RDMA/rxe: Fix data copy for IB_SEND_INLINE

For RDMA Send and Write with IB_SEND_INLINE, the memory buffers
specified in sge list will be placed inline in the Send Request.

The data should be copied by CPU from the virtual addresses of
corresponding sge list DMA addresses.

Cc: stable@kernel.org
Fixes: 8d7c7c0eeb ("RDMA: Add ib_virt_dma_to_page()")
Signed-off-by: Honggang LI <honggangli@163.com>
Link: https://lore.kernel.org/r/20240516095052.542767-1-honggangli@163.com
Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
This commit is contained in:
Honggang LI
2024-05-16 17:50:52 +08:00
committed by Leon Romanovsky
parent 056620da89
commit 03fa18a992
+1 -1
View File
@@ -812,7 +812,7 @@ static void copy_inline_data_to_wqe(struct rxe_send_wqe *wqe,
int i;
for (i = 0; i < ibwr->num_sge; i++, sge++) {
memcpy(p, ib_virt_dma_to_page(sge->addr), sge->length);
memcpy(p, ib_virt_dma_to_ptr(sge->addr), sge->length);
p += sge->length;
}
}