Re: [PATCH] Revert "nvme: verify MNAN value if ANA is enabled"

From: Daniel Wagner <dwagner@suse.de>
To: Chaitanya Kulkarni <Chaitanya.Kulkarni@wdc.com>
Cc: "linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>
Subject: Re: [PATCH] Revert "nvme: verify MNAN value if ANA is enabled"
Date: Fri, 11 Jun 2021 11:08:19 +0200	[thread overview]
Message-ID: <20210611090819.d2ths7e4swfuj2s6@beryllium.lan> (raw)
In-Reply-To: <BYAPR04MB4965781915C1D4AFF2830E9586359@BYAPR04MB4965.namprd04.prod.outlook.com>

Hi Chaitanay,

On Thu, Jun 10, 2021 at 08:32:56PM +0000, Chaitanya Kulkarni wrote:
> Let us look into the host side issue first then we can move to target side.

Sure

> Consider a scenario where ANA enabled subsys with 0 namespaces on the target
> side. When host issues connect command to such a controller ctrl->mnan
> should be 0 and ctrl->nn should be 0 which should be valid.
> 
> With original check in the code :-
> 
> if (!ctrl->max_namespaces ||
>     ctrl->max_namespaces > le32_to_cpu(id->nn))

In the scenario you describe this is obvious wrong. I suspect we can't
decide if there are namespaces on the target side enabled or not at this
point. Maybe later in the discovery path we could check max_namespace
again. But I don't know if this is the trouble worth.

So yes I agree with your proposed change. In the end it's not important
anyway.

Thanks,
Daniel

_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme