[PATCH net-next 0/3] be2net: patch set

[PATCH net-next 0/3] be2net: patch set

[Sathya Perla]

Patch 1 fixes a regression caused by a previous commit on net-next.
Old versions of BE3 FW may not support cmds to re-provision (and hence
optimize) resources/queues in SR-IOV config. Do not treat this FW cmd
failure as fatal and fail the function initialization. Instead, just
enable SR-IOV with the resources provided by the FW.

Patch 2 ignores a VF mac address setting if the new mac is already active
on the VF.

Patch 3 adds support to delete a FW-dump via ethtool on Lancer adapters.

Please consider applying this patch-set to the net-next tree. Thanks!

Kalesh AP (1):
  be2net: support deleting FW dump via ethtool (only for Lancer)

Sathya Perla (1):
  be2net: ignore get/set profile FW cmd failures

Vasundhara Volam (1):
  be2net: ignore VF mac address setting for the same mac

 drivers/net/ethernet/emulex/benet/be.h         |    2 +
 drivers/net/ethernet/emulex/benet/be_cmds.c    |   48 ++++++++++++++-
 drivers/net/ethernet/emulex/benet/be_cmds.h    |   10 +++
 drivers/net/ethernet/emulex/benet/be_ethtool.c |   17 +++---
 drivers/net/ethernet/emulex/benet/be_main.c    |   75 ++++++++++++++----------
 5 files changed, 109 insertions(+), 43 deletions(-)

[PATCH net-next 1/3] be2net: ignore get/set profile FW cmd failures

[Sathya Perla]

Old versions of BE3 FW may not support cmds to re-provision (and hence
optimize) resources/queues in SR-IOV config. Do not treat this FW cmd
failure as fatal and fail the function initialization. Instead, just
enable SR-IOV with the resources provided by the FW.

Prior to the "create optimal number of queues on SR-IOV config" patch
such failures were ignored.
Fixes: bec84e6b2 ("create optimal number of queues on SR-IOV config")

Reported-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
---
 drivers/net/ethernet/emulex/benet/be_main.c |   67 +++++++++++++++------------
 1 files changed, 37 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 9c50814..da4d386 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -3342,22 +3342,17 @@ static int be_get_sriov_config(struct be_adapter *adapter)
 {
 	struct device *dev = &adapter->pdev->dev;
 	struct be_resources res = {0};
-	int status, max_vfs, old_vfs;
-
-	status = be_cmd_get_profile_config(adapter, &res, 0);
-	if (status)
-		return status;
-
-	adapter->pool_res = res;
+	int max_vfs, old_vfs;
 
 	/* Some old versions of BE3 FW don't report max_vfs value */
+	be_cmd_get_profile_config(adapter, &res, 0);
+
 	if (BE3_chip(adapter) && !res.max_vfs) {
 		max_vfs = pci_sriov_get_totalvfs(adapter->pdev);
 		res.max_vfs = max_vfs > 0 ? min(MAX_VFS, max_vfs) : 0;
 	}
 
-	adapter->pool_res.max_vfs = res.max_vfs;
-	pci_sriov_set_totalvfs(adapter->pdev, be_max_vfs(adapter));
+	adapter->pool_res = res;
 
 	if (!be_max_vfs(adapter)) {
 		if (num_vfs)
@@ -3366,6 +3361,8 @@ static int be_get_sriov_config(struct be_adapter *adapter)
 		return 0;
 	}
 
+	pci_sriov_set_totalvfs(adapter->pdev, be_max_vfs(adapter));
+
 	/* validate num_vfs module param */
 	old_vfs = pci_num_vf(adapter->pdev);
 	if (old_vfs) {
@@ -3423,6 +3420,35 @@ static int be_get_resources(struct be_adapter *adapter)
 	return 0;
 }
 
