Re: [avr-libc-dev] Release philosofy questions

[Theodore A. Roth]

On Wed, 23 Oct 2002, Joerg Wunsch wrote:

:) As Theodore A. Roth wrote:
:)
:) > OT: Joerg, don't forget to update the NEWS file with the new interfaces
:) > you've added.
:)
:) OK.  Well, probably rather a verbal summary, and a pointer to the
:) documentation.

That's fine. I think just a simple list of the new functions and a note to
see the dox for details will suffice.

:)
:) > Also, could you make a simple test program available so I
:) > can start reviewing the new stuff? (I've been a bad boy and haven't looked
:) > at it yet, sorry :-( )
:)
:) You mean, something to test printf() & Co with?
:)
:) Here's my current test case.  Of course, the float formats are not
:) supposed to work with the CVS version of vfprintf(), but they should
:) at least be parsed correctly.  IIRC, they are supposed to print a
:) single `?'.
:)
:) Also, linking against -lprintf_min is supposed to work, see the
:) documentation.
:)
:) As you can see from the #defines :), this is currently being tested on
:) an ATmega128.  Other chips are supposed to work as well, except you
:) need quite a bit of flash for all my constant strings here.  My
:) floating point test is now already close to 8 KB flash in total, but
:) the non-FP version shouldn't use much more than 5 KB or so.
:)
:) I'd appreciate if people could stress-test all this a bit.  All the
:) documented functionality is supposed to be functioning.

Thanks. I'll add it to my (evergrowing ;) TODO file. I still need to get
my stk500 and friends setup at home again...

:)
:) If you're going to emulate it in simulavr, put an #if 0 around the
:) loop_until_bit_is_set() call in uart_putchar().
:)
:) You'll probably be as surprised as i've been when seeing how much
:) random garbage there is in your uninitialized RAM. :-)

Not sure what you mean, but I bet it will be clear when I actually run it.
;-)

<code snipped>

Ted Roth