Skip to contentSkip to conversation

USNIC and RAW enhancements series

PATCH In review Leon Romanovsky 5 participants 14 messages Jan 19, 2017 – Jan 25, 2017 7 messages not in the archive

Subsystems (12)
InfiniBand Mellanox MLX5 IB driver Mellanox MLX4 IB driver Soft-RoCE driver (rxe) +8 more subsystems
lore
Share & tools
lore

Leon Romanovsky

author
#0 lore raw

cover USNIC and RAW enhancements series

Hi Doug,

Please find below USNIC and RAW protocols enhancement series from Or.

from: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>

This series adds the USNIC and RAW protocols to the core and relevant devices (usnic, and mlx5/mlx4 over Eth ports) such that now the protocol part of the caps covers all in-tree HW/SW devices.

It also contains a cleanup patch to the HW/SW drivers such that they use ib_query_port when needed to query their own ports and not apply their internal callback. The patch doesn't add new functionality but makes the different drivers to be behave the same in that respect and makes the IB stack code a little bit more robust and clear.

Thanks
---
CC: Steve Wise <[email protected]>
CC: Dennis Dalessandro <[email protected]>
CC: Lijun Ou <[email protected]>
CC: Faisal Latif <[email protected]>
CC: Shiraz Saleem <[email protected]>
CC: Selvin Xavier <selvin.xavier-1wcpHE2jlwO1Z/[email protected]>
CC: Christian Benvenuti <[email protected]>
CC: Dave Goodell <[email protected]>
CC: Adit Ranadive <[email protected]>
CC: Michal Kalderon <Michal.Kalderon-YGCgFSpz5w/[email protected]>
CC: Rajesh Borundia <rajesh.borundia-YGCgFSpz5w/[email protected]>
---
This patchset was generated against commit
f502d834950a ("net/mlx5: Activate support for 4K UARs") and it applies cleanly on k.o/for-4.11 branch.

Available in the "topic/raw-qp-query-enhance" topic branch of this git repo:
git://git.kernel.org/pub/scm/linux/kernel/git/leon/linux-rdma.git

Or for browsing:
https://git.kernel.org/cgit/linux/kernel/git/leon/linux-rdma.git/log/?h=topic/raw-qp-query-enhance

Or Gerlitz (5):
IB/core: Add raw packet protocol
IB/mlx5: Support raw packet protocol
IB/mlx4: Support raw packet protocol
IB: Add protocol for USNIC
IB: Query port through the core instead of directly calling the driver
handler

 drivers/infiniband/hw/cxgb3/iwch_provider.c     |  7 ++++---
 drivers/infiniband/hw/cxgb4/provider.c          |  8 ++++----
 drivers/infiniband/hw/hfi1/verbs.c              |  1 +
 drivers/infiniband/hw/hns/hns_roce_main.c       |  7 ++++---
 drivers/infiniband/hw/i40iw/i40iw_verbs.c       |  8 ++++----
 drivers/infiniband/hw/mlx4/alias_GUID.c         |  1 +
 drivers/infiniband/hw/mlx4/main.c               | 23 +++++++++++++----------
 drivers/infiniband/hw/mlx4/sysfs.c              |  1 +
 drivers/infiniband/hw/mlx5/mad.c                |  2 +-
 drivers/infiniband/hw/mlx5/main.c               | 17 +++++++++++------
 drivers/infiniband/hw/mthca/mthca_provider.c    |  9 +++++----
 drivers/infiniband/hw/nes/nes_verbs.c           |  5 +++--
 drivers/infiniband/hw/ocrdma/ocrdma_main.c      |  9 +++++----
 drivers/infiniband/hw/ocrdma/ocrdma_verbs.c     |  1 +
 drivers/infiniband/hw/qedr/verbs.c              |  9 +++++----
 drivers/infiniband/hw/qib/qib_verbs.c           |  1 +
 drivers/infiniband/hw/usnic/usnic_ib_main.c     |  4 +++-
 drivers/infiniband/hw/usnic/usnic_ib_verbs.c    |  2 +-
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c  |  5 +++--
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c |  4 ++--
 drivers/infiniband/sw/rdmavt/vt.c               |  7 ++++---
 drivers/infiniband/sw/rxe/rxe_verbs.c           |  6 ++++--
 include/rdma/ib_verbs.h                         | 16 ++++++++++++++++
 23 files changed, 97 insertions(+), 56 deletions(-)

--
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): 20170119054102.20135-1-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org

author
#2 lore raw

PATCH 1/5 IB/core: Add raw packet protocol

From: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>

Define raw packet protocol which comes to denote this port supports working with raw ethernet frames, e.g as done with RAW_PACKET QPs.

Signed-off-by Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>
Reviewed-by Matan Barak <matanb-VPRAkNaXOzVWk0Htik3J/[email protected]>
Signed-off-by Leon Romanovsky <[email protected]>
---
 include/rdma/ib_verbs.h | 8 ++++++++
 1 file changed, 8 insertions(+)
include/rdma/ib_verbs.h +8 -0
index 958a24d..68fff34 100644
@@ -485,6 +485,7 @@ static inline struct rdma_hw_stats *rdma_alloc_hw_stats_struct(
 #define RDMA_CORE_CAP_PROT_ROCE         0x00200000
 #define RDMA_CORE_CAP_PROT_IWARP        0x00400000
 #define RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP 0x00800000
+#define RDMA_CORE_CAP_PROT_RAW_PACKET   0x01000000
 
 #define RDMA_CORE_PORT_IBA_IB          (RDMA_CORE_CAP_PROT_IB  \
 					| RDMA_CORE_CAP_IB_MAD \

@@ -508,6 +509,8 @@ static inline struct rdma_hw_stats *rdma_alloc_hw_stats_struct(
 #define RDMA_CORE_PORT_INTEL_OPA       (RDMA_CORE_PORT_IBA_IB  \
 					| RDMA_CORE_CAP_OPA_MAD)
 
+#define RDMA_CORE_PORT_RAW_PACKET	(RDMA_CORE_CAP_PROT_RAW_PACKET)
+
 struct ib_port_attr {
 	u64			subnet_prefix;
 	enum ib_port_state	state;

@@ -2307,6 +2310,11 @@ static inline bool rdma_ib_or_roce(const struct ib_device *device, u8 port_num)
 		rdma_protocol_roce(device, port_num);
 }
 
+static inline bool rdma_protocol_raw_packet(const struct ib_device *device, u8 port_num)
+{
+	return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_RAW_PACKET;
+}
+
 /**
  * rdma_cap_ib_mad - Check if the port of a device supports Infiniband
  * Management Datagrams.
Signature · 6 lines
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author
#3 lore raw

PATCH 2/5 IB/mlx5: Support raw packet protocol

In reply to a message not in the archive

From: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>

Mark support for the new raw packet protocol on Eth ports.

Signed-off-by Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>
Reviewed-by Matan Barak <matanb-VPRAkNaXOzVWk0Htik3J/[email protected]>
Signed-off-by Leon Romanovsky <[email protected]>
---
 drivers/infiniband/hw/mlx5/main.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/infiniband/hw/mlx5/main.c +4 -2
index a191b93..4e8e78f 100644
@@ -2858,11 +2858,13 @@ static u32 get_core_cap_flags(struct ib_device *ibdev)
 	if (ll == IB_LINK_LAYER_INFINIBAND)
 		return RDMA_CORE_PORT_IBA_IB;
 
+	ret = RDMA_CORE_PORT_RAW_PACKET;
+
 	if (!(l3_type_cap & MLX5_ROCE_L3_TYPE_IPV4_CAP))
-		return 0;
+		return ret;
 
 	if (!(l3_type_cap & MLX5_ROCE_L3_TYPE_IPV6_CAP))
-		return 0;
+		return ret;
 
 	if (roce_version_cap & MLX5_ROCE_VERSION_1_CAP)
 		ret |= RDMA_CORE_PORT_IBA_ROCE;
Signature · 6 lines
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author
#4 lore raw

PATCH 3/5 IB/mlx4: Support raw packet protocol

In reply to a message not in the archive

From: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>

Mark support for the new raw packet protocol on Eth ports.

Signed-off-by Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>
Reviewed-by Matan Barak <matanb-VPRAkNaXOzVWk0Htik3J/[email protected]>
Signed-off-by Leon Romanovsky <[email protected]>
---
 drivers/infiniband/hw/mlx4/main.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/infiniband/hw/mlx4/main.c +5 -2
index c8413fc..7019b94 100644
@@ -2536,16 +2536,19 @@ static int mlx4_port_immutable(struct ib_device *ibdev, u8 port_num,
 
 	if (mlx4_ib_port_link_layer(ibdev, port_num) == IB_LINK_LAYER_INFINIBAND) {
 		immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB;
+		immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 	} else {
 		if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_IBOE)
 			immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
 		if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2)
 			immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE |
 				RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
+		immutable->core_cap_flags |= RDMA_CORE_PORT_RAW_PACKET;
+		if (immutable->core_cap_flags & (RDMA_CORE_PORT_IBA_ROCE |
+		    RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP))
+			immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 	}
 
-	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
-
 	return 0;
 }
Signature · 6 lines
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author
#5 lore raw

PATCH 4/5 IB: Add protocol for USNIC

In reply to a message not in the archive

From: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>

Add protocol definition for the proprietary the USNIC driver.

Signed-off-by Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>
Reviewed-by Matan Barak <matanb-VPRAkNaXOzVWk0Htik3J/[email protected]>
Signed-off-by Leon Romanovsky <[email protected]>
---
 drivers/infiniband/hw/usnic/usnic_ib_main.c | 1 +
 include/rdma/ib_verbs.h                     | 8 ++++++++
 2 files changed, 9 insertions(+)
drivers/infiniband/hw/usnic/usnic_ib_main.c +1 -0
index 0a89a95..dde0b23 100644
@@ -325,6 +325,7 @@ static int usnic_port_immutable(struct ib_device *ibdev, u8 port_num,
 	if (err)
 		return err;
 
+	immutable->core_cap_flags = RDMA_CORE_PORT_USNIC;
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
include/rdma/ib_verbs.h +8 -0
index 68fff34..d5109e6 100644
@@ -486,6 +486,7 @@ static inline struct rdma_hw_stats *rdma_alloc_hw_stats_struct(
 #define RDMA_CORE_CAP_PROT_IWARP        0x00400000
 #define RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP 0x00800000
 #define RDMA_CORE_CAP_PROT_RAW_PACKET   0x01000000
+#define RDMA_CORE_CAP_PROT_USNIC        0x02000000
 
 #define RDMA_CORE_PORT_IBA_IB          (RDMA_CORE_CAP_PROT_IB  \
 					| RDMA_CORE_CAP_IB_MAD \

@@ -511,6 +512,8 @@ static inline struct rdma_hw_stats *rdma_alloc_hw_stats_struct(
 
 #define RDMA_CORE_PORT_RAW_PACKET	(RDMA_CORE_CAP_PROT_RAW_PACKET)
 
+#define RDMA_CORE_PORT_USNIC		(RDMA_CORE_CAP_PROT_USNIC)
+
 struct ib_port_attr {
 	u64			subnet_prefix;
 	enum ib_port_state	state;

@@ -2315,6 +2318,11 @@ static inline bool rdma_protocol_raw_packet(const struct ib_device *device, u8 p
 	return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_RAW_PACKET;
 }
 
+static inline bool rdma_protocol_usnic(const struct ib_device *device, u8 port_num)
+{
+	return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_USNIC;
+}
+
 /**
  * rdma_cap_ib_mad - Check if the port of a device supports Infiniband
  * Management Datagrams.
Signature · 6 lines
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author
#6 lore raw

PATCH 5/5 IB: Query port through the core instead of directly calling the driver handler

In reply to a message not in the archive

From: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>

Change the drivers to call ib_query_port in their get port immutable handler instead of their own query port handler.

Doing this required to set the core cap flags of this device before the ib_query_port call is made, since the IB core might need these caps to serve the port query.

Drivers are ensured by the IB core that the port attributes passed to the port query verb implementation are zero, and hence we removed the zeroing from the drivers.

This patch doesn't add any new functionality.

Signed-off-by Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>
Reviewed-by Matan Barak <matanb-VPRAkNaXOzVWk0Htik3J/[email protected]>
Signed-off-by Leon Romanovsky <[email protected]>
---
 drivers/infiniband/hw/cxgb3/iwch_provider.c     |  7 ++++---
 drivers/infiniband/hw/cxgb4/provider.c          |  8 ++++----
 drivers/infiniband/hw/hfi1/verbs.c              |  1 +
 drivers/infiniband/hw/hns/hns_roce_main.c       |  7 ++++---
 drivers/infiniband/hw/i40iw/i40iw_verbs.c       |  8 ++++----
 drivers/infiniband/hw/mlx4/alias_GUID.c         |  1 +
 drivers/infiniband/hw/mlx4/main.c               | 18 +++++++++---------
 drivers/infiniband/hw/mlx4/sysfs.c              |  1 +
 drivers/infiniband/hw/mlx5/mad.c                |  2 +-
 drivers/infiniband/hw/mlx5/main.c               | 11 +++++++----
 drivers/infiniband/hw/mthca/mthca_provider.c    |  9 +++++----
 drivers/infiniband/hw/nes/nes_verbs.c           |  5 +++--
 drivers/infiniband/hw/ocrdma/ocrdma_main.c      |  9 +++++----
 drivers/infiniband/hw/ocrdma/ocrdma_verbs.c     |  1 +
 drivers/infiniband/hw/qedr/verbs.c              |  9 +++++----
 drivers/infiniband/hw/qib/qib_verbs.c           |  1 +
 drivers/infiniband/hw/usnic/usnic_ib_main.c     |  5 +++--
 drivers/infiniband/hw/usnic/usnic_ib_verbs.c    |  2 +-
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c  |  5 +++--
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c |  4 ++--
 drivers/infiniband/sw/rdmavt/vt.c               |  7 ++++---
 drivers/infiniband/sw/rxe/rxe_verbs.c           |  6 ++++--
 22 files changed, 73 insertions(+), 54 deletions(-)
drivers/infiniband/hw/cxgb3/iwch_provider.c +4 -3
index 9d5fe18..9e39252 100644
@@ -1133,7 +1133,7 @@ static int iwch_query_port(struct ib_device *ibdev,
 	dev = to_iwch_dev(ibdev);
 	netdev = dev->rdev.port_info.lldevs[port-1];
 
-	memset(props, 0, sizeof(struct ib_port_attr));
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->max_mtu = IB_MTU_4096;
 	if (netdev->mtu >= 4096)
 		props->active_mtu = IB_MTU_4096;

@@ -1338,13 +1338,14 @@ static int iwch_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = iwch_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
 
 	return 0;
 }
drivers/infiniband/hw/cxgb4/provider.c +4 -4
index 49b51b7..5b00b50a 100644
@@ -358,8 +358,7 @@ static int c4iw_query_port(struct ib_device *ibdev, u8 port,
 
 	dev = to_c4iw_dev(ibdev);
 	netdev = dev->rdev.lldi.ports[port-1];
-
-	memset(props, 0, sizeof(struct ib_port_attr));
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->max_mtu = IB_MTU_4096;
 	if (netdev->mtu >= 4096)
 		props->active_mtu = IB_MTU_4096;

