Received-SPF: pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 5 May 2022 17:10:00 -0700
From:   Jakub Kicinski <kuba@kernel.org>
To:     Marc Kleine-Budde <mkl@pengutronix.de>
Cc:     Steffen Trumtrar <s.trumtrar@pengutronix.de>,
        linux-crypto@vger.kernel.org, io-uring@vger.kernel.org,
        kernel@pengutronix.de,
        Horia =?UTF-8?B?R2VhbnTEgw==?= <horia.geanta@nxp.com>,
        Pankaj Gupta <pankaj.gupta@nxp.com>,
        Gaurav Jain <gaurav.jain@nxp.com>,
        Jens Axboe <axboe@kernel.dk>,
        Pavel Begunkov <asml.silence@gmail.com>,
        Boris Pismenny <borisp@nvidia.com>,
        John Fastabend <john.fastabend@gmail.com>,
        Daniel Borkmann <daniel@iogearbox.net>, netdev@vger.kernel.org
Subject: Re: [BUG] Layerscape CAAM+kTLS+io_uring
Message-ID: <20220505171000.48a9155b@kernel.org>
In-Reply-To: <20220505192046.hczmzg7k6tz2rjv3@pengutronix.de>
References: <878rrqrgaj.fsf@pengutronix.de>
        <20220505192046.hczmzg7k6tz2rjv3@pengutronix.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Thu, 5 May 2022 21:20:46 +0200 Marc Kleine-Budde wrote:
> Hello,
> 
> no one seems to care about this problem. :/
> 
> Maybe too many components are involved, I'm the respective maintainers
> on Cc.
> 
> Cc += the CAAM maintainers
> Cc += the io_uring maintainers
> Cc += the kTLS maintainers
> 
> On 27.04.2022 10:20:40, Steffen Trumtrar wrote:
> > Hi all,
> > 
> > I have a Layerscape-1046a based board where I'm trying to use a
> > combination of liburing (v2.0) with splice, kTLS and CAAM (kernel
> > v5.17). The problem I see is that on shutdown the last bytes are
> > missing. It looks like io_uring is not waiting for all completions
> > from the CAAM driver.
> > 
> > With ARM-ASM instead of the CAAM, the setup works fine.  
> 
> What's the difference between the CAAM and ARM-ASM crypto? Without
> looking into the code I think the CAAM is asynchron while ARM-ASM is
> synchron. Is this worth investigating?

Sounds like
20ffc7adf53a ("net/tls: missing received data after fast remote close")