phc-discussions - Re: [PHC] Overwriting early hashing memory

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20140228155728.GA17032@openwall.com>
Date: Fri, 28 Feb 2014 19:57:28 +0400
From: Solar Designer <solar@...nwall.com>
To: discussions@...sword-hashing.net
Subject: Re: [PHC] Overwriting early hashing memory

On Fri, Feb 28, 2014 at 10:16:14AM -0500, Bill Cox wrote:
> This is a super-simple upgrade to my algorithm, and I think I'm going
> to do this.  I just don't know how much overwriting initial memory is
> worth.  Is it worth running 50% longer, or only 3% longer?

I think it depends on use case.  You shouldn't universally say that
leaking memory contents is the primary threat; for many (most?) use
cases and threat models, it is not.

When I added runtime self-test and zeroization to crypt_blowfish, I
measured its cost at 0.6% at the $2a$08 bcrypt cost setting.  I think
this is a good tradeoff.  3% might be (barely) acceptable too.  50%
definitely not, although it could be OK as a non-default option (but
then having that option might not be worth the complexity).

Alexander