@@ -505,13 +504,14 @@ static int c4iw_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = c4iw_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
 
 	return 0;
 }
drivers/infiniband/hw/hfi1/verbs.c +1 -0
index 95ed4d6..f2d4e04 100644
@@ -1384,6 +1384,7 @@ static int query_port(struct rvt_dev_info *rdi, u8 port_num,
 	struct hfi1_pportdata *ppd = &dd->pport[port_num - 1];
 	u16 lid = ppd->lid;
 
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->lid = lid ? lid : 0;
 	props->lmc = ppd->lmc;
 	/* OPA logical states match IB logical states */
drivers/infiniband/hw/hns/hns_roce_main.c +4 -3
index 4953d9c..7be8158 100644
@@ -249,7 +249,7 @@ static int hns_roce_query_port(struct ib_device *ib_dev, u8 port_num,
 	assert(port_num > 0);
 	port = port_num - 1;
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->max_mtu = hr_dev->caps.max_mtu;
 	props->gid_tbl_len = hr_dev->caps.gid_table_len[port];

@@ -400,14 +400,15 @@ static int hns_roce_port_immutable(struct ib_device *ib_dev, u8 port_num,
 	struct ib_port_attr attr;
 	int ret;
 
-	ret = hns_roce_query_port(ib_dev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
+
+	ret = ib_query_port(ib_dev, port_num, &attr);
 	if (ret)
 		return ret;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
 
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
drivers/infiniband/hw/i40iw/i40iw_verbs.c +4 -4
index 29e97df..3e83048 100644
@@ -97,8 +97,7 @@ static int i40iw_query_port(struct ib_device *ibdev,
 	struct i40iw_device *iwdev = to_iwdev(ibdev);
 	struct net_device *netdev = iwdev->netdev;
 
-	memset(props, 0, sizeof(*props));
-
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->max_mtu = IB_MTU_4096;
 	if (netdev->mtu >= 4096)
 		props->active_mtu = IB_MTU_4096;

@@ -2506,14 +2505,15 @@ static int i40iw_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = i40iw_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
 
 	return 0;
 }
drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
index 06020c5..4648374 100644
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
 	struct list_head *head =
 		&dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
 
+	memset(&attr, 0, sizeof(struct ib_port_attr));
 	err = __mlx4_ib_query_port(ibdev, port, &attr, 1);
 	if (err) {
 		pr_debug("mlx4_ib_query_port failed (err: %d), port: %d\n",
drivers/infiniband/hw/mlx4/main.c +9 -9
index 7019b94..646dcbf 100644
@@ -741,7 +741,7 @@ int __mlx4_ib_query_port(struct ib_device *ibdev, u8 port,
 {
 	int err;
 
-	memset(props, 0, sizeof *props);
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	err = mlx4_ib_port_link_layer(ibdev, port) == IB_LINK_LAYER_INFINIBAND ?
 		ib_link_query_port(ibdev, port, props, netw_view) :

@@ -1014,7 +1014,7 @@ static int mlx4_ib_modify_port(struct ib_device *ibdev, u8 port, int mask,
 
 	mutex_lock(&mdev->cap_mask_mutex);
 
-	err = mlx4_ib_query_port(ibdev, port, &attr);
+	err = ib_query_port(ibdev, port, &attr);
 	if (err)
 		goto out;
 

@@ -2527,13 +2527,6 @@ static int mlx4_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct mlx4_ib_dev *mdev = to_mdev(ibdev);
 	int err;
 
-	err = mlx4_ib_query_port(ibdev, port_num, &attr);
-	if (err)
-		return err;
-
-	immutable->pkey_tbl_len = attr.pkey_tbl_len;
-	immutable->gid_tbl_len = attr.gid_tbl_len;
-
 	if (mlx4_ib_port_link_layer(ibdev, port_num) == IB_LINK_LAYER_INFINIBAND) {
 		immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB;
 		immutable->max_mad_size = IB_MGMT_MAD_SIZE;

@@ -2549,6 +2542,13 @@ static int mlx4_port_immutable(struct ib_device *ibdev, u8 port_num,
 			immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 	}
 
+	err = ib_query_port(ibdev, port_num, &attr);
+	if (err)
+		return err;
+
+	immutable->pkey_tbl_len = attr.pkey_tbl_len;
+	immutable->gid_tbl_len = attr.gid_tbl_len;
+
 	return 0;
 }
drivers/infiniband/hw/mlx4/sysfs.c +1 -0
index 69fb5ba..5835165 100644
@@ -226,6 +226,7 @@ static int add_port_entries(struct mlx4_ib_dev *device, int port_num)
 	int ret = 0 ;
 	struct ib_port_attr attr;
 
+	memset(&attr, 0, sizeof(struct ib_port_attr));
 	/* get the physical gid and pkey table sizes.*/
 	ret = __mlx4_ib_query_port(&device->ib_dev, port_num, &attr, 1);
 	if (ret)
drivers/infiniband/hw/mlx5/mad.c +1 -1
index 39e5848..4e58b8f 100644
@@ -515,7 +515,7 @@ int mlx5_query_mad_ifc_port(struct ib_device *ibdev, u8 port,
 	if (!in_mad || !out_mad)
 		goto out;
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	init_query_mad(in_mad);
 	in_mad->attr_id  = IB_SMP_ATTR_PORT_INFO;
drivers/infiniband/hw/mlx5/main.c +7 -4
index 4e8e78f..176417b 100644
@@ -174,7 +174,7 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
 	enum ib_mtu ndev_ib_mtu;
 	u16 qkey_viol_cntr;
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->port_cap_flags  |= IB_PORT_CM_SUP;
 	props->port_cap_flags  |= IB_PORT_IP_BASED_GIDS;

@@ -830,7 +830,7 @@ static int mlx5_query_hca_port(struct ib_device *ibdev, u8 port,
 		goto out;
 	}
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	err = mlx5_query_hca_vport_context(mdev, 0, port, 0, rep);
 	if (err)

@@ -978,7 +978,7 @@ static int mlx5_ib_modify_port(struct ib_device *ibdev, u8 port, int mask,
 
 	mutex_lock(&dev->cap_mask_mutex);
 
-	err = mlx5_ib_query_port(ibdev, port, &attr);
+	err = ib_query_port(ibdev, port, &attr);
 	if (err)
 		goto out;
 

@@ -2564,6 +2564,7 @@ static int get_port_caps(struct mlx5_ib_dev *dev)
 	}
 
 	for (port = 1; port <= MLX5_CAP_GEN(dev->mdev, num_ports); port++) {
+		memset(pprops, 0, sizeof(struct ib_port_attr));
 		err = mlx5_ib_query_port(&dev->ib_dev, port, pprops);
 		if (err) {
 			mlx5_ib_warn(dev, "query_port %d failed %d\n",

@@ -2883,7 +2884,9 @@ static int mlx5_port_immutable(struct ib_device *ibdev, u8 port_num,
 	enum rdma_link_layer ll = mlx5_ib_port_link_layer(ibdev, port_num);
 	int err;
 
-	err = mlx5_ib_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = get_core_cap_flags(ibdev);
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
drivers/infiniband/hw/mthca/mthca_provider.c +5 -4
index d317087..ce16318 100644
@@ -146,7 +146,7 @@ static int mthca_query_port(struct ib_device *ibdev,
 	if (!in_mad || !out_mad)
 		goto out;
 
-	memset(props, 0, sizeof *props);
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	init_query_mad(in_mad);
 	in_mad->attr_id  = IB_SMP_ATTR_PORT_INFO;

@@ -212,7 +212,7 @@ static int mthca_modify_port(struct ib_device *ibdev,
 	if (mutex_lock_interruptible(&to_mdev(ibdev)->cap_mask_mutex))
 		return -ERESTARTSYS;
 
