From d98995b4bf981519dde4af0a081c393d62474039 Mon Sep 17 00:00:00 2001 From: Jianbo Liu Date: Mon, 3 Jun 2024 13:26:37 +0300 Subject: net/mlx5: Reimplement write combining test The test of write combining was added before in mlx5_ib driver. It opens UD QP and posts NOP WQEs, and uses BlueFlame doorbell. When BlueFlame is used, WQEs get written directly to a PCI BAR of the device (in addition to memory) so that the device handles them without having to access memory. In this test, the WQEs written in memory are different from the ones written to the BlueFlame which request CQE update. By checking the completion reports posted on CQ, we can know if BlueFlame succeeds or not. The write combining must be supported if BlueFlame succeeds as its register is written using write combining. This patch reimplements the test in the same way, but using a pair of SQ and CQ only. It is moved to mlx5_core as a general feature used by both mlx5_core and mlx5_ib. Besides, save write combine test result of the PCI function, so that its thousands of child functions such as SF can query without paying the time and resource penalty by itself. The test function is called only after failing to get the cached result. With this enhancement, all thousands of SFs of the PF attached to same driver no longer need to perform WC check explicitly, which is already done in the system. This saves several commands per SF, thereby speeds up SF creation and also saves completion EQ creation. Signed-off-by: Jianbo Liu Reviewed-by: Tariq Toukan Link: https://lore.kernel.org/r/4ff5a8cc4c5b5b0d98397baa45a5019bcdbf096e.1717409369.git.leon@kernel.org Signed-off-by: Leon Romanovsky --- include/linux/mlx5/driver.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/linux/mlx5') diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 779cfdf2e9d6..0d31f77396fc 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -766,6 +766,12 @@ struct mlx5_hca_cap { u32 max[MLX5_UN_SZ_DW(hca_cap_union)]; }; +enum mlx5_wc_state { + MLX5_WC_STATE_UNINITIALIZED, + MLX5_WC_STATE_UNSUPPORTED, + MLX5_WC_STATE_SUPPORTED, +}; + struct mlx5_core_dev { struct device *device; enum mlx5_coredev_type coredev_type; @@ -824,6 +830,9 @@ struct mlx5_core_dev { #endif u64 num_ipsec_offloads; struct mlx5_sd *sd; + enum mlx5_wc_state wc_state; + /* sync write combining state */ + struct mutex wc_state_lock; }; struct mlx5_db { @@ -1375,4 +1384,6 @@ static inline bool mlx5_is_macsec_roce_supported(struct mlx5_core_dev *mdev) enum { MLX5_OCTWORD = 16, }; + +bool mlx5_wc_support_get(struct mlx5_core_dev *mdev); #endif /* MLX5_DRIVER_H */ -- cgit v1.2.3 From b339e0a39dc37726712b9f0485d78fe4306d1667 Mon Sep 17 00:00:00 2001 From: Patrisious Haddad Date: Thu, 13 Jun 2024 21:00:04 +0300 Subject: RDMA/mlx5: Add Qcounters req_transport_retries_exceeded/req_rnr_retries_exceeded The req_transport_retries_exceeded counter shows the number of times requester detected transport retries exceed error. The req_rnr_retries_exceeded counter show the number of times the requester detected RNR NAKs retries exceed error. Signed-off-by: Patrisious Haddad Link: https://lore.kernel.org/r/250466af94f4989d638fab168e246035530e912f.1718301543.git.leon@kernel.org Reviewed-by: Simon Horman Signed-off-by: Leon Romanovsky --- include/linux/mlx5/mlx5_ifc.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux/mlx5') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 5df52e15f7d6..09d9d87d62c6 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -5629,7 +5629,11 @@ struct mlx5_ifc_query_q_counter_out_bits { u8 local_ack_timeout_err[0x20]; - u8 reserved_at_320[0xa0]; + u8 reserved_at_320[0x60]; + + u8 req_rnr_retries_exceeded[0x20]; + + u8 reserved_at_3a0[0x20]; u8 resp_local_length_error[0x20]; -- cgit v1.2.3