<div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Jan 24, 2022 at 12:22 PM Dávid Bolvanský via Phabricator <<a href="mailto:reviews@reviews.llvm.org">reviews@reviews.llvm.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">xbolva00 added a comment.<br>
<br>
In D116821#3266620 <<a href="https://reviews.llvm.org/D116821#3266620" rel="noreferrer" target="_blank">https://reviews.llvm.org/D116821#3266620</a>>, @jmorse wrote:<br>
<br>
> Pushed up 74db5c8c95 <<a href="https://reviews.llvm.org/rG74db5c8c95e2aed40d288bb2df92eb859b87c827" rel="noreferrer" target="_blank">https://reviews.llvm.org/rG74db5c8c95e2aed40d288bb2df92eb859b87c827</a>> that effectively reverts this patch, through disabling the on-by-default-for-x86 portion of code.<br>
><br>
> I've dug through the CTMark SPASS sample and found some now-obvious problems. I reckon they can be addressed, but not immediately, so it's best to revert for the moment.<br>
><br>
> This raises some uncertainty regarding the LLVM14 branch date; I'll post back here closer to the time.<br>
<br>
You could default this option to just false, instead of ifdefing it out, so anybody could enable it explicitly.<br></blockquote><div><br></div><div>Agreed if you wouldn't mind :)</div><div><br></div><div>-eric </div></div></div>