[En-Nut-Discussion] Recent NutOS Changes Broke _P

[Harald Kipp]

Guess, Jan already fall asleep. :-)

The CVS HEAD compiles again, but printf_P and similar
routines produce garbage. I'm sure, that's something
very easy to fix, but can't find it.

If anyone is faster than me, please let me know.

Btw. looking to vfprintf_p.c

     rl = strlen_P(fmt) + 1;
     if ((rp = NutHeapAlloc(rl)) == 0)
         return -1;
     memcpy_P(rp, fmt, rl);
     rc = _putf(_write, stream->iob_fd, rp, ap);
     NutHeapFree(rp);

reminds me again, that this was a bad hack, which
saved me from implementing _putf_P(). It breaks the
Nut/OS rule of moving as least data as possible.
So, please tell me that this is rubbish, I knew
already. :-)

Sigh,

Harald