Date: Thu, 1 Feb 2018 19:00:05 +0000 (UTC) From: Alexander Motin <mav@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-stable@freebsd.org, svn-src-stable-11@freebsd.org Subject: svn commit: r328698 - stable/11/sys/dev/nvme Message-ID: <201802011900.w11J05Ed001820@repo.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: mav Date: Thu Feb 1 19:00:05 2018 New Revision: 328698 URL: https://svnweb.freebsd.org/changeset/base/328698 Log: MFC r323834 (by imp): Fix queue depth for nda. 1/4 of the number of queues times queue entries is too limiting. It works up to about 4k IOPS / 3.0GB/s for hardware that can do 4.4k/3.2GB/s with nvd. 3/4 works better, though it highlights issues in the fairness of nda's choice of TRIM vs READ. That will be fixed separately. Modified: stable/11/sys/dev/nvme/nvme_ctrlr.c Directory Properties: stable/11/ (props changed) Modified: stable/11/sys/dev/nvme/nvme_ctrlr.c ============================================================================== --- stable/11/sys/dev/nvme/nvme_ctrlr.c Thu Feb 1 18:59:03 2018 (r328697) +++ stable/11/sys/dev/nvme/nvme_ctrlr.c Thu Feb 1 19:00:05 2018 (r328698) @@ -151,7 +151,7 @@ nvme_ctrlr_construct_io_qpairs(struct nvme_controller * not a hard limit and will need to be revisitted when the upper layers * of the storage system grows multi-queue support. */ - ctrlr->max_hw_pend_io = num_trackers * ctrlr->num_io_queues / 4; + ctrlr->max_hw_pend_io = num_trackers * ctrlr->num_io_queues * 3 / 4; /* * This was calculated previously when setting up interrupts, but
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201802011900.w11J05Ed001820>