-	err = mthca_query_port(ibdev, port, &attr);
+	err = ib_query_port(ibdev, port, &attr);
 	if (err)
 		goto out;
 

@@ -1166,13 +1166,14 @@ static int mthca_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = mthca_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
drivers/infiniband/hw/nes/nes_verbs.c +3 -2
index aff9fb1..b7179f4 100644
@@ -475,7 +475,7 @@ static int nes_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr
 	struct nes_vnic *nesvnic = to_nesvnic(ibdev);
 	struct net_device *netdev = nesvnic->netdev;
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->max_mtu = IB_MTU_4096;
 

@@ -3670,13 +3670,14 @@ static int nes_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
+	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
 	err = nes_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
 
 	return 0;
 }
drivers/infiniband/hw/ocrdma/ocrdma_main.c +5 -4
index 8960715..3e43bdc 100644
@@ -93,15 +93,16 @@ static int ocrdma_port_immutable(struct ib_device *ibdev, u8 port_num,
 	int err;
 
 	dev = get_ocrdma_dev(ibdev);
-	err = ocrdma_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
+	if (ocrdma_is_udp_encap_supported(dev))
+		immutable->core_cap_flags |= RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
-	if (ocrdma_is_udp_encap_supported(dev))
-		immutable->core_cap_flags |= RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +1 -0
index 6af44f8..013d15c 100644
@@ -210,6 +210,7 @@ int ocrdma_query_port(struct ib_device *ibdev,
 	struct ocrdma_dev *dev;
 	struct net_device *netdev;
 
+	/* props being zeroed by the caller, avoid zeroing it here */
 	dev = get_ocrdma_dev(ibdev);
 	if (port > 1) {
 		pr_err("%s(%d) invalid_port=0x%x\n", __func__,
drivers/infiniband/hw/qedr/verbs.c +5 -4
index 57c8de2..91a16d3 100644
@@ -238,8 +238,8 @@ int qedr_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr *attr)
 	}
 
 	rdma_port = dev->ops->rdma_query_port(dev->rdma_ctx);
-	memset(attr, 0, sizeof(*attr));
 
+	/* *attr being zeroed by the caller, avoid zeroing it here */
 	if (rdma_port->port_state == QED_RDMA_PORT_UP) {
 		attr->state = IB_PORT_ACTIVE;
 		attr->phys_state = 5;

@@ -3549,14 +3549,15 @@ int qedr_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = qedr_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE |
+				    RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE |
-				    RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
drivers/infiniband/hw/qib/qib_verbs.c +1 -0
index 4b54c0d..9161574 100644
@@ -1303,6 +1303,7 @@ static int qib_query_port(struct rvt_dev_info *rdi, u8 port_num,
 	enum ib_mtu mtu;
 	u16 lid = ppd->lid;
 
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->lid = lid ? lid : be16_to_cpu(IB_LID_PERMISSIVE);
 	props->lmc = ppd->lmc;
 	props->state = dd->f_iblink_state(ppd->lastibcstat);
drivers/infiniband/hw/usnic/usnic_ib_main.c +3 -2
index dde0b23..4f5a45d 100644
@@ -321,11 +321,12 @@ static int usnic_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = usnic_ib_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_USNIC;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
-	immutable->core_cap_flags = RDMA_CORE_PORT_USNIC;
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
drivers/infiniband/hw/usnic/usnic_ib_verbs.c +1 -1
index 74819a7..0ba274f 100644
@@ -330,7 +330,7 @@ int usnic_ib_query_port(struct ib_device *ibdev, u8 port,
 
 	mutex_lock(&us_ibdev->usdev_lock);
 	__ethtool_get_link_ksettings(us_ibdev->netdev, &cmd);
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->lid = 0;
 	props->lmc = 1;
drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c +3 -2
index 231a1ce..b8b0081 100644
@@ -132,13 +132,14 @@ static int pvrdma_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = pvrdma_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 	return 0;
 }
drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c +2 -2
index 5489137..b3a982b 100644
@@ -135,7 +135,7 @@ int pvrdma_query_port(struct ib_device *ibdev, u8 port,
 		return err;
 	}
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->state = pvrdma_port_state_to_ib(resp->attrs.state);
 	props->max_mtu = pvrdma_mtu_to_ib(resp->attrs.max_mtu);

@@ -275,7 +275,7 @@ int pvrdma_modify_port(struct ib_device *ibdev, u8 port, int mask,
 	}
 
 	mutex_lock(&vdev->port_mutex);
-	ret = pvrdma_query_port(ibdev, port, &attr);
+	ret = ib_query_port(ibdev, port, &attr);
 	if (ret)
 		goto out;
drivers/infiniband/sw/rdmavt/vt.c +4 -3
index d430c2f..1165639 100644
@@ -165,7 +165,7 @@ static int rvt_query_port(struct ib_device *ibdev, u8 port_num,
 		return -EINVAL;
 
 	rvp = rdi->ports[port_index];
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->sm_lid = rvp->sm_lid;
 	props->sm_sl = rvp->sm_sl;
 	props->port_cap_flags = rvp->port_cap_flags;

@@ -326,13 +326,14 @@ static int rvt_get_port_immutable(struct ib_device *ibdev, u8 port_num,
 	if (port_index < 0)
 		return -EINVAL;
 
-	err = rvt_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = rdi->dparms.core_cap_flags;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = rdi->dparms.core_cap_flags;
 	immutable->max_mad_size = rdi->dparms.max_mad_size;
 
 	return 0;
drivers/infiniband/sw/rxe/rxe_verbs.c +4 -2
index beb7021..371ef3b 100644
@@ -86,6 +86,7 @@ static int rxe_query_port(struct ib_device *dev,
 
 	port = &rxe->port;
 
+	/* *attr being zeroed by the caller, avoid zeroing it here */
 	*attr = port->attr;
 
 	mutex_lock(&rxe->usdev_lock);

@@ -261,13 +262,14 @@ static int rxe_port_immutable(struct ib_device *dev, u8 port_num,
 	int err;
 	struct ib_port_attr attr;
 
-	err = rxe_query_port(dev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
+
+	err = ib_query_port(dev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
Signature · 6 lines
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): 20170119054102.20135-6-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org

#8 lore raw

RE: [PATCH rdma-next 5/5] IB: Query port through the core instead of directly calling the driver handler

Minor nit for safer code.

drivers/infiniband/hw/mlx4/alias_GUID.c static int set_guid_rec(struct ib_device *ibdev, · line 502

Quoted 12 lines

-----Original Message-----

drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
index 06020c5..4648374 100644
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
 	struct list_head *head =
 		&dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
 
+	memset(&attr, 0, sizeof(struct ib_port_attr));

Can you please change above code to,

	memset(&attr, 0, sizeof(attr));

drivers/infiniband/hw/mlx4/sysfs.c static int add_port_entries(struct mlx4_ib_dev · line 230

Quoted 11 lines
drivers/infiniband/hw/mlx4/sysfs.c +1 -0
index 69fb5ba..5835165 100644
@@ -226,6 +226,7 @@ static int add_port_entries(struct mlx4_ib_dev
 *device, int port_num)
 	int ret = 0 ;
 	struct ib_port_attr attr;
 
+	memset(&attr, 0, sizeof(struct ib_port_attr));

Ditto.

	memset(&attr, 0, sizeof(attr));

drivers/infiniband/hw/mlx5/main.c static int mlx5_query_port_roce(struct ib_device · line 177

Quoted 9 lines

a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 4e8e78f..176417b 100644

drivers/infiniband/hw/mlx5/main.c +1 -0
@@ -174,7 +174,7 @@ static int mlx5_query_port_roce(struct ib_device
 
 	for (port = 1; port <= MLX5_CAP_GEN(dev->mdev, num_ports);
 port++) {
+		memset(pprops, 0, sizeof(struct ib_port_attr));

Here too.

		memset(pprops, 0, sizeof(*pprops));

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): HE1PR0502MB3004E6D28AED46720C428576D17E0-692Kmc8YnlL9PhveBwpv4cDSnupUy6xnnBOFsp37pqbUKgpGm//BTAC/[email protected]

author
#10 lore raw

Re: [PATCH rdma-next 5/5] IB: Query port through the core instead of directly calling the driver handler

Quoted 13 lines · Parav Pandit

Minor nit for safer code.

