[pull request][rdma-next v2 00/23] RDMA core, drivers and IPoIB fixes

[pull request][rdma-next v2 00/23] RDMA core, drivers and IPoIB fixes

[Leon Romanovsky]

Hi,

This is bunch of fixes to the RDMA stack which we found during code review,
static checkers cleanups and by various code inspections.

There is one exception to this batch - patch from Feras to add
statistics support to SRIOV VF in IPoIB.

However, the diff stat of that change is so small and I didn't find
other batch to attach it.
 drivers/infiniband/ulp/ipoib/ipoib_main.c | 1 +
 1 file changed, 1 insertion(+)

Thanks

Changelog
v1->v2:
 * Dropped patch to remove empty gid function from hns, till proper
   solution will be found.
 * Added two new patches to rxe and mlx5 - add static annotation.

v0->v1:
 * Patch 20: Reused memchr_inv function instead of open-coded variant
 * Added Dennis's ROB tags

----------------------------------------------------------------

The following changes since commit 3e5f0881f17525e3b49835947a5e0cf2d681b1e2:

  IB/hns: Avoid compile test under non 64bit environments (2017-08-14 11:16:53 -0400)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/leon/linux-rdma.git tags/rdma-next-2017-08-15

for you to fetch changes up to 189964826465f24557a46c2c86673ab6e1056c39:

  RDMA/mlx5: Limit scope of get vector affinity local function (2017-08-15 11:46:32 +0300)

----------------------------------------------------------------
RDMA core, drivers and IPoIB fixes

----------------------------------------------------------------
Erez Shitrit (1):
      IB/ipoib: Sync between remove_one to sysfs calls that use rtnl_lock

Feras Daoud (1):
      IB/ipoib: Add get statistics support to SRIOV VF

Guy Levi (4):
      IB/mlx4: Fix RSS QP type in creation verb
      IB/mlx4: Fix struct mlx4_ib_create_wq alignment
      IB/mlx4: Remove redundant attribute in mlx4_ib_create_qp_rss struct
      IB/mlx4: Check that reserved fields in mlx4_ib_create_qp_rss are zero

Kamal Heib (1):
      IB/rxe: Make rxe_counter_name static

Leon Romanovsky (11):
      IB/cma: Fix erroneous validation of supported default GID type
      RDMA/mlx4: Don't use uninitialized variable
      RDMA/(core,ulp): Convert register/unregister event handler to be void
      RDMA/core: Cleanup device capability enum
      RDMA/core: Remove unimplemented node_types and node transport
      RDMA/core: Delete BUG() from unreachable flow
      RDMA/core: Refactor get link layer wrapper
      RDMA/mlx4: Remove gfp_mask argument from acquire_group call
      RDMA/usnic: Fix remove address space warning
      RDMA/mthca: Make explicit conversion to 64bit value
      RDMA/mlx5: Limit scope of get vector affinity local function

Maor Gottlieb (2):
      RDMA/mlx4: Fix create qp command alignment
      IB/mlx5: Add necessary delay drop assignment

Parav Pandit (1):
      IB/uverbs: Introduce and use helper functions to copy ah attributes

Talat Batheesh (2):
      IB/mlx4: Fix some spelling mistakes
      IB/mlx5: Fix some spelling mistakes

 drivers/infiniband/core/cache.c                 |  23 ++---
 drivers/infiniband/core/cma.c                   |   4 +-
 drivers/infiniband/core/device.c                |   8 +-
 drivers/infiniband/core/sa_query.c              |   3 +-
 drivers/infiniband/core/sysfs.c                 |   3 -
 drivers/infiniband/core/uverbs_cmd.c            | 124 ++++++++++--------------
 drivers/infiniband/core/uverbs_main.c           |  13 +--
 drivers/infiniband/core/verbs.c                 |  32 ++----
 drivers/infiniband/hw/mlx4/alias_GUID.c         |   2 +-
 drivers/infiniband/hw/mlx4/cq.c                 |   2 +-
 drivers/infiniband/hw/mlx4/mcg.c                |   9 +-
 drivers/infiniband/hw/mlx4/qp.c                 |  16 +--
 drivers/infiniband/hw/mlx5/cq.c                 |   2 +-
 drivers/infiniband/hw/mlx5/mad.c                |   2 +-
 drivers/infiniband/hw/mlx5/main.c               |   6 +-
 drivers/infiniband/hw/mthca/mthca_cmd.c         |   2 +-
 drivers/infiniband/hw/usnic/usnic_fwd.c         |  12 +--
 drivers/infiniband/hw/usnic/usnic_fwd.h         |   2 +-
 drivers/infiniband/hw/usnic/usnic_ib_main.c     |  10 +-
 drivers/infiniband/sw/rxe/rxe_hw_counters.c     |   2 +-
 drivers/infiniband/ulp/ipoib/ipoib.h            |   1 +
 drivers/infiniband/ulp/ipoib/ipoib_cm.c         |   8 +-
 drivers/infiniband/ulp/ipoib/ipoib_main.c       |  16 ++-
 drivers/infiniband/ulp/ipoib/ipoib_vlan.c       |  22 ++++-
 drivers/infiniband/ulp/iser/iser_verbs.c        |   6 +-
 drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c |   7 +-
 drivers/infiniband/ulp/srpt/ib_srpt.c           |   5 +-
 include/rdma/ib_verbs.h                         |  26 +++--
 include/uapi/rdma/mlx4-abi.h                    |   6 +-
 29 files changed, 156 insertions(+), 218 deletions(-)
--
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

[rdma-next v2 01/23] IB/cma: Fix erroneous validation of supported default GID type

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

When rdma_cm is initializing a cma_device it checks if this device
supports the preferred default GID type. This check was done in a wrong way
and therefore sometimes rdma_cm is coming up with default GID type that is
not supported by the device.

Fix that by checking for supported GID type properly.

Fixes: 3c7f67d1880d ("IB/cma: Fix default RoCE type setting")
Signed-off-by: Moni Shoua <monis-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
 drivers/infiniband/core/cma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index d8edd8b11561..f5f88b427328 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -72,7 +72,7 @@ MODULE_LICENSE("Dual BSD/GPL");
 #define CMA_MAX_CM_RETRIES 15
 #define CMA_CM_MRA_SETTING (IB_CM_MRA_FLAG_DELAY | 24)
 #define CMA_IBOE_PACKET_LIFETIME 18
-#define CMA_PREFERRED_ROCE_GID_TYPE (1 << IB_GID_TYPE_ROCE_UDP_ENCAP)
+#define CMA_PREFERRED_ROCE_GID_TYPE IB_GID_TYPE_ROCE_UDP_ENCAP
 
 static const char * const cma_events[] = {
 	[RDMA_CM_EVENT_ADDR_RESOLVED]	 = "address resolved",
@@ -4282,7 +4282,7 @@ static void cma_add_one(struct ib_device *device)
 	for (i = rdma_start_port(device); i <= rdma_end_port(device); i++) {
 		supported_gids = roce_gid_type_mask_support(device, i);
 		WARN_ON(!supported_gids);
-		if (supported_gids & CMA_PREFERRED_ROCE_GID_TYPE)
+		if (supported_gids & (1 << CMA_PREFERRED_ROCE_GID_TYPE))
 			cma_dev->default_gid_type[i - rdma_start_port(device)] =
 				CMA_PREFERRED_ROCE_GID_TYPE;
 		else
-- 
2.14.0

--
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

[rdma-next v2 02/23] IB/uverbs: Introduce and use helper functions to copy ah attributes

[Leon Romanovsky]

From: Parav Pandit <parav-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

This patch introduces two helper functions to copy ah attributes
from uverbs to internal ib_ah_attr structure and the other way
during modify qp and query qp respectively.

Signed-off-by: Parav Pandit <parav-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/core/uverbs_cmd.c | 124 ++++++++++++++---------------------
 1 file changed, 49 insertions(+), 75 deletions(-)

diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index 670176b670a0..515425a50059 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -1810,6 +1810,28 @@ ssize_t ib_uverbs_open_qp(struct ib_uverbs_file *file,
 	return ret;
 }
 
+static void copy_ah_attr_to_uverbs(struct ib_uverbs_qp_dest *uverb_attr,
+				   struct rdma_ah_attr *rdma_attr)
+{
+	const struct ib_global_route   *grh;
+
+	uverb_attr->dlid              = rdma_ah_get_dlid(rdma_attr);
+	uverb_attr->sl                = rdma_ah_get_sl(rdma_attr);
+	uverb_attr->src_path_bits     = rdma_ah_get_path_bits(rdma_attr);
+	uverb_attr->static_rate       = rdma_ah_get_static_rate(rdma_attr);
+	uverb_attr->is_global         = !!(rdma_ah_get_ah_flags(rdma_attr) &
+					 IB_AH_GRH);
+	if (uverb_attr->is_global) {
+		grh = rdma_ah_read_grh(rdma_attr);
+		memcpy(uverb_attr->dgid, grh->dgid.raw, 16);
+		uverb_attr->flow_label        = grh->flow_label;
+		uverb_attr->sgid_index        = grh->sgid_index;
+		uverb_attr->hop_limit         = grh->hop_limit;
+		uverb_attr->traffic_class     = grh->traffic_class;
+	}
+	uverb_attr->port_num          = rdma_ah_get_port_num(rdma_attr);
+}
+
 ssize_t ib_uverbs_query_qp(struct ib_uverbs_file *file,
 			   struct ib_device *ib_dev,
 			   const char __user *buf, int in_len,
@@ -1820,7 +1842,6 @@ ssize_t ib_uverbs_query_qp(struct ib_uverbs_file *file,
 	struct ib_qp                   *qp;
 	struct ib_qp_attr              *attr;
 	struct ib_qp_init_attr         *init_attr;
-	const struct ib_global_route   *grh;
 	int                            ret;
 
 	if (copy_from_user(&cmd, buf, sizeof cmd))
@@ -1870,39 +1891,8 @@ ssize_t ib_uverbs_query_qp(struct ib_uverbs_file *file,
 	resp.alt_port_num           = attr->alt_port_num;
 	resp.alt_timeout            = attr->alt_timeout;
 
