<div dir="ltr">First batch of patches for review:<div><a href="http://reviews.llvm.org/D4958">http://reviews.llvm.org/D4958</a><br></div><div><a href="http://reviews.llvm.org/D4959">http://reviews.llvm.org/D4959</a><br></div>
<div><a href="http://reviews.llvm.org/D4960">http://reviews.llvm.org/D4960</a><br></div><div><a href="http://reviews.llvm.org/D4961">http://reviews.llvm.org/D4961</a></div><div>These fix the problem exposed above (and another I found since), and impact exclusively with the ARM backend.</div>
<div><br></div><div>I have another patch coming soon offering more refactoring, and merging with X86AtomicExpandPass. Do you want to be added as subscribers/reviewers ? (I only put Tim Northover as subscriber on the first batch because he looked interested in that bug).</div>
<div><br></div><div>Cheers,</div><div>Robin</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Aug 18, 2014 at 8:55 AM, Robin Morisset <span dir="ltr"><<a href="mailto:morisset@google.com" target="_blank">morisset@google.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Thanks for taking the time to look at the patches.<div><br></div><div>I will redo them in the way you suggest and put them on Phabricator for review.</div>
<div><br></div><div>Cheers,</div><div>Robin</div></div><div class="HOEnZb"><div class="h5">
<div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Aug 18, 2014 at 6:16 AM, Tim Northover <span dir="ltr"><<a href="mailto:t.p.northover@gmail.com" target="_blank">t.p.northover@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Robin,<br>
<div><br>
On 15 August 2014 21:49, Robin Morisset <<a href="mailto:morisset@google.com" target="_blank">morisset@google.com</a>> wrote:<br>
> I have<br>
> attached the corresponding patches; they are not completely ready for review<br>
> (mostly missing tests), I would just like to make sure that the general<br>
> approach seems reasonable.<br>
<br>
</div>I've had a quick glance at the patches, and the code seems fairly sane.<br>
<br>
But I'm not so sure about starting with a new pass then deleting the<br>
other two. I think it's likely to muddy the revision control history.<br>
I'd prefer to see a gradual evolution of the<br>
AtomicExpandLoadLinked.cpp (perhaps starting with a "git mv" and some<br>
internal renaming to stake out the intent, followed by adding and<br>
using the extra hooks).<br>
<br>
Cheers.<br>
<span><font color="#888888"><br>
Tim.<br>
</font></span></blockquote></div><br></div>
</div></div></blockquote></div><br></div>