Quoted 11 lines

-----Original Message-----

drivers/infiniband/hw/mlx4/alias_GUID.c +0 -0
index 06020c5..4648374 100644
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
 	struct list_head *head =
 		&dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
Quoted 4 lines

drivers/infiniband/hw/mlx4/alias_GUID.c static int set_guid_rec(struct ib_device *ibdev, · line 502

Quoted 1 line
drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
+	memset(&attr, 0, sizeof(struct ib_port_attr));

Can you please change above code to,

	memset(&attr, 0, sizeof(attr));

Sure,
We will wait a little bit more to gather additional comments if any and resubmit the series.

Thanks for the feedback.

Quoted 32 lines

drivers/infiniband/hw/mlx4/sysfs.c static int add_port_entries(struct mlx4_ib_dev · line 230

Quoted 12 lines
drivers/infiniband/hw/mlx4/sysfs.c +1 -0
index 69fb5ba..5835165 100644
@@ -226,6 +226,7 @@ static int add_port_entries(struct mlx4_ib_dev
 *device, int port_num)
 	int ret = 0 ;
 	struct ib_port_attr attr;
 
+	memset(&attr, 0, sizeof(struct ib_port_attr));

Ditto.

	memset(&attr, 0, sizeof(attr));

drivers/infiniband/hw/mlx5/main.c static int mlx5_query_port_roce(struct ib_device · line 177

Quoted 9 lines

a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 4e8e78f..176417b 100644

drivers/infiniband/hw/mlx5/main.c +1 -0
@@ -174,7 +174,7 @@ static int mlx5_query_port_roce(struct ib_device
 
 	for (port = 1; port <= MLX5_CAP_GEN(dev->mdev, num_ports);
 port++) {
+		memset(pprops, 0, sizeof(struct ib_port_attr));

Here too.

		memset(pprops, 0, sizeof(*pprops));

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#11 lore raw

Re: [PATCH rdma-next 5/5] IB: Query port through the core instead of directly calling the driver handler

In reply to a message not in the archive
Quoted 8 lines · Parav Pandit

Minor nit for safer code.

Quoted 6 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +0 -0
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
      struct list_head *head =
              &dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
Quoted 4 lines

drivers/infiniband/hw/mlx4/alias_GUID.c static int set_guid_rec(struct ib_device *ibdev, · line 502

Quoted 1 line
drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
+     memset(&attr, 0, sizeof(struct ib_port_attr));

Can you please change above code to,

        memset(&attr, 0, sizeof(attr));

sure, same for the rest of the places

Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#12 lore raw

Re: [PATCH rdma-next 5/5] IB: Query port through the core instead of directly calling the driver handler

In reply to a message not in the archive
Quoted 184 lines · Leon Romanovsky

From: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>

Change the drivers to call ib_query_port in their get port immutable handler instead of their own query port handler.

Doing this required to set the core cap flags of this device before the ib_query_port call is made, since the IB core might need these caps to serve the port query.

Drivers are ensured by the IB core that the port attributes passed to the port query verb implementation are zero, and hence we removed the zeroing from the drivers.

This patch doesn't add any new functionality.

Signed-off-by Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>
Reviewed-by Matan Barak <matanb-VPRAkNaXOzVWk0Htik3J/[email protected]>
Signed-off-by Leon Romanovsky <[email protected]>
---
 drivers/infiniband/hw/cxgb3/iwch_provider.c     |  7 ++++---
 drivers/infiniband/hw/cxgb4/provider.c          |  8 ++++----
 drivers/infiniband/hw/hfi1/verbs.c              |  1 +
 drivers/infiniband/hw/hns/hns_roce_main.c       |  7 ++++---
 drivers/infiniband/hw/i40iw/i40iw_verbs.c       |  8 ++++----
 drivers/infiniband/hw/mlx4/alias_GUID.c         |  1 +
 drivers/infiniband/hw/mlx4/main.c               | 18 +++++++++---------
 drivers/infiniband/hw/mlx4/sysfs.c              |  1 +
 drivers/infiniband/hw/mlx5/mad.c                |  2 +-
 drivers/infiniband/hw/mlx5/main.c               | 11 +++++++----
 drivers/infiniband/hw/mthca/mthca_provider.c    |  9 +++++----
 drivers/infiniband/hw/nes/nes_verbs.c           |  5 +++--
 drivers/infiniband/hw/ocrdma/ocrdma_main.c      |  9 +++++----
 drivers/infiniband/hw/ocrdma/ocrdma_verbs.c     |  1 +
 drivers/infiniband/hw/qedr/verbs.c              |  9 +++++----
 drivers/infiniband/hw/qib/qib_verbs.c           |  1 +
 drivers/infiniband/hw/usnic/usnic_ib_main.c     |  5 +++--
 drivers/infiniband/hw/usnic/usnic_ib_verbs.c    |  2 +-
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c  |  5 +++--
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c |  4 ++--
 drivers/infiniband/sw/rdmavt/vt.c               |  7 ++++---
 drivers/infiniband/sw/rxe/rxe_verbs.c           |  6 ++++--
 22 files changed, 73 insertions(+), 54 deletions(-)
drivers/infiniband/hw/cxgb3/iwch_provider.c +4 -3
index 9d5fe18..9e39252 100644
@@ -1133,7 +1133,7 @@ static int iwch_query_port(struct ib_device *ibdev,
 	dev = to_iwch_dev(ibdev);
 	netdev = dev->rdev.port_info.lldevs[port-1];
 
-	memset(props, 0, sizeof(struct ib_port_attr));
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->max_mtu = IB_MTU_4096;
 	if (netdev->mtu >= 4096)
 		props->active_mtu = IB_MTU_4096;

@@ -1338,13 +1338,14 @@ static int iwch_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = iwch_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
 
 	return 0;
 }
drivers/infiniband/hw/cxgb4/provider.c +4 -4
index 49b51b7..5b00b50a 100644
@@ -358,8 +358,7 @@ static int c4iw_query_port(struct ib_device *ibdev, u8 port,
 
 	dev = to_c4iw_dev(ibdev);
 	netdev = dev->rdev.lldi.ports[port-1];
-
-	memset(props, 0, sizeof(struct ib_port_attr));
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->max_mtu = IB_MTU_4096;
 	if (netdev->mtu >= 4096)
 		props->active_mtu = IB_MTU_4096;

@@ -505,13 +504,14 @@ static int c4iw_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = c4iw_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
 
 	return 0;
 }
drivers/infiniband/hw/hfi1/verbs.c +1 -0
index 95ed4d6..f2d4e04 100644
@@ -1384,6 +1384,7 @@ static int query_port(struct rvt_dev_info *rdi, u8 port_num,
 	struct hfi1_pportdata *ppd = &dd->pport[port_num - 1];
 	u16 lid = ppd->lid;
 
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->lid = lid ? lid : 0;
 	props->lmc = ppd->lmc;
 	/* OPA logical states match IB logical states */
drivers/infiniband/hw/hns/hns_roce_main.c +4 -3
index 4953d9c..7be8158 100644
@@ -249,7 +249,7 @@ static int hns_roce_query_port(struct ib_device *ib_dev, u8 port_num,
 	assert(port_num > 0);
 	port = port_num - 1;
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->max_mtu = hr_dev->caps.max_mtu;
 	props->gid_tbl_len = hr_dev->caps.gid_table_len[port];