-	resp.dest.dlid              = rdma_ah_get_dlid(&attr->ah_attr);
-	resp.dest.sl                = rdma_ah_get_sl(&attr->ah_attr);
-	resp.dest.src_path_bits     = rdma_ah_get_path_bits(&attr->ah_attr);
-	resp.dest.static_rate       = rdma_ah_get_static_rate(&attr->ah_attr);
-	resp.dest.is_global         = !!(rdma_ah_get_ah_flags(&attr->ah_attr) &
-					 IB_AH_GRH);
-	if (resp.dest.is_global) {
-		grh = rdma_ah_read_grh(&attr->ah_attr);
-		memcpy(resp.dest.dgid, grh->dgid.raw, 16);
-		resp.dest.flow_label        = grh->flow_label;
-		resp.dest.sgid_index        = grh->sgid_index;
-		resp.dest.hop_limit         = grh->hop_limit;
-		resp.dest.traffic_class     = grh->traffic_class;
-	}
-	resp.dest.port_num          = rdma_ah_get_port_num(&attr->ah_attr);
-
-	resp.alt_dest.dlid          = rdma_ah_get_dlid(&attr->alt_ah_attr);
-	resp.alt_dest.sl            = rdma_ah_get_sl(&attr->alt_ah_attr);
-	resp.alt_dest.src_path_bits = rdma_ah_get_path_bits(&attr->alt_ah_attr);
-	resp.alt_dest.static_rate
-			= rdma_ah_get_static_rate(&attr->alt_ah_attr);
-	resp.alt_dest.is_global
-			= !!(rdma_ah_get_ah_flags(&attr->alt_ah_attr) &
-						  IB_AH_GRH);
-	if (resp.alt_dest.is_global) {
-		grh = rdma_ah_read_grh(&attr->alt_ah_attr);
-		memcpy(resp.alt_dest.dgid, grh->dgid.raw, 16);
-		resp.alt_dest.flow_label    = grh->flow_label;
-		resp.alt_dest.sgid_index    = grh->sgid_index;
-		resp.alt_dest.hop_limit     = grh->hop_limit;
-		resp.alt_dest.traffic_class = grh->traffic_class;
-	}
-	resp.alt_dest.port_num      = rdma_ah_get_port_num(&attr->alt_ah_attr);
+	copy_ah_attr_to_uverbs(&resp.dest, &attr->ah_attr);
+	copy_ah_attr_to_uverbs(&resp.alt_dest, &attr->alt_ah_attr);
 
 	resp.max_send_wr            = init_attr->cap.max_send_wr;
 	resp.max_recv_wr            = init_attr->cap.max_recv_wr;
@@ -1936,6 +1926,28 @@ static int modify_qp_mask(enum ib_qp_type qp_type, int mask)
 	}
 }
 
+static void copy_ah_attr_from_uverbs(struct ib_device *dev,
+				     struct rdma_ah_attr *rdma_attr,
+				     struct ib_uverbs_qp_dest *uverb_attr)
+{
+	rdma_attr->type = rdma_ah_find_type(dev, uverb_attr->port_num);
+	if (uverb_attr->is_global) {
+		rdma_ah_set_grh(rdma_attr, NULL,
+				uverb_attr->flow_label,
+				uverb_attr->sgid_index,
+				uverb_attr->hop_limit,
+				uverb_attr->traffic_class);
+		rdma_ah_set_dgid_raw(rdma_attr, uverb_attr->dgid);
+	} else {
+		rdma_ah_set_ah_flags(rdma_attr, 0);
+	}
+	rdma_ah_set_dlid(rdma_attr, uverb_attr->dlid);
+	rdma_ah_set_sl(rdma_attr, uverb_attr->sl);
+	rdma_ah_set_path_bits(rdma_attr, uverb_attr->src_path_bits);
+	rdma_ah_set_static_rate(rdma_attr, uverb_attr->static_rate);
+	rdma_ah_set_port_num(rdma_attr, uverb_attr->port_num);
+}
+
 static int modify_qp(struct ib_uverbs_file *file,
 		     struct ib_uverbs_ex_modify_qp *cmd, struct ib_udata *udata)
 {
@@ -1982,47 +1994,9 @@ static int modify_qp(struct ib_uverbs_file *file,
 	attr->alt_timeout	  = cmd->base.alt_timeout;
 	attr->rate_limit	  = cmd->rate_limit;
 
-	attr->ah_attr.type = rdma_ah_find_type(qp->device,
-					       cmd->base.dest.port_num);
-	if (cmd->base.dest.is_global) {
-		rdma_ah_set_grh(&attr->ah_attr, NULL,
-				cmd->base.dest.flow_label,
-				cmd->base.dest.sgid_index,
-				cmd->base.dest.hop_limit,
-				cmd->base.dest.traffic_class);
-		rdma_ah_set_dgid_raw(&attr->ah_attr, cmd->base.dest.dgid);
-	} else {
-		rdma_ah_set_ah_flags(&attr->ah_attr, 0);
-	}
-	rdma_ah_set_dlid(&attr->ah_attr, cmd->base.dest.dlid);
-	rdma_ah_set_sl(&attr->ah_attr, cmd->base.dest.sl);
-	rdma_ah_set_path_bits(&attr->ah_attr, cmd->base.dest.src_path_bits);
-	rdma_ah_set_static_rate(&attr->ah_attr, cmd->base.dest.static_rate);
-	rdma_ah_set_port_num(&attr->ah_attr,
-			     cmd->base.dest.port_num);
-
-	attr->alt_ah_attr.type = rdma_ah_find_type(qp->device,
-						   cmd->base.dest.port_num);
-	if (cmd->base.alt_dest.is_global) {
-		rdma_ah_set_grh(&attr->alt_ah_attr, NULL,
-				cmd->base.alt_dest.flow_label,
-				cmd->base.alt_dest.sgid_index,
-				cmd->base.alt_dest.hop_limit,
-				cmd->base.alt_dest.traffic_class);
-		rdma_ah_set_dgid_raw(&attr->alt_ah_attr,
-				     cmd->base.alt_dest.dgid);
-	} else {
-		rdma_ah_set_ah_flags(&attr->alt_ah_attr, 0);
-	}
-
-	rdma_ah_set_dlid(&attr->alt_ah_attr, cmd->base.alt_dest.dlid);
-	rdma_ah_set_sl(&attr->alt_ah_attr, cmd->base.alt_dest.sl);
-	rdma_ah_set_path_bits(&attr->alt_ah_attr,
-			      cmd->base.alt_dest.src_path_bits);
-	rdma_ah_set_static_rate(&attr->alt_ah_attr,
-				cmd->base.alt_dest.static_rate);
-	rdma_ah_set_port_num(&attr->alt_ah_attr,
-			     cmd->base.alt_dest.port_num);
+	copy_ah_attr_from_uverbs(qp->device, &attr->ah_attr, &cmd->base.dest);
+	copy_ah_attr_from_uverbs(qp->device, &attr->alt_ah_attr,
+				 &cmd->base.alt_dest);
 
 	ret = ib_modify_qp_with_udata(qp, attr,
 				      modify_qp_mask(qp->qp_type,
-- 
2.14.0

--
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

[rdma-next v2 03/23] RDMA/mlx4: Don't use uninitialized variable

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Avoid usage of uninitialized variable.

Fixes: 3078f5f1bd8b ("IB/mlx4: Add support for RSS QP")
Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
 drivers/infiniband/hw/mlx4/qp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index e42acfb20588..793d77f29d3e 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -2034,8 +2034,8 @@ static u8 gid_type_to_qpc(enum ib_gid_type gid_type)
  */
 static int bringup_rss_rwqs(struct ib_rwq_ind_table *ind_tbl, u8 port_num)
 {
+	int err = 0;
 	int i;
-	int err;
 
 	for (i = 0; i < (1 << ind_tbl->log_ind_tbl_size); i++) {
 		struct ib_wq *ibwq = ind_tbl->ind_tbl[i];
-- 
2.14.0

--
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

[rdma-next v2 04/23] RDMA/mlx4: Fix create qp command alignment

[Leon Romanovsky]

From: Maor Gottlieb <maorg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Avoid extra padding by replacing the order of inl_recv_sz and reserved,
otherwise 'mlx4_ib_create_qp' structure might be larger than legacy user
input leading to copy of some garbage data from the user space buffer.

Fixes: ea30b966f7dd ('IB/mlx4: Add inline-receive support')
Signed-off-by: Maor Gottlieb <maorg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 include/uapi/rdma/mlx4-abi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/uapi/rdma/mlx4-abi.h b/include/uapi/rdma/mlx4-abi.h
index d915cab37ec3..21cce1a4c4dd 100644
--- a/include/uapi/rdma/mlx4-abi.h
+++ b/include/uapi/rdma/mlx4-abi.h
@@ -111,8 +111,8 @@ struct mlx4_ib_create_qp {
 	__u8	log_sq_bb_count;
 	__u8	log_sq_stride;
 	__u8	sq_no_prefetch;
-	__u32	inl_recv_sz;
 	__u8	reserved;
+	__u32	inl_recv_sz;
 };
 
 struct mlx4_ib_create_wq {
-- 
2.14.0

--
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

[rdma-next v2 05/23] RDMA/(core,ulp): Convert register/unregister event handler to be void

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

The functions ib_register_event_handler() and
ib_unregister_event_handler() always returned success and they can't fail.

Let's convert those functions to be void, remove redundant checks and
cleanup tons of goto statements.

Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/core/cache.c                 | 23 ++++++++---------------
 drivers/infiniband/core/device.c                |  8 ++------
 drivers/infiniband/core/sa_query.c              |  3 +--
 drivers/infiniband/core/uverbs_main.c           | 13 +------------
 drivers/infiniband/ulp/ipoib/ipoib_main.c       | 10 +---------
 drivers/infiniband/ulp/iser/iser_verbs.c        |  6 ++----
 drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c |  7 +------
 drivers/infiniband/ulp/srpt/ib_srpt.c           |  5 ++---
 include/rdma/ib_verbs.h                         |  4 ++--
 9 files changed, 20 insertions(+), 59 deletions(-)

diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c
index efc94304dee3..77515638c55c 100644
--- a/drivers/infiniband/core/cache.c
+++ b/drivers/infiniband/core/cache.c
@@ -1199,30 +1199,23 @@ int ib_cache_setup_one(struct ib_device *device)
 	device->cache.ports =
 		kzalloc(sizeof(*device->cache.ports) *
 			(rdma_end_port(device) - rdma_start_port(device) + 1), GFP_KERNEL);
-	if (!device->cache.ports) {
-		err = -ENOMEM;
-		goto out;
-	}
+	if (!device->cache.ports)
+		return -ENOMEM;
 
 	err = gid_table_setup_one(device);
-	if (err)
-		goto out;
+	if (err) {
+		kfree(device->cache.ports);
+		device->cache.ports = NULL;
+		return err;
+	}
 
 	for (p = 0; p <= rdma_end_port(device) - rdma_start_port(device); ++p)
 		ib_cache_update(device, p + rdma_start_port(device), true);
 
 	INIT_IB_EVENT_HANDLER(&device->cache.event_handler,
 			      device, ib_cache_event);
-	err = ib_register_event_handler(&device->cache.event_handler);
-	if (err)
-		goto err;
-
+	ib_register_event_handler(&device->cache.event_handler);
 	return 0;
-
-err:
-	gid_table_cleanup_one(device);
-out:
-	return err;
 }
 
 void ib_cache_release_one(struct ib_device *device)
diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index fbc92c649be8..475b93d62748 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -746,7 +746,7 @@ EXPORT_SYMBOL(ib_set_client_data);
  * chapter 11 of the InfiniBand Architecture Specification).  This
  * callback may occur in interrupt context.
  */
-int ib_register_event_handler  (struct ib_event_handler *event_handler)
+void ib_register_event_handler(struct ib_event_handler *event_handler)
 {
 	unsigned long flags;
 
@@ -754,8 +754,6 @@ int ib_register_event_handler  (struct ib_event_handler *event_handler)
 	list_add_tail(&event_handler->list,
 		      &event_handler->device->event_handler_list);
 	spin_unlock_irqrestore(&event_handler->device->event_handler_lock, flags);
-
-	return 0;
 }
 EXPORT_SYMBOL(ib_register_event_handler);
 
