net/mlx5: Introduce blue flame register allocator

Here is an implementation of an allocator that allocates blue flame
registers. A blue flame register is used for generating send doorbells.
A blue flame register can be used to generate either a regular doorbell
or a blue flame doorbell where the data to be sent is written to the
device's I/O memory hence saving the need to read the data from memory.
For blue flame kind of doorbells to succeed, the blue flame register
need to be mapped as write combining. The user can specify what kind of
send doorbells she wishes to use. If she requested write combining
mapping but that failed, the allocator will fall back to non write
combining mapping and will indicate that to the user.
Subsequent patches in this series will make use of this allocator.

Signed-off-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

3 files changed, 44 insertions, 2 deletions

diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index aa851c51ab59..db1b9287012f 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -215,6 +215,8 @@ enum {
 	MLX5_BFREGS_PER_UAR		= 4,
 	MLX5_MAX_UARS			= 1 << 8,
 	MLX5_NON_FP_BFREGS_PER_UAR	= 2,
+	MLX5_FP_BFREGS_PER_UAR		= MLX5_BFREGS_PER_UAR -
+					  MLX5_NON_FP_BFREGS_PER_UAR,
 	MLX5_MAX_BFREGS			= MLX5_MAX_UARS *
 					  MLX5_NON_FP_BFREGS_PER_UAR,
 };
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 3d07e25b3bf1..969aa1fe17e2 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -452,6 +452,39 @@ struct mlx5_eq_table {
 	spinlock_t		lock;
 };
 
+struct mlx5_uars_page {
+	void __iomem	       *map;
+	bool			wc;
+	u32			index;
+	struct list_head	list;
+	unsigned int		bfregs;
+	unsigned long	       *reg_bitmap; /* for non fast path bf regs */
+	unsigned long	       *fp_bitmap;
+	unsigned int		reg_avail;
+	unsigned int		fp_avail;
+	struct kref		ref_count;
+	struct mlx5_core_dev   *mdev;
+};
+
+struct mlx5_bfreg_head {
+	/* protect blue flame registers allocations */
+	struct mutex		lock;
+	struct list_head	list;
+};
+
+struct mlx5_bfreg_data {
+	struct mlx5_bfreg_head	reg_head;
+	struct mlx5_bfreg_head	wc_head;
+};
+
+struct mlx5_sq_bfreg {
+	void __iomem	       *map;
+	struct mlx5_uars_page  *up;
+	bool			wc;
+	u32			index;
+	unsigned int		offset;
+};
+
 struct mlx5_uar {
 	u32			index;
 	struct list_head	bf_list;
@@ -645,6 +678,7 @@ struct mlx5_priv {
 	void		       *pfault_ctx;
 	struct srcu_struct      pfault_srcu;
 #endif
+	struct mlx5_bfreg_data		bfregs;
 };
 
 enum mlx5_device_state {
@@ -1022,6 +1056,9 @@ void mlx5_cleanup_rl_table(struct mlx5_core_dev *dev);
 int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u32 rate, u16 *index);
 void mlx5_rl_remove_rate(struct mlx5_core_dev *dev, u32 rate);
 bool mlx5_rl_is_in_range(struct mlx5_core_dev *dev, u32 rate);
+int mlx5_alloc_bfreg(struct mlx5_core_dev *mdev, struct mlx5_sq_bfreg *bfreg,
+		     bool map_wc, bool fast_path);
+void mlx5_free_bfreg(struct mlx5_core_dev *mdev, struct mlx5_sq_bfreg *bfreg);
 
 static inline int fw_initializing(struct mlx5_core_dev *dev)
 {
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 15f896781966..1223feff0ea4 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -905,7 +905,8 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         uc[0x1];
 	u8         rc[0x1];
 
-	u8         reserved_at_240[0xa];
+	u8         uar_4k[0x1];
+	u8         reserved_at_241[0x9];
 	u8         uar_sz[0x6];
 	u8         reserved_at_250[0x8];
 	u8         log_pg_sz[0x8];
@@ -997,7 +998,9 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         device_frequency_mhz[0x20];
 	u8         device_frequency_khz[0x20];
 
-	u8         reserved_at_500[0x80];
+	u8         reserved_at_500[0x20];
+	u8	   num_of_uars_per_page[0x20];
+	u8         reserved_at_540[0x40];
 
 	u8         reserved_at_580[0x3f];
 	u8         cqe_compression[0x1];