<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Mar 14, 2017, at 6:00 AM, Nema, Ashutosh <<a href="mailto:Ashutosh.Nema@amd.com" class="">Ashutosh.Nema@amd.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div class="WordSection1" style="page: WordSection1; font-family: Helvetica; font-size: 10px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: 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;"><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Summarizing the discussion on the implementation approaches.<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Discussed about two approaches, first running ‘InnerLoopVectorizer’ again on the epilog loop immediately after vectorizing the original loop within the same vectorization pass, the second approach where re-running vectorization pass and limiting vectorization factor of epilog loop by metadata.<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><Approach-2><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Challenges with re-running the vectorizer pass:<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif; text-indent: -0.25in;" class=""><span class="">1)<span style="font-style: normal; font-variant-caps: normal; font-weight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New Roman';" class=""> <span class="Apple-converted-space"> </span></span></span>Reusing alias check result:<span class="Apple-converted-space"> </span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">When vectorizer pass runs again it finds the epilog loop as a new loop and it may generates alias check, this new alias check may overkill the gains of epilog vectorization.<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">We should use the already computed alias check result instead of re computing again.<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif; text-indent: -0.25in;" class=""><span class="">2)<span style="font-style: normal; font-variant-caps: normal; font-weight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New Roman';" class=""> <span class="Apple-converted-space"> </span></span></span>Rerun the vectorizer and hoist the new alias check:<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">It’s not possible to hoist alias checks as its not fully redundant (not dominated by other checks), it’s not getting execute in all paths.<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><img width="567" height="156" id="Picture_x0020_1" class="" apple-inline="yes" apple-width="yes" apple-height="yes" src="cid:image003.jpg@01D29CF1.122287F0"></div></div></div></blockquote><div><br class=""></div><div><br class=""></div><div><div>I don’t understand. Looks like you have the same alias checks for the epilog loop too. How is this CFG different from the re-vectorization of the scalar loop? Would be good to have both CFGs here and highlighting the difference.</div><div><br class=""></div><div>I thought that the whole point was that *if* you reached the epilog vector loop via the first vector loop, you want to bypass the alias checks before the epilog vector.</div><div><br class=""></div><div>I still don’t understand why that’s not possible with some sophisticated predicate propagation independent from the vectorizer. I am not saying it’s already possible but it should be.</div><div><br class=""></div><div>Adam</div><div class=""><br class=""></div></div><br class=""><blockquote type="cite" class=""><div class=""><div class="WordSection1" style="page: WordSection1; font-family: Helvetica; font-size: 10px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: 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;"><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">NOTE: We cannot prepone alias check as its expensive compared to other checks.<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><Approach-1><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif; text-indent: -0.25in;" class=""><span class="">1)<span style="font-style: normal; font-variant-caps: normal; font-weight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New Roman';" class=""> <span class="Apple-converted-space"> </span></span></span>Current patch depends on the existing functionality of LoopVectorizer, it uses ‘InnerLoopVectorizer’ again to vectorize the epilog loop, as it happens in the same vectorization pass we have flexibility to reuse already computed alias result check & limit vectorization factor for the epilog loop.