+		      struct isert_cmd *isert_cmd)
 {
 	struct isert_device *device = isert_conn->device;
 	struct se_cmd *se_cmd = &isert_cmd->iscsi_cmd->se_cmd;
 	int ret;
 
-	if (!wr->fr_desc->pi_ctx) {
-		ret = isert_create_pi_ctx(wr->fr_desc,
+	if (!isert_cmd->fr_desc->pi_ctx) {
+		ret = isert_create_pi_ctx(isert_cmd->fr_desc,
 					  device->ib_device,
 					  device->pd);
 		if (ret) {
@@ -2659,16 +2651,20 @@ isert_handle_prot_cmd(struct isert_conn *isert_conn,
 					 se_cmd->t_prot_sg,
 					 se_cmd->t_prot_nents,
 					 se_cmd->prot_length,
-					 0, wr->iser_ib_op, &wr->prot);
+					 0,
+					 isert_cmd->iser_ib_op,
+					 &isert_cmd->prot);
 		if (ret) {
 			isert_err("conn %p failed to map protection buffer\n",
 				  isert_conn);
 			return ret;
 		}
 
-		memset(&wr->ib_sg[PROT], 0, sizeof(wr->ib_sg[PROT]));
-		ret = isert_fast_reg_mr(isert_conn, wr->fr_desc, &wr->prot,
-					ISERT_PROT_KEY_VALID, &wr->ib_sg[PROT]);
+		memset(&isert_cmd->ib_sg[PROT], 0, sizeof(isert_cmd->ib_sg[PROT]));
+		ret = isert_fast_reg_mr(isert_conn, isert_cmd->fr_desc,
+					&isert_cmd->prot,
+					ISERT_PROT_KEY_VALID,
+					&isert_cmd->ib_sg[PROT]);
 		if (ret) {
 			isert_err("conn %p failed to fast reg mr\n",
 				  isert_conn);
@@ -2676,29 +2672,28 @@ isert_handle_prot_cmd(struct isert_conn *isert_conn,
 		}
 	}
 
-	ret = isert_reg_sig_mr(isert_conn, se_cmd, wr, wr->fr_desc);
+	ret = isert_reg_sig_mr(isert_conn, isert_cmd, isert_cmd->fr_desc);
 	if (ret) {
 		isert_err("conn %p failed to fast reg mr\n",
 			  isert_conn);
 		goto unmap_prot_cmd;
 	}
-	wr->fr_desc->ind |= ISERT_PROTECTED;
+	isert_cmd->fr_desc->ind |= ISERT_PROTECTED;
 
 	return 0;
 
 unmap_prot_cmd:
 	if (se_cmd->t_prot_sg)
-		isert_unmap_data_buf(isert_conn, &wr->prot);
+		isert_unmap_data_buf(isert_conn, &isert_cmd->prot);
 
 	return ret;
 }
 
 static int
-isert_reg_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
-	       struct isert_rdma_wr *wr)
+isert_reg_rdma(struct isert_cmd *isert_cmd, struct iscsi_conn *conn)
 {
+	struct iscsi_cmd *cmd = isert_cmd->iscsi_cmd;
 	struct se_cmd *se_cmd = &cmd->se_cmd;
-	struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd);
 	struct isert_conn *isert_conn = conn->context;
 	struct fast_reg_descriptor *fr_desc = NULL;
 	struct ib_rdma_wr *rdma_wr;
@@ -2709,49 +2704,51 @@ isert_reg_rdma(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 
 	isert_cmd->tx_desc.isert_cmd = isert_cmd;
 
-	offset = wr->iser_ib_op == ISER_IB_RDMA_READ ? cmd->write_data_done : 0;
+	offset = isert_cmd->iser_ib_op == ISER_IB_RDMA_READ ?
+			cmd->write_data_done : 0;
 	ret = isert_map_data_buf(isert_conn, isert_cmd, se_cmd->t_data_sg,
 				 se_cmd->t_data_nents, se_cmd->data_length,
-				 offset, wr->iser_ib_op, &wr->data);
+				 offset, isert_cmd->iser_ib_op,
+				 &isert_cmd->data);
 	if (ret)
 		return ret;
 
-	if (wr->data.dma_nents != 1 || isert_prot_cmd(isert_conn, se_cmd)) {
+	if (isert_cmd->data.dma_nents != 1 ||
+	    isert_prot_cmd(isert_conn, se_cmd)) {
 		spin_lock_irqsave(&isert_conn->pool_lock, flags);
 		fr_desc = list_first_entry(&isert_conn->fr_pool,
 					   struct fast_reg_descriptor, list);
 		list_del(&fr_desc->list);
 		spin_unlock_irqrestore(&isert_conn->pool_lock, flags);
-		wr->fr_desc = fr_desc;
+		isert_cmd->fr_desc = fr_desc;
 	}
 