@@ -766,15 +764,13 @@ EXPORT_SYMBOL(ib_register_event_handler);
  * Unregister an event handler registered with
  * ib_register_event_handler().
  */
-int ib_unregister_event_handler(struct ib_event_handler *event_handler)
+void ib_unregister_event_handler(struct ib_event_handler *event_handler)
 {
 	unsigned long flags;
 
 	spin_lock_irqsave(&event_handler->device->event_handler_lock, flags);
 	list_del(&event_handler->list);
 	spin_unlock_irqrestore(&event_handler->device->event_handler_lock, flags);
-
-	return 0;
 }
 EXPORT_SYMBOL(ib_unregister_event_handler);
 
diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
index da29e2863c84..14aef0dd8625 100644
--- a/drivers/infiniband/core/sa_query.c
+++ b/drivers/infiniband/core/sa_query.c
@@ -2408,8 +2408,7 @@ static void ib_sa_add_one(struct ib_device *device)
 	 */
 
 	INIT_IB_EVENT_HANDLER(&sa_dev->event_handler, device, ib_sa_event);
-	if (ib_register_event_handler(&sa_dev->event_handler))
-		goto err;
+	ib_register_event_handler(&sa_dev->event_handler);
 
 	for (i = 0; i <= e - s; ++i) {
 		if (rdma_cap_ib_sa(device, i + 1))
diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c
index 3d2609608f58..1ac99aec3d4e 100644
--- a/drivers/infiniband/core/uverbs_main.c
+++ b/drivers/infiniband/core/uverbs_main.c
@@ -594,7 +594,6 @@ struct file *ib_uverbs_alloc_async_event_file(struct ib_uverbs_file *uverbs_file
 {
 	struct ib_uverbs_async_event_file *ev_file;
 	struct file *filp;
-	int ret;
 
 	ev_file = kzalloc(sizeof(*ev_file), GFP_KERNEL);
 	if (!ev_file)
@@ -620,21 +619,11 @@ struct file *ib_uverbs_alloc_async_event_file(struct ib_uverbs_file *uverbs_file
 	INIT_IB_EVENT_HANDLER(&uverbs_file->event_handler,
 			      ib_dev,
 			      ib_uverbs_event_handler);
-	ret = ib_register_event_handler(&uverbs_file->event_handler);
-	if (ret)
-		goto err_put_file;
-
+	ib_register_event_handler(&uverbs_file->event_handler);
 	/* At that point async file stuff was fully set */
 
 	return filp;
 
-err_put_file:
-	fput(filp);
-	kref_put(&uverbs_file->async_file->ref,
-		 ib_uverbs_release_async_event_file);
-	uverbs_file->async_file = NULL;
-	return ERR_PTR(ret);
-
 err_put_refs:
 	kref_put(&ev_file->uverbs_file->ref, ib_uverbs_release_file);
 	kref_put(&ev_file->ref, ib_uverbs_release_async_event_file);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 7dcdbbacbf46..645217232250 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -2223,13 +2223,7 @@ static struct net_device *ipoib_add_port(const char *format,
 
 	INIT_IB_EVENT_HANDLER(&priv->event_handler,
 			      priv->ca, ipoib_event);
-	result = ib_register_event_handler(&priv->event_handler);
-	if (result < 0) {
-		printk(KERN_WARNING "%s: ib_register_event_handler failed for "
-		       "port %d (ret = %d)\n",
-		       hca->name, port, result);
-		goto event_failed;
-	}
+	ib_register_event_handler(&priv->event_handler);
 
 	result = register_netdev(priv->dev);
 	if (result) {
@@ -2262,8 +2256,6 @@ static struct net_device *ipoib_add_port(const char *format,
 	set_bit(IPOIB_STOP_NEIGH_GC, &priv->flags);
 	cancel_delayed_work(&priv->neigh_reap_task);
 	flush_workqueue(priv->wq);
-
-event_failed:
 	ipoib_dev_cleanup(priv->dev);
 
 device_init_failed:
diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c
index 26a004e97ae0..55a73b0ed4c6 100644
--- a/drivers/infiniband/ulp/iser/iser_verbs.c
+++ b/drivers/infiniband/ulp/iser/iser_verbs.c
@@ -106,9 +106,7 @@ static int iser_create_device_ib_res(struct iser_device *device)
 
 	INIT_IB_EVENT_HANDLER(&device->event_handler, ib_dev,
 			      iser_event_handler);
-	if (ib_register_event_handler(&device->event_handler))
-		goto cq_err;
-
+	ib_register_event_handler(&device->event_handler);
 	return 0;
 
 cq_err:
@@ -141,7 +139,7 @@ static void iser_free_device_ib_res(struct iser_device *device)
 		comp->cq = NULL;
 	}
 
-	(void)ib_unregister_event_handler(&device->event_handler);
+	ib_unregister_event_handler(&device->event_handler);
 	ib_dealloc_pd(device->pd);
 
 	kfree(device->comps);
diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c
index 57b862b94dca..21f0b481edcc 100644
--- a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c
+++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c
@@ -954,12 +954,7 @@ static int vema_register(struct opa_vnic_ctrl_port *cport)
 
 		INIT_IB_EVENT_HANDLER(&port->event_handler,
 				      cport->ibdev, opa_vnic_event);
-		ret = ib_register_event_handler(&port->event_handler);
-		if (ret) {
-			c_err("port %d: event handler register failed\n", i);
-			vema_unregister(cport);
-			return ret;
-		}
+		ib_register_event_handler(&port->event_handler);
 
 		idr_init(&port->vport_idr);
 		mutex_init(&port->lock);
diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c
index 402275be0931..9e8e9220f816 100644
--- a/drivers/infiniband/ulp/srpt/ib_srpt.c
+++ b/drivers/infiniband/ulp/srpt/ib_srpt.c
@@ -2238,7 +2238,7 @@ static int srpt_write_pending(struct se_cmd *se_cmd)
 				cqe, first_wr);
 		cqe = NULL;
 	}
-	
+
 	ret = ib_post_send(ch->qp, first_wr, &bad_wr);
 	if (ret) {
 		pr_err("%s: ib_post_send() returned %d for %d (avail: %d)\n",
@@ -2530,8 +2530,7 @@ static void srpt_add_one(struct ib_device *device)
 
 	INIT_IB_EVENT_HANDLER(&sdev->event_handler, sdev->device,
 			      srpt_event_handler);
-	if (ib_register_event_handler(&sdev->event_handler))
-		goto err_cm;
+	ib_register_event_handler(&sdev->event_handler);
 
 	sdev->ioctx_ring = (struct srpt_recv_ioctx **)
 		srpt_alloc_ioctx_ring(sdev, sdev->srq_size,
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 4db4ad56ace6..3c9c514bcd7d 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -2411,8 +2411,8 @@ int ib_modify_qp_is_ok(enum ib_qp_state cur_state, enum ib_qp_state next_state,
 		       enum ib_qp_type type, enum ib_qp_attr_mask mask,
 		       enum rdma_link_layer ll);
 
-int ib_register_event_handler  (struct ib_event_handler *event_handler);
-int ib_unregister_event_handler(struct ib_event_handler *event_handler);
+void ib_register_event_handler(struct ib_event_handler *event_handler);
+void ib_unregister_event_handler(struct ib_event_handler *event_handler);
 void ib_dispatch_event(struct ib_event *event);
 
 int ib_query_port(struct ib_device *device,
-- 
2.14.0

--
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

[rdma-next v2 06/23] RDMA/core: Cleanup device capability enum

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Cleanup patch prior exporting the ib_device_cap_flags
to the user space. In this patch, we are aligning the
indentation, removing IB_DEVICE_INIT_TYPE and IB_DEVICE_RESERVED
fields, because it is not used in the kernel.

Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 include/rdma/ib_verbs.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 3c9c514bcd7d..4ce188128aa9 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -170,7 +170,7 @@ enum ib_device_cap_flags {
 	IB_DEVICE_UD_AV_PORT_ENFORCE		= (1 << 6),
 	IB_DEVICE_CURR_QP_STATE_MOD		= (1 << 7),
 	IB_DEVICE_SHUTDOWN_PORT			= (1 << 8),
-	IB_DEVICE_INIT_TYPE			= (1 << 9),
+	/* Not in use, former INIT_TYPE		= (1 << 9),*/
 	IB_DEVICE_PORT_ACTIVE_EVENT		= (1 << 10),
 	IB_DEVICE_SYS_IMAGE_GUID		= (1 << 11),
 	IB_DEVICE_RC_RNR_NAK_GEN		= (1 << 12),
@@ -185,7 +185,7 @@ enum ib_device_cap_flags {
 	 * which will always contain a usable lkey.
 	 */
 	IB_DEVICE_LOCAL_DMA_LKEY		= (1 << 15),
-	IB_DEVICE_RESERVED /* old SEND_W_INV */	= (1 << 16),
+	/* Reserved, old SEND_W_INV		= (1 << 16),*/
 	IB_DEVICE_MEM_WINDOW			= (1 << 17),
 	/*
 	 * Devices should set IB_DEVICE_UD_IP_SUM if they support
@@ -220,7 +220,7 @@ enum ib_device_cap_flags {
 	 * of I/O operations with single completion queue managed
 	 * by hardware.
 	 */
-	IB_DEVICE_CROSS_CHANNEL		= (1 << 27),
+	IB_DEVICE_CROSS_CHANNEL			= (1 << 27),
 	IB_DEVICE_MANAGED_FLOW_STEERING		= (1 << 29),
 	IB_DEVICE_SIGNATURE_HANDOVER		= (1 << 30),
 	IB_DEVICE_ON_DEMAND_PAGING		= (1ULL << 31),
-- 
2.14.0

--
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

[rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

There is no need to carry code, which is not implemented in any
underlying hardware. This patch removes unimplemented node_types and
their respective node transport layers.

Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/core/sysfs.c |  3 ---
 drivers/infiniband/core/verbs.c |  5 -----
 include/rdma/ib_verbs.h         | 16 ++++++----------
 3 files changed, 6 insertions(+), 18 deletions(-)

diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c
index abc5ab581f82..c43280f8d5b3 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -1146,10 +1146,7 @@ static ssize_t show_node_type(struct device *device,
 	switch (dev->node_type) {
 	case RDMA_NODE_IB_CA:	  return sprintf(buf, "%d: CA\n", dev->node_type);
 	case RDMA_NODE_RNIC:	  return sprintf(buf, "%d: RNIC\n", dev->node_type);
-	case RDMA_NODE_USNIC:	  return sprintf(buf, "%d: usNIC\n", dev->node_type);
 	case RDMA_NODE_USNIC_UDP: return sprintf(buf, "%d: usNIC UDP\n", dev->node_type);
-	case RDMA_NODE_IB_SWITCH: return sprintf(buf, "%d: switch\n", dev->node_type);
-	case RDMA_NODE_IB_ROUTER: return sprintf(buf, "%d: router\n", dev->node_type);
 	default:		  return sprintf(buf, "%d: <unknown>\n", dev->node_type);
 	}
 }
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 3d1de62de839..21fef6b6d4f3 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -182,13 +182,9 @@ rdma_node_get_transport(enum rdma_node_type node_type)
 {
 	switch (node_type) {
 	case RDMA_NODE_IB_CA:
-	case RDMA_NODE_IB_SWITCH:
-	case RDMA_NODE_IB_ROUTER:
 		return RDMA_TRANSPORT_IB;
 	case RDMA_NODE_RNIC:
 		return RDMA_TRANSPORT_IWARP;
-	case RDMA_NODE_USNIC:
-		return RDMA_TRANSPORT_USNIC;
 	case RDMA_NODE_USNIC_UDP:
 		return RDMA_TRANSPORT_USNIC_UDP;
 	default:
@@ -207,7 +203,6 @@ enum rdma_link_layer rdma_port_get_link_layer(struct ib_device *device, u8 port_
 	case RDMA_TRANSPORT_IB:
 		return IB_LINK_LAYER_INFINIBAND;
 	case RDMA_TRANSPORT_IWARP:
-	case RDMA_TRANSPORT_USNIC:
 	case RDMA_TRANSPORT_USNIC_UDP:
 		return IB_LINK_LAYER_ETHERNET;
 	default:
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 4ce188128aa9..6aec3971628b 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -95,12 +95,9 @@ struct ib_gid_attr {
 
 enum rdma_node_type {
 	/* IB values map to NodeInfo:NodeType. */
-	RDMA_NODE_IB_CA 	= 1,
-	RDMA_NODE_IB_SWITCH,
-	RDMA_NODE_IB_ROUTER,
-	RDMA_NODE_RNIC,
-	RDMA_NODE_USNIC,
-	RDMA_NODE_USNIC_UDP,
+	RDMA_NODE_IB_CA		= 1,
+	RDMA_NODE_RNIC		= 4,
+	RDMA_NODE_USNIC_UDP	= 6,
 };
 
 enum {
@@ -109,10 +106,9 @@ enum {
 };
 
 enum rdma_transport_type {
-	RDMA_TRANSPORT_IB,
-	RDMA_TRANSPORT_IWARP,
-	RDMA_TRANSPORT_USNIC,
-	RDMA_TRANSPORT_USNIC_UDP
+	RDMA_TRANSPORT_IB		= 0,
+	RDMA_TRANSPORT_IWARP		= 1,
+	RDMA_TRANSPORT_USNIC_UDP	= 3,
 };
 
 enum rdma_protocol_type {
-- 
2.14.0

--
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: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

[Suri Shelvapille]

Leon:

We do have IB_SWITCH and IB_ROUTER functionality and we do use the SWITCH capability provided by the "core" in our product. The "core" has a wrapper function which Identifies the device as a switch to direct smp packets. I am glad that has not been removed. And I hope it won't be removed as it would be a burden to us. Although we prefer to have the IB_SWITCH/IB_ROUTER definitions in the core as well, it is not a big deal to move these definitions inside our driver.

Thanks,
Suri

-----Original Message-----
From: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org [mailto:linux-rdma-owner-u79uwXL29TZUIDd8j+nm9g@public.gmane.org.org] On Behalf Of Leon Romanovsky
Sent: Tuesday, August 15, 2017 4:55 AM
To: Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org; Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>; Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Subject: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

There is no need to carry code, which is not implemented in any underlying hardware. This patch removes unimplemented node_types and their respective node transport layers.

Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/core/sysfs.c |  3 ---  drivers/infiniband/core/verbs.c |  5 -----
 include/rdma/ib_verbs.h         | 16 ++++++----------
 3 files changed, 6 insertions(+), 18 deletions(-)

diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c index abc5ab581f82..c43280f8d5b3 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -1146,10 +1146,7 @@ static ssize_t show_node_type(struct device *device,
 switch (dev->node_type) {
 case RDMA_NODE_IB_CA:  return sprintf(buf, "%d: CA\n", dev->node_type);
 case RDMA_NODE_RNIC:  return sprintf(buf, "%d: RNIC\n", dev->node_type);
-case RDMA_NODE_USNIC:  return sprintf(buf, "%d: usNIC\n", dev->node_type);
 case RDMA_NODE_USNIC_UDP: return sprintf(buf, "%d: usNIC UDP\n", dev->node_type);
-case RDMA_NODE_IB_SWITCH: return sprintf(buf, "%d: switch\n", dev->node_type);
-case RDMA_NODE_IB_ROUTER: return sprintf(buf, "%d: router\n", dev->node_type);
 default:  return sprintf(buf, "%d: <unknown>\n", dev->node_type);
 }
 }
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 3d1de62de839..21fef6b6d4f3 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -182,13 +182,9 @@ rdma_node_get_transport(enum rdma_node_type node_type)  {
 switch (node_type) {
 case RDMA_NODE_IB_CA:
-case RDMA_NODE_IB_SWITCH:
-case RDMA_NODE_IB_ROUTER:
 return RDMA_TRANSPORT_IB;
 case RDMA_NODE_RNIC:
 return RDMA_TRANSPORT_IWARP;
-case RDMA_NODE_USNIC:
-return RDMA_TRANSPORT_USNIC;
 case RDMA_NODE_USNIC_UDP:
 return RDMA_TRANSPORT_USNIC_UDP;
 default:
@@ -207,7 +203,6 @@ enum rdma_link_layer rdma_port_get_link_layer(struct ib_device *device, u8 port_
 case RDMA_TRANSPORT_IB:
 return IB_LINK_LAYER_INFINIBAND;
 case RDMA_TRANSPORT_IWARP:
-case RDMA_TRANSPORT_USNIC:
 case RDMA_TRANSPORT_USNIC_UDP:
 return IB_LINK_LAYER_ETHERNET;
 default:
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 4ce188128aa9..6aec3971628b 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -95,12 +95,9 @@ struct ib_gid_attr {

 enum rdma_node_type {
 /* IB values map to NodeInfo:NodeType. */
-RDMA_NODE_IB_CA = 1,
-RDMA_NODE_IB_SWITCH,
-RDMA_NODE_IB_ROUTER,
-RDMA_NODE_RNIC,
-RDMA_NODE_USNIC,
-RDMA_NODE_USNIC_UDP,
+RDMA_NODE_IB_CA= 1,
+RDMA_NODE_RNIC= 4,
+RDMA_NODE_USNIC_UDP= 6,
 };

 enum {
@@ -109,10 +106,9 @@ enum {
 };

 enum rdma_transport_type {
-RDMA_TRANSPORT_IB,
-RDMA_TRANSPORT_IWARP,
-RDMA_TRANSPORT_USNIC,
-RDMA_TRANSPORT_USNIC_UDP
+RDMA_TRANSPORT_IB= 0,
+RDMA_TRANSPORT_IWARP= 1,
+RDMA_TRANSPORT_USNIC_UDP= 3,
 };

 enum rdma_protocol_type {
--
2.14.0

--
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

This correspondence, and any attachments or files transmitted with this correspondence, contains information which may be confidential and privileged and is intended solely for the use of the addressee. Unless you are the addressee or are authorized to receive messages for the addressee, you may not use, copy, disseminate, or disclose this correspondence or any information contained in this correspondence to any third party. If you have received this correspondence in error, please notify the sender immediately and delete this correspondence and any attachments or files transmitted with this correspondence from your system, and destroy any and all copies thereof, electronic or otherwise. Your cooperation and understanding are greatly appreciated.
--
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: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 673 bytes --]

On Tue, Aug 15, 2017 at 06:25:54PM +0000, Suri Shelvapille wrote:
> Leon:
>
> We do have IB_SWITCH and IB_ROUTER functionality and we do use the SWITCH capability provided by the "core" in our product. The "core" has a wrapper function which Identifies the device as a switch to direct smp packets. I am glad that has not been removed. And I hope it won't be removed as it would be a burden to us. Although we prefer to have the IB_SWITCH/IB_ROUTER definitions in the core as well, it is not a big deal to move these definitions inside our driver.
>

Can you please point me to IN-KERNEL hardware/driver where these values are being set? Because I didn't find any.

Thanks

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

RE: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

[Suri Shelvapille]

Leon:
Our Drivers are not in Kernel (being a small company we don't have resources to make it in-kernel). When I and Hal worked on adding the switch capability into the kernel, we thought it would be useful for the community (as there was interest in a few others besides us). Philosophically, since SWITCH and ROUTER definitions are part of the IB spec, don't you think having these definitions makes the kernel close to spec and hence useful? I know these definitions can be easily added into our drivers.
Just treat this as a request, if it is not an onus to the community, please keep the SWITCH and ROUTER (only) definitions. If you have strong reasons I have no objections to your patches.

Thanks,
Suri

-----Original Message-----
From: Leon Romanovsky [mailto:leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org]
Sent: Wednesday, August 16, 2017 1:38 AM
To: Suri Shelvapille <suri-lFb5ksp6isWhEniVeURVKkEOCMrvLtNR@public.gmane.org>
Cc: Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>; linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

On Tue, Aug 15, 2017 at 06:25:54PM +0000, Suri Shelvapille wrote:
> Leon:
>
> We do have IB_SWITCH and IB_ROUTER functionality and we do use the SWITCH capability provided by the "core" in our product. The "core" has a wrapper function which Identifies the device as a switch to direct smp packets. I am glad that has not been removed. And I hope it won't be removed as it would be a burden to us. Although we prefer to have the IB_SWITCH/IB_ROUTER definitions in the core as well, it is not a big deal to move these definitions inside our driver.
>

Can you please point me to IN-KERNEL hardware/driver where these values are being set? Because I didn't find any.

Thanks

This correspondence, and any attachments or files transmitted with this correspondence, contains information which may be confidential and privileged and is intended solely for the use of the addressee. Unless you are the addressee or are authorized to receive messages for the addressee, you may not use, copy, disseminate, or disclose this correspondence or any information contained in this correspondence to any third party. If you have received this correspondence in error, please notify the sender immediately and delete this correspondence and any attachments or files transmitted with this correspondence from your system, and destroy any and all copies thereof, electronic or otherwise. Your cooperation and understanding are greatly appreciated.
--
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: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

[Devesh Sharma]

Hi Leon,

We probably should be okay to retain SWITCH and ROUTER devices, lets
say if someone wants to port entire linux kernel IB stack on a
IB-switch or Router. Probably portability was the motivation to add
these since the birth of kernel IB stack.