<span class="Apple-converted-space"> </span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif; text-indent: -0.25in;" class=""><span class="">2)<span style="font-style: normal; font-variant-caps: normal; font-weight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New Roman';" class=""> <span class="Apple-converted-space"> </span></span></span>It does not generate the blocks for new block layout explicitly, rather it depends on ‘InnerLoopVectorizer::createEmptyLoop’ to generate new block layout. The new block layout get automatically generated by calling the ‘InnerLoopVectorizer:: vectorize’ again.<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif; text-indent: -0.25in;" class=""><span class="">3)<span style="font-style: normal; font-variant-caps: normal; font-weight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New Roman';" class=""> <span class="Apple-converted-space"> </span></span></span>Block layout description with epilog loop vectorization is available at<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><a href="https://reviews.llvm.org/file/data/fxg5vx3capyj257rrn5j/PHID-FILE-x6thnbf6ub55ep5yhalu/LayoutDescription.png" style="color: purple; text-decoration: underline;" class="">https://reviews.llvm.org/file/data/fxg5vx3capyj257rrn5j/PHID-FILE-x6thnbf6ub55ep5yhalu/LayoutDescription.png</a><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Approach-1 looks feasible, please comment if any objections.<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Regards,<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Ashutosh<o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div class=""><div style="border-style: solid none none; border-top-color: rgb(225, 225, 225); border-top-width: 1pt; padding: 3pt 0in 0in;" class=""><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><b class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class="">From:</span></b><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span class="Apple-converted-space"> </span>Nema, Ashutosh<span class="Apple-converted-space"> </span><br class=""><b class="">Sent:</b><span class="Apple-converted-space"> </span>Wednesday, March 1, 2017 10:42 AM<br class=""><b class="">To:</b><span class="Apple-converted-space"> </span>'Daniel Berlin' <<a href="mailto:dberlin@dberlin.org" style="color: purple; text-decoration: underline;" class="">dberlin@dberlin.org</a>><br class=""><b class="">Cc:</b><span class="Apple-converted-space"> </span><a href="mailto:anemet@apple.com" style="color: purple; text-decoration: underline;" class="">anemet@apple.com</a>; Hal Finkel <<a href="mailto:hfinkel@anl.gov" style="color: purple; text-decoration: underline;" class="">hfinkel@anl.gov</a>>; Zaks, Ayal <<a href="mailto:ayal.zaks@intel.com" style="color: purple; text-decoration: underline;" class="">ayal.zaks@intel.com</a>>; Renato Golin <<a href="mailto:renato.golin@linaro.org" style="color: purple; text-decoration: underline;" class="">renato.golin@linaro.org</a>>;<span class="Apple-converted-space"> </span><a href="mailto:mkuper@google.com" style="color: purple; text-decoration: underline;" class="">mkuper@google.com</a>; Mehdi Amini <<a href="mailto:mehdi.amini@apple.com" style="color: purple; text-decoration: underline;" class="">mehdi.amini@apple.com</a>>; llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" style="color: purple; text-decoration: underline;" class="">llvm-dev@lists.llvm.org</a>><br class=""><b class="">Subject:</b><span class="Apple-converted-space"> </span>RE: [llvm-dev] [Proposal][RFC] Epilog loop vectorization<o:p class=""></o:p></span></div></div></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Sorry I misunderstood, gvn/newgvn/gvnhoist cannot help here as these checks are not dominated by all paths.<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Regards,<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Ashutosh<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt 0.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><b class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class="">From:</span></b><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span class="Apple-converted-space"> </span>Daniel Berlin [<a href="mailto:dberlin@dberlin.org" style="color: purple; text-decoration: underline;" class="">mailto:dberlin@dberlin.org</a>]<span class="Apple-converted-space"> </span><br class=""><b class="">Sent:</b><span class="Apple-converted-space"> </span>Tuesday, February 28, 2017 6:58 PM<br class=""><b class="">To:</b><span class="Apple-converted-space"> </span>Nema, Ashutosh <<a href="mailto:Ashutosh.Nema@amd.com" style="color: purple; text-decoration: underline;" class="">Ashutosh.Nema@amd.com</a>><br class=""><b class="">Cc:</b><span class="Apple-converted-space"> </span><a href="mailto:anemet@apple.com" style="color: purple; text-decoration: underline;" class="">anemet@apple.com</a>; Hal Finkel <<a href="mailto:hfinkel@anl.gov" style="color: