Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 6 Aug 2015 21:09:15 +0300
From: Solar Designer <>
Subject: Re: auditing our use of FMT_* flags (was: more robustness)

On Tue, Aug 04, 2015 at 08:57:29AM +0800, Kai Zhao wrote:
> 1. FMT_8_BIT
>     1.1 formats have not set FMT_8_BIT but there is at least one
>           password which does not ignore the 8th bit
>         bsdicrypt, has-160, pomelo, pufferfish, Stribog-256, wpapsk

I've just fixed bsdicrypt's code.  The rest should have the flag set.
Kai, you may commit that change.

>     1.2 formats have set FMT_8_BIT but all passwords ignore 8th bit
>         VNC, crypt

Looks like VNC's bit_flip[] table is such that it in fact ignores the
8th bit.  Kai, please re-format the table in vnc_fmt_plug.c to keep 8
values per line, so that it's symmetry is easily seen.  Once confirmed,
please drop the FMT_8_BIT flag.

For crypt, this is on purpose, and no change is needed.  It "can't"
reliably know if the underlying system's crypt() is 8-bit or not, and in
fact this will vary by actual hash type, of which multiple ones may be
loaded at once (with that one format).  The default test vectors are
such that they're 7-bit, hence the "spurious" problem detection by the
extended test.

There's at least one more:

Testing: skein-512, Skein 512 [Skein 32/64]... (32xOMP) FAILED (format:skein-512 have set FMT_8_BIT but all passwords ignore the 8th bit)

This is weird.  I think it's a bug in the code, to be found and fixed.
Kai, please try to find the bug.

BTW, the messages have a grammar error - maybe should use "has" in place
of "have", since they are referring to just one format.  So maybe instead
of "skein-512 have set FMT_8_BIT" report "skein-512 has FMT_8_BIT set,").


Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.