On Wed, Aug 16, 2017 at 8:00 PM, Suri Shelvapille
<suri-lFb5ksp6isWhEniVeURVKkEOCMrvLtNR@public.gmane.org> wrote:
> Leon:
> Our Drivers are not in Kernel (being a small company we don't have resources to make it in-kernel). When I and Hal worked on adding the switch capability into the kernel, we thought it would be useful for the community (as there was interest in a few others besides us). Philosophically, since SWITCH and ROUTER definitions are part of the IB spec, don't you think having these definitions makes the kernel close to spec and hence useful? I know these definitions can be easily added into our drivers.
> Just treat this as a request, if it is not an onus to the community, please keep the SWITCH and ROUTER (only) definitions. If you have strong reasons I have no objections to your patches.
>
> Thanks,
> Suri
>
> -----Original Message-----
> From: Leon Romanovsky [mailto:leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org]
> Sent: Wednesday, August 16, 2017 1:38 AM
> To: Suri Shelvapille <suri-lFb5ksp6isWhEniVeURVKkEOCMrvLtNR@public.gmane.org>
> Cc: Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>; linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> Subject: Re: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport
>
> On Tue, Aug 15, 2017 at 06:25:54PM +0000, Suri Shelvapille wrote:
>> Leon:
>>
>> We do have IB_SWITCH and IB_ROUTER functionality and we do use the SWITCH capability provided by the "core" in our product. The "core" has a wrapper function which Identifies the device as a switch to direct smp packets. I am glad that has not been removed. And I hope it won't be removed as it would be a burden to us. Although we prefer to have the IB_SWITCH/IB_ROUTER definitions in the core as well, it is not a big deal to move these definitions inside our driver.
>>
>
> Can you please point me to IN-KERNEL hardware/driver where these values are being set? Because I didn't find any.
>
> Thanks
>
> This correspondence, and any attachments or files transmitted with this correspondence, contains information which may be confidential and privileged and is intended solely for the use of the addressee. Unless you are the addressee or are authorized to receive messages for the addressee, you may not use, copy, disseminate, or disclose this correspondence or any information contained in this correspondence to any third party. If you have received this correspondence in error, please notify the sender immediately and delete this correspondence and any attachments or files transmitted with this correspondence from your system, and destroy any and all copies thereof, electronic or otherwise. Your cooperation and understanding are greatly appreciated.
> --
> 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
--
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: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 1699 bytes --]

On Wed, Aug 16, 2017 at 02:30:05PM +0000, Suri Shelvapille wrote:
> Leon:
> Our Drivers are not in Kernel (being a small company we don't have resources to make it in-kernel). When I and Hal worked on adding the switch capability into the kernel, we thought it would be useful for the community (as there was interest in a few others besides us). Philosophically, since SWITCH and ROUTER definitions are part of the IB spec, don't you think having these definitions makes the kernel close to spec and hence useful? I know these definitions can be easily added into our drivers.
> Just treat this as a request, if it is not an onus to the community, please keep the SWITCH and ROUTER (only) definitions. If you have strong reasons I have no objections to your patches.

It looks like I need to explain the rationale why I did this patch.

RDMAtool presents various information from the kernel, one of such info
is the node_type. In order to correctly present it, I was asked to expose
possible node_types through UAPI files.

These files are done with extra care and my goal was to provide the minimal
set and the most cleanest exposure, so I cleaned everything in those paths,
from the lowest possible layer to highest possible layer.

One of such cleanups were removal of node_types fields, which don't have
in-kernel users and against software development principles - don't
leave dead code.

So for now, I'm leaving this patch as is.

It is far below my lowest quality bar to leave those fields in place,
but if members of RDMA community prefer to go such low, they should
speak and explain publicly why RDMA subsystem is different from the
rest of the kernel.

Thanks

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 2019 bytes --]

On Wed, Aug 16, 2017 at 07:21:03PM +0300, Leon Romanovsky wrote:
> On Wed, Aug 16, 2017 at 02:30:05PM +0000, Suri Shelvapille wrote:
> > Leon:
> > Our Drivers are not in Kernel (being a small company we don't have resources to make it in-kernel). When I and Hal worked on adding the switch capability into the kernel, we thought it would be useful for the community (as there was interest in a few others besides us). Philosophically, since SWITCH and ROUTER definitions are part of the IB spec, don't you think having these definitions makes the kernel close to spec and hence useful? I know these definitions can be easily added into our drivers.
> > Just treat this as a request, if it is not an onus to the community, please keep the SWITCH and ROUTER (only) definitions. If you have strong reasons I have no objections to your patches.
>
> It looks like I need to explain the rationale why I did this patch.
>
> RDMAtool presents various information from the kernel, one of such info
> is the node_type. In order to correctly present it, I was asked to expose
> possible node_types through UAPI files.
>
> These files are done with extra care and my goal was to provide the minimal
> set and the most cleanest exposure, so I cleaned everything in those paths,
> from the lowest possible layer to highest possible layer.
>
> One of such cleanups were removal of node_types fields, which don't have
> in-kernel users and against software development principles - don't
> leave dead code.
>
> So for now, I'm leaving this patch as is.
>
> It is far below my lowest quality bar to leave those fields in place,
> but if members of RDMA community prefer to go such low, they should
> speak and explain publicly why RDMA subsystem is different from the
> rest of the kernel.

I found the elegant way how to code the RDMAtool without dependency on
those UAPI files. I'll post tomorrow morning new version of the tool,
and we will end with exposing of two enums only - rdma_dev_cap and
rdma_port_cap.

Thanks

>
> Thanks



[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [rdma-next v2 07/23] RDMA/core: Remove unimplemented node_types and node transport

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 496 bytes --]

On Wed, Aug 16, 2017 at 09:33:18PM +0530, Devesh Sharma wrote:
> Hi Leon,
>
> We probably should be okay to retain SWITCH and ROUTER devices, lets say if
> someone wants to port entire linux kernel IB stack on a IB-switch or
> Router. Probably portability was the motivation to add these since the
> birth of kernel IB stack.
>

It was in the happy days of coding of the IBTA spec, in 2017, the
reality is different. I don't want to support people who doesn't
contribute their code back.

Thanks

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[rdma-next v2 08/23] RDMA/core: Delete BUG() from unreachable flow

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Remove call to BUG() in case wrong node_type was provided.
This flow is unreachable, because node_types are supplied
from specific enum.

Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/core/verbs.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 21fef6b6d4f3..0d359a482b13 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -180,17 +180,13 @@ EXPORT_SYMBOL(ib_rate_to_mbps);
 __attribute_const__ enum rdma_transport_type
 rdma_node_get_transport(enum rdma_node_type node_type)
 {
-	switch (node_type) {
-	case RDMA_NODE_IB_CA:
+	if (node_type == RDMA_NODE_IB_CA)
 		return RDMA_TRANSPORT_IB;
-	case RDMA_NODE_RNIC:
+
+	if (node_type == RDMA_NODE_RNIC)
 		return RDMA_TRANSPORT_IWARP;
-	case RDMA_NODE_USNIC_UDP:
-		return RDMA_TRANSPORT_USNIC_UDP;
-	default:
-		BUG();
-		return 0;
-	}
+
+	return RDMA_TRANSPORT_USNIC_UDP;
 }
 EXPORT_SYMBOL(rdma_node_get_transport);
 
-- 
2.14.0

--
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

[rdma-next v2 09/23] RDMA/core: Refactor get link layer wrapper

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

The return values from rdma_node_get_transport() are strict
and IB_LINK_LAYER_UNSPECIFIED is unreachable in this flow.

Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/core/verbs.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 0d359a482b13..e3aa9c60bcc5 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -192,18 +192,15 @@ EXPORT_SYMBOL(rdma_node_get_transport);
 
 enum rdma_link_layer rdma_port_get_link_layer(struct ib_device *device, u8 port_num)
 {
+	enum rdma_transport_type lt;
 	if (device->get_link_layer)
 		return device->get_link_layer(device, port_num);
 
-	switch (rdma_node_get_transport(device->node_type)) {
-	case RDMA_TRANSPORT_IB:
+	lt = rdma_node_get_transport(device->node_type);
+	if (lt == RDMA_TRANSPORT_IB)
 		return IB_LINK_LAYER_INFINIBAND;
-	case RDMA_TRANSPORT_IWARP:
-	case RDMA_TRANSPORT_USNIC_UDP:
-		return IB_LINK_LAYER_ETHERNET;
-	default:
-		return IB_LINK_LAYER_UNSPECIFIED;
-	}
+
+	return IB_LINK_LAYER_ETHERNET;
 }
 EXPORT_SYMBOL(rdma_port_get_link_layer);
 
-- 
2.14.0

--
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

[rdma-next v2 10/23] RDMA/mlx4: Remove gfp_mask argument from acquire_group call

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

All callers of acquire_group() passed the same gfp_mask to it
and it is safe to remove it.

Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/hw/mlx4/mcg.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c
index b73f89700ef9..70eb9f917303 100644
--- a/drivers/infiniband/hw/mlx4/mcg.c
+++ b/drivers/infiniband/hw/mlx4/mcg.c
@@ -808,8 +808,7 @@ static ssize_t sysfs_show_group(struct device *dev,
 		struct device_attribute *attr, char *buf);
 
 static struct mcast_group *acquire_group(struct mlx4_ib_demux_ctx *ctx,
-					 union ib_gid *mgid, int create,
-					 gfp_t gfp_mask)
+					 union ib_gid *mgid, int create)
 {
 	struct mcast_group *group, *cur_group;
 	int is_mgid0;
@@ -825,7 +824,7 @@ static struct mcast_group *acquire_group(struct mlx4_ib_demux_ctx *ctx,
 	if (!create)
 		return ERR_PTR(-ENOENT);
 
-	group = kzalloc(sizeof *group, gfp_mask);
+	group = kzalloc(sizeof(*group), GFP_KERNEL);
 	if (!group)
 		return ERR_PTR(-ENOMEM);
 
@@ -892,7 +891,7 @@ int mlx4_ib_mcg_demux_handler(struct ib_device *ibdev, int port, int slave,
 	case IB_MGMT_METHOD_GET_RESP:
 	case IB_SA_METHOD_DELETE_RESP:
 		mutex_lock(&ctx->mcg_table_lock);
-		group = acquire_group(ctx, &rec->mgid, 0, GFP_KERNEL);
+		group = acquire_group(ctx, &rec->mgid, 0);
 		mutex_unlock(&ctx->mcg_table_lock);
 		if (IS_ERR(group)) {
 			if (mad->mad_hdr.method == IB_MGMT_METHOD_GET_RESP) {
@@ -954,7 +953,7 @@ int mlx4_ib_mcg_multiplex_handler(struct ib_device *ibdev, int port,
 		req->sa_mad = *sa_mad;
 
 		mutex_lock(&ctx->mcg_table_lock);
-		group = acquire_group(ctx, &rec->mgid, may_create, GFP_KERNEL);
+		group = acquire_group(ctx, &rec->mgid, may_create);
 		mutex_unlock(&ctx->mcg_table_lock);
 		if (IS_ERR(group)) {
 			kfree(req);
-- 
2.14.0

--
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

[rdma-next v2 11/23] RDMA/usnic: Fix remove address space warning

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Sparse tool complains with the following error:
drivers/infiniband/hw/usnic/usnic_ib_main.c:445:16: warning: cast removes
	address space of expression

Fix it by doing casting on correct field and convert function helper which
sets ifaddr to be void, because there are no users who are interested in
returned value.

