Testing alternatives to functions from lib1funcs.S

Wolfgang Hospital <[email protected]> Sun, 21 Apr 2024 10:08:11 +0200
Newsgroups	gmane.comp.hardware.avr.gcc
Message-ID	<[email protected]>
This is a multi-part message in MIME format.
--------------4KPx4S2jfUekx0b5UPlgdxzV
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

  Dear all,

Is there a test scaffold for the functions from lib1funcs.S,
correctness, size&speed over the variety of 8-bit AVR cores?

Is there a more comprehensive statement of calling conventions than 
https://gcc.gnu.org/wiki/avr-gcc#Exceptions_to_the_Calling_Convention, 
in particular explicitly stating which functions are guaranteed to have 
__zero_reg__ 0 on entry/where it suffices to have __zero_reg__ 0 on 
return as opposed to preserving its value?

I've been tinkeringaround, the "ldi  r_cnt, 9""rjmp entry point" in 
__udivmodqi4 instead of "ldi  r_cnt, 8""lsl  r_arg1" annoying me for 
years. (Biggest relative strict improvement I found, FWIW.)

Recommendations for a platform to vent such ideas welcome (I know of 
stackoverflow.com).

regards

W. Hospital

-- 
Wolfgang Hospital

--------------4KPx4S2jfUekx0b5UPlgdxzV
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<!DOCTYPE html>
<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p><font face="tt"> Dear all,</font></p>
    <p><font face="tt">Is there a test scaffold for the functions from
        lib1funcs.S,<br>
        correctness, size&amp;speed over the variety of 8-bit AVR cores?<br>
        <br>
        Is there a more comprehensive statement of calling conventions
        than
        <a class="moz-txt-link-freetext" href="https://gcc.gnu.org/wiki/avr-gcc#Exceptions_to_the_Calling_Convention">https://gcc.gnu.org/wiki/avr-gcc#Exceptions_to_the_Calling_Convention</a>,
        in particular explicitly stating which functions are guaranteed
        to have __zero_reg__ 0 on </font><font face="tt">entry/where it
        suffices to have </font><font face="tt">__zero_reg__ 0 on
        return as opposed to preserving its value?<br>
      </font></p>
    <p><font face="tt">I've been tinkering</font><span class="HwtZe"
        lang="en"><span class="jCAhz ChMk0b"><span class="ryNqvb">
            around</span></span></span><font face="tt">, the "</font><font
        face="tt">ldi  r_cnt, 9""</font><font face="tt">rjmp entry
        point" in __udivmodqi4 instead of "ldi  r_cnt, 8""lsl  r_arg1"
        annoying me for years. (Biggest relative strict improvement I
        found, FWIW.)<br>
      </font></p>
    <p><font face="tt">Recommendations for a platform to vent such ideas
        welcome (I know of stackoverflow.com).<br>
        <br>
        regards<br>
        <br>
        W. Hospital<br>
      </font></p>
    <pre class="moz-signature" cols="72">-- 
Wolfgang Hospital</pre>
  </body>
</html>

--------------4KPx4S2jfUekx0b5UPlgdxzV--