[En-Nut-Discussion] Next release 4.10?

[Thiago A. Corrêa]

Hi Harald,

     Thanks for looking into it. I'm still surprised that it didn't
affect other 32 bit platforms nor did it trigger an unaligned access
exception. So perhaps this is really a compiler bug.
     I guess it should be safe to fix it as:

s = (cp[2] << 8) & 0xFF00;
s |= cp[3];

Kind Regards,
    Thiago A. Correa

On Sat, May 7, 2011 at 2:00 PM, Harald Kipp <harald.kipp at egnite.de> wrote:
> Hi Thiago,
>
> On 4/27/2011 6:15 PM, Thiago A. Corrêa wrote:
>>     No, I mean this one:
>> http://lists.egnite.de/pipermail/en-nut-discussion/2011-January/012457.html
>
> Michael and I located the problem. It was introduced by Henrik's patch r3170. Note, that this fixes a severe bug, but the way it was done may break 32-bit systems. See Henrik's excellent description of the initial problem and my comment on bug #3006749:
>
> http://sourceforge.net/tracker/index.php?func=detail&aid=3006749&group_id=34079&atid=410687
>
> Regards,
>
> Harald
> _______________________________________________
> http://lists.egnite.de/mailman/listinfo/en-nut-discussion
>