<html><head><meta http-equiv="Content-Type" content="text/html charset=windows-1252"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">I don’t have an opinion on this.  :) <div><br><div><div>On Nov 21, 2013, at 6:39 AM, Arnold Schwaighofer <<a href="mailto:aschwaighofer@apple.com">aschwaighofer@apple.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div style="font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><br>On Nov 21, 2013, at 7:29 AM, Hal Finkel <<a href="mailto:hfinkel@anl.gov">hfinkel@anl.gov</a>> wrote:<br><br><blockquote type="cite">----- Original Message -----<br><blockquote type="cite">From: "Pekka Jääskeläinen" <<a href="mailto:pekka.jaaskelainen@tut.fi">pekka.jaaskelainen@tut.fi</a>><br>To: "Craig Topper" <<a href="mailto:craig.topper@gmail.com">craig.topper@gmail.com</a>><br>Cc: "Commit Messages and Patches for LLVM" <<a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a>><br>Sent: Thursday, November 21, 2013 4:15:12 AM<br>Subject: Re: [llvm] r190916 - Lift alignment restrictions for load/store<span class="Apple-tab-span" style="white-space: pre;">       </span>folding on VINSERTF128/VEXTRACTF128. Fixes<br>PR17268. [PATCH]<br><br>Hi,<br><br>The attached patch reverts the line that forces the SLPVectorizer<br>always on in opt. Passes "make check" in LLVM 3.4.<br></blockquote><br>Nadav, Do you have an opinion about this? Generally, I'd say to disable it in opt (since we already have a command-line flag in the PMB to enable it). On the other hand, we also seem to force on the loop vectorizer (in a way that matches Clang's defaults) in the line above.<br></blockquote><br>Yes, I added this when I moved the logic at which optimization level to optimize out of the pass manager.<br><br>We should do the same for the slp vectorizer. I think it is a good thing that opt -Owhatever tries to match clang -Owhatever as close as possible.<br><br><br><br><blockquote type="cite"><br>-Hal<br><br><blockquote type="cite"><br>OK to commit?<br><br>On 11/19/2013 09:51 AM, Nick Lewycky wrote:<br><blockquote type="cite">On 17 September 2013 20:55, Craig Topper <<a href="mailto:craig.topper@gmail.com">craig.topper@gmail.com</a><br><<a href="mailto:craig.topper@gmail.com">mailto:craig.topper@gmail.com</a>>> wrote:<br><br>  Author: ctopper<br>  Date: Tue Sep 17 22:55:53 2013<br>  New Revision: 190916<br><br>  URL: <a href="http://llvm.org/viewvc/llvm-project?rev=190916&view=rev">http://llvm.org/viewvc/llvm-project?rev=190916&view=rev</a><br>  Log:<br>  Lift alignment restrictions for load/store folding on<br>  VINSERTF128/VEXTRACTF128. Fixes PR17268.<br><br><br>  Modified: llvm/trunk/tools/opt/opt.cpp<br>  URL:<br>  <a href="http://llvm.org/viewvc/llvm-project/llvm/trunk/tools/opt/opt.cpp?rev=190916&r1=190915&r2=190916&view=diff">http://llvm.org/viewvc/llvm-project/llvm/trunk/tools/opt/opt.cpp?rev=190916&r1=190915&r2=190916&view=diff</a><br>  ==============================================================================<br>  --- llvm/trunk/tools/opt/opt.cpp (original)<br>  +++ llvm/trunk/tools/opt/opt.cpp Tue Sep 17 22:55:53 2013<br>  @@ -462,6 +462,7 @@ static void AddOptimizationPasses(PassMa<br>                                   DisableLoopUnrolling :<br>                                   OptLevel == 0;<br><br>      Builder.LoopVectorize = OptLevel > 1 && SizeLevel < 2;<br>  +  Builder.SLPVectorize = true;<br><br><br>This doesn't match the commit log. Was this intentional?<br><br>Nick<br></blockquote><br>--<br>Pekka<br><br>_______________________________________________<br>llvm-commits mailing list<br><a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits<br><br></blockquote><br>--<span class="Apple-converted-space"> </span><br>Hal Finkel<br>Assistant Computational Scientist<br>Leadership Computing Facility<br>Argonne National Laboratory<br><br>_______________________________________________<br>llvm-commits mailing list<br><a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br><a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits">http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits</a></blockquote></div></blockquote></div><br></div></body></html>