[tex-k] mf segfaults if compiled with -g3 (bug in mfmpi386.asm)

[Karl Berry]

    It is clear we don't exactly _need_ the code, but it _is_ subtantially
    better than the compiler-generated code, because the default
    implementation is complicated by being restricted to 32 bit arithmetic
    and the other alternative uses floating point.

I guess what I was wondering is whether "substantially better" is
actually measurable these days, on modern machines.  I can easily
believe that that assembly code is still much cleaner than what current
compilers generate.  If it speeds up a TeX run by a few hundred
milliseconds (a whole lot of cycles nowadays), who cares, it's not worth
the trouble to maintain.

Would someone like to try compiling with and without the assembler and
see what difference it makes?

Thanks,
karl