linux-kernel - Re: [PATCH v2 1/2] crypto: engine: permit to enqueue ashash

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: <20160602093835.GB2521@Red>
Date:	Thu, 2 Jun 2016 11:38:35 +0200
From:	LABBE Corentin <clabbe.montjoie@...il.com>
To:	Herbert Xu <herbert@...dor.apana.org.au>
Cc:	davem@...emloft.net, baolin.wang@...aro.org,
	linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/2] crypto: engine: permit to enqueue ashash_request

On Thu, Jun 02, 2016 at 05:19:40PM +0800, Herbert Xu wrote:
> On Thu, Jun 02, 2016 at 11:12:13AM +0200, LABBE Corentin wrote:
> > On Thu, Jun 02, 2016 at 04:32:59PM +0800, Herbert Xu wrote:
> > > On Mon, May 30, 2016 at 03:32:01PM +0200, LABBE Corentin wrote:
> > > > The current crypto engine allow only ablkcipher_request to be enqueued.
> > > > Thus denying any use of it for hardware that also handle hash algo.
> > > > 
> > > > This patch convert all ablkcipher_request references to the
> > > > more general crypto_async_request.
> > > > 
> > > > Signed-off-by: LABBE Corentin <clabbe.montjoie@...il.com>
> > > 
> > > First of all your patches break bisection which is unacceptable.
> > > 
> > 
> > How do I break bisection ?
> 
> Because the kernel won't compile after your first patch.
> 
> Either do it as one single patch or use the more elaborate "new
> interafce" + "switchover" + "delete old interface" ritual.
> 

Since my patch is small and easy (and only one client is modified), do you mind if I choose the first one ?

> > So, if my hwcrypto can handle hash and ciphers, I need to have two engine and each crypt_one_request()/hash_one_request()
> > need to lock the engine.
> > Having only one engine that handle all types permit to avoid this locking.
> 
> OK then we should add some type-checking as you suggested.  What
> I don't want is just blind casting by the user of crypto_engine.

I will add this type checking on my patch against omap-aes/des.

Regards