Fixes: c7845bcafe4d ("IB/usnic: Add UDP support in u*verbs.c, u*main.c and u*util.h")
Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
 drivers/infiniband/hw/usnic/usnic_fwd.c     | 12 ++----------
 drivers/infiniband/hw/usnic/usnic_fwd.h     |  2 +-
 drivers/infiniband/hw/usnic/usnic_ib_main.c | 10 ++++++----
 3 files changed, 9 insertions(+), 15 deletions(-)

diff --git a/drivers/infiniband/hw/usnic/usnic_fwd.c b/drivers/infiniband/hw/usnic/usnic_fwd.c
index 3c37dd59c04e..995a26b65156 100644
--- a/drivers/infiniband/hw/usnic/usnic_fwd.c
+++ b/drivers/infiniband/hw/usnic/usnic_fwd.c
@@ -110,20 +110,12 @@ void usnic_fwd_set_mac(struct usnic_fwd_dev *ufdev, char mac[ETH_ALEN])
 	spin_unlock(&ufdev->lock);
 }
 
-int usnic_fwd_add_ipaddr(struct usnic_fwd_dev *ufdev, __be32 inaddr)
+void usnic_fwd_add_ipaddr(struct usnic_fwd_dev *ufdev, __be32 inaddr)
 {
-	int status;
-
 	spin_lock(&ufdev->lock);
-	if (ufdev->inaddr == 0) {
+	if (!ufdev->inaddr)
 		ufdev->inaddr = inaddr;
-		status = 0;
-	} else {
-		status = -EFAULT;
-	}
 	spin_unlock(&ufdev->lock);
-
-	return status;
 }
 
 void usnic_fwd_del_ipaddr(struct usnic_fwd_dev *ufdev)
diff --git a/drivers/infiniband/hw/usnic/usnic_fwd.h b/drivers/infiniband/hw/usnic/usnic_fwd.h
index b2ac22be0731..0b2cc4e79707 100644
--- a/drivers/infiniband/hw/usnic/usnic_fwd.h
+++ b/drivers/infiniband/hw/usnic/usnic_fwd.h
@@ -75,7 +75,7 @@ struct usnic_fwd_dev *usnic_fwd_dev_alloc(struct pci_dev *pdev);
 void usnic_fwd_dev_free(struct usnic_fwd_dev *ufdev);
 
 void usnic_fwd_set_mac(struct usnic_fwd_dev *ufdev, char mac[ETH_ALEN]);
-int usnic_fwd_add_ipaddr(struct usnic_fwd_dev *ufdev, __be32 inaddr);
+void usnic_fwd_add_ipaddr(struct usnic_fwd_dev *ufdev, __be32 inaddr);
 void usnic_fwd_del_ipaddr(struct usnic_fwd_dev *ufdev);
 void usnic_fwd_carrier_up(struct usnic_fwd_dev *ufdev);
 void usnic_fwd_carrier_down(struct usnic_fwd_dev *ufdev);
diff --git a/drivers/infiniband/hw/usnic/usnic_ib_main.c b/drivers/infiniband/hw/usnic/usnic_ib_main.c
index e86700f994cb..f45e99a938e0 100644
--- a/drivers/infiniband/hw/usnic/usnic_ib_main.c
+++ b/drivers/infiniband/hw/usnic/usnic_ib_main.c
@@ -351,7 +351,7 @@ static void *usnic_ib_device_add(struct pci_dev *dev)
 {
 	struct usnic_ib_dev *us_ibdev;
 	union ib_gid gid;
-	struct in_ifaddr *in;
+	struct in_device *ind;
 	struct net_device *netdev;
 
 	usnic_dbg("\n");
@@ -441,9 +441,11 @@ static void *usnic_ib_device_add(struct pci_dev *dev)
 	if (netif_carrier_ok(us_ibdev->netdev))
 		usnic_fwd_carrier_up(us_ibdev->ufdev);
 
-	in = ((struct in_device *)(netdev->ip_ptr))->ifa_list;
-	if (in != NULL)
-		usnic_fwd_add_ipaddr(us_ibdev->ufdev, in->ifa_address);
+	ind = in_dev_get(netdev);
+	if (ind->ifa_list)
+		usnic_fwd_add_ipaddr(us_ibdev->ufdev,
+				     ind->ifa_list->ifa_address);
+	in_dev_put(ind);
 
 	usnic_mac_ip_to_gid(us_ibdev->netdev->perm_addr,
 				us_ibdev->ufdev->inaddr, &gid.raw[0]);
-- 
2.14.0

--
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

[rdma-next v2 12/23] RDMA/mthca: Make explicit conversion to 64bit value

[Leon Romanovsky]

From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

The "lg" variable is declared as int so in all places where
this variable is used as a shift operand, the output will be
int too.

This produces the following smatch warning:
drivers/infiniband/hw/mthca/mthca_cmd.c:701 mthca_map_cmd() warn:
	should '1 << lg' be a 64 bit type?

Simple declaration of "1" to be "1ULL" will fix the issue.

Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Bart Van Assche <bart.vanassche-Sjgp3cTcYWE@public.gmane.org>
---
 drivers/infiniband/hw/mthca/mthca_cmd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c
index e19ae0b9b439..89529d204178 100644
--- a/drivers/infiniband/hw/mthca/mthca_cmd.c
+++ b/drivers/infiniband/hw/mthca/mthca_cmd.c
@@ -698,7 +698,7 @@ static int mthca_map_cmd(struct mthca_dev *dev, u16 op, struct mthca_icm *icm,
 		for (i = 0; i < mthca_icm_size(&iter) >> lg; ++i) {
 			if (virt != -1) {
 				pages[nent * 2] = cpu_to_be64(virt);
-				virt += 1 << lg;
+				virt += 1ULL << lg;
 			}
 
 			pages[nent * 2 + 1] =
-- 
2.14.0

--
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

[rdma-next v2 13/23] IB/mlx4: Fix some spelling mistakes

[Leon Romanovsky]

From: Talat Batheesh <talatb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Fix spelling mistakes in remarks
    "retrun"->"return"
    "cancell"->"cancel"

Signed-off-by: Talat Batheesh <talatb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/hw/mlx4/alias_GUID.c | 2 +-
 drivers/infiniband/hw/mlx4/cq.c         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/alias_GUID.c b/drivers/infiniband/hw/mlx4/alias_GUID.c
index 5a897b0106a9..9ed41f56e141 100644
--- a/drivers/infiniband/hw/mlx4/alias_GUID.c
+++ b/drivers/infiniband/hw/mlx4/alias_GUID.c
@@ -781,7 +781,7 @@ void mlx4_ib_init_alias_guid_work(struct mlx4_ib_dev *dev, int port)
 	spin_lock_irqsave(&dev->sriov.going_down_lock, flags);
 	spin_lock_irqsave(&dev->sriov.alias_guid.ag_work_lock, flags1);
 	if (!dev->sriov.is_going_down) {
-		/* If there is pending one should cancell then run, otherwise
+		/* If there is pending one should cancel then run, otherwise
 		  * won't run till previous one is ended as same work
 		  * struct is used.
 		  */
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 95382faa7ad1..cab796341697 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -637,7 +637,7 @@ static void mlx4_ib_poll_sw_comp(struct mlx4_ib_cq *cq, int num_entries,
 	struct mlx4_ib_qp *qp;
 
 	*npolled = 0;
-	/* Find uncompleted WQEs belonging to that cq and retrun
+	/* Find uncompleted WQEs belonging to that cq and return
 	 * simulated FLUSH_ERR completions
 	 */
 	list_for_each_entry(qp, &cq->send_qp_list, cq_send_list) {
-- 
2.14.0

--
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

[rdma-next v2 14/23] IB/mlx5: Fix some spelling mistakes

[Leon Romanovsky]

From: Talat Batheesh <talatb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Fix spelling mistakes in remarks
    "retrun"->"return"
    "Decalring"->"Declaring"

Signed-off-by: Talat Batheesh <talatb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/hw/mlx5/cq.c  | 2 +-
 drivers/infiniband/hw/mlx5/mad.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index a384d72ea3cd..f7fcfc8c22bf 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -499,7 +499,7 @@ static void mlx5_ib_poll_sw_comp(struct mlx5_ib_cq *cq, int num_entries,
 	struct mlx5_ib_qp *qp;
 
 	*npolled = 0;
-	/* Find uncompleted WQEs belonging to that cq and retrun mmics ones */
+	/* Find uncompleted WQEs belonging to that cq and return mmics ones */
 	list_for_each_entry(qp, &cq->list_send_qp, cq_send_list) {
 		sw_send_comp(qp, num_entries, wc + *npolled, npolled);
 		if (*npolled >= num_entries)
diff --git a/drivers/infiniband/hw/mlx5/mad.c b/drivers/infiniband/hw/mlx5/mad.c
index cd2264ac88ae..98e51d51bb04 100644
--- a/drivers/infiniband/hw/mlx5/mad.c
+++ b/drivers/infiniband/hw/mlx5/mad.c
@@ -204,7 +204,7 @@ static int process_pma_cmd(struct ib_device *ibdev, u8 port_num,
 	int err;
 	void *out_cnt;
 
-	/* Decalring support of extended counters */
+	/* Declaring support of extended counters */
 	if (in_mad->mad_hdr.attr_id == IB_PMA_CLASS_PORT_INFO) {
 		struct ib_class_port_info cpi = {};
 
-- 
2.14.0

--
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

[rdma-next v2 15/23] IB/mlx5: Add necessary delay drop assignment

[Leon Romanovsky]

From: Maor Gottlieb <maorg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Assign the statistics and configuration structure pointer on success.

Fixes: fe248c3a5837 ('IB/mlx5: Add delay drop configuration and statistics')
Signed-off-by: Maor Gottlieb <maorg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/hw/mlx5/main.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index cde7d1ce4a3c..986404827848 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -3792,6 +3792,8 @@ static int delay_drop_debugfs_init(struct mlx5_ib_dev *dev)
 	if (!dbg->timeout_debugfs)
 		goto out_debugfs;
 
+	dev->delay_drop.dbg = dbg;
+
 	return 0;
 
 out_debugfs:
-- 
2.14.0

--
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

[rdma-next v2 16/23] IB/mlx4: Fix RSS QP type in creation verb

[Leon Romanovsky]

From: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

The mlx4 was designed to support QP type of MLX4_IB_QPT_RAW_PACKET.

Fixes: 3078f5f1bd8b ("IB/mlx4: Add support for RSS QP")
Signed-off-by: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/hw/mlx4/qp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 793d77f29d3e..603a0bf49576 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -749,7 +749,7 @@ static int create_qp_rss(struct mlx4_ib_dev *dev, struct ib_pd *ibpd,
 	INIT_LIST_HEAD(&qp->gid_list);
 	INIT_LIST_HEAD(&qp->steering_rules);
 
-	qp->mlx4_ib_qp_type = MLX4_IB_QPT_RAW_ETHERTYPE;
+	qp->mlx4_ib_qp_type = MLX4_IB_QPT_RAW_PACKET;
 	qp->state = IB_QPS_RESET;
 
 	/* Set dummy send resources to be compatible with HV and PRM */
-- 
2.14.0

--
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

[rdma-next v2 17/23] IB/mlx4: Fix struct mlx4_ib_create_wq alignment

[Leon Romanovsky]

From: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

The mlx4 ABI defines to have structures with alignment of 64B.

