[PATCH] common/mlx5: fix MR range calculation
Dmitry Kozlyuk
dkozlyuk at nvidia.com
Thu Mar 31 16:33:16 CEST 2022
MR end for a mempool chunk may be calculated incorrectly.
For example, for chunk with addr=1.5M and len=1M with 2M page size
the range would be [0, 2M), while the proper result is [0, 4M).
Fix the calculation.
Fixes: 690b2a88c2f7 ("common/mlx5: add mempool registration facilities")
Cc: stable at dpdk.org
Signed-off-by: Dmitry Kozlyuk <dkozlyuk at nvidia.com>
Acked-by: Matan Azrad <matan at nvidia.com>
---
drivers/common/mlx5/mlx5_common_mr.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/common/mlx5/mlx5_common_mr.c b/drivers/common/mlx5/mlx5_common_mr.c
index fa27bd98de..06e4c8f187 100644
--- a/drivers/common/mlx5/mlx5_common_mr.c
+++ b/drivers/common/mlx5/mlx5_common_mr.c
@@ -1289,11 +1289,12 @@ mlx5_range_from_mempool_chunk(struct rte_mempool *mp, void *opaque,
unsigned int idx)
{
struct mlx5_range *ranges = opaque, *range = &ranges[idx];
+ uintptr_t start = (uintptr_t)memhdr->addr;
uint64_t page_size = rte_mem_page_size();
RTE_SET_USED(mp);
- range->start = RTE_ALIGN_FLOOR((uintptr_t)memhdr->addr, page_size);
- range->end = RTE_ALIGN_CEIL(range->start + memhdr->len, page_size);
+ range->start = RTE_ALIGN_FLOOR(start, page_size);
+ range->end = RTE_ALIGN_CEIL(start + memhdr->len, page_size);
}
/**
--
2.25.1
More information about the stable
mailing list