Commit 62638991 authored Aug 23, 2024 by Alexey Marchuk Committed by Ben Walker Dec 10, 2024

nvme/rdma: Don't limit max_sge if UMR is used



Since UMR creates a vurtually contig memory buffer,
we can always support up to 16 SGEs regardless of
MSDBD reported by target

Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: Ibd339f71ad35d355783993f777fcf8009ea68466
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24710


Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <ben@nvidia.com>
Reviewed-by: Jim Harris <jim.harris@nvidia.com>
Community-CI: Community CI Samsung <spdk.community.ci.samsung@gmail.com>
Community-CI: Mellanox Build Bot

parent cec5ba28

lib/nvme/nvme_rdma.c

+6 −2

Original line number	Diff line number	Diff line
		@@ -3077,8 +3077,12 @@ nvme_rdma_ctrlr_get_max_sges(struct spdk_nvme_ctrlr *ctrlr)

		/* Max SGE is limited by capsule size */
		max_sge = spdk_min(max_sge, max_in_capsule_sge);
		/* Max SGE may be limited by MSDBD */
		if (ctrlr->cdata.nvmf_specific.msdbd != 0) {
		/* Max SGE may be limited by MSDBD.
		* If umr_per_io is enabled and supported, we always use virtually contig buffer, we don't limit max_sge by
		* MSDBD in that case */
		if (!(g_spdk_nvme_transport_opts.rdma_umr_per_io &&
		spdk_rdma_provider_accel_sequence_supported()) &&
		ctrlr->cdata.nvmf_specific.msdbd != 0) {
		max_sge = spdk_min(max_sge, ctrlr->cdata.nvmf_specific.msdbd);
		}