From: Daniel Wagner <dwagner@suse.de> To: Sagi Grimberg <sagi@grimberg.me> Cc: linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, James Smart <james.smart@broadcom.com>, Keith Busch <kbusch@kernel.org>, Ming Lei <ming.lei@redhat.com>, Hannes Reinecke <hare@suse.de>, Wen Xiong <wenxiong@us.ibm.com> Subject: Re: [PATCH v4 2/8] nvme-tcp: Update number of hardware queues before using them Date: Mon, 9 Aug 2021 10:52:50 +0200 [thread overview] Message-ID: <20210809085250.xguvx5qiv2gxcoqk@carbon> (raw) In-Reply-To: <8373c07f-f5df-1ec6-9fda-d0262fc1b377@grimberg.me> Hi Sagi, On Fri, Aug 06, 2021 at 12:57:17PM -0700, Sagi Grimberg wrote: > > - ret = nvme_tcp_start_io_queues(ctrl); > > - if (ret) > > - goto out_cleanup_connect_q; > > - > > - if (!new) { > > - nvme_start_queues(ctrl); > > + } else if (prior_q_cnt != ctrl->queue_count) { > > So if the queue count did not change we don't wait to make sure > the queue g_usage_counter ref made it to zero? What guarantees that it > did? Hmm, good point. we should always call nvme_wait_freeze_timeout() for !new queues. Is this what you are implying? > > if (!nvme_wait_freeze_timeout(ctrl, NVME_IO_TIMEOUT)) { > > /* > > * If we timed out waiting for freeze we are likely to > > @@ -1828,6 +1822,10 @@ static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new) > > nvme_unfreeze(ctrl); > > } > > + ret = nvme_tcp_start_io_queues(ctrl); > > + if (ret) > > + goto out_cleanup_connect_q; > > + > > Did you test this with both heavy I/O, reset loop and ifdown/ifup > loop? Not sure if this classifies as heavy I/O (on 80 CPU machine) fio --rw=readwrite --name=test --filename=/dev/nvme16n1 --size=50M \ --direct=1 --bs=4k --numjobs=40 --group_reporting --runtime=4h \ --time_based and then I installed iptables rules to block the traffic on the controller side. With this test it is pretty easily to get the host hanging. Let me know what test you would like to see from me. I am glad to try to get them running. > If we unquiesce and unfreeze before we start the queues the pending I/Os > may resume before the connect and not allow the connect to make forward > progress. So the unfreeze should happen after the connect call? What about the newly created queues? Do they not suffer from the same problem? Isn't the NVME_TCP_Q_LIVE flag not enough? Daniel
WARNING: multiple messages have this Message-ID (diff)
From: Daniel Wagner <dwagner@suse.de> To: Sagi Grimberg <sagi@grimberg.me> Cc: linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, James Smart <james.smart@broadcom.com>, Keith Busch <kbusch@kernel.org>, Ming Lei <ming.lei@redhat.com>, Hannes Reinecke <hare@suse.de>, Wen Xiong <wenxiong@us.ibm.com> Subject: Re: [PATCH v4 2/8] nvme-tcp: Update number of hardware queues before using them Date: Mon, 9 Aug 2021 10:52:50 +0200 [thread overview] Message-ID: <20210809085250.xguvx5qiv2gxcoqk@carbon> (raw) In-Reply-To: <8373c07f-f5df-1ec6-9fda-d0262fc1b377@grimberg.me> Hi Sagi, On Fri, Aug 06, 2021 at 12:57:17PM -0700, Sagi Grimberg wrote: > > - ret = nvme_tcp_start_io_queues(ctrl); > > - if (ret) > > - goto out_cleanup_connect_q; > > - > > - if (!new) { > > - nvme_start_queues(ctrl); > > + } else if (prior_q_cnt != ctrl->queue_count) { > > So if the queue count did not change we don't wait to make sure > the queue g_usage_counter ref made it to zero? What guarantees that it > did? Hmm, good point. we should always call nvme_wait_freeze_timeout() for !new queues. Is this what you are implying? > > if (!nvme_wait_freeze_timeout(ctrl, NVME_IO_TIMEOUT)) { > > /* > > * If we timed out waiting for freeze we are likely to > > @@ -1828,6 +1822,10 @@ static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new) > > nvme_unfreeze(ctrl); > > } > > + ret = nvme_tcp_start_io_queues(ctrl); > > + if (ret) > > + goto out_cleanup_connect_q; > > + > > Did you test this with both heavy I/O, reset loop and ifdown/ifup > loop? Not sure if this classifies as heavy I/O (on 80 CPU machine) fio --rw=readwrite --name=test --filename=/dev/nvme16n1 --size=50M \ --direct=1 --bs=4k --numjobs=40 --group_reporting --runtime=4h \ --time_based and then I installed iptables rules to block the traffic on the controller side. With this test it is pretty easily to get the host hanging. Let me know what test you would like to see from me. I am glad to try to get them running. > If we unquiesce and unfreeze before we start the queues the pending I/Os > may resume before the connect and not allow the connect to make forward > progress. So the unfreeze should happen after the connect call? What about the newly created queues? Do they not suffer from the same problem? Isn't the NVME_TCP_Q_LIVE flag not enough? Daniel _______________________________________________ Linux-nvme mailing list Linux-nvme@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-nvme
next prev parent reply other threads:[~2021-08-09 8:52 UTC|newest] Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-08-02 11:26 [PATCH RESEND v4 0/8] Handle update hardware queues and queue freeze more carefully Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-02 11:26 ` [PATCH v4 1/8] nvme-fc: Update hardware queues before using them Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-02 19:34 ` Himanshu Madhani 2021-08-02 19:34 ` Himanshu Madhani 2021-08-02 11:26 ` [PATCH v4 2/8] nvme-tcp: Update number of " Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-06 19:57 ` Sagi Grimberg 2021-08-06 19:57 ` Sagi Grimberg 2021-08-09 8:52 ` Daniel Wagner [this message] 2021-08-09 8:52 ` Daniel Wagner 2021-08-11 1:00 ` Sagi Grimberg 2021-08-11 1:00 ` Sagi Grimberg 2021-08-11 1:07 ` Keith Busch 2021-08-11 1:07 ` Keith Busch 2021-08-11 5:57 ` Sagi Grimberg 2021-08-11 5:57 ` Sagi Grimberg 2021-08-11 10:25 ` Daniel Wagner 2021-08-11 10:25 ` Daniel Wagner 2021-08-02 11:26 ` [PATCH v4 3/8] nvme-rdma: " Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-02 11:26 ` [PATCH v4 4/8] nvme-fc: Wait with a timeout for queue to freeze Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-02 19:36 ` Himanshu Madhani 2021-08-02 19:36 ` Himanshu Madhani 2021-08-02 11:26 ` [PATCH v4 5/8] nvme-fc: avoid race between time out and tear down Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-02 19:38 ` Himanshu Madhani 2021-08-02 19:38 ` Himanshu Madhani 2021-08-02 11:26 ` [PATCH v4 6/8] nvme-fc: fix controller reset hang during traffic Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-02 19:39 ` Himanshu Madhani 2021-08-02 19:39 ` Himanshu Madhani 2021-08-04 7:23 ` Hannes Reinecke 2021-08-04 7:23 ` Hannes Reinecke 2021-08-04 8:08 ` Daniel Wagner 2021-08-04 8:08 ` Daniel Wagner 2021-08-11 1:05 ` Sagi Grimberg 2021-08-11 1:05 ` Sagi Grimberg 2021-08-11 10:30 ` Daniel Wagner 2021-08-11 10:30 ` Daniel Wagner 2021-08-12 20:03 ` James Smart 2021-08-12 20:03 ` James Smart 2021-08-18 11:43 ` Daniel Wagner 2021-08-18 11:43 ` Daniel Wagner 2021-08-18 11:49 ` Daniel Wagner 2021-08-18 11:49 ` Daniel Wagner 2021-08-02 11:26 ` [PATCH v4 7/8] nvme-tcp: Unfreeze queues on reconnect Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-02 11:26 ` [PATCH v4 8/8] nvme-rdma: " Daniel Wagner 2021-08-02 11:26 ` Daniel Wagner 2021-08-04 7:25 ` Hannes Reinecke 2021-08-04 7:25 ` Hannes Reinecke 2021-08-06 19:59 ` Sagi Grimberg 2021-08-06 19:59 ` Sagi Grimberg 2021-08-09 8:58 ` Daniel Wagner 2021-08-09 8:58 ` Daniel Wagner -- strict thread matches above, loose matches on Subject: below -- 2021-08-02 9:14 [PATCH v4 0/8] Handle update hardware queues and queue freeze more carefully Daniel Wagner 2021-08-02 9:14 ` [PATCH v4 2/8] nvme-tcp: Update number of hardware queues before using them Daniel Wagner
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20210809085250.xguvx5qiv2gxcoqk@carbon \ --to=dwagner@suse.de \ --cc=hare@suse.de \ --cc=james.smart@broadcom.com \ --cc=kbusch@kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-nvme@lists.infradead.org \ --cc=ming.lei@redhat.com \ --cc=sagi@grimberg.me \ --cc=wenxiong@us.ibm.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.