#1 floating point error after calling JPEG_Decode()

[Anonymous]

Hello,

We use your library in our application build using MS
Visual C++ 6 on a P4 cumputer. But after a call to
the decode function (JPEG_Decode) all further
floating point computing with "double" in our
application give always unexpected false result.

To workaround this bug we just add the EMMS
Instruction at the end of the JPEG_Decode asm
function.

What do you think about ?

Nicolas.

nicolas.courtiller@free.fr