-	ret = isert_fast_reg_mr(isert_conn, fr_desc, &wr->data,
-				ISERT_DATA_KEY_VALID, &wr->ib_sg[DATA]);
+	ret = isert_fast_reg_mr(isert_conn, fr_desc, &isert_cmd->data,
+				ISERT_DATA_KEY_VALID, &isert_cmd->ib_sg[DATA]);
 	if (ret)
 		goto unmap_cmd;
 
 	if (isert_prot_cmd(isert_conn, se_cmd)) {
-		ret = isert_handle_prot_cmd(isert_conn, isert_cmd, wr);
+		ret = isert_handle_prot_cmd(isert_conn, isert_cmd);
 		if (ret)
 			goto unmap_cmd;
 
-		ib_sg = &wr->ib_sg[SIG];
+		ib_sg = &isert_cmd->ib_sg[SIG];
 	} else {
-		ib_sg = &wr->ib_sg[DATA];
+		ib_sg = &isert_cmd->ib_sg[DATA];
 	}
 
-	memcpy(&wr->s_ib_sge, ib_sg, sizeof(*ib_sg));
-	wr->ib_sge = &wr->s_ib_sge;
-	wr->rdma_wr_num = 1;
-	memset(&wr->s_rdma_wr, 0, sizeof(wr->s_rdma_wr));
-	wr->rdma_wr = &wr->s_rdma_wr;
-	wr->isert_cmd = isert_cmd;
+	memcpy(&isert_cmd->s_ib_sge, ib_sg, sizeof(*ib_sg));
+	isert_cmd->ib_sge = &isert_cmd->s_ib_sge;
+	isert_cmd->rdma_wr_num = 1;
+	memset(&isert_cmd->s_rdma_wr, 0, sizeof(isert_cmd->s_rdma_wr));
+	isert_cmd->rdma_wr = &isert_cmd->s_rdma_wr;
 
-	rdma_wr = &isert_cmd->rdma_wr.s_rdma_wr;
-	rdma_wr->wr.sg_list = &wr->s_ib_sge;
+	rdma_wr = &isert_cmd->s_rdma_wr;
+	rdma_wr->wr.sg_list = &isert_cmd->s_ib_sge;
 	rdma_wr->wr.num_sge = 1;
 	rdma_wr->wr.wr_cqe = &isert_cmd->tx_desc.tx_cqe;
-	if (wr->iser_ib_op == ISER_IB_RDMA_WRITE) {
+	if (isert_cmd->iser_ib_op == ISER_IB_RDMA_WRITE) {
 		isert_cmd->tx_desc.tx_cqe.done = isert_rdma_write_done;
 
 		rdma_wr->wr.opcode = IB_WR_RDMA_WRITE;
@@ -2776,7 +2773,7 @@ unmap_cmd:
 		list_add_tail(&fr_desc->list, &isert_conn->fr_pool);
 		spin_unlock_irqrestore(&isert_conn->pool_lock, flags);
 	}
-	isert_unmap_data_buf(isert_conn, &wr->data);
+	isert_unmap_data_buf(isert_conn, &isert_cmd->data);
 
 	return ret;
 }
@@ -2786,7 +2783,6 @@ isert_put_datain(struct iscsi_conn *conn, struct iscsi_cmd *cmd)
 {
 	struct se_cmd *se_cmd = &cmd->se_cmd;
 	struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd);
-	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
 	struct isert_conn *isert_conn = conn->context;
 	struct isert_device *device = isert_conn->device;
 	struct ib_send_wr *wr_failed;
@@ -2795,8 +2791,8 @@ isert_put_datain(struct iscsi_conn *conn, struct iscsi_cmd *cmd)
 	isert_dbg("Cmd: %p RDMA_WRITE data_length: %u\n",
 		 isert_cmd, se_cmd->data_length);
 
-	wr->iser_ib_op = ISER_IB_RDMA_WRITE;
-	rc = device->reg_rdma_mem(conn, cmd, wr);
+	isert_cmd->iser_ib_op = ISER_IB_RDMA_WRITE;
+	rc = device->reg_rdma_mem(isert_cmd, conn);
 	if (rc) {
 		isert_err("Cmd: %p failed to prepare RDMA res\n", isert_cmd);
 		return rc;
@@ -2813,8 +2809,8 @@ isert_put_datain(struct iscsi_conn *conn, struct iscsi_cmd *cmd)
 		isert_init_tx_hdrs(isert_conn, &isert_cmd->tx_desc);
 		isert_init_send_wr(isert_conn, isert_cmd,
 				   &isert_cmd->tx_desc.send_wr);
-		isert_cmd->rdma_wr.s_rdma_wr.wr.next = &isert_cmd->tx_desc.send_wr;
-		wr->rdma_wr_num += 1;
+		isert_cmd->s_rdma_wr.wr.next = &isert_cmd->tx_desc.send_wr;
+		isert_cmd->rdma_wr_num += 1;
 
 		rc = isert_post_recv(isert_conn, isert_cmd->rx_desc);
 		if (rc) {
@@ -2823,7 +2819,7 @@ isert_put_datain(struct iscsi_conn *conn, struct iscsi_cmd *cmd)
 		}
 	}
 
