diff options
| author | Jack Morgenstein <jackm@dev.mellanox.co.il> | 2007-08-01 12:28:53 +0300 |
|---|---|---|
| committer | Roland Dreier <rolandd@cisco.com> | 2007-10-09 19:59:16 -0700 |
| commit | d7bb58fb1c0e7264a7261c7d0304121ef9402e94 (patch) | |
| tree | aefbae7ee4b4b84022a7646372ddfcdb0a99e72f /drivers/net/mlx4/icm.c | |
| parent | 121964ec38d3e17b5ea6183b3c0571df2f4b0eb6 (diff) | |
mlx4_core: Write MTTs from CPU instead with of WRITE_MTT FW command
Write MTT entries directly to ICM from the driver (eliminating use of
WRITE_MTT command). This reduces the number of FW commands needed to
register an MR by at least a factor of 2 and speeds up memory
registration significantly. This code will also be used to implement
FMRs.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Diffstat (limited to 'drivers/net/mlx4/icm.c')
| -rw-r--r-- | drivers/net/mlx4/icm.c | 21 |
1 files changed, 16 insertions, 5 deletions
diff --git a/drivers/net/mlx4/icm.c b/drivers/net/mlx4/icm.c index 250e24887578..4b3c109d5eae 100644 --- a/drivers/net/mlx4/icm.c +++ b/drivers/net/mlx4/icm.c @@ -301,9 +301,9 @@ void mlx4_table_put(struct mlx4_dev *dev, struct mlx4_icm_table *table, int obj) mutex_unlock(&table->mutex); } -void *mlx4_table_find(struct mlx4_icm_table *table, int obj) +void *mlx4_table_find(struct mlx4_icm_table *table, int obj, dma_addr_t *dma_handle) { - int idx, offset, i; + int idx, offset, dma_offset, i; struct mlx4_icm_chunk *chunk; struct mlx4_icm *icm; struct page *page = NULL; @@ -313,15 +313,26 @@ void *mlx4_table_find(struct mlx4_icm_table *table, int obj) mutex_lock(&table->mutex); - idx = obj & (table->num_obj - 1); - icm = table->icm[idx / (MLX4_TABLE_CHUNK_SIZE / table->obj_size)]; - offset = idx % (MLX4_TABLE_CHUNK_SIZE / table->obj_size); + idx = (obj & (table->num_obj - 1)) * table->obj_size; + icm = table->icm[idx / MLX4_TABLE_CHUNK_SIZE]; + dma_offset = offset = idx % MLX4_TABLE_CHUNK_SIZE; if (!icm) goto out; list_for_each_entry(chunk, &icm->chunk_list, list) { for (i = 0; i < chunk->npages; ++i) { + if (dma_handle && dma_offset >= 0) { + if (sg_dma_len(&chunk->mem[i]) > dma_offset) + *dma_handle = sg_dma_address(&chunk->mem[i]) + + dma_offset; + dma_offset -= sg_dma_len(&chunk->mem[i]); + } + /* + * DMA mapping can merge pages but not split them, + * so if we found the page, dma_handle has already + * been assigned to. + */ if (chunk->mem[i].length > offset) { page = chunk->mem[i].page; goto out; |