All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Shay Drory <shayd@nvidia.com>,
	Michael Guralnik <michaelgur@nvidia.com>,
	Saeed Mahameed <saeedm@nvidia.com>,
	Maher Sanalla <msanalla@nvidia.com>,
	Leon Romanovsky <leon@kernel.org>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.19 04/38] RDMA/mlx5: Rely on RoCE fw cap instead of devlink when setting profile
Date: Fri, 16 Sep 2022 12:08:38 +0200	[thread overview]
Message-ID: <20220916100448.622064836@linuxfoundation.org> (raw)
In-Reply-To: <20220916100448.431016349@linuxfoundation.org>

From: Maher Sanalla <msanalla@nvidia.com>

[ Upstream commit 9ca05b0f27de928be121cccf07735819dc9e1ed3 ]

When the RDMA auxiliary driver probes, it sets its profile based on
devlink driverinit value. The latter might not be in sync with FW yet
(In case devlink reload is not performed), thus causing a mismatch
between RDMA driver and FW. This results in the following FW syndrome
when the RDMA driver tries to adjust RoCE state, which fails the probe:

"0xC1F678 | modify_nic_vport_context: roce_en set on a vport that
doesn't support roce"

To prevent this, select the PF profile based on FW RoCE capability
instead of relying on devlink driverinit value.
To provide backward compatibility of the RoCE disable feature, on older
FW's where roce_rw is not set (FW RoCE capability is read-only), keep
the current behavior e.g., rely on devlink driverinit value.

Fixes: fbfa97b4d79f ("net/mlx5: Disable roce at HCA level")
Reviewed-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Link: https://lore.kernel.org/r/cb34ce9a1df4a24c135cb804db87f7d2418bd6cc.1661763459.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/hw/mlx5/main.c             |  2 +-
 .../net/ethernet/mellanox/mlx5/core/main.c    | 23 +++++++++++++++++--
 include/linux/mlx5/driver.h                   | 19 +++++++--------
 3 files changed, 32 insertions(+), 12 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 63c89a72cc352..bb13164124fdb 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -4336,7 +4336,7 @@ static int mlx5r_probe(struct auxiliary_device *adev,
 	dev->mdev = mdev;
 	dev->num_ports = num_ports;
 
-	if (ll == IB_LINK_LAYER_ETHERNET && !mlx5_is_roce_init_enabled(mdev))
+	if (ll == IB_LINK_LAYER_ETHERNET && !mlx5_get_roce_state(mdev))
 		profile = &raw_eth_profile;
 	else
 		profile = &pf_profile;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 64d54bba91f69..6c8bb74bd8fc6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -501,6 +501,24 @@ static int max_uc_list_get_devlink_param(struct mlx5_core_dev *dev)
 	return err;
 }
 
+bool mlx5_is_roce_on(struct mlx5_core_dev *dev)
+{
+	struct devlink *devlink = priv_to_devlink(dev);
+	union devlink_param_value val;
+	int err;
+
+	err = devlink_param_driverinit_value_get(devlink,
+						 DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
+						 &val);
+
+	if (!err)
+		return val.vbool;
+
+	mlx5_core_dbg(dev, "Failed to get param. err = %d\n", err);
+	return MLX5_CAP_GEN(dev, roce);
+}
+EXPORT_SYMBOL(mlx5_is_roce_on);
+
 static int handle_hca_cap_2(struct mlx5_core_dev *dev, void *set_ctx)
 {
 	void *set_hca_cap;
@@ -604,7 +622,8 @@ static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx)
 			 MLX5_CAP_GEN_MAX(dev, num_total_dynamic_vf_msix));
 
 	if (MLX5_CAP_GEN(dev, roce_rw_supported))
-		MLX5_SET(cmd_hca_cap, set_hca_cap, roce, mlx5_is_roce_init_enabled(dev));
+		MLX5_SET(cmd_hca_cap, set_hca_cap, roce,
+			 mlx5_is_roce_on(dev));
 
 	max_uc_list = max_uc_list_get_devlink_param(dev);
 	if (max_uc_list > 0)
@@ -630,7 +649,7 @@ static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx)
  */
 static bool is_roce_fw_disabled(struct mlx5_core_dev *dev)
 {
-	return (MLX5_CAP_GEN(dev, roce_rw_supported) && !mlx5_is_roce_init_enabled(dev)) ||
+	return (MLX5_CAP_GEN(dev, roce_rw_supported) && !mlx5_is_roce_on(dev)) ||
 		(!MLX5_CAP_GEN(dev, roce_rw_supported) && !MLX5_CAP_GEN(dev, roce));
 }
 
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 0015a08ddbd24..b3ea245faa515 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -1275,16 +1275,17 @@ enum {
 	MLX5_TRIGGERED_CMD_COMP = (u64)1 << 32,
 };
 
