[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <2a620e3a-f08c-6251-7dbb-46e843118bf4@grimberg.me>
Date: Tue, 20 Nov 2018 16:10:21 -0800
From: Sagi Grimberg <sagi@...mberg.me>
To: Narayan Ayalasomayajula
<narayan.ayalasomayajula@...an-networks.com>,
"linux-nvme@...ts.infradead.org" <linux-nvme@...ts.infradead.org>
Cc: "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
Keith Busch <keith.busch@...el.com>,
"David S. Miller" <davem@...emloft.net>,
Christoph Hellwig <hch@....de>
Subject: Re: [PATCH v2 14/14] nvme-tcp: add NVMe over TCP host driver
> Hi Sagi,
>
>> + icreq->pfv = cpu_to_le16(NVME_TCP_PFV_1_0);
>> + icreq->maxr2t = cpu_to_le16(1); /* single inflight r2t supported */
>> + icreq->hpda = 0; /* no alignment constraint */
>
> The NVMe-TCP spec indicates that MAXR2T is a 0's-based value. To support a single inflight R2T as indicated in the comment above, icreq->maxr2t should be set to 0, right?
Correct, will fix in the next spin.
Thanks
Powered by blists - more mailing lists