From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: <t.lamprecht@proxmox.com> Received: from firstgate.proxmox.com (firstgate.proxmox.com [212.224.123.68]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by lists.proxmox.com (Postfix) with ESMTPS id 4C6B691B52 for <pve-devel@lists.proxmox.com>; Sat, 11 Mar 2023 10:02:01 +0100 (CET) Received: from firstgate.proxmox.com (localhost [127.0.0.1]) by firstgate.proxmox.com (Proxmox) with ESMTP id 2FFC5285E9 for <pve-devel@lists.proxmox.com>; Sat, 11 Mar 2023 10:01:31 +0100 (CET) Received: from proxmox-new.maurer-it.com (proxmox-new.maurer-it.com [94.136.29.106]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by firstgate.proxmox.com (Proxmox) with ESMTPS for <pve-devel@lists.proxmox.com>; Sat, 11 Mar 2023 10:01:30 +0100 (CET) Received: from proxmox-new.maurer-it.com (localhost.localdomain [127.0.0.1]) by proxmox-new.maurer-it.com (Proxmox) with ESMTP id F1E2441CB8; Sat, 11 Mar 2023 10:01:29 +0100 (CET) Message-ID: <ab594136-845f-bcf9-c1b6-8aaeea8ef486@proxmox.com> Date: Sat, 11 Mar 2023 10:01:28 +0100 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:111.0) Gecko/20100101 Thunderbird/111.0 Content-Language: en-GB, de-AT To: "DERUMIER, Alexandre" <Alexandre.DERUMIER@groupe-cyllene.com>, "pve-devel@lists.proxmox.com" <pve-devel@lists.proxmox.com> References: <20201210152338.19423-1-s.reiter@proxmox.com> <3dd51877-142c-55b2-e2ed-33f48ecb348f@proxmox.com> <5d774074596e5430faaa26146c70fdd13b513598.camel@groupe-cyllene.com> From: Thomas Lamprecht <t.lamprecht@proxmox.com> In-Reply-To: <5d774074596e5430faaa26146c70fdd13b513598.camel@groupe-cyllene.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-SPAM-LEVEL: Spam detection results: 0 AWL -0.050 Adjusted score from AWL reputation of From: address BAYES_00 -1.9 Bayes spam probability is 0 to 1% KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment NICE_REPLY_A -0.001 Looks like a legit reply (A) SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record SPF_PASS -0.001 SPF: sender matches SPF record Subject: Re: [pve-devel] applied: [RFC pve-qemu] disable jemalloc X-BeenThere: pve-devel@lists.proxmox.com X-Mailman-Version: 2.1.29 Precedence: list List-Id: Proxmox VE development discussion <pve-devel.lists.proxmox.com> List-Unsubscribe: <https://lists.proxmox.com/cgi-bin/mailman/options/pve-devel>, <mailto:pve-devel-request@lists.proxmox.com?subject=unsubscribe> List-Archive: <http://lists.proxmox.com/pipermail/pve-devel/> List-Post: <mailto:pve-devel@lists.proxmox.com> List-Help: <mailto:pve-devel-request@lists.proxmox.com?subject=help> List-Subscribe: <https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel>, <mailto:pve-devel-request@lists.proxmox.com?subject=subscribe> X-List-Received-Date: Sat, 11 Mar 2023 09:02:01 -0000 Hi, Am 10/03/2023 um 19:05 schrieb DERUMIER, Alexandre: > I'm currently benching again qemu with librbd and memory allocator. > > > It's seem that they are still performance problem with default glibc > allocator, around 20-25% less iops and bigger latency. Are those numbers compared to jemalloc or tcmalloc? Also, a key problem with allocator tuning is that its heavily dependent on the workload of each specific library (i.e., not only QEMU itself but also the specific block backend (library). > > From my bench, i'm around 60k iops vs 80-90k iops with 4k randread. > > Redhat have also notice it > https://bugzilla.redhat.com/show_bug.cgi?id=1717414 > https://sourceware.org/bugzilla/show_bug.cgi?id=28050 > > > I known than jemalloc was buggy with rust lib && pbs block driver, > but did you have evaluated tcmalloc ? Yes, for PBS once - was way worse in how it generally worked than either jemalloc and default glibc IIRC, but I don't think I checked for latency, as then we tracked down freed memory that the allocator did not give back to the system to how they internally try to keep a pool of available memory around. So for latency it might be a win, but IMO not to sure if the other effects it has are worth that. > > Note that it's possible to load it dynamically with LD_PRELOAD, > so maybe could we add an option in vm config to enable it ? > I'm not 100% sure if QEMU copes well with preloading it via the dynlinker as is, or if we need to hard-disable malloc_trim support for it then. As currently with the "system" allocator (glibc) there's malloc_trim called (semi-) periodically via call_rcu_thread - and at least qemu's meson build system config disables malloc_trim for tcmalloc or jemalloc. Or did you already test this directly on QEMU, not just rbd bench? As then I'd be open to add some tuning config with a allocator sub-property in there to our CFGs.