linux-kernel - Re: [Patch V5 1/7] crypto: Multi-buffer encryption infrastructure support

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] [day] [month] [year] [list]

Message-ID: <20170623083200.GA29882@gondor.apana.org.au>
Date:   Fri, 23 Jun 2017 16:32:00 +0800
From:   Herbert Xu <herbert@...dor.apana.org.au>
To:     Megha Dey <megha.dey@...el.com>
Cc:     tim.c.chen@...ux.intel.com, davem@...emloft.net,
        linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [Patch V5 1/7] crypto: Multi-buffer encryption infrastructure
 support

On Thu, Jun 08, 2017 at 12:52:54PM -0700, Megha Dey wrote:
>
> I will move this code to the mcryptd.c.
> 
> About the naming scheme, could you give me an example where the internal
> and external algorithm have the same name? I tried searching but did not
> find any.
> 
> When the outer and inner algorithm have the same name, I see a crash
> when testing using tcrypt. This is because the wrong algortihm (with a
> higher priority) is being picked up in __crypto_alg_lookup.  
> 
> Inner alg:
> Currently:
> alg name:__cbc(aes), driver name:__cbc-aes-aesni-mb
> 
> expected:
> alg name:cbc(aes), driver name: cbc-aes-aesni-mb
> 
> Outer alg:
> Currently:
> alg name:cbc(aes), driver name:cbc-aes-aesni-mb
> 
> expected:
> alg name:cbc(aes), driver name:mcryptd-cbc-aes-aesni-mb

This all looks right.  So I'm not sure why you're getting the crash.
We're relying on the INTERNAL flag to ensure the internal algorithm
is not picked up except when we strictly ask for it.

In fact I see something fishy in your testmgr code (the last patch
in the series I think).  It's setting the INTERNAL bit when
allocating tfms, that does not look right.

The only one that should be setting this is mcryptd.

Cheers,
-- 
Email: Herbert Xu <herbert@...dor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt