<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Renato,<div><br></div><div><div><div>On Mar 15, 2011, at 10:13 AM, Renato Golin wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div>On 14/03/11 19:45, Rafael Avila de Espindola wrote:<br><blockquote type="cite">We might need a virtual method or similar. Devang, for testing on linux<br></blockquote><blockquote type="cite">I normally use a current version of the gdb testsuite. I assume that for<br></blockquote><blockquote type="cite">darwin it is better to use an older one. Which testsuite do you normally<br></blockquote><blockquote type="cite">use?<br></blockquote><br>Hi all,<br><br>I've just compiled LLVM on an old Mac box (9.1.0, 2007) and compiled the<br>same example in the test case and everything (symbols, stepping,<br>debugging experience) was ok. All tests, including my new one, pass.<br><br>I don't know how to dump mach-o objects nor how to check for the<br>relocation sections, though. Somebody with more experience in Mach-o<br>should take a look.<font class="Apple-style-span" color="#000000"><font class="Apple-style-span" color="#144FAE"><br></font></font></div></blockquote><br></div><div>We had a bug in older assembler, which we still support, where it could not handle constructs like,</div><div><br></div><div> .long (Ltmp5-Ltmp4)-4</div><div><br></div><div>This required use of .set</div><div><br></div><div>However in this case we emitting reference to label directly without doing any arithmetic so it should be OK. If all testings looks OK, please go ahead and commit. If someone complains from our side, I'll let you know!</div><div><br></div><div>-</div><div>Devang</div><br></div></body></html>