From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-9.9 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,NICE_REPLY_A,SIGNED_OFF_BY, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 54697C433E2 for ; Tue, 8 Sep 2020 08:44:45 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 2B72C215A4 for ; Tue, 8 Sep 2020 08:44:45 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729911AbgIHIoo (ORCPT ); Tue, 8 Sep 2020 04:44:44 -0400 Received: from mx2.suse.de ([195.135.220.15]:32868 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729390AbgIHIon (ORCPT ); Tue, 8 Sep 2020 04:44:43 -0400 X-Virus-Scanned: by amavisd-new at test-mx.suse.de Received: from relay2.suse.de (unknown [195.135.221.27]) by mx2.suse.de (Postfix) with ESMTP id 3054BAEE9; Tue, 8 Sep 2020 08:44:42 +0000 (UTC) Subject: Re: [PATCH V3 2/4] blk-mq: implement queue quiesce via percpu_ref for BLK_MQ_F_BLOCKING To: Ming Lei , Jens Axboe , linux-block@vger.kernel.org, linux-nvme@lists.infradead.org, Christoph Hellwig , Keith Busch Cc: Sagi Grimberg , Bart Van Assche , Johannes Thumshirn , Chao Leng References: <20200908081538.1434936-1-ming.lei@redhat.com> <20200908081538.1434936-3-ming.lei@redhat.com> From: Hannes Reinecke Openpgp: preference=signencrypt Autocrypt: addr=hare@suse.de; prefer-encrypt=mutual; keydata= mQINBE6KyREBEACwRN6XKClPtxPiABx5GW+Yr1snfhjzExxkTYaINHsWHlsLg13kiemsS6o7 qrc+XP8FmhcnCOts9e2jxZxtmpB652lxRB9jZE40mcSLvYLM7S6aH0WXKn8bOqpqOGJiY2bc 6qz6rJuqkOx3YNuUgiAxjuoYauEl8dg4bzex3KGkGRuxzRlC8APjHlwmsr+ETxOLBfUoRNuE b4nUtaseMPkNDwM4L9+n9cxpGbdwX0XwKFhlQMbG3rWA3YqQYWj1erKIPpgpfM64hwsdk9zZ QO1krgfULH4poPQFpl2+yVeEMXtsSou915jn/51rBelXeLq+cjuK5+B/JZUXPnNDoxOG3j3V VSZxkxLJ8RO1YamqZZbVP6jhDQ/bLcAI3EfjVbxhw9KWrh8MxTcmyJPn3QMMEp3wpVX9nSOQ tzG72Up/Py67VQe0x8fqmu7R4MmddSbyqgHrab/Nu+ak6g2RRn3QHXAQ7PQUq55BDtj85hd9 W2iBiROhkZ/R+Q14cJkWhzaThN1sZ1zsfBNW0Im8OVn/J8bQUaS0a/NhpXJWv6J1ttkX3S0c QUratRfX4D1viAwNgoS0Joq7xIQD+CfJTax7pPn9rT////hSqJYUoMXkEz5IcO+hptCH1HF3 qz77aA5njEBQrDRlslUBkCZ5P+QvZgJDy0C3xRGdg6ZVXEXJOQARAQABtCpIYW5uZXMgUmVp bmVja2UgKFN1U0UgTGFicykgPGhhcmVAc3VzZS5kZT6JAkEEEwECACsCGwMFCRLMAwAGCwkI BwMCBhUIAgkKCwQWAgMBAh4BAheABQJOisquAhkBAAoJEGz4yi9OyKjPOHoQAJLeLvr6JNHx GPcHXaJLHQiinz2QP0/wtsT8+hE26dLzxb7hgxLafj9XlAXOG3FhGd+ySlQ5wSbbjdxNjgsq FIjqQ88/Lk1NfnqG5aUTPmhEF+PzkPogEV7Pm5Q17ap22VK623MPaltEba+ly6/pGOODbKBH ak3gqa7Gro5YCQzNU0QVtMpWyeGF7xQK76DY/atvAtuVPBJHER+RPIF7iv5J3/GFIfdrM+wS