phc-discussions - RE: [PHC] C99 in reference implementations

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

Date: Tue, 13 Aug 2013 02:08:03 +0000
From: Marsh Ray <maray@...rosoft.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: RE: [PHC] C99 in reference implementations

*Proposed* guidance for developers/submitters:

Developers ought to be able to count on ...

#include <stdint.h>

... as making available ...

intN_t, uintN_t, for 8 <= N <= 32 (and N a power of 2 obviously)

... in the global namespace.

So you should be able to avoid having to work this out yourself or make assumptions about the size of int/long/longlong.
[Rationale: It's 2013. Pre-stdint.h compiler users can adapt.]

Calls to malloc()/free() ought to be the preferred method of acquiring and releasing memory for memory-hard functions.
[Rationale: They are the standard method and easy to swap out if needed]

Developers should use a function like secure_wipe(p, n) to wipe sensitive data from memory before free()ing. This can be defined like:
void secure_wipe(void * p, size_t cnt_bytes)
{
    // TODO use SecureZeroMemory or equivalent on platforms which provide it.
    memset(p, 0, cnt_bytes);
}
[Rationale: see http://blogs.msdn.com/b/oldnewthing/archive/2013/05/29/10421912.aspx ]

Will this work for everyone?

- Marsh