Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=91.218.175.84 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1794764593; b=4BDAlsKitq8r+jJvQZf4X0MRbV7nd5ZRTv+KKWAA7vx1pNjmTj/EggaawdUtadVfAHTs 9+zqkKsSJ+D8RJIYx3gkoYZxKszXEXK8qZ2vqWibT0vqaE724mBnti8Gv60DTSYXEl6TX x6H7p36Z2SKsOFYBDAPN4c+u6AV131iEKlxWuniJxEehgMxkGDqOr13GG2ApxyeoA536j TbK7wYEQOHHfaPPiQaRoRpZTyL4QXjpwxMuLJBHpqlGArlnhSuKeHgQdYuhoc5hpvHSba 4dETt1rGchRBF5q70wsvuy9N+XK1zPeuTYzy4S4g8quAWNORVXDsDtTzcehhvTNCnoQ== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1794764593; h=DMARC-Filter:X-Envelope-To:DKIM-Signature:X-Envelope-To:Received: X-Mizu-Trace-ID:X-Migadu-Flow:Message-ID:Date:MIME-Version:User-Agent: Subject:To:Cc:References:Content-Language:From:In-Reply-To: Content-Type:Content-Transfer-Encoding; bh=woOfyy6+u6Bpq89IupWYIeQk6EVkwaxCSpWmgn+r9TE=; b=ZxlhQgudrG3LE22iqyg/Us432KEGIZVtvacrOwLU/oNo80LH1b5tkGelCtwNttkTzB5B DTlZLjE3KatxPB6lxU3fcA3KBqhslb7XT4SNPwK09zcRuwTt4tTKXMQRz+8S7qlLWtlV8 LHYMhbacSa0D7tapyvkW618Go0qx5uOmLhyApivpjZacaWTJGylxSrxeQLIoNxrSHnydG HuJCdiFDMpMdrBWp07r0RBTTrEdy9z0FoJ1e4S5K2TfN0oWG6k80XK8ObLTtiGlWWU5Un dR27zbpTi6Kv5cMiCeiWU6O2li93+nmrmwySHFFNWO/jDJbVC1AOI9ZPWuxxMLPcpYA== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass header.d=linux.dev header.i=@linux.dev header.a=rsa-sha256 header.s=key1 header.b="LDXK/GA8"; arc=none smtp.remote-ip=91.218.175.84 Received: from mta0.migadu.com (out-84.mta0.migadu.com [91.218.175.84]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature ECDSA (prime256v1) server-digest SHA256) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 41421606A32F for ; Sun, 15 Nov 2026 17:43:10 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.2 smtp.subspace.kernel.org 41421606A32F Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key, unprotected) header.d=linux.dev header.i=@linux.dev header.a=rsa-sha256 header.s=key1 header.b=LDXK/GA8 X-Envelope-To: dm-devel@lists.linux.dev DKIM-Signature: a=rsa-sha256; bh=au0odLyb7/2WgYODnIjx2j64U1vQfxh6IkxY78a99H4=; c=simple/simple; d=linux.dev; h=from:to:subject:date:message-id:mime-version:content-type; s=key1; t=1794764589; v=1; x=1795369389; b=LDXK/GA8qatl4AxoC5doVanT0Gmp/wCId8BkMLaopRin/HeDRiUhXMGlOQ2c5ZNXJo4MHuoT O0DyofxGhOCisLoh11BNVCTIAat9kdWV6hD8NMViUR7CHrisFonDyEyQs1iK+cWudD+BpnkEIQe QrND9Fs4aghv+0lBg9WbwCU4= X-Envelope-To: dm-devel@lists.linux.dev Received: by mta10.migadu.com with ESMTPS id d3995290662ba302; Sun, 15 Nov 2026 17:43:08 +0000 X-Mizu-Trace-ID: d3995290662ba302 X-Migadu-Flow: FLOW_OUT Message-ID: <87d9b34e-1d63-41d2-9c01-595f0f1c7f23@linux.dev> Date: Sun, 15 Nov 2026 17:43:02 +0000 X-Mailing-List: dm-devel@lists.linux.dev List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v9 00/29] Native SCSI Multipath support To: Xose Vazquez Perez , Laurence Oberman Cc: hare@suse.com, linux-nvme@lists.infradead.org, linux-scsi@vger.kernel.org, snitzer@kernel.org, jmeneghi@redhat.com, dm-devel@lists.linux.dev, hch@lst.de, michael.christie@oracle.com, "James E.J. Bottomley" , mkp@kernel.org, Benjamin Marzinski , Martin Wilck References: <20261108204733.557995-1-john.garry@linux.dev> <9b90edc29d5b3be49f5bd0e0ecec5f1701e089bf.camel@redhat.com> <3332329d-b922-45a4-8333-e8353e983885@linux.dev> <00e03246-95dc-46dd-a2f9-89ad88f177d1@gmail.com> Content-Language: en-US From: John Garry In-Reply-To: <00e03246-95dc-46dd-a2f9-89ad88f177d1@gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit On 11/14/26 13:03, Xose Vazquez Perez wrote: > On 11/10/26 10:05 AM, John Garry wrote: > >> On 11/9/26 20:19, Laurence Oberman wrote: >>> 3. Smaller points >>> >>> - numa with equal NUMA distance uses one path per LUN, so with two AO >>>    paths through two HBAs, half the optimized bandwidth is idle (sdl:2 >>>    above under numa). That matches NVMe, but users coming from >>>    dm-multipath service-time will notice. Would round-robin or >>>    queue-depth be a better default for SCSI? >> >> Maybe, but round-robin may not be always ideal considering only a bio >> mode of operation is supported and no merging like dm-mpath RQ mode. > > iopolicy can easily be customized per array/device via udev rules. > Just like it's done for NVMe: > https://github.com/linux-nvme/nvme-cli/blob/master/nvmf-autoconnect/ > udev-rules/71-nvmf-hpe.rules.in Sure > > Additionally, "service-time" and "latency" I/O policies are currently under > development: > https://lore.kernel.org/linux-nvme/20261026111019.3724465-1- > kanie@linux.alibaba.com/T/ > https://lore.kernel.org/linux-nvme/20260815173502.1185929-1- > nilay@linux.ibm.com/T/ I already did an experimental port of the latency iopolicy to libmultipath. Porting iopolicy schemes to libmultipath is not too challenging, really. > >>> - There is no userspace tool yet to show heads, paths and ALUA state >>>    (the equivalent of multipath -ll or nvme list-subsys). I used a >>>    small sysfs script for this testing. Ben, is native SCSI multipath >>>    support in multipath-tools planned? >> >> Well, the whole point is that we don't need userspace tools. However, >> I appreciate that there is no general SCSI tooling which could help >> here. I know that Himanshu M did work on a SCSI version of nvme-cli >> tool previously. Martin may know the status of that. > > Agreed that userspace tools are not strictly needed for configuration, but > observability is key. > A standard utility to visualize the device topology, path health and ALUA > states at a glance, similar to what multipath-tools provides, would save a > lot of friction for sysadmins migrating away from dm-multipath. > Relying solely on ad-hoc scripts or manual sysfs parsing can become quite > tedious in production environments. Right, and any info required here should be available in sysfs. Thanks, John