+static void be_sriov_config(struct be_adapter *adapter)
+{
+	struct device *dev = &adapter->pdev->dev;
+	int status;
+
+	status = be_get_sriov_config(adapter);
+	if (status) {
+		dev_err(dev, "Failed to query SR-IOV configuration\n");
+		dev_err(dev, "SR-IOV cannot be enabled\n");
+		return;
+	}
+
+	/* When the HW is in SRIOV capable configuration, the PF-pool
+	 * resources are equally distributed across the max-number of
+	 * VFs. The user may request only a subset of the max-vfs to be
+	 * enabled. Based on num_vfs, redistribute the resources across
+	 * num_vfs so that each VF will have access to more number of
+	 * resources. This facility is not available in BE3 FW.
+	 * Also, this is done by FW in Lancer chip.
+	 */
+	if (be_max_vfs(adapter) && !pci_num_vf(adapter->pdev)) {
+		status = be_cmd_set_sriov_config(adapter,
+						 adapter->pool_res,
+						 adapter->num_vfs);
+		if (status)
+			dev_err(dev, "Failed to optimize SR-IOV resources\n");
+	}
+}
+
 static int be_get_config(struct be_adapter *adapter)
 {
 	u16 profile_id;
@@ -3439,27 +3465,8 @@ static int be_get_config(struct be_adapter *adapter)
 				 "Using profile 0x%x\n", profile_id);
 	}
 
-	if (!BE2_chip(adapter) && be_physfn(adapter)) {
-		status = be_get_sriov_config(adapter);
-		if (status)
-			return status;
-
-		/* When the HW is in SRIOV capable configuration, the PF-pool
-		 * resources are equally distributed across the max-number of
-		 * VFs. The user may request only a subset of the max-vfs to be
-		 * enabled. Based on num_vfs, redistribute the resources across
-		 * num_vfs so that each VF will have access to more number of
-		 * resources. This facility is not available in BE3 FW.
-		 * Also, this is done by FW in Lancer chip.
-		 */
-		if (!pci_num_vf(adapter->pdev)) {
-			status = be_cmd_set_sriov_config(adapter,
-							 adapter->pool_res,
-							 adapter->num_vfs);
-			if (status)
-				return status;
-		}
-	}
+	if (!BE2_chip(adapter) && be_physfn(adapter))
+		be_sriov_config(adapter);
 
 	status = be_get_resources(adapter);
 	if (status)
-- 
1.7.1

[PATCH net-next 2/3] be2net: ignore VF mac address setting for the same mac

[Sathya Perla]

From: Vasundhara Volam <vasundhara.volam@emulex.com>

ndo_set_vf_mac() call may be issued for a mac-addr that is already
active on a VF. If so, silently ignore the request.

Signed-off-by: Vasundhara Volam <vasundhara.volam@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
---
 drivers/net/ethernet/emulex/benet/be_main.c |    6 ++++++
 1 files changed, 6 insertions(+), 0 deletions(-)

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index da4d386..db4ff14 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1270,6 +1270,12 @@ static int be_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
 	if (!is_valid_ether_addr(mac) || vf >= adapter->num_vfs)
 		return -EINVAL;
 
+	/* Proceed further only if user provided MAC is different
+	 * from active MAC
+	 */
+	if (ether_addr_equal(mac, vf_cfg->mac_addr))
+		return 0;
+
 	if (BEx_chip(adapter)) {
 		be_cmd_pmac_del(adapter, vf_cfg->if_handle, vf_cfg->pmac_id,
 				vf + 1);
-- 
1.7.1

[PATCH net-next 3/3] be2net: support deleting FW dump via ethtool (only for Lancer)

[Sathya Perla]

From: Kalesh AP <kalesh.purayil@emulex.com>

This patch adds support to delete an existing FW-dump in Lancer via ethtool.
Initiating a new dump is not allowed if a FW dump is already present in the
adapter. The existing dump has to be first explicitly deleted.