Fixes: 400b1ebcfe31 ("IB/mlx4: Add support for WQ related verbs")
Signed-off-by: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/hw/mlx4/qp.c | 9 ++++-----
 include/uapi/rdma/mlx4-abi.h    | 1 -
 2 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 603a0bf49576..44d2d0a2d5e0 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1047,9 +1047,8 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 		}
 
 		if (src == MLX4_IB_RWQ_SRC) {
-			if (ucmd.wq.comp_mask || ucmd.wq.reserved1 ||
-			    ucmd.wq.reserved[0] || ucmd.wq.reserved[1] ||
-			    ucmd.wq.reserved[2]) {
+			if (ucmd.wq.comp_mask || ucmd.wq.reserved[0] ||
+			    ucmd.wq.reserved[1] || ucmd.wq.reserved[2]) {
 				pr_debug("user command isn't supported\n");
 				err = -EOPNOTSUPP;
 				goto err;
@@ -4153,8 +4152,8 @@ struct ib_wq *mlx4_ib_create_wq(struct ib_pd *pd,
 	if (!(udata && pd->uobject))
 		return ERR_PTR(-EINVAL);
 
-	required_cmd_sz = offsetof(typeof(ucmd), reserved) +
-			  sizeof(ucmd.reserved);
+	required_cmd_sz = offsetof(typeof(ucmd), comp_mask) +
+			  sizeof(ucmd.comp_mask);
 	if (udata->inlen < required_cmd_sz) {
 		pr_debug("invalid inlen\n");
 		return ERR_PTR(-EINVAL);
diff --git a/include/uapi/rdma/mlx4-abi.h b/include/uapi/rdma/mlx4-abi.h
index 21cce1a4c4dd..0e10102861b5 100644
--- a/include/uapi/rdma/mlx4-abi.h
+++ b/include/uapi/rdma/mlx4-abi.h
@@ -121,7 +121,6 @@ struct mlx4_ib_create_wq {
 	__u8	log_range_size;
 	__u8	reserved[3];
 	__u32   comp_mask;
-	__u32   reserved1;
 };
 
 struct mlx4_ib_modify_wq {
-- 
2.14.0

--
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

[rdma-next v2 18/23] IB/mlx4: Remove redundant attribute in mlx4_ib_create_qp_rss struct

[Leon Romanovsky]

From: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

rss_key_len is not in use and need to be removed.

Fixes: 3078f5f1bd8b ("IB/mlx4: Add support for RSS QP")
Signed-off-by: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 include/uapi/rdma/mlx4-abi.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/include/uapi/rdma/mlx4-abi.h b/include/uapi/rdma/mlx4-abi.h
index 0e10102861b5..c55f60e05f86 100644
--- a/include/uapi/rdma/mlx4-abi.h
+++ b/include/uapi/rdma/mlx4-abi.h
@@ -98,8 +98,7 @@ struct mlx4_ib_create_srq_resp {
 struct mlx4_ib_create_qp_rss {
 	__u64   rx_hash_fields_mask;
 	__u8    rx_hash_function;
-	__u8    rx_key_len;
-	__u8    reserved[6];
+	__u8    reserved[7];
 	__u8    rx_hash_key[40];
 	__u32   comp_mask;
 	__u32   reserved1;
-- 
2.14.0

--
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: [rdma-next v2 18/23] IB/mlx4: Remove redundant attribute in mlx4_ib_create_qp_rss struct

[Doug Ledford]

On Tue, 2017-08-15 at 11:54 +0300, Leon Romanovsky wrote:
> From: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> 
> rss_key_len is not in use and need to be removed.
> 
> Fixes: 3078f5f1bd8b ("IB/mlx4: Add support for RSS QP")
> Signed-off-by: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
> ---
>  include/uapi/rdma/mlx4-abi.h | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/include/uapi/rdma/mlx4-abi.h b/include/uapi/rdma/mlx4-
> abi.h
> index 0e10102861b5..c55f60e05f86 100644
> --- a/include/uapi/rdma/mlx4-abi.h
> +++ b/include/uapi/rdma/mlx4-abi.h
> @@ -98,8 +98,7 @@ struct mlx4_ib_create_srq_resp {
>  struct mlx4_ib_create_qp_rss {
>  	__u64   rx_hash_fields_mask;
>  	__u8    rx_hash_function;
> -	__u8    rx_key_len;
> -	__u8    reserved[6];
> +	__u8    reserved[7];

I believe Yuval's point, which has not been fixed, was not that you
could remove the variable without increasing the reserved size.  That
makes sense.  He was referring to the fact that you reference
rss_key_len in the commit message, but what you removed was rx_key_len
instead.

>  	__u8    rx_hash_key[40];
>  	__u32   comp_mask;
>  	__u32   reserved1;
-- 
Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
    GPG KeyID: B826A3330E572FDD
    Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333 0E57 2FDD

--
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: [rdma-next v2 18/23] IB/mlx4: Remove redundant attribute in mlx4_ib_create_qp_rss struct

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 1875 bytes --]

On Tue, Aug 22, 2017 at 04:09:50PM -0400, Doug Ledford wrote:
> On Tue, 2017-08-15 at 11:54 +0300, Leon Romanovsky wrote:
> > From: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> >
> > rss_key_len is not in use and need to be removed.
> >
> > Fixes: 3078f5f1bd8b ("IB/mlx4: Add support for RSS QP")
> > Signed-off-by: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> > Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
> > ---
> >  include/uapi/rdma/mlx4-abi.h | 3 +--
> >  1 file changed, 1 insertion(+), 2 deletions(-)
> >
> > diff --git a/include/uapi/rdma/mlx4-abi.h b/include/uapi/rdma/mlx4-
> > abi.h
> > index 0e10102861b5..c55f60e05f86 100644
> > --- a/include/uapi/rdma/mlx4-abi.h
> > +++ b/include/uapi/rdma/mlx4-abi.h
> > @@ -98,8 +98,7 @@ struct mlx4_ib_create_srq_resp {
> >  struct mlx4_ib_create_qp_rss {
> >  	__u64   rx_hash_fields_mask;
> >  	__u8    rx_hash_function;
> > -	__u8    rx_key_len;
> > -	__u8    reserved[6];
> > +	__u8    reserved[7];
>
> I believe Yuval's point, which has not been fixed, was not that you
> could remove the variable without increasing the reserved size.  That
> makes sense.  He was referring to the fact that you reference
> rss_key_len in the commit message, but what you removed was rx_key_len
> instead.

Ahh, I didn't understand it from his comment back then.

How do you want me to proceed here? Do you want me to resend all patches?
Do you want me to resend this patch only? Do you prefer to fix it by yourself?

It looks like we can declare working with pull-requests in RDMA as a failed
experiment.

>
> >  	__u8    rx_hash_key[40];
> >  	__u32   comp_mask;
> >  	__u32   reserved1;
> --
> Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
>     GPG KeyID: B826A3330E572FDD
>     Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333 0E57 2FDD
>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [rdma-next v2 18/23] IB/mlx4: Remove redundant attribute in mlx4_ib_create_qp_rss struct

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 2045 bytes --]

On Wed, Aug 23, 2017 at 07:43:54AM +0300, Leon Romanovsky wrote:
> On Tue, Aug 22, 2017 at 04:09:50PM -0400, Doug Ledford wrote:
> > On Tue, 2017-08-15 at 11:54 +0300, Leon Romanovsky wrote:
> > > From: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> > >
> > > rss_key_len is not in use and need to be removed.
> > >
> > > Fixes: 3078f5f1bd8b ("IB/mlx4: Add support for RSS QP")
> > > Signed-off-by: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> > > Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
> > > ---
> > >  include/uapi/rdma/mlx4-abi.h | 3 +--
> > >  1 file changed, 1 insertion(+), 2 deletions(-)
> > >
> > > diff --git a/include/uapi/rdma/mlx4-abi.h b/include/uapi/rdma/mlx4-
> > > abi.h
> > > index 0e10102861b5..c55f60e05f86 100644
> > > --- a/include/uapi/rdma/mlx4-abi.h
> > > +++ b/include/uapi/rdma/mlx4-abi.h
> > > @@ -98,8 +98,7 @@ struct mlx4_ib_create_srq_resp {
> > >  struct mlx4_ib_create_qp_rss {
> > >  	__u64   rx_hash_fields_mask;
> > >  	__u8    rx_hash_function;
> > > -	__u8    rx_key_len;
> > > -	__u8    reserved[6];
> > > +	__u8    reserved[7];
> >
> > I believe Yuval's point, which has not been fixed, was not that you
> > could remove the variable without increasing the reserved size.  That
> > makes sense.  He was referring to the fact that you reference
> > rss_key_len in the commit message, but what you removed was rx_key_len
> > instead.
>
> Ahh, I didn't understand it from his comment back then.
>
> How do you want me to proceed here? Do you want me to resend all patches?
> Do you want me to resend this patch only? Do you prefer to fix it by yourself?


Doug?

>
> It looks like we can declare working with pull-requests in RDMA as a failed
> experiment.
>
> >
> > >  	__u8    rx_hash_key[40];
> > >  	__u32   comp_mask;
> > >  	__u32   reserved1;
> > --
> > Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
> >     GPG KeyID: B826A3330E572FDD
> >     Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333 0E57 2FDD
> >



[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [rdma-next v2 18/23] IB/mlx4: Remove redundant attribute in mlx4_ib_create_qp_rss struct

[Doug Ledford]

On Thu, 2017-08-24 at 12:22 +0300, Leon Romanovsky wrote:
> On Wed, Aug 23, 2017 at 07:43:54AM +0300, Leon Romanovsky wrote:
> > On Tue, Aug 22, 2017 at 04:09:50PM -0400, Doug Ledford wrote:
> > > On Tue, 2017-08-15 at 11:54 +0300, Leon Romanovsky wrote:
> > > > From: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> > > > 
> > > > rss_key_len is not in use and need to be removed.
> > > > 
> > > > Fixes: 3078f5f1bd8b ("IB/mlx4: Add support for RSS QP")
> > > > Signed-off-by: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> > > > Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
> > > > ---
> > > >  include/uapi/rdma/mlx4-abi.h | 3 +--
> > > >  1 file changed, 1 insertion(+), 2 deletions(-)
> > > > 
> > > > diff --git a/include/uapi/rdma/mlx4-abi.h
> > > > b/include/uapi/rdma/mlx4-
> > > > abi.h
> > > > index 0e10102861b5..c55f60e05f86 100644
> > > > --- a/include/uapi/rdma/mlx4-abi.h
> > > > +++ b/include/uapi/rdma/mlx4-abi.h
> > > > @@ -98,8 +98,7 @@ struct mlx4_ib_create_srq_resp {
> > > >  struct mlx4_ib_create_qp_rss {
> > > >  	__u64   rx_hash_fields_mask;
> > > >  	__u8    rx_hash_function;
> > > > -	__u8    rx_key_len;
> > > > -	__u8    reserved[6];
> > > > +	__u8    reserved[7];
> > > 
> > > I believe Yuval's point, which has not been fixed, was not that
> > > you
> > > could remove the variable without increasing the reserved
> > > size.  That
> > > makes sense.  He was referring to the fact that you reference
> > > rss_key_len in the commit message, but what you removed was
> > > rx_key_len
> > > instead.
> > 
> > Ahh, I didn't understand it from his comment back then.
> > 
> > How do you want me to proceed here? Do you want me to resend all
> > patches?
> > Do you want me to resend this patch only? Do you prefer to fix it
> > by yourself?
> 
> 
> Doug?

I'll fix it up when I pull the patch in.