@@ -400,14 +400,15 @@ static int hns_roce_port_immutable(struct ib_device *ib_dev, u8 port_num,
 	struct ib_port_attr attr;
 	int ret;
 
-	ret = hns_roce_query_port(ib_dev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
+
+	ret = ib_query_port(ib_dev, port_num, &attr);
 	if (ret)
 		return ret;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
 
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
drivers/infiniband/hw/i40iw/i40iw_verbs.c +4 -4
index 29e97df..3e83048 100644
@@ -97,8 +97,7 @@ static int i40iw_query_port(struct ib_device *ibdev,
 	struct i40iw_device *iwdev = to_iwdev(ibdev);
 	struct net_device *netdev = iwdev->netdev;
 
-	memset(props, 0, sizeof(*props));
-
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->max_mtu = IB_MTU_4096;
 	if (netdev->mtu >= 4096)
 		props->active_mtu = IB_MTU_4096;

@@ -2506,14 +2505,15 @@ static int i40iw_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = i40iw_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
 
 	return 0;
 }
drivers/infiniband/hw/mlx4/alias_GUID.c +0 -0
index 06020c5..4648374 100644

drivers/infiniband/hw/mlx4/alias_GUID.c static int set_guid_rec(struct ib_device *ibdev, · line 503

Quoted 5 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
 	struct list_head *head =
 		&dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
 
+	memset(&attr, 0, sizeof(struct ib_port_attr));
 	err = __mlx4_ib_query_port(ibdev, port, &attr, 1);

So why not call ib_query_port here as opposed to driver handler? As stated in your subject line.

Quoted 57 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +0 -0
 	if (err) {
 		pr_debug("mlx4_ib_query_port failed (err: %d), port: %d\n",
drivers/infiniband/hw/mlx4/main.c +9 -9
index 7019b94..646dcbf 100644
@@ -741,7 +741,7 @@ int __mlx4_ib_query_port(struct ib_device *ibdev, u8 port,
 {
 	int err;
 
-	memset(props, 0, sizeof *props);
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	err = mlx4_ib_port_link_layer(ibdev, port) == IB_LINK_LAYER_INFINIBAND ?
 		ib_link_query_port(ibdev, port, props, netw_view) :

@@ -1014,7 +1014,7 @@ static int mlx4_ib_modify_port(struct ib_device *ibdev, u8 port, int mask,
 
 	mutex_lock(&mdev->cap_mask_mutex);
 
-	err = mlx4_ib_query_port(ibdev, port, &attr);
+	err = ib_query_port(ibdev, port, &attr);
 	if (err)
 		goto out;
 

@@ -2527,13 +2527,6 @@ static int mlx4_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct mlx4_ib_dev *mdev = to_mdev(ibdev);
 	int err;
 
-	err = mlx4_ib_query_port(ibdev, port_num, &attr);
-	if (err)
-		return err;
-
-	immutable->pkey_tbl_len = attr.pkey_tbl_len;
-	immutable->gid_tbl_len = attr.gid_tbl_len;
-
 	if (mlx4_ib_port_link_layer(ibdev, port_num) == IB_LINK_LAYER_INFINIBAND) {
 		immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB;
 		immutable->max_mad_size = IB_MGMT_MAD_SIZE;

@@ -2549,6 +2542,13 @@ static int mlx4_port_immutable(struct ib_device *ibdev, u8 port_num,
 			immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 	}
 
+	err = ib_query_port(ibdev, port_num, &attr);
+	if (err)
+		return err;
+
+	immutable->pkey_tbl_len = attr.pkey_tbl_len;
+	immutable->gid_tbl_len = attr.gid_tbl_len;
+
 	return 0;
 }
drivers/infiniband/hw/mlx4/sysfs.c +0 -0
index 69fb5ba..5835165 100644

drivers/infiniband/hw/mlx4/sysfs.c static int add_port_entries(struct mlx4_ib_dev *device, int port_num) · line 231

Quoted 6 lines
drivers/infiniband/hw/mlx4/sysfs.c +1 -0
@@ -226,6 +226,7 @@ static int add_port_entries(struct mlx4_ib_dev *device, int port_num)
 	int ret = 0 ;
 	struct ib_port_attr attr;
 
+	memset(&attr, 0, sizeof(struct ib_port_attr));
 	/* get the physical gid and pkey table sizes.*/
 	ret = __mlx4_ib_query_port(&device->ib_dev, port_num, &attr, 1);

Ditto as previous comment.

Quoted 355 lines
drivers/infiniband/hw/mlx4/sysfs.c +0 -0
 	if (ret)
drivers/infiniband/hw/mlx5/mad.c +1 -1
index 39e5848..4e58b8f 100644
@@ -515,7 +515,7 @@ int mlx5_query_mad_ifc_port(struct ib_device *ibdev, u8 port,
 	if (!in_mad || !out_mad)
 		goto out;
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	init_query_mad(in_mad);
 	in_mad->attr_id  = IB_SMP_ATTR_PORT_INFO;
drivers/infiniband/hw/mlx5/main.c +7 -4
index 4e8e78f..176417b 100644
@@ -174,7 +174,7 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
 	enum ib_mtu ndev_ib_mtu;
 	u16 qkey_viol_cntr;
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->port_cap_flags  |= IB_PORT_CM_SUP;
 	props->port_cap_flags  |= IB_PORT_IP_BASED_GIDS;

@@ -830,7 +830,7 @@ static int mlx5_query_hca_port(struct ib_device *ibdev, u8 port,
 		goto out;
 	}
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	err = mlx5_query_hca_vport_context(mdev, 0, port, 0, rep);
 	if (err)

@@ -978,7 +978,7 @@ static int mlx5_ib_modify_port(struct ib_device *ibdev, u8 port, int mask,
 
 	mutex_lock(&dev->cap_mask_mutex);
 
-	err = mlx5_ib_query_port(ibdev, port, &attr);
+	err = ib_query_port(ibdev, port, &attr);
 	if (err)
 		goto out;
 

@@ -2564,6 +2564,7 @@ static int get_port_caps(struct mlx5_ib_dev *dev)
 	}
 
 	for (port = 1; port <= MLX5_CAP_GEN(dev->mdev, num_ports); port++) {
+		memset(pprops, 0, sizeof(struct ib_port_attr));
 		err = mlx5_ib_query_port(&dev->ib_dev, port, pprops);
 		if (err) {
 			mlx5_ib_warn(dev, "query_port %d failed %d\n",

@@ -2883,7 +2884,9 @@ static int mlx5_port_immutable(struct ib_device *ibdev, u8 port_num,
 	enum rdma_link_layer ll = mlx5_ib_port_link_layer(ibdev, port_num);
 	int err;
 
-	err = mlx5_ib_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = get_core_cap_flags(ibdev);
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
drivers/infiniband/hw/mthca/mthca_provider.c +5 -4
index d317087..ce16318 100644
@@ -146,7 +146,7 @@ static int mthca_query_port(struct ib_device *ibdev,
 	if (!in_mad || !out_mad)
 		goto out;
 
-	memset(props, 0, sizeof *props);
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	init_query_mad(in_mad);
 	in_mad->attr_id  = IB_SMP_ATTR_PORT_INFO;

@@ -212,7 +212,7 @@ static int mthca_modify_port(struct ib_device *ibdev,
 	if (mutex_lock_interruptible(&to_mdev(ibdev)->cap_mask_mutex))
 		return -ERESTARTSYS;
 
-	err = mthca_query_port(ibdev, port, &attr);
+	err = ib_query_port(ibdev, port, &attr);
 	if (err)
 		goto out;
 

