<div dir="ltr"><div>Buzz Aldrin?<br><br></div>(Yeah, I'm still on this list. :) I dread to think how long it is since I last did any work on the project.)<br></div><div class="gmail_extra"><br><div class="gmail_quote">On 26 January 2016 at 11:14, TeeEmCee <span dir="ltr"><<a href="mailto:teeemcee@users.sf.net" target="_blank">teeemcee@users.sf.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><p>Wonferful. You ovelooked this line in the log:<br>
01-25 08:19:39.838 200 200 F DEBUG : signal 7 (SIGBUS), code 1 (BUS_ADRALN), fault addr 0x9fc7106a</p>
<p>BUS_ADRALN clearly means misaligned address.</p>
<p>Funny, as soon as I read that this was related to attacks I guessed this had to do with the animation stack. On a little inspection, it's obvious what the problem is: the stack contains a mix of 32 bit and 16 bit values, leading to misalignment (until only a few months ago it only had 16 bit values). I'm not sure why it doesn't crash with debug=2, but it might be that temporarily storing the pointer in another variable causes different instructions to be emitted which can tolerate unaligned access.</p>
<p>Anyway the solution is easy: I replaced each popw/pushw with popdw/pushdw, and simplified the stack code. But I managed to break battles.Can't figure out what I've done wrong, will try again later.</p><span class="">
<hr>
<p><strong> <a href="http://sourceforge.net/p/ohrrpgce/bugs/2017/" target="_blank">[bugs:#2017]</a> When Android port is compiled with debug=0 battles crash the whole app at the end of any attack</strong></p>
</span><p><span class=""><strong>Status:</strong> open<br>
<strong>Milestone:</strong> OHRRPGCE<br>
<strong>Created:</strong> Sun Jan 24, 2016 04:22 AM UTC by Bob the Hamster<br>
</span><strong>Last Updated:</strong> Mon Jan 25, 2016 04:23 PM UTC<br>
<strong>Owner:</strong> nobody</p><span class="">
<p>The Wobbler noticed a serious bug in the Android port. It seems that when it is compiled with debug=0 the battles always crash at the end of the first attack.</p>
<p>I have no idea why this is happening, but I temporarily went back to compiling with debug=2 for Android games that use the battle system until we figure out what is going on.</p>
<p>Here is a copy of Wandering Hamster build with debug=0 that reproduces the crash <a href="http://hamsterrepublic.com/tmp/wandering-hamster-crash.apk" rel="nofollow" target="_blank">http://hamsterrepublic.com/tmp/wandering-hamster-crash.apk</a></p>
<hr>
<p>Sent from <a href="http://sourceforge.net" target="_blank">sourceforge.net</a> because <a href="mailto:ohrrpgce@lists.motherhamster.org" target="_blank">ohrrpgce@lists.motherhamster.org</a> is subscribed to <a href="https://sourceforge.net/p/ohrrpgce/bugs/" target="_blank">https://sourceforge.net/p/ohrrpgce/bugs/</a></p>
<p>To unsubscribe from further messages, a project admin can change settings at <a href="https://sourceforge.net/p/ohrrpgce/admin/bugs/options." target="_blank">https://sourceforge.net/p/ohrrpgce/admin/bugs/options.</a> Or, if this is a mailing list, you can unsubscribe from the mailing list.</p></span></div><br>_______________________________________________<br>
Ohrrpgce mailing list<br>
<a href="mailto:ohrrpgce@lists.motherhamster.org">ohrrpgce@lists.motherhamster.org</a><br>
<a href="http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org" rel="noreferrer" target="_blank">http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org</a><br>
<br></blockquote></div><br></div>