> > 
> > It looks like we can declare working with pull-requests in RDMA as
> > a failed
> > experiment.
> > 
> > > 
> > > >  	__u8    rx_hash_key[40];
> > > >  	__u32   comp_mask;
> > > >  	__u32   reserved1;
> > > 
> > > --
> > > Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
> > >     GPG KeyID: B826A3330E572FDD
> > >     Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333
> > > 0E57 2FDD
> > > 
> 
> 
-- 
Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
    GPG KeyID: B826A3330E572FDD
    Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333 0E57 2FDD

--
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

[rdma-next v2 19/23] IB/mlx4: Check that reserved fields in mlx4_ib_create_qp_rss are zero

[Leon Romanovsky]

From: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

According to mlx4 convention, need to fail the command due to a non-zero
value in the user data which is expected to be zero.

Fixes: 3078f5f1bd8b ("IB/mlx4: Add support for RSS QP")
Signed-off-by: Guy Levi <guyle-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/hw/mlx4/qp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 44d2d0a2d5e0..2874d8790570 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -813,6 +813,9 @@ static struct ib_qp *_mlx4_ib_create_qp_rss(struct ib_pd *pd,
 		return ERR_PTR(-EFAULT);
 	}
 
+	if (memchr_inv(ucmd.reserved, 0, sizeof(ucmd.reserved)))
+		return ERR_PTR(-EOPNOTSUPP);
+
 	if (ucmd.comp_mask || ucmd.reserved1)
 		return ERR_PTR(-EOPNOTSUPP);
 
-- 
2.14.0

--
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

[rdma-next v2 20/23] IB/ipoib: Sync between remove_one to sysfs calls that use rtnl_lock

[Leon Romanovsky]

From: Erez Shitrit <erezsh-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

In order to avoid deadlock between sysfs functions (like create/delete
child) and remove_one (both of them are using the sysfs lock and
rtnl_lock) the driver will use a state mutex for sync.

That will fix traces as the following:
schedule+0x3e/0x90
kernfs_drain+0x75/0xf0
? wait_woken+0x90/0x90
__kernfs_remove+0x12e/0x1c0
kernfs_remove+0x25/0x40
sysfs_remove_dir+0x57/0x90
kobject_del+0x22/0x60
device_del+0x195/0x230
 pm_runtime_set_memalloc_noio+0xac/0xf0
netdev_unregister_kobject+0x71/0x80
rollback_registered_many+0x205/0x2f0
rollback_registered+0x31/0x40
unregister_netdevice_queue+0x58/0xb0
unregister_netdev+0x20/0x30
ipoib_remove_one+0xb7/0x240 [ib_ipoib]
ib_unregister_device+0xbc/0x1b0 [ib_core]
ib_unregister_mad_agent+0x29/0x30 [ib_core]
mlx4_ib_remove+0x67/0x280 [mlx4_ib]
INFO: task echo:24082 blocked for more than 120 seconds.
Tainted: G           OE   4.1.12-37.5.1.el6uek.x86_64 #2
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this
message.
Call Trace:
schedule+0x3e/0x90
schedule_preempt_disabled+0xe/0x10
__mutex_lock_slowpath+0x95/0x110
? _rcu_barrier+0x177/0x220
mutex_lock+0x23/0x40
rtnl_lock+0x15/0x20
netdev_run_todo+0x81/0x1f0
rtnl_unlock+0xe/0x10
ipoib_vlan_delete+0x12f/0x1c0 [ib_ipoib]
delete_child+0x69/0x80 [ib_ipoib]
dev_attr_store+0x20/0x30
sysfs_kf_write+0x41/0x50

Signed-off-by: Erez Shitrit <erezsh-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Alex Vesker <valex-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/ulp/ipoib/ipoib.h      |  1 +
 drivers/infiniband/ulp/ipoib/ipoib_cm.c   |  8 +++++++-
 drivers/infiniband/ulp/ipoib/ipoib_main.c |  5 +++++
 drivers/infiniband/ulp/ipoib/ipoib_vlan.c | 22 ++++++++++++++++++----
 4 files changed, 31 insertions(+), 5 deletions(-)

diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index 9e738104c2a1..7aa114cf53e3 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -336,6 +336,7 @@ struct ipoib_dev_priv {
 	unsigned long flags;
 
 	struct rw_semaphore vlan_rwsem;
+	struct mutex        sysfs_lock; /* Protect sysfs running*/
 
 	struct rb_root  path_tree;
 	struct list_head path_list;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
index f87d104837dc..85ef9f10bdee 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
@@ -1507,9 +1507,14 @@ static ssize_t set_mode(struct device *d, struct device_attribute *attr,
 	if (test_bit(IPOIB_FLAG_GOING_DOWN, &priv->flags))
 		return -EPERM;
 
-	if (!rtnl_trylock())
+	if (!mutex_trylock(&priv->sysfs_lock))
 		return restart_syscall();
 
+	if (!rtnl_trylock()) {
+		mutex_unlock(&priv->sysfs_lock);
+		return restart_syscall();
+	}
+
 	ret = ipoib_set_mode(dev, buf);
 
 	/* The assumption is that the function ipoib_set_mode returned
@@ -1518,6 +1523,7 @@ static ssize_t set_mode(struct device *d, struct device_attribute *attr,
 	 */
 	if (ret != -EBUSY)
 		rtnl_unlock();
+	mutex_unlock(&priv->sysfs_lock);
 
 	return (!ret || ret == -EBUSY) ? count : ret;
 }
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 645217232250..627ff8fb59b0 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -1876,6 +1876,7 @@ static void ipoib_build_priv(struct net_device *dev)
 	priv->dev = dev;
 	spin_lock_init(&priv->lock);
 	init_rwsem(&priv->vlan_rwsem);
+	mutex_init(&priv->sysfs_lock);
 
 	INIT_LIST_HEAD(&priv->path_list);
 	INIT_LIST_HEAD(&priv->child_intfs);
@@ -2325,7 +2326,11 @@ static void ipoib_remove_one(struct ib_device *device, void *client_data)
 		cancel_delayed_work(&priv->neigh_reap_task);
 		flush_workqueue(priv->wq);
 
+		/* Wrap rtnl_lock/unlock with mutex to protect sysfs calls */
+		mutex_lock(&priv->sysfs_lock);
 		unregister_netdev(priv->dev);
+		mutex_unlock(&priv->sysfs_lock);
+
 		rn->free_rdma_netdev(priv->dev);
 
 		list_for_each_entry_safe(cpriv, tcpriv, &priv->child_intfs, list)
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
index 081b33deff1b..dd46abcf53c2 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
@@ -133,12 +133,20 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
 	snprintf(intf_name, sizeof intf_name, "%s.%04x",
 		 ppriv->dev->name, pkey);
 
-	if (!rtnl_trylock())
+	if (!mutex_trylock(&ppriv->sysfs_lock))
 		return restart_syscall();
 
+	if (!rtnl_trylock()) {
+		mutex_unlock(&ppriv->sysfs_lock);
+		return restart_syscall();
+	}
+
 	priv = ipoib_intf_alloc(ppriv->ca, ppriv->port, intf_name);
-	if (!priv)
+	if (!priv) {
+		rtnl_unlock();
+		mutex_unlock(&ppriv->sysfs_lock);
 		return -ENOMEM;
+	}
 
 	down_write(&ppriv->vlan_rwsem);
 
@@ -164,8 +172,8 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
 
 out:
 	up_write(&ppriv->vlan_rwsem);
-
 	rtnl_unlock();
+	mutex_unlock(&ppriv->sysfs_lock);
 
 	if (result) {
 		free_netdev(priv->dev);
@@ -188,8 +196,13 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey)
 	if (test_bit(IPOIB_FLAG_GOING_DOWN, &ppriv->flags))
 		return -EPERM;
 
-	if (!rtnl_trylock())
+	if (!mutex_trylock(&ppriv->sysfs_lock))
+		return restart_syscall();
+
+	if (!rtnl_trylock()) {
+		mutex_unlock(&ppriv->sysfs_lock);
 		return restart_syscall();
+	}
 
 	down_write(&ppriv->vlan_rwsem);
 	list_for_each_entry_safe(priv, tpriv, &ppriv->child_intfs, list) {
@@ -208,6 +221,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey)
 	}
 
 	rtnl_unlock();
+	mutex_unlock(&ppriv->sysfs_lock);
 
 	if (dev) {
 		free_netdev(dev);
-- 
2.14.0

--
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

[rdma-next v2 21/23] IB/ipoib: Add get statistics support to SRIOV VF

[Leon Romanovsky]

From: Feras Daoud <ferasda-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

Add SRIOV VF support to get traffic statistics.

Signed-off-by: Feras Daoud <ferasda-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/ulp/ipoib/ipoib_main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 627ff8fb59b0..9e8dce6e9fc2 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -1846,6 +1846,7 @@ static const struct net_device_ops ipoib_netdev_ops_vf = {
 	.ndo_tx_timeout		 = ipoib_timeout,
 	.ndo_set_rx_mode	 = ipoib_set_mcast_list,
 	.ndo_get_iflink		 = ipoib_get_iflink,
+	.ndo_get_stats64	 = ipoib_get_stats,
 };
 
 void ipoib_setup_common(struct net_device *dev)
-- 
2.14.0

--
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

[rdma-next v2 22/23] IB/rxe: Make rxe_counter_name static

[Leon Romanovsky]

From: Kamal Heib <kamalh-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

rxe_counter_name is used in rxe_hw_counters.c only. Make it static.

Fixes: 0b1e5b99a48b ('IB/rxe: Add port protocol stats')
Signed-off-by: Kamal Heib <kamalh-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Reviewed-by: Yonatan Cohen <yonatanc-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/sw/rxe/rxe_hw_counters.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/sw/rxe/rxe_hw_counters.c b/drivers/infiniband/sw/rxe/rxe_hw_counters.c
index 7ef90aad7dfd..6aeb7a165e46 100644
--- a/drivers/infiniband/sw/rxe/rxe_hw_counters.c
+++ b/drivers/infiniband/sw/rxe/rxe_hw_counters.c
@@ -33,7 +33,7 @@
 #include "rxe.h"
 #include "rxe_hw_counters.h"
 
-const char * const rxe_counter_name[] = {
+static const char * const rxe_counter_name[] = {
 	[RXE_CNT_SENT_PKTS]           =  "sent_pkts",
 	[RXE_CNT_RCVD_PKTS]           =  "rcvd_pkts",
 	[RXE_CNT_DUP_REQ]             =  "duplicate_request",
-- 
2.14.0

--
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

[rdma-next v2 23/23] RDMA/mlx5: Limit scope of get vector affinity local function

[Leon Romanovsky]

The mlx5_ib_get_vector_affinity() call is local to main.c file and there
is no need to be declared globally visible.

Fixes: 40b24403f33e ("mlx5: support ->get_vector_affinity")
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/hw/mlx5/main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 986404827848..47cf622c6d24 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -3818,8 +3818,8 @@ static void init_delay_drop(struct mlx5_ib_dev *dev)
 		mlx5_ib_warn(dev, "Failed to init delay drop debugfs\n");
 }
 
-const struct cpumask *mlx5_ib_get_vector_affinity(struct ib_device *ibdev,
-		int comp_vector)
+static const struct cpumask *
+mlx5_ib_get_vector_affinity(struct ib_device *ibdev, int comp_vector)
 {
 	struct mlx5_ib_dev *dev = to_mdev(ibdev);
 
-- 
2.14.0

--
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