purple; text-decoration: underline;" class="">hfinkel@anl.gov</a>>; Zaks, Ayal <<a href="mailto:ayal.zaks@intel.com" style="color: purple; text-decoration: underline;" class="">ayal.zaks@intel.com</a>>; Renato Golin <<a href="mailto:renato.golin@linaro.org" style="color: purple; text-decoration: underline;" class="">renato.golin@linaro.org</a>>;<span class="Apple-converted-space"> </span><a href="mailto:mkuper@google.com" style="color: purple; text-decoration: underline;" class="">mkuper@google.com</a>; Mehdi Amini <<a href="mailto:mehdi.amini@apple.com" style="color: purple; text-decoration: underline;" class="">mehdi.amini@apple.com</a>>; llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" style="color: purple; text-decoration: underline;" class="">llvm-dev@lists.llvm.org</a>><br class=""><b class="">Subject:</b><span class="Apple-converted-space"> </span>Re: [llvm-dev] [Proposal][RFC] Epilog loop vectorization<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Hoisting or removing?<br class="">Neither pass does hoisting, you'd need gvnhoist for that.<o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Even then:<br class="">Staring at your example, none of the checks are fully redundant (IE they are not dominated by other checks) or execute on all paths.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Thus, hoisting them would be purely speculative code motion, which none of our passes do.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">If you would like these sets of checks to be removed, you would need to place them in a place that they execute unconditionally.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Otherwise, this is not a standard code hoisting/removal transform.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">The only redundancy i can see here at all is the repeated getelementptr computation.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">If you move it to the preheader, it will be eliminated.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Otherwise, none of the checks are redundant.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><br class="">What would you hope to happen in this case?<o:p class=""></o:p></div></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Tue, Feb 28, 2017 at 5:09 AM, Nema, Ashutosh <<a href="mailto:Ashutosh.Nema@amd.com" target="_blank" style="color: purple; text-decoration: underline;" class="">Ashutosh.Nema@amd.com</a>> wrote:<o:p class=""></o:p></div><blockquote style="border-style: none none none solid; border-left-color: rgb(204, 204, 204); border-left-width: 1pt; padding: 0in 0in 0in 6pt; margin: 5pt 0in 5pt 4.8pt;" class=""><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">I have tried running both gvn and newgvn but it did not helped in hoisting the alias checks:</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Please check, maybe I have missed something.</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><TestCase></span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">void foo (char *A, char *B, char *C, int len) {</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> int i = 0;</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> for (i=0 ; i< len; i++)</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> A[i] = B[i] + C[i];</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">}</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><Command></span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> $ opt –O3 –gvn test.ll –o test.opt.ll</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> $ opt –O3 –newgvn test.ll –o test.opt.ll</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">“test.ll” is attached, it got already vectorized by the approach running vectorizer twice by annotate the remainder loop with metadata to limit the vectorization factor for epilog vector loop.</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Regards,</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Ashutosh</span><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> </span><o:p class=""></o:p></div><div class=""><div style="border-style: solid none none; border-top-color: rgb(225, 225, 225); border-top-width: 1pt; padding: 3pt 0in 0in;" class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><b class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class="">From:</span></b><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span class="Apple-converted-space"> </span><a href="mailto:anemet@apple.com" target="_blank" style="color: purple; text-decoration: underline;" class="">anemet@apple.com</a><span class="Apple-converted-space"> </span>[mailto:<a href="mailto:anemet@apple.com" target="_blank" style="color: purple; text-decoration: underline;" class="">anemet@apple.com</a>]<span class="Apple-converted-space"> </span><br class=""><b class="">Sent:</b><span class="Apple-converted-space"> </span>Tuesday, February 28, 2017 1:33 AM<br class=""><b class="">To:</b><span class="Apple-converted-space"> </span>Hal Finkel <<a href="mailto:hfinkel@anl.gov" target="_blank" style="color: purple; text-decoration: underline;" class="">hfinkel@anl.gov</a>><br class=""><b class="">Cc:</b><span class="Apple-converted-space"> </span>Daniel Berlin <<a href="mailto:dberlin@dberlin.org" target="_blank" style="color: purple; text-decoration: underline;" class="">dberlin@dberlin.org</a>>; Nema, Ashutosh <<a href="mailto:Ashutosh.Nema@amd.com" target="_blank" style="color: purple; text-decoration: underline;" class="">Ashutosh.Nema@amd.com</a>>; Zaks, Ayal <<a href="mailto:ayal.zaks@intel.com" target="_blank" style="color: purple; text-decoration: underline;" class="">ayal.zaks@intel.com</a>>; Renato Golin <<a href="mailto:renato.golin@linaro.org" target="_blank" style="color: purple; text-decoration: underline;" class="">renato.golin@linaro.org</a>>;<span class="Apple-converted-space"> </span><a href="mailto:mkuper@google.com" target="_blank" style="color: purple; text-decoration: underline;" class="">mkuper@google.com</a>; Mehdi Amini <<a href="mailto:mehdi.amini@apple.com" target="_blank" style="color: purple; text-decoration: underline;" class="">mehdi.amini@apple.com</a>>; llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank" style="color: purple; text-decoration: underline;" class="">llvm-dev@lists.llvm.org</a>><br class=""><b class="">Subject:</b><span class="Apple-converted-space"> </span>Re: [llvm-dev] [Proposal][RFC] Epilog loop vectorization</span><o:p class=""></o:p></div></div></div><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Feb 27, 2017, at 12:01 PM, Hal Finkel <<a href="mailto:hfinkel@anl.gov" target="_blank" style="color: purple; text-decoration: underline;" class="">hfinkel@anl.gov</a>> wrote:<o:p class=""></o:p></div></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On 02/27/2017 01:47 PM, Daniel Berlin wrote:<o:p class=""></o:p></div></div><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Mon, Feb 27, 2017 at 11:29 AM, Adam Nemet <<a href="mailto:anemet@apple.com" target="_blank" style="color: purple; text-decoration: underline;" class="">anemet@apple.com</a>> wrote:<o:p class=""></o:p></div><blockquote style="border-style: none none none solid; border-left-color: rgb(204, 204, 204); border-left-width: 1pt; padding: 0in 0in 0in 6pt; margin: 5pt 0in 5pt 4.8pt;" class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div class=""><div class=""><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Feb 27, 2017, at 10:11 AM, Hal Finkel <<a href="mailto:hfinkel@anl.gov" target="_blank" style="color: purple; text-decoration: underline;" class="">hfinkel@anl.gov</a>> wrote:<o:p class=""></o:p></div></div><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On 02/27/2017 11:47 AM, Adam Nemet wrote:<o:p class=""></o:p></div></div><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Feb 27, 2017, at 9:39 AM, Daniel Berlin <<a href="mailto:dberlin@dberlin.org" target="_blank" style="color: purple; text-decoration: underline;" class="">dberlin@dberlin.org</a>> wrote:<o:p class=""></o:p></div></div><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Mon, Feb 27, 2017 at 9:29 AM, Adam Nemet <<a href="mailto:anemet@apple.com" target="_blank" style="color: purple; text-decoration: underline;" class="">anemet@apple.com</a>> wrote:<o:p class=""></o:p></div><blockquote style="border-style: none none none solid; border-left-color: rgb(204, 204, 204); border-left-width: 1pt; padding: 0in 0in 0in 6pt; margin: 5pt 0in 5pt 4.8pt;" class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Feb 27, 2017, at 7:27 AM, Hal Finkel <<a href="mailto:hfinkel@anl.gov" target="_blank" style="color: purple; text-decoration: underline;" class="">hfinkel@anl.gov</a>> wrote:<o:p class=""></o:p></div></div><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class=""><br class="">On 02/27/2017 06:29 AM, Nema, Ashutosh wrote:</span><o:p class=""></o:p></div></div><blockquote style="margin-top: 5pt; margin-bottom: 5pt; font-variant-caps: normal; text-align: start; word-spacing: 0px;" class=""><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class="">Thanks for looking into this.</span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class=""> </span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class="">1) Issues with re running vectorizer:</span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class="">Vectorizer might generