Re: Please help: do lua sources assume a 32-bit int?

[Ben Sunshine-Hill <sneftel@...>]

That code looks fine. It'll work with int sizes up to 32 bits, but it
doesn't assume them. That part of the conditional handles ints bigger
than the 16-bit maxint, and simply won't ever be executed on a
16-bit-int machine.

IIRC, there are people using Lua on 16-bit embedded systems with no
problems at all.

Ben


On Mon, 31 Jan 2005 09:52:27 +0200, Bogdan Marinescu
<bogdan.marinescu@gmail.com> wrote:
> (This was first posted on the forum; i'm reposting it here with the
> hope of a shorter response time)
> 
> Hello all,
> 
> I'm trying to compile lua 5.0.2 for a CPU that has sizeof(int)==2,
> instead of 4. While trying to compile the lua core, I got the
> following warning:
> 
> [Warning(ccom):lobject.c,line 62] this comparison is always false
> ===> if (x >= 0x00010000) {
> [Warning(ccom):lobject.c,line 63] this comparison is always false
> ===> if (x >= 0x01000000) return log_8[((x>>24) & 0xff) - 1]+24;
> 
> Looking in lobject.c, this is what I found:
> 
> Code:
> int luaO_log2 (unsigned int x) {
>  static const lu_byte log_8[255] = {
>    0,
>    1,1,
>    2,2,2,2,
>    3,3,3,3,3,3,3,3,
>    4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,
>    5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,
>    6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,
>    6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,
>    7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
>    7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
>    7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
>    7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7
>  };
>  if (x >= 0x00010000) {
>    if (x >= 0x01000000) return log_8[((x>>24) & 0xff) - 1]+24;
>    else return log_8[((x>>16) & 0xff) - 1]+16;
>  }
>  else {
>    if (x >= 0x00000100) return log_8[((x>>8) & 0xff) - 1]+8;
>    else if (x) return log_8[(x & 0xff) - 1];
>    return -1;  /* special `log' for 0 */
>  }
> }
> 
> Ar the first glance, this piece of code seems to indicate that the
> size of int is always 32 bits. Is this the case? If so, are there more
> places in the source where this happens? If they are, I'll have to
> define some custom types and modify the code (write everywhere 'u32'
> instead of 'unsigned int', for example, where 'u32' is an user defined
> type).
> Thank you very much for your help,
> 
> Bogdan M.
>