-static inline bool mlx5_is_roce_init_enabled(struct mlx5_core_dev *dev)
+bool mlx5_is_roce_on(struct mlx5_core_dev *dev);
+
+static inline bool mlx5_get_roce_state(struct mlx5_core_dev *dev)
 {
-	struct devlink *devlink = priv_to_devlink(dev);
-	union devlink_param_value val;
-	int err;
-
-	err = devlink_param_driverinit_value_get(devlink,
-						 DEVLINK_PARAM_GENERIC_ID_ENABLE_ROCE,
-						 &val);
-	return err ? MLX5_CAP_GEN(dev, roce) : val.vbool;
+	if (MLX5_CAP_GEN(dev, roce_rw_supported))
+		return MLX5_CAP_GEN(dev, roce);
+
+	/* If RoCE cap is read-only in FW, get RoCE state from devlink
+	 * in order to support RoCE enable/disable feature
+	 */
+	return mlx5_is_roce_on(dev);
 }
 
 #endif /* MLX5_DRIVER_H */
-- 
2.35.1




  parent reply	other threads:[~2022-09-16 10:25 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-16 10:08 [PATCH 5.19 00/38] 5.19.10-rc1 review Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 01/38] iommu/vt-d: Fix kdump kernels boot failure with scalable mode Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 02/38] net/mlx5: Introduce ifc bits for using software vhca id Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 03/38] net/mlx5: Use software VHCA id when its supported Greg Kroah-Hartman
2022-09-16 10:08 ` Greg Kroah-Hartman [this message]
2022-09-16 10:08 ` [PATCH 5.19 05/38] RDMA/mlx5: Add a umr recovery flow Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 06/38] RDMA/mlx5: Fix UMR cleanup on error flow of driver init Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 07/38] ACPI: resource: skip IRQ override on AMD Zen platforms Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 08/38] Input: goodix - add support for GT1158 Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 09/38] platform/surface: aggregator_registry: Add support for Surface Laptop Go 2 Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 10/38] drm/msm/rd: Fix FIFO-full deadlock Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 11/38] peci: cpu: Fix use-after-free in adev_release() Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 12/38] hwmon: (pmbus) Use dev_err_probe() to filter -EPROBE_DEFER error messages Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 13/38] kvm: x86: mmu: Always flush TLBs when enabling dirty logging Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 14/38] dt-bindings: iio: gyroscope: bosch,bmg160: correct number of pins Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 15/38] HID: ishtp-hid-clientHID: ishtp-hid-client: Fix comment typo Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 16/38] hid: intel-ish-hid: ishtp: Fix ishtp client sending disordered message Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 17/38] Bluetooth: MGMT: Fix Get Device Flags Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 18/38] tg3: Disable tg3 device on system reboot to avoid triggering AER Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 19/38] r8152: add PID for the Lenovo OneLink+ Dock Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 20/38] gpio: mockup: remove gpio debugfs when remove device Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 21/38] ieee802154: cc2520: add rc code in cc2520_tx() Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 22/38] Input: iforce - add support for Boeder Force Feedback Wheel Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 23/38] drm/amdgpu: disable FRU access on special SIENNA CICHLID card Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 24/38] drm/amd/pm: use vbios carried pptable for all SMU13.0.7 SKUs Greg Kroah-Hartman
2022-09-16 10:08 ` [PATCH 5.19 25/38] nvme-pci: add NVME_QUIRK_BOGUS_NID for Lexar NM610 Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 26/38] nvmet-tcp: fix unhandled tcp states in nvmet_tcp_state_change() Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 27/38] drm/amd/amdgpu: skip ucode loading if ucode_size == 0 Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 28/38] net: dsa: hellcreek: Print warning only once Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 29/38] perf/arm_pmu_platform: fix tests for platform_get_irq() failure Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 30/38] platform/x86: acer-wmi: Acer Aspire One AOD270/Packard Bell Dot keymap fixes Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 31/38] usb: storage: Add ASUS <0x0b05:0x1932> to IGNORE_UAS Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 32/38] platform/x86: asus-wmi: Increase FAN_CURVE_BUF_LEN to 32 Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 33/38] LoongArch: Fix section mismatch due to acpi_os_ioremap() Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 34/38] LoongArch: Fix arch_remove_memory() undefined build error Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 35/38] gpio: 104-dio-48e: Make irq_chip immutable Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 36/38] gpio: 104-idio-16: " Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 37/38] RDMA/irdma: Use s/g array in post send only when its valid Greg Kroah-Hartman
2022-09-16 10:09 ` [PATCH 5.19 38/38] Input: goodix - add compatible string for GT1158 Greg Kroah-Hartman
2022-09-16 21:49 ` [PATCH 5.19 00/38] 5.19.10-rc1 review Guenter Roeck
2022-09-16 22:35 ` Ron Economos
2022-09-17  7:26 ` Fenil Jain
2022-09-17  8:16 ` Bagas Sanjaya
2022-09-17 14:29 ` Sudip Mukherjee (Codethink)
2022-09-17 15:03 ` Naresh Kamboju
2022-09-19  0:12 ` Justin Forbes
2022-09-19  1:06 ` Florian Fainelli

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=20220916100448.622064836@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=leon@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=michaelgur@nvidia.com \
    --cc=msanalla@nvidia.com \
    --cc=saeedm@nvidia.com \
    --cc=sashal@kernel.org \
    --cc=shayd@nvidia.com \
    --cc=stable@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.