Signed-off-by: Kalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
---
 drivers/net/ethernet/emulex/benet/be.h         |    2 +
 drivers/net/ethernet/emulex/benet/be_cmds.c    |   48 ++++++++++++++++++++++-
 drivers/net/ethernet/emulex/benet/be_cmds.h    |   10 +++++
 drivers/net/ethernet/emulex/benet/be_ethtool.c |   17 ++++----
 4 files changed, 65 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index 0048ef8..811f135 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -383,8 +383,10 @@ enum vf_state {
 
 #define BE_UC_PMAC_COUNT			30
 #define BE_VF_UC_PMAC_COUNT			2
+
 /* Ethtool set_dump flags */
 #define LANCER_INITIATE_FW_DUMP			0x1
+#define LANCER_DELETE_FW_DUMP			0x2
 
 struct phy_info {
 	u8 transceiver;
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 791094c..4370ec1 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -2240,6 +2240,34 @@ err_unlock:
 	return status;
 }
 
+int lancer_cmd_delete_object(struct be_adapter *adapter, const char *obj_name)
+{
+	struct lancer_cmd_req_delete_object *req;
+	struct be_mcc_wrb *wrb;
+	int status;
+
+	spin_lock_bh(&adapter->mcc_lock);
+
+	wrb = wrb_from_mccq(adapter);
+	if (!wrb) {
+		status = -EBUSY;
+		goto err;
+	}
+
+	req = embedded_payload(wrb);
+
+	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+			       OPCODE_COMMON_DELETE_OBJECT,
+			       sizeof(*req), wrb, NULL);
+
+	strcpy(req->object_name, obj_name);
+
+	status = be_mcc_notify_wait(adapter);
+err:
+	spin_unlock_bh(&adapter->mcc_lock);
+	return status;
+}
+
 int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 			   u32 data_size, u32 data_offset, const char *obj_name,
 			   u32 *data_read, u32 *eof, u8 *addn_status)
@@ -3805,13 +3833,19 @@ bool dump_present(struct be_adapter *adapter)
 
 int lancer_initiate_dump(struct be_adapter *adapter)
 {
+	struct device *dev = &adapter->pdev->dev;
 	int status;
 
+	if (dump_present(adapter)) {
+		dev_info(dev, "Previous dump not cleared, not forcing dump\n");
+		return -EEXIST;
+	}
+
 	/* give firmware reset and diagnostic dump */
 	status = lancer_physdev_ctrl(adapter, PHYSDEV_CONTROL_FW_RESET_MASK |
 				     PHYSDEV_CONTROL_DD_MASK);
 	if (status < 0) {
-		dev_err(&adapter->pdev->dev, "Firmware reset failed\n");
+		dev_err(dev, "FW reset failed\n");
 		return status;
 	}
 
@@ -3820,13 +3854,21 @@ int lancer_initiate_dump(struct be_adapter *adapter)
 		return status;
 
 	if (!dump_present(adapter)) {
-		dev_err(&adapter->pdev->dev, "Dump image not present\n");
-		return -1;
+		dev_err(dev, "FW dump not generated\n");
+		return -EIO;
 	}
 
 	return 0;
 }
 
+int lancer_delete_dump(struct be_adapter *adapter)
+{
+	int status;
+
+	status = lancer_cmd_delete_object(adapter, LANCER_FW_DUMP_FILE);
+	return be_cmd_status(status);
+}
+
 /* Uses sync mcc */
 int be_cmd_enable_vf(struct be_adapter *adapter, u8 domain)
 {
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index 03e8a15..5284b82 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -231,6 +231,7 @@ struct be_mcc_mailbox {
 #define OPCODE_COMMON_GET_FN_PRIVILEGES			170
 #define OPCODE_COMMON_READ_OBJECT			171
 #define OPCODE_COMMON_WRITE_OBJECT			172
+#define OPCODE_COMMON_DELETE_OBJECT			174
 #define OPCODE_COMMON_MANAGE_IFACE_FILTERS		193
 #define OPCODE_COMMON_GET_IFACE_LIST			194
 #define OPCODE_COMMON_ENABLE_DISABLE_VF			196
@@ -1253,6 +1254,13 @@ struct lancer_cmd_resp_read_object {
 	u32 eof;
 };
 
