netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Saeed Mahameed <saeedm@mellanox.com>
To: "David S. Miller" <davem@davemloft.net>
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	Jiri Pirko <jiri@mellanox.com>,
	Feras Daoud <ferasda@mellanox.com>,
	Saeed Mahameed <saeedm@mellanox.com>
Subject: [net-next v2 07/15] net/mlx5: Control CR-space access by different PFs
Date: Thu, 13 Jun 2019 20:39:27 +0000	[thread overview]
Message-ID: <20190613203825.31049-8-saeedm@mellanox.com> (raw)
In-Reply-To: <20190613203825.31049-1-saeedm@mellanox.com>

From: Feras Daoud <ferasda@mellanox.com>

Since the FW can be shared between different PFs/VFs it is common
that more than one health poll will detected a failure, this can
lead to multiple resets which are unneeded.

The solution is to use a FW locking mechanism using semaphore space
to provide a way to allow only one device to collect the cr-dump and
to issue a sw-reset.

Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 .../ethernet/mellanox/mlx5/core/lib/pci_vsc.c | 40 ++++++++++++++++---
 .../ethernet/mellanox/mlx5/core/lib/pci_vsc.h |  8 ++++
 .../ethernet/mellanox/mlx5/core/mlx5_core.h   |  4 ++
 3 files changed, 47 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c
index a27b0119b3d6..6b774e0c2766 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c
@@ -24,11 +24,6 @@
 	pci_write_config_dword((dev)->pdev, (dev)->vsc_addr + (offset), (val))
 #define VSC_MAX_RETRIES 2048
 
-enum mlx5_vsc_state {
-	MLX5_VSC_UNLOCK,
-	MLX5_VSC_LOCK,
-};
-
 enum {
 	VSC_CTRL_OFFSET = 0x4,
 	VSC_COUNTER_OFFSET = 0x8,
@@ -284,3 +279,38 @@ int mlx5_vsc_gw_read_block_fast(struct mlx5_core_dev *dev, u32 *data,
 	}
 	return length;
 }
+
+int mlx5_vsc_sem_set_space(struct mlx5_core_dev *dev, u16 space,
+			   enum mlx5_vsc_state state)
+{
+	u32 data, id = 0;
+	int ret;
+
+	ret = mlx5_vsc_gw_set_space(dev, MLX5_SEMAPHORE_SPACE_DOMAIN, NULL);
+	if (ret) {
+		mlx5_core_warn(dev, "Failed to set gw space %d\n", ret);
+		return ret;
+	}
+
+	if (state == MLX5_VSC_LOCK) {
+		/* Get a unique ID based on the counter */
+		ret = vsc_read(dev, VSC_COUNTER_OFFSET, &id);
+		if (ret)
+			return ret;
+	}
+
+	/* Try to modify lock */
+	ret = mlx5_vsc_gw_write(dev, space, id);
+	if (ret)
+		return ret;
+
+	/* Verify lock was modified */
+	ret = mlx5_vsc_gw_read(dev, space, &data);
+	if (ret)
+		return -EINVAL;
+
+	if (data != id)
+		return -EBUSY;
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.h
index 28ea6bfa439f..64272a6d7754 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.h
@@ -4,6 +4,11 @@
 #ifndef __MLX5_PCI_VSC_H__
 #define __MLX5_PCI_VSC_H__
 
+enum mlx5_vsc_state {
+	MLX5_VSC_UNLOCK,
+	MLX5_VSC_LOCK,
+};
+
 enum {
 	MLX5_VSC_SPACE_SCAN_CRSPACE = 0x7,
 };
@@ -21,4 +26,7 @@ static inline bool mlx5_vsc_accessible(struct mlx5_core_dev *dev)
 	return !!dev->vsc_addr;
 }
 
+int mlx5_vsc_sem_set_space(struct mlx5_core_dev *dev, u16 space,
+			   enum mlx5_vsc_state state);
+
 #endif /* __MLX5_PCI_VSC_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index 97f8cf67ced0..8593c8183d87 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -111,6 +111,10 @@ enum {
 	MLX5_DRIVER_SYND = 0xbadd00de,
 };
 
+enum mlx5_semaphore_space_address {
+	MLX5_SEMAPHORE_SPACE_DOMAIN     = 0xA,
+};
+
 int mlx5_query_hca_caps(struct mlx5_core_dev *dev);
 int mlx5_query_board_id(struct mlx5_core_dev *dev);
 int mlx5_cmd_init_hca(struct mlx5_core_dev *dev, uint32_t *sw_owner_id);
-- 
2.21.0


  parent reply	other threads:[~2019-06-13 20:39 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-13 20:39 [pull request][net-next v2 00/15] Mellanox, mlx5 Firmware devlink health and sw reset Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 01/15] devlink: Hang reporter's dump method on a dumpit cb Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 02/15] Documentation: net: mlx5: Add mlx5 initial documentation Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 03/15] net/mlx5: Move all devlink related functions calls to devlink.c Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 04/15] net/mlx5: Add Vendor Specific Capability access gateway Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 05/15] net/mlx5: Add Crdump support Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 06/15] net/mlx5: Handle SW reset of FW in error flow Saeed Mahameed
2019-06-13 20:39 ` Saeed Mahameed [this message]
2019-06-13 20:39 ` [net-next v2 08/15] net/mlx5: Issue SW reset on FW assert Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 09/15] net/mlx5: Create FW devlink_health_reporter Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 10/15] net/mlx5: Add support for FW reporter dump Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 11/15] net/mlx5: Report devlink health on FW issues Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 12/15] net/mlx5: Add fw fatal devlink_health_reporter Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 13/15] net/mlx5: Add support for FW fatal reporter dump Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 14/15] net/mlx5: Report devlink health on FW fatal issues Saeed Mahameed
2019-06-13 20:39 ` [net-next v2 15/15] Documentation: net: mlx5: Devlink health documentation Saeed Mahameed
2019-06-15  2:46 ` [pull request][net-next v2 00/15] Mellanox, mlx5 Firmware devlink health and sw reset David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190613203825.31049-8-saeedm@mellanox.com \
    --to=saeedm@mellanox.com \
    --cc=davem@davemloft.net \
    --cc=ferasda@mellanox.com \
    --cc=jiri@mellanox.com \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).