BubFVDOibgM7UBnpa7aohZ9RgPkzJpzECsbmbttxYaiv8+EOwark4VjvOne8dRaj50qeyJH6 HLpBXZDJH5ZcYJPMgunghSqghgfuUsd5fHmjFr3hDb5EoqAfgiRMSDom7wLZ9TGtT6viDldv hfWaIOD5UhpNYxfNgH6Y102gtMmN4o2P6g3UbZK1diH13s9DA5vI2mO2krGz2c5BOBmcctE5 iS+JWiCizOqia5Op+B/tUNye/YIXSC4oMR++Fgt30OEafB8twxydMAE3HmY+foawCpGq06yM vAguLzvm7f6wAPesDAO9vxRNC5y7JeN4Kytl561ciTICmBR80Pdgs/Obj2DwM6dvHquQbQrU Op4XtD3eGUW4qgD99DrMXqCcSXX/uay9kOG+fQBfK39jkPKZEuEV2QdpE4Pry36SUGfohSNq xXW+bMc6P+irTT39VWFUJMcSuQINBE6KyREBEACvEJggkGC42huFAqJcOcLqnjK83t4TVwEn JRisbY/VdeZIHTGtcGLqsALDzk+bEAcZapguzfp7cySzvuR6Hyq7hKEjEHAZmI/3IDc9nbdh EgdCiFatah0XZ/p4vp7KAelYqbv8YF/ORLylAdLh9rzLR6yHFqVaR4WL4pl4kEWwFhNSHLxe 55G56/dxBuoj4RrFoX3ynerXfbp4dH2KArPc0NfoamqebuGNfEQmDbtnCGE5zKcR0zvmXsRp qU7+caufueZyLwjTU+y5p34U4PlOO2Q7/bdaPEdXfpgvSpWk1o3H36LvkPV/PGGDCLzaNn04 BdiiiPEHwoIjCXOAcR+4+eqM4TSwVpTn6SNgbHLjAhCwCDyggK+3qEGJph+WNtNU7uFfscSP k4jqlxc8P+hn9IqaMWaeX9nBEaiKffR7OKjMdtFFnBRSXiW/kOKuuRdeDjL5gWJjY+IpdafP KhjvUFtfSwGdrDUh3SvB5knSixE3qbxbhbNxmqDVzyzMwunFANujyyVizS31DnWC6tKzANkC k15CyeFC6sFFu+WpRxvC6fzQTLI5CRGAB6FAxz8Hu5rpNNZHsbYs9Vfr/BJuSUfRI/12eOCL IvxRPpmMOlcI4WDW3EDkzqNAXn5Onx/b0rFGFpM4GmSPriEJdBb4M4pSD6fN6Y/Jrng/Bdwk SQARAQABiQIlBBgBAgAPBQJOiskRAhsMBQkSzAMAAAoJEGz4yi9OyKjPgEwQAIP/gy/Xqc1q OpzfFScswk3CEoZWSqHxn/fZasa4IzkwhTUmukuIvRew+BzwvrTxhHcz9qQ8hX7iDPTZBcUt ovWPxz+3XfbGqE+q0JunlIsP4N+K/I10nyoGdoFpMFMfDnAiMUiUatHRf9Wsif/nT6oRiPNJ T0EbbeSyIYe+ZOMFfZBVGPqBCbe8YMI+JiZeez8L9JtegxQ6O3EMQ//1eoPJ5mv5lWXLFQfx f4rAcKseM8DE6xs1+1AIsSIG6H+EE3tVm+GdCkBaVAZo2VMVapx9k8RMSlW7vlGEQsHtI0FT c1XNOCGjaP4ITYUiOpfkh+N0nUZVRTxWnJqVPGZ2Nt7xCk7eoJWTSMWmodFlsKSgfblXVfdM 9qoNScM3u0b9iYYuw/ijZ7VtYXFuQdh0XMM/V6zFrLnnhNmg0pnK6hO1LUgZlrxHwLZk5X8F uD/0MCbPmsYUMHPuJd5dSLUFTlejVXIbKTSAMd0tDSP5Ms8Ds84z5eHreiy1ijatqRFWFJRp ZtWlhGRERnDH17PUXDglsOA08HCls0PHx8itYsjYCAyETlxlLApXWdVl9YVwbQpQ+i693t/Y PGu8jotn0++P19d3JwXW8t6TVvBIQ1dRZHx1IxGLMn+CkDJMOmHAUMWTAXX2rf5tUjas8/v2 azzYF4VRJsdl+d0MCaSy8mUh Message-ID: <54fce956-4e8f-6e4b-0ae1-e0905a06f3e4@suse.de> Date: Tue, 8 Sep 2020 10:44:40 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.7.2 MIME-Version: 1.0 In-Reply-To: <20200908081538.1434936-3-ming.lei@redhat.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Sender: linux-block-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org On 9/8/20 10:15 AM, Ming Lei wrote: > In case of BLK_MQ_F_BLOCKING, blk-mq uses SRCU to mark read critical > section during dispatching request, then request queue quiesce is based on > SRCU. What we want to get is low cost added in fast path. > > With percpu-ref, it is cleaner and simpler & enough for implementing queue > quiesce. The main requirement is to make sure all read sections to observe > QUEUE_FLAG_QUIESCED once blk_mq_quiesce_queue() returns. > > Also it becomes much easier to add interface of async queue quiesce. > > Meantime memory footprint can be reduced with per-request-queue percpu-ref. > > From implementation viewpoint, in fast path, not see percpu_ref is > slower than SRCU, and srcu tree(default option in most distributions) > could be slower since memory barrier is required in both lock & unlock, > and rcu_read_lock()/rcu_read_unlock() should be much cheap than > smp_mb(). > > 1) percpu_ref just hold the rcu_read_lock, then run a check & > increase/decrease on the percpu variable: > > rcu_read_lock() > if (__ref_is_percpu(ref, &percpu_count)) > this_cpu_inc(*percpu_count); > rcu_read_unlock() > > 2) srcu tree: > idx = READ_ONCE(ssp->srcu_idx) & 0x1; > this_cpu_inc(ssp->sda->srcu_lock_count[idx]); > smp_mb(); /* B */ /* Avoid leaking the critical section. */ > > Also from my test on null_blk(blocking), not observe percpu-ref performs > worse than srcu, see the following test: > > 1) test steps: > > rmmod null_blk > /dev/null 2>&1 > modprobe null_blk nr_devices=1 submit_queues=1 blocking=1 > fio --bs=4k --size=512G --rw=randread --norandommap --direct=1 --ioengine=libaio \ > --iodepth=64 --runtime=60 --group_reporting=1 --name=nullb0 \ > --filename=/dev/nullb0 --numjobs=32 > > test machine: HP DL380, 16 cpu cores, 2 threads per core, dual > sockets/numa, Intel(R) Xeon(R) Silver 4110 CPU @ 2.10GHz > > 2) test result: > - srcu quiesce: 6063K IOPS > - percpu-ref quiesce: 6113K IOPS > > Signed-off-by: Ming Lei > Cc: Sagi Grimberg > Cc: Bart Van Assche > Cc: Johannes Thumshirn > Cc: Chao Leng > --- > block/blk-mq-sysfs.c | 2 - > block/blk-mq.c | 130 +++++++++++++++++++++-------------------- > block/blk-sysfs.c | 6 +- > include/linux/blk-mq.h | 8 --- > include/linux/blkdev.h | 4 ++ > 5 files changed, 77 insertions(+), 73 deletions(-) > Reviewed-by: Hannes Reinecke Cheers, Hannes -- Dr. Hannes Reinecke Kernel Storage Architect hare@suse.de +49 911 74053 688 SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nürnberg HRB 36809 (AG Nürnberg), GF: Felix Imendörffer