nvme_rdma: Allow the SGL to be larger than the req_length
The upper levels of the stack allow for this, so we should follow that pattern so I/O don't break here. Change-Id: Ia862f14975a551b0675bafd7709fb7897d0d567e Signed-off-by: Seth Howell <seth.howell@intel.com> Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/450685 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
This commit is contained in:
parent
ae694fced8
commit
3f2553a3f7
@ -1156,7 +1156,9 @@ nvme_rdma_build_sgl_inline_request(struct nvme_rdma_qpair *rqpair,
|
|||||||
return -1;
|
return -1;
|
||||||
}
|
}
|
||||||
|
|
||||||
assert(length <= remaining_payload);
|
if (length > remaining_payload) {
|
||||||
|
length = remaining_payload;
|
||||||
|
}
|
||||||
|
|
||||||
requested_size = length;
|
requested_size = length;
|
||||||
mr = (struct ibv_mr *)spdk_mem_map_translate(rqpair->mr_map->map, (uint64_t)virt_addr,
|
mr = (struct ibv_mr *)spdk_mem_map_translate(rqpair->mr_map->map, (uint64_t)virt_addr,
|
||||||
|
Loading…
Reference in New Issue
Block a user