The problem is, I can't get it to segfault when compiled with --compile-type=debug. I'll see if I can't add some debugging of my own..<br><br>- Mark.<br><br><div><span class="gmail_quote">On 2/13/07, <b class="gmail_sendername">
Alan Nisota</b> <<a href="mailto:alannisota@gmail.com">alannisota@gmail.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Alan Nisota wrote:<br>> It appears that the VirtualAlloc is failing. This has nothing to do<br>> with the # of CPUs but does have to do with your CPU type and compile<br>> options. All I can say is compile with debug enabled, then startup
<br>> mythfrontend inside gdb and set a breakpoint at ext.c:445<br>> Then step through and find out where it fails. If you get a<br>> stack-trace before you get to the breakpoint, give me a 'bt' but I'd
<br>> be surprised if that were the case. The VirtualAlloc function changed<br>> significantly between the 1.0 and 1.1 versions of my patch.<br>I forgot to add, you'll get a break because of the clone() call<br>
(depending on your gdb options, this may not happen). Once you select a<br>recording which starts CoreAVC you should get a break in VirtualAlloc<br>(due to the breakpoint I told you to add). If you can get past that<br>
(which you probably can't based on the logs), you'll get a segmentation<br>fault that looks like:<br><br>Program received signal SIGSEGV, Segmentation fault.<br>0x100bd887 in ?? ()<br><br>This is normal. you just need to 'continue'. This is a triggered
<br>segfault used to swap the compressed image with the uncompressed one,<br>and is one of the nastier sides of PECompact2. The SEGV will be caught<br>and handled properly by the relevant handler in this case so it is a red
<br>herring as far as crashes go.<br><br>_______________________________________________<br>mythtv-dev mailing list<br><a href="mailto:mythtv-dev@mythtv.org">mythtv-dev@mythtv.org</a><br><a href="http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev">
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev</a><br></blockquote></div><br>