<div dir="ltr">Hi all,<div><br></div><div>Sorry I can't attend the conference call on Wednesday, so I wrapped up my survey into a blog post that is easier to digest, here is the link:</div><div><a href="https://medium.com/@mshockwave/llvm-loop-lr-survey-6df55f8dff74">https://medium.com/@mshockwave/llvm-loop-lr-survey-6df55f8dff74</a> </div><div><br></div><div>Here is the take away:</div><div>1. I'm not sure LoopRotation will cause any problem if the user Passes require the property of exit block predecessors (i.e. predecessors of exit blocks are always in the loop)<br></div><div>2. I do a simple experiment on top of LLVM's existing regression testsuite by running LoopRotation before every Passes that use LoopSimplify loops as inputs. About 10% of the selected test cases are failed.</div><div>3. As the follow up of the experiment, I pick part of the failures from LoopFusion and LoopStrengthReduction to find the root cause. All of them are actually producing expecting and correct results. The failures are caused by false negatives coming from the testing infrastructure.</div><div><br></div><div>As a personal verdicts, I think migrating canonical loops to LoopRotation loops seem to require little of works on existing loop Passes, I’m pretty positive on this migration. But of course, we definitely need feedback from the community and real-field test results.</div><div><br></div><div>B.R.</div><div>- Min</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Wed, Jul 17, 2019 at 1:28 PM Bardia Mahjour via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.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"><div><p><font size="2">Hi all,</font><br><br><font size="2">Apparently some people had trouble joining today. I also had trouble using the attendee ID provided by webex. I've canceled the meeting series and will be scheduling a new one with an updated link. Please remember to update your calendars with the new invite.</font><br><br><font size="2">My apologies to those who couldn't join today, and to everyone for the churn. </font><br><br><br><font size="2">Today's Meeting Minutes:</font><br><br><font size="2">---------------------------</font><br><font size="2">Wed, July 17th, 2019:</font><br><font size="2">---------------------------</font><br><font size="2">- Data Dependence Graph Presentation</font><br><font size="2"> - Minimal vs full graph: Start with the full graph implementation and </font><br><font size="2"> leave optimizations until later when we have more imperical evidence on</font><br><font size="2"> where the bottlenecks are in this implementation.</font><br><font size="2"> - Trade-offs of the current proposal discussed with no objections to the </font><br><font size="2"> overall design from the audience at this time. More feedback is welcome!</font><br><font size="2"> - The approximate sequence of patches to be posted for review will be: </font><br><font size="2"> 1. DDG (basic), </font><br><font size="2"> 2. DDG (with simplification)</font><br><font size="2"> 3. DDG (with root node and pi-blocks)</font><br><font size="2"> 4. Control Dependence Graph (CDG)</font><br><font size="2"> 5. Program Dependence Graph (PDG)</font><br><br><font size="2">- Status Updates:</font><br><font size="2"> - Directed Graph (</font><a href="https://reviews.llvm.org/D64088" target="_blank"><font size="2">https://reviews.llvm.org/D64088</font></a><font size="2">)</font><br><font size="2"> - Patch is approved and ready to land.</font><br><font size="2"> - First patch of DDG to be posted soon.</font><br><font size="2"> - Loop Cache Cost Analysis (</font><a href="https://reviews.llvm.org/D63459" target="_blank"><font size="2">https://reviews.llvm.org/D63459</font></a><font size="2">)</font><br><font size="2"> - Ready and awaiting approval.</font><br><font size="2"> - Loop Guard API (</font><a href="https://reviews.llvm.org/D63885" target="_blank"><font size="2">https://reviews.llvm.org/D63885</font></a><font size="2">)</font><br><font size="2"> - Need to clarify comments from Philip to agree on the strength of the </font><br><font size="2"> definition of a loop guard</font><br><font size="2"> - Impact of Loop Rotation on existing passes</font><br><font size="2"> - Min unable to join today, will send a follow up email and/or discuss</font><br><font size="2"> next meeting.</font><br><font size="2"> - Bugzilla bugs update</font><br><font size="2"> - Ran out of time. Postponed to next meeting.</font><br><br><br><font size="2">------------------------------------------------------------</font><br><font size="2">Tentative Agenda for next meeting (July 31 @ 11:00 am EST):</font><br><font size="2">------------------------------------------------------------</font><br><font size="2">- Status Updates</font><br><font size="2"> - Impact of Loop Rotation on existing passes (Min-Yih)</font><br><font size="2"> - Bugzilla bugs update (Vivek)</font><br><font size="2"> - Loop Cache Cost Analysis (</font><a href="https://reviews.llvm.org/D63459" target="_blank"><font size="2">https://reviews.llvm.org/D63459</font></a><font size="2">) (Ettore)</font><br><font size="2"> - Loop Guard API (</font><a href="https://reviews.llvm.org/D63885" target="_blank"><font size="2">https://reviews.llvm.org/D63885</font></a><font size="2">) (Whitney)</font><br><font size="2"> - Data Dependence Graph (Bardia)</font><br><br><font size="2">- Delinearization (Gary Elsesser)</font><br><font size="2">- Loop vs Function pass (Jamie Schmeiser)</font><br><font size="2">- Loop Guard API (Whitney)</font><br><br><br><font size="2">Bardia Mahjour<br>Compiler Optimizations<br>IBM Toronto Software Lab<br><a href="mailto:bmahjour@ca.ibm.com" target="_blank">bmahjour@ca.ibm.com</a> (905) 413-2336</font><br>
</p></div>
_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
<a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
</blockquote></div><br clear="all"><div><br></div>-- <br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div>Min-Yih Hsu</div><div>Ph.D Student in ICS Department, University of California, Irvine (UCI).<br></div></div></div>