-	rc = ib_post_send(isert_conn->qp, &wr->rdma_wr->wr, &wr_failed);
+	rc = ib_post_send(isert_conn->qp, &isert_cmd->rdma_wr->wr, &wr_failed);
 	if (rc)
 		isert_warn("ib_post_send() failed for IB_WR_RDMA_WRITE\n");
 
@@ -2842,7 +2838,6 @@ isert_get_dataout(struct iscsi_conn *conn, struct iscsi_cmd *cmd, bool recovery)
 {
 	struct se_cmd *se_cmd = &cmd->se_cmd;
 	struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd);
-	struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
 	struct isert_conn *isert_conn = conn->context;
 	struct isert_device *device = isert_conn->device;
 	struct ib_send_wr *wr_failed;
@@ -2850,14 +2845,14 @@ isert_get_dataout(struct iscsi_conn *conn, struct iscsi_cmd *cmd, bool recovery)
 
 	isert_dbg("Cmd: %p RDMA_READ data_length: %u write_data_done: %u\n",
 		 isert_cmd, se_cmd->data_length, cmd->write_data_done);
-	wr->iser_ib_op = ISER_IB_RDMA_READ;
-	rc = device->reg_rdma_mem(conn, cmd, wr);
+	isert_cmd->iser_ib_op = ISER_IB_RDMA_READ;
+	rc = device->reg_rdma_mem(isert_cmd, conn);
 	if (rc) {
 		isert_err("Cmd: %p failed to prepare RDMA res\n", isert_cmd);
 		return rc;
 	}
 
-	rc = ib_post_send(isert_conn->qp, &wr->rdma_wr->wr, &wr_failed);
+	rc = ib_post_send(isert_conn->qp, &isert_cmd->rdma_wr->wr, &wr_failed);
 	if (rc)
 		isert_warn("ib_post_send() failed for IB_WR_RDMA_READ\n");
 
diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h
index df494c7..6103e68 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.h
+++ b/drivers/infiniband/ulp/isert/ib_isert.h
@@ -154,20 +154,6 @@ enum {
 	SIG = 2,
 };
 
-struct isert_rdma_wr {
-	struct isert_cmd	*isert_cmd;
-	enum iser_ib_op_code	iser_ib_op;
-	struct ib_sge		*ib_sge;
-	struct ib_sge		s_ib_sge;
-	int			rdma_wr_num;
-	struct ib_rdma_wr	*rdma_wr;
-	struct ib_rdma_wr	s_rdma_wr;
-	struct ib_sge		ib_sg[3];
-	struct isert_data_buf	data;
-	struct isert_data_buf	prot;
-	struct fast_reg_descriptor *fr_desc;
-};
-
 struct isert_cmd {
 	uint32_t		read_stag;
 	uint32_t		write_stag;
@@ -180,7 +166,16 @@ struct isert_cmd {
 	struct iscsi_cmd	*iscsi_cmd;
 	struct iser_tx_desc	tx_desc;
 	struct iser_rx_desc	*rx_desc;
-	struct isert_rdma_wr	rdma_wr;
+	enum iser_ib_op_code	iser_ib_op;
+	struct ib_sge		*ib_sge;
+	struct ib_sge		s_ib_sge;
+	int			rdma_wr_num;
+	struct ib_rdma_wr	*rdma_wr;
+	struct ib_rdma_wr	s_rdma_wr;
+	struct ib_sge		ib_sg[3];
+	struct isert_data_buf	data;
+	struct isert_data_buf	prot;
+	struct fast_reg_descriptor *fr_desc;
 	struct work_struct	comp_work;
 	struct scatterlist	sg;
 };
@@ -248,9 +243,8 @@ struct isert_device {
 	struct isert_comp	*comps;
 	int                     comps_used;
 	struct list_head	dev_node;
-	int			(*reg_rdma_mem)(struct iscsi_conn *conn,
-						    struct iscsi_cmd *cmd,
-						    struct isert_rdma_wr *wr);
+	int			(*reg_rdma_mem)(struct isert_cmd *isert_cmd,
+						struct iscsi_conn *conn);
 	void			(*unreg_rdma_mem)(struct isert_cmd *isert_cmd,
 						  struct isert_conn *isert_conn);
 };
-- 
2.1.4

Re: [PATCH 4/5] IB/isert: Kill struct isert_rdma_wr

[Sagi Grimberg]

> There is exactly one instance per struct isert_cmd, so merge the two to
> simplify everyones life.

Agreed,

Reviewed-by: Sagi Grimberg <sagig@mellanox.com>