@@ -1166,13 +1166,14 @@ static int mthca_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = mthca_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
drivers/infiniband/hw/nes/nes_verbs.c +3 -2
index aff9fb1..b7179f4 100644
@@ -475,7 +475,7 @@ static int nes_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr
 	struct nes_vnic *nesvnic = to_nesvnic(ibdev);
 	struct net_device *netdev = nesvnic->netdev;
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->max_mtu = IB_MTU_4096;
 

@@ -3670,13 +3670,14 @@ static int nes_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
+	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
 	err = nes_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
 
 	return 0;
 }
drivers/infiniband/hw/ocrdma/ocrdma_main.c +5 -4
index 8960715..3e43bdc 100644
@@ -93,15 +93,16 @@ static int ocrdma_port_immutable(struct ib_device *ibdev, u8 port_num,
 	int err;
 
 	dev = get_ocrdma_dev(ibdev);
-	err = ocrdma_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
+	if (ocrdma_is_udp_encap_supported(dev))
+		immutable->core_cap_flags |= RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
-	if (ocrdma_is_udp_encap_supported(dev))
-		immutable->core_cap_flags |= RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +1 -0
index 6af44f8..013d15c 100644
@@ -210,6 +210,7 @@ int ocrdma_query_port(struct ib_device *ibdev,
 	struct ocrdma_dev *dev;
 	struct net_device *netdev;
 
+	/* props being zeroed by the caller, avoid zeroing it here */
 	dev = get_ocrdma_dev(ibdev);
 	if (port > 1) {
 		pr_err("%s(%d) invalid_port=0x%x\n", __func__,
drivers/infiniband/hw/qedr/verbs.c +5 -4
index 57c8de2..91a16d3 100644
@@ -238,8 +238,8 @@ int qedr_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr *attr)
 	}
 
 	rdma_port = dev->ops->rdma_query_port(dev->rdma_ctx);
-	memset(attr, 0, sizeof(*attr));
 
+	/* *attr being zeroed by the caller, avoid zeroing it here */
 	if (rdma_port->port_state == QED_RDMA_PORT_UP) {
 		attr->state = IB_PORT_ACTIVE;
 		attr->phys_state = 5;

@@ -3549,14 +3549,15 @@ int qedr_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = qedr_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE |
+				    RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE |
-				    RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
drivers/infiniband/hw/qib/qib_verbs.c +1 -0
index 4b54c0d..9161574 100644
@@ -1303,6 +1303,7 @@ static int qib_query_port(struct rvt_dev_info *rdi, u8 port_num,
 	enum ib_mtu mtu;
 	u16 lid = ppd->lid;
 
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->lid = lid ? lid : be16_to_cpu(IB_LID_PERMISSIVE);
 	props->lmc = ppd->lmc;
 	props->state = dd->f_iblink_state(ppd->lastibcstat);
drivers/infiniband/hw/usnic/usnic_ib_main.c +3 -2
index dde0b23..4f5a45d 100644
@@ -321,11 +321,12 @@ static int usnic_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = usnic_ib_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_USNIC;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
-	immutable->core_cap_flags = RDMA_CORE_PORT_USNIC;
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
drivers/infiniband/hw/usnic/usnic_ib_verbs.c +1 -1
index 74819a7..0ba274f 100644
@@ -330,7 +330,7 @@ int usnic_ib_query_port(struct ib_device *ibdev, u8 port,
 
 	mutex_lock(&us_ibdev->usdev_lock);
 	__ethtool_get_link_ksettings(us_ibdev->netdev, &cmd);
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->lid = 0;
 	props->lmc = 1;
drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c +3 -2
index 231a1ce..b8b0081 100644
@@ -132,13 +132,14 @@ static int pvrdma_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	int err;
 
-	err = pvrdma_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 	return 0;
 }
drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c +2 -2
index 5489137..b3a982b 100644
@@ -135,7 +135,7 @@ int pvrdma_query_port(struct ib_device *ibdev, u8 port,
 		return err;
 	}
 
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	props->state = pvrdma_port_state_to_ib(resp->attrs.state);
 	props->max_mtu = pvrdma_mtu_to_ib(resp->attrs.max_mtu);

@@ -275,7 +275,7 @@ int pvrdma_modify_port(struct ib_device *ibdev, u8 port, int mask,
 	}
 
 	mutex_lock(&vdev->port_mutex);
-	ret = pvrdma_query_port(ibdev, port, &attr);
+	ret = ib_query_port(ibdev, port, &attr);
 	if (ret)
 		goto out;
drivers/infiniband/sw/rdmavt/vt.c +4 -3
index d430c2f..1165639 100644
@@ -165,7 +165,7 @@ static int rvt_query_port(struct ib_device *ibdev, u8 port_num,
 		return -EINVAL;
 
 	rvp = rdi->ports[port_index];
-	memset(props, 0, sizeof(*props));
+	/* props being zeroed by the caller, avoid zeroing it here */
 	props->sm_lid = rvp->sm_lid;
 	props->sm_sl = rvp->sm_sl;
 	props->port_cap_flags = rvp->port_cap_flags;

@@ -326,13 +326,14 @@ static int rvt_get_port_immutable(struct ib_device *ibdev, u8 port_num,
 	if (port_index < 0)
 		return -EINVAL;
 
-	err = rvt_query_port(ibdev, port_num, &attr);
+	immutable->core_cap_flags = rdi->dparms.core_cap_flags;
+
+	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = rdi->dparms.core_cap_flags;
 	immutable->max_mad_size = rdi->dparms.max_mad_size;
 
 	return 0;
drivers/infiniband/sw/rxe/rxe_verbs.c +4 -2
index beb7021..371ef3b 100644
@@ -86,6 +86,7 @@ static int rxe_query_port(struct ib_device *dev,
 
 	port = &rxe->port;
 
+	/* *attr being zeroed by the caller, avoid zeroing it here */
 	*attr = port->attr;
 
 	mutex_lock(&rxe->usdev_lock);

@@ -261,13 +262,14 @@ static int rxe_port_immutable(struct ib_device *dev, u8 port_num,
 	int err;
 	struct ib_port_attr attr;
 
-	err = rxe_query_port(dev, port_num, &attr);
+	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
+
+	err = ib_query_port(dev, port_num, &attr);
 	if (err)
 		return err;
 
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;
 	immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
 	return 0;
Signature · 6 lines
2.10.2

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
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 [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): 20170124205938.GA5524-GOXS9JX10wfOxmVO0tvppfooFf0ArEBIu+b9c/[email protected]

author
#14 lore raw

Re: [PATCH rdma-next 5/5] IB: Query port through the core instead of directly calling the driver handler

Quoted 20 lines · Shiraz Saleem
Quoted 19 lines · Leon Romanovsky

From: Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>

Change the drivers to call ib_query_port in their get port immutable handler instead of their own query port handler.

Doing this required to set the core cap flags of this device before the ib_query_port call is made, since the IB core might need these caps to serve the port query.

Drivers are ensured by the IB core that the port attributes passed to the port query verb implementation are zero, and hence we removed the zeroing from the drivers.

This patch doesn't add any new functionality.

Signed-off-by Or Gerlitz <ogerlitz-VPRAkNaXOzVWk0Htik3J/[email protected]>
Reviewed-by Matan Barak <matanb-VPRAkNaXOzVWk0Htik3J/[email protected]>
Signed-off-by Leon Romanovsky <[email protected]>

---

<snip>

Quoted 10 lines

drivers/infiniband/hw/mlx4/alias_GUID.c static int set_guid_rec(struct ib_device *ibdev, · line 503

Quoted 8 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
 	struct list_head *head =
 		&dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
 
+	memset(&attr, 0, sizeof(struct ib_port_attr));
 	err = __mlx4_ib_query_port(ibdev, port, &attr, 1);

So why not call ib_query_port here as opposed to driver handler? As stated in your subject line.

