<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div><br></div><div>On Mar 16, 2015, at 6:51 PM, David Blaikie <<a href="mailto:dblaikie@gmail.com">dblaikie@gmail.com</a>> wrote:<br><br></div><blockquote type="cite"><div><div dir="ltr">I'm all for it :)</div></div></blockquote><div><br></div><div>Great! I'll commit soon. </div><div><br><blockquote type="cite"><div><div dir="ltr"><br>So this is enabled for any .ll file parsing? (I don't know exactly when/where the normal non-debug verifier is run)</div></div></blockquote><div><br></div><div>There, and also in a number of spots in the optimization pipeline. IIRC, the clang driver passes -disable-verify to -cc1 by default when you've built in Release mode to skip it. </div><br><blockquote type="cite"><div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Mar 16, 2015 at 5:38 PM, Duncan P. N. Exon Smith <span dir="ltr"><<a href="mailto:dexonsmith@apple.com" target="_blank">dexonsmith@apple.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">r186634 started verifying debug info, and r194986 disabled it by default<br>
because it was too expensive to run the checks on every function (since<br>
most of the graph was reachable from each function).<br>
<br>
r206300 moved the checks to module-level to make it cheaper, but there<br>
was already quite a bit of testcase bitrot (and the verifier would only<br>
print `<badref>`) so I guess no one had time to turn it back on.<br>
<br>
This does just that. Upgrade scripts this past autumn and winter<br>
probably fixed some of the bitrot, and this weekend I fixed the verifier<br>
output (r232275, r232417, r232418) and thusly the remaining failing<br>
testcases (r232290, r232415).<br>
<br>
<br>
</blockquote></div><br></div>
</div></blockquote></div></body></html>