linux-kernel - Re: [PATCH] nvme: remove multipath module parameter

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ccc171d1-8de3-4632-9483-71fd5ea63d46@grimberg.me>
Date: Tue, 18 Feb 2025 10:19:50 +0200
From: Sagi Grimberg <sagi@...mberg.me>
To: John Meneghini <jmeneghi@...hat.com>, kbusch@...nel.org, hch@....de
Cc: bmarzins@...hat.com, Bryan Gurney <bgurney@...hat.com>,
 linux-nvme@...ts.infradead.org, linux-kernel@...r.kernel.org,
 Marco Patalano <mpatalan@...hat.com>, axboe@...nel.dk
Subject: Re: [PATCH] nvme: remove multipath module parameter




On 17/02/2025 18:14, John Meneghini wrote:
>
> On 2/17/25 3:08 AM, Sagi Grimberg wrote:
>>>
>>> We plan to ship this patch with RHEL-10. So it would be really good 
>>> if we could get this
>>> change accepted and merged upstream, perhaps into v6.15.
>>
>> Hey John,
>>
>> This looks fine to me, I'm assuming this was also tested with 
>> CONFIG_NVME_MULTIPATH=n ?
>
> Yes, everything has been tested with CONFIG_NVME_MULTIPATH both 
> enabled (Y) and disabled (N).
>
> As we discussed in the previous email thread[1] there is an anomaly 
> seen when you build a kernel with CONFIG_NVME_MULTIPATH=n.
> and the host discovers a multipath capable nvme device (CMIC/NMIC=1). 
> You will see exactly the same thing that you do with
> CONFIG_NVME_MULTIPATH=y when the nvme_core.multipath parameter is N. 
> You see a separate /dev/nvmeNN entry for
> every namespace/controller path.
>
> We can send send a separate patch to address that problem, but this 
> patch, which simply removes the nvme_core.multipath parameter
> has beeen fully tested and is ready to go.

I think that we want to print a warning in this case though. Can you 
resubmit with logging a warning in this case?