This is call to mlx4 internal implementation of ib_query_port. This implementation takes an additional parameter. In this case it is "1", while call to ib_query_port will call to this function with "0".

Quoted 61 lines
Quoted 61 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +0 -0
 	if (err) {
 		pr_debug("mlx4_ib_query_port failed (err: %d), port: %d\n",
drivers/infiniband/hw/mlx4/main.c +9 -9
index 7019b94..646dcbf 100644
@@ -741,7 +741,7 @@ int __mlx4_ib_query_port(struct ib_device *ibdev, u8 port,
 {
 	int err;
 
-	memset(props, 0, sizeof *props);
+	/* props being zeroed by the caller, avoid zeroing it here */
 
 	err = mlx4_ib_port_link_layer(ibdev, port) == IB_LINK_LAYER_INFINIBAND ?
 		ib_link_query_port(ibdev, port, props, netw_view) :

@@ -1014,7 +1014,7 @@ static int mlx4_ib_modify_port(struct ib_device *ibdev, u8 port, int mask,
 
 	mutex_lock(&mdev->cap_mask_mutex);
 
-	err = mlx4_ib_query_port(ibdev, port, &attr);
+	err = ib_query_port(ibdev, port, &attr);
 	if (err)
 		goto out;
 

@@ -2527,13 +2527,6 @@ static int mlx4_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct mlx4_ib_dev *mdev = to_mdev(ibdev);
 	int err;
 
-	err = mlx4_ib_query_port(ibdev, port_num, &attr);
-	if (err)
-		return err;
-
-	immutable->pkey_tbl_len = attr.pkey_tbl_len;
-	immutable->gid_tbl_len = attr.gid_tbl_len;
-
 	if (mlx4_ib_port_link_layer(ibdev, port_num) == IB_LINK_LAYER_INFINIBAND) {
 		immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB;
 		immutable->max_mad_size = IB_MGMT_MAD_SIZE;

@@ -2549,6 +2542,13 @@ static int mlx4_port_immutable(struct ib_device *ibdev, u8 port_num,
 			immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 	}
 
+	err = ib_query_port(ibdev, port_num, &attr);
+	if (err)
+		return err;
+
+	immutable->pkey_tbl_len = attr.pkey_tbl_len;
+	immutable->gid_tbl_len = attr.gid_tbl_len;
+
 	return 0;
 }
drivers/infiniband/hw/mlx4/sysfs.c +0 -0
index 69fb5ba..5835165 100644
@@ -226,6 +226,7 @@ static int add_port_entries(struct mlx4_ib_dev *device, int port_num)
 	int ret = 0 ;
 	struct ib_port_attr attr;
Quoted 5 lines

drivers/infiniband/hw/mlx4/sysfs.c static int add_port_entries(struct mlx4_ib_dev *device, int port_num) · line 231

Quoted 3 lines
drivers/infiniband/hw/mlx4/sysfs.c +1 -0
+	memset(&attr, 0, sizeof(struct ib_port_attr));
 	/* get the physical gid and pkey table sizes.*/
 	ret = __mlx4_ib_query_port(&device->ib_dev, port_num, &attr, 1);

Ditto as previous comment.

The same answer as before.

Quoted 2 lines

A message that is not in the archive (sent off-list or not yet received): 20170125123920.GO6005-U/[email protected]

#16 lore raw

Re: [PATCH rdma-next 5/5] IB: Query port through the core instead of directly calling the driver handler

Quoted 10 lines · Leon Romanovsky
Quoted 6 lines · Shiraz Saleem
Quoted 6 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +0 -0
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
    struct list_head *head =
            &dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
Quoted 8 lines
Quoted 4 lines

drivers/infiniband/hw/mlx4/alias_GUID.c static int set_guid_rec(struct ib_device *ibdev, · line 503

Quoted 2 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
+   memset(&attr, 0, sizeof(struct ib_port_attr));
    err = __mlx4_ib_query_port(ibdev, port, &attr, 1);

So why not call ib_query_port here as opposed to driver handler? As stated in your subject line.

This is call to mlx4 internal implementation of ib_query_port. This implementation takes an additional parameter. In this case it is "1", while call to ib_query_port will call to this function with "0".

The mlx4 IB driver is used all of native, SRIOV PF and SRIOV VF instances.

Under SRIOV, the port has network view and host view. This internal query port call (and the other one you commented on) is done in a network view by the PF.

The calls that come from the IB core are done with host view.

As such I had to retain these two internal calls.

Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): CAJ3xEMjfxPs85OQP2v2MvYzU9ajUELs7_3r0nA-=TTa-tp-hSQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org

#18 lore raw

Re: [PATCH rdma-next 5/5] IB: Query port through the core instead of directly calling the driver handler

Quoted 20 lines · Or Gerlitz
Quoted 10 lines · Leon Romanovsky
Quoted 6 lines · Shiraz Saleem
Quoted 6 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +0 -0
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
    struct list_head *head =
            &dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
Quoted 8 lines
Quoted 4 lines

drivers/infiniband/hw/mlx4/alias_GUID.c static int set_guid_rec(struct ib_device *ibdev, · line 503

Quoted 2 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
+   memset(&attr, 0, sizeof(struct ib_port_attr));
    err = __mlx4_ib_query_port(ibdev, port, &attr, 1);

So why not call ib_query_port here as opposed to driver handler? As stated in your subject line.

This is call to mlx4 internal implementation of ib_query_port. This implementation takes an additional parameter. In this case it is "1", while call to ib_query_port will call to this function with "0".

Quoted 8 lines

The mlx4 IB driver is used all of native, SRIOV PF and SRIOV VF instances.

Under SRIOV, the port has network view and host view. This internal query port call (and the other one you commented on) is done in a network view by the PF.

The calls that come from the IB core are done with host view.

As such I had to retain these two internal calls.

Ok. Understand it better now. BTW this parameter 'netw_view' doesnt appear to be used in one of the functions eth_link_query_port called from mlx4 query port implementation.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): 20170125180551.GA8964-GOXS9JX10wfOxmVO0tvppfooFf0ArEBIu+b9c/[email protected]

#20 lore raw

Re: [PATCH rdma-next 5/5] IB: Query port through the core instead of directly calling the driver handler

Quoted 26 lines · Shiraz Saleem
Quoted 25 lines · Or Gerlitz
Quoted 10 lines · Leon Romanovsky
Quoted 6 lines · Shiraz Saleem
Quoted 6 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +0 -0
@@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev,
    struct list_head *head =
            &dev->sriov.alias_guid.ports_guid[port - 1].cb_list;
Quoted 8 lines
Quoted 4 lines

drivers/infiniband/hw/mlx4/alias_GUID.c static int set_guid_rec(struct ib_device *ibdev, · line 503

Quoted 2 lines
drivers/infiniband/hw/mlx4/alias_GUID.c +1 -0
+   memset(&attr, 0, sizeof(struct ib_port_attr));
    err = __mlx4_ib_query_port(ibdev, port, &attr, 1);

So why not call ib_query_port here as opposed to driver handler? As stated in your subject line.

This is call to mlx4 internal implementation of ib_query_port. This implementation takes an additional parameter. In this case it is "1", while call to ib_query_port will call to this function with "0".

The mlx4 IB driver is used all of native, SRIOV PF and SRIOV VF instances.

Under SRIOV, the port has network view and host view. This internal query port call (and the other one you commented on) is done in a network view by the PF.

Quoted 7 lines
Quoted 3 lines

The calls that come from the IB core are done with host view.

As such I had to retain these two internal calls.

Ok. Understand it better now. BTW this parameter 'netw_view' doesnt appear to be used in one of the functions eth_link_query_port called from mlx4 query port implementation.

Thanks for pointing it.
I sent cleanup patch to ML.

Quoted 4 lines

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html