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]

Message-Id: <05549388-30E6-44BB-BE96-01D2122E1F9E@gmail.com>
Date: Mon, 12 Aug 2013 12:46:24 -0400
From: Mansour Moufid <mansourmoufid@...il.com>
To: discussions@...sword-hashing.net
Subject: Re: [PHC] C99 in reference implementations

On 2013-08-10, at 1:45 PM, Daniel Franke wrote:

> I have a reference implementation of a prospective PHC entry written in
> "portable" C99. By "portable", I mean that it uses only standard C99
> language features, has no external library dependencies, and should
> produce identical output regardless of host CPU architecture. However,
> it makes extensive use of C99 language features, including
> <stdint.h>/<stdbool.h>, mixed declarations and code, and variable-length
> arrays. Taking advantage of these features significantly improves
> readability, but will prevent the code from compiling on MSVC and any
> other compilers with poor C99 support. Does/should this pass muster for
> PHC submission requirements?


The only problematic compiler is MSVC, and its single biggest problem is
the header <stdint.h>. MSVC implements the LLP64 data model, the other
popular data model being LP64:[1]

            short   int     long    long long   size_t
     LP64   16      32      64      64          64
    LLP64   16      32      32      64          64

The common denominator is: 'int' is 32-bit and 'long long' is 64-bit.

So for MSVC one can minimally replace the <stdint.h> header with:

    typedef int i32;
    typedef long long int i64;
    typedef unsigned int u32;
    typedef unsigned long long int u64;

I propose the following solution: that submissions conform to C99, using
the types int32_t, int64_t, uint32_t and uint64_t defined in the header
<stdint.h>; and that a patch for MSVC be generated automatically using
Coccinelle[2] and the semantic patch listed at the end of this message.

The reason being that there are as many definitions of "Clean C" as
there are developers. C99 however is unambiguous.

For example:

    $ cat << EOF > foo.c
    > #include <stdint.h>
    > int main(void)
    > {
    >     uint32_t foo;
    >     int64_t bar;
    >     return 0;
    > }
    > EOF
    $ spatch --sp-file rm-stdint.cocci foo.c | tee foo-msvc.patch
    ...
    @@ -1,7 +1,10 @@
    -#include <stdint.h>
    +typedef int i32;
    +typedef long long int i64;
    +typedef unsigned int u32;
    +typedef unsigned long long int u64;
     int main(void)
     {
    -    uint32_t foo;
    -    int64_t bar;
    +    u32 foo;
    +    i64 bar;
         return 0;
     }

Of course other issues like <stdbool.h> are similarly easily solved.

[1]: <http://www.unix.org/version2/whatsnew/lp64_wp.html>
[2]: <http://coccinelle.lip6.fr/>

The semantic patch referred to above, rm-stdint.cocci:

@typedefs@
@@
-#include <stdint.h>
+typedef int i32;
+typedef long long int i64;
+typedef unsigned int u32;
+typedef unsigned long long int u64;

@replace_types depends on typedefs@
identifier x;
@@
(
- int32_t x;
+ i32 x;
|
- int64_t x;
+ i64 x;
|
- uint32_t x;
+ u32 x;
|
- uint64_t x;
+ u64 x;
)