Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   "Fabio M. De Francesco" <fmdefrancesco@gmail.com>
To:     Christoph Hellwig <hch@lst.de>, Sagi Grimberg <sagi@grimberg.me>
Cc:     Chaitanya Kulkarni <kch@nvidia.com>,
        linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org,
        Chaitanya Kulkarni <chaitanyak@nvidia.com>,
        Keith Busch <kbusch@kernel.org>,
        Ira Weiny <ira.weiny@intel.com>
Subject: Re: [PATCH v2] nvmet-tcp: Don't kmap() pages which can't come from HIGHMEM
Date:   Sun, 21 Aug 2022 21:12:30 +0200
Message-ID: <8107517.T7Z3S40VBb@opensuse>
In-Reply-To: <20220821054602.GA25754@lst.de>
References: <20220817184519.25141-1-fmdefrancesco@gmail.com> <20220821054602.GA25754@lst.de>
MIME-Version: 1.0
Content-Transfer-Encoding: 7Bit
Content-Type: text/plain; charset="us-ascii"
Precedence: bulk

On domenica 21 agosto 2022 07:46:02 CEST Christoph Hellwig wrote:
> This introduced tons of pointless over 80 character lines, please stick
> to a proper coding style.

I'm sorry for this mistake. I'll break those long lines in the next version.
 
> Also or in-kernel I/O wouldn't it make more sense to use a BVEC iter
> insted of a KVEC if we have the pages anyway?  Or does networking not
> support them properly?

I must admit I knew practically nothing about this code until I met it while 
doing my long journey towards kmap() and kmap_atomic() removals. I can say the 
same for regard to the differences between BVEC iter and KVEC iter. 

Since you are asking this, today I did some research and read code from other 
drivers/subsystems, despite this particular issue is out of the scope of my 
patch.

I also read an interesting article in LWN: "The iov_iter interface" at 
https://lwn.net/Articles/625077/

I may very well be wrong, however I think that iov_iter_kvec() is better 
suited here.

@Sagi, anything to add to this discussion?

Thanks,

Fabio