+struct lancer_cmd_req_delete_object {
+	struct be_cmd_req_hdr hdr;
+	u32 rsvd1;
+	u32 rsvd2;
+	u8 object_name[104];
+};
+
 /************************ WOL *******************************/
 struct be_cmd_req_acpi_wol_magic_config{
 	struct be_cmd_req_hdr hdr;
@@ -2067,6 +2075,7 @@ int lancer_cmd_write_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd,
 			   u32 data_size, u32 data_offset, const char *obj_name,
 			   u32 *data_read, u32 *eof, u8 *addn_status);
+int lancer_cmd_delete_object(struct be_adapter *adapter, const char *obj_name);
 int be_cmd_get_flash_crc(struct be_adapter *adapter, u8 *flashed_crc,
 			  u16 optype, int offset);
 int be_cmd_enable_magic_wol(struct be_adapter *adapter, u8 *mac,
@@ -2120,6 +2129,7 @@ int be_cmd_set_ext_fat_capabilites(struct be_adapter *adapter,
 				   struct be_fat_conf_params *cfgs);
 int lancer_physdev_ctrl(struct be_adapter *adapter, u32 mask);
 int lancer_initiate_dump(struct be_adapter *adapter);
+int lancer_delete_dump(struct be_adapter *adapter);
 bool dump_present(struct be_adapter *adapter);
 int lancer_test_and_set_rdy_state(struct be_adapter *adapter);
 int be_cmd_query_port_name(struct be_adapter *adapter, u8 *port_name);
diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c
index 25f516d..0cd3311 100644
--- a/drivers/net/ethernet/emulex/benet/be_ethtool.c
+++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c
@@ -681,22 +681,21 @@ static int be_set_dump(struct net_device *netdev, struct ethtool_dump *dump)
 	struct device *dev = &adapter->pdev->dev;
 	int status;
 
-	if (!lancer_chip(adapter)) {
-		dev_err(dev, "FW dump not supported\n");
+	if (!lancer_chip(adapter) ||
+	    !check_privilege(adapter, MAX_PRIVILEGES))
 		return -EOPNOTSUPP;
-	}
-
-	if (dump_present(adapter)) {
-		dev_err(dev, "Previous dump not cleared, not forcing dump\n");
-		return 0;
-	}
 
 	switch (dump->flag) {
 	case LANCER_INITIATE_FW_DUMP:
 		status = lancer_initiate_dump(adapter);
 		if (!status)
-			dev_info(dev, "F/w dump initiated successfully\n");
+			dev_info(dev, "FW dump initiated successfully\n");
 		break;
+	case LANCER_DELETE_FW_DUMP:
+		status = lancer_delete_dump(adapter);
+		if (!status)
+			dev_info(dev, "FW dump deleted successfully\n");
+	break;
 	default:
 		dev_err(dev, "Invalid dump level: 0x%x\n", dump->flag);
 		return -EINVAL;
-- 
1.7.1

Re: [PATCH net-next 0/3] be2net: patch set

[David Miller]

From: Sathya Perla <sathya.perla@emulex.com>
Date: Fri, 1 Aug 2014 17:47:29 +0530

> Patch 1 fixes a regression caused by a previous commit on net-next.
> Old versions of BE3 FW may not support cmds to re-provision (and hence
> optimize) resources/queues in SR-IOV config. Do not treat this FW cmd
> failure as fatal and fail the function initialization. Instead, just
> enable SR-IOV with the resources provided by the FW.
> 
> Patch 2 ignores a VF mac address setting if the new mac is already active
> on the VF.
> 
> Patch 3 adds support to delete a FW-dump via ethtool on Lancer adapters.
> 
> Please consider applying this patch-set to the net-next tree. Thanks!

Series applied, thank you.

[PATCH net-next 0/3] be2net: patch set

[Sathya Perla]

Hi Dave, the following patch set includes three feature additions relating
to SR-IOV to be2net. 

Patch 1 avoid creating a non-RSS default RXQ when FW allows it.
This prevents wasting one RXQ for each VF.

Patch 2 adds support for evenly distributing all queue & filter resources
across VFs. The FW informs the driver as to which resources are distributable.

Patch 3 implements the sriov_configure PCI method to allow runtime
enablement of VFs via sysfs.

Pls consider applying this patch-set to the net-next tree. Thanks!

Vasundhara Volam (3):
  be2net: avoid creating the non-RSS default RXQ if FW allows to
  be2net: re-distribute SRIOV resources allowed by FW
  be2net: implement .sriov_configure() PCI callback

 drivers/net/ethernet/emulex/benet/be.h         |   9 +-
 drivers/net/ethernet/emulex/benet/be_cmds.c    | 114 +++++++----
 drivers/net/ethernet/emulex/benet/be_cmds.h    |  23 ++-
 drivers/net/ethernet/emulex/benet/be_ethtool.c |   2 +-
 drivers/net/ethernet/emulex/benet/be_main.c    | 260 +++++++++++++++++--------
 5 files changed, 285 insertions(+), 123 deletions(-)

-- 
2.2.0

Re: [PATCH net-next 0/3] be2net: patch set

[David Miller]

From: Sathya Perla <sathya.perla@emulex.com>
Date: Wed, 4 Mar 2015 00:44:31 -0500

> Hi Dave, the following patch set includes three feature additions relating
> to SR-IOV to be2net. 
> 
> Patch 1 avoid creating a non-RSS default RXQ when FW allows it.
> This prevents wasting one RXQ for each VF.
> 
> Patch 2 adds support for evenly distributing all queue & filter resources
> across VFs. The FW informs the driver as to which resources are distributable.
> 
> Patch 3 implements the sriov_configure PCI method to allow runtime
> enablement of VFs via sysfs.
> 
> Pls consider applying this patch-set to the net-next tree. Thanks!

Series applied, thanks.

[PATCH net-next 0/3] be2net: patch set

[Sathya Perla]

Hi David, this patch set includes 2 feature additions to the be2net driver:

Patch 1 sets up cpu affinity hints for be2net irqs using the
cpumask_set_cpu_local_first() API that first picks the near numa cores
and when they are exhausted, selects the far numa cores.

Patch 2 setups up xps queue mapping for be2net's TXQs to avoid,
by default, TX lock contention.

Patch 3 just bumps up the driver version.

Pls consider applying this patch set to the net-next queue. Thanks!

Padmanabh Ratnakar (1):
  be2net: assign CPU affinity hints to be2net IRQs

Sathya Perla (2):
  be2net: setup xps queue mapping
  be2net: bump up the driver version to 10.6.0.1

 drivers/net/ethernet/emulex/benet/be.h      |  4 +++-
 drivers/net/ethernet/emulex/benet/be_main.c | 27 +++++++++++++++++++++------
 2 files changed, 24 insertions(+), 7 deletions(-)

-- 
2.2.0

Re: [PATCH net-next 0/3] be2net: patch set

[David Miller]

From: Sathya Perla <sathya.perla@emulex.com>
Date: Thu, 26 Mar 2015 03:05:07 -0400

> Hi David, this patch set includes 2 feature additions to the be2net driver:
> 
> Patch 1 sets up cpu affinity hints for be2net irqs using the
> cpumask_set_cpu_local_first() API that first picks the near numa cores
> and when they are exhausted, selects the far numa cores.
> 
> Patch 2 setups up xps queue mapping for be2net's TXQs to avoid,
> by default, TX lock contention.
> 
> Patch 3 just bumps up the driver version.
> 
> Pls consider applying this patch set to the net-next queue. Thanks!

Series applied, thanks.