From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-io1-f66.google.com ([209.85.166.66]:33469 "EHLO mail-io1-f66.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1731721AbeIXVGt (ORCPT ); Mon, 24 Sep 2018 17:06:49 -0400 Received: by mail-io1-f66.google.com with SMTP id l25-v6so4743742ioj.0 for ; Mon, 24 Sep 2018 08:04:14 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <8222c235-ff07-b16b-3389-0b9780145abd@virtuozzo.com> References: <153666041612.19117.14667042009014596105.stgit@localhost.localdomain> <8222c235-ff07-b16b-3389-0b9780145abd@virtuozzo.com> From: Miklos Szeredi Date: Mon, 24 Sep 2018 17:04:13 +0200 Message-ID: Subject: Re: [PATCH 0/3] fuse: Solve request_find() bottleneck To: Kirill Tkhai Cc: kuznet@virtuozzo.com, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org Content-Type: text/plain; charset="UTF-8" Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Mon, Sep 24, 2018 at 2:08 PM, Kirill Tkhai wrote: > 2 weeks passed, so ping. > > Miklos, any reaction on this? > > Thanks, > Kirill > > On 11.09.2018 13:11, Kirill Tkhai wrote: >> Hi, >> >> We noticed the performance bottle neck in FUSE running our >> Virtuozzo storage over rdma. On some types of workload >> we observe 20% of time spent in request_find() in profiler. >> This function is iterating over long list of requests, and it >> scales bad. >> >> The patch introduces hash table to reduce the number >> of iterations, we do in this function. Also, algorithm >> of generating IDs for interrupt requests is changed, >> simplified request_find() function and killed >> fuse_req::intr_unique field. Concept looks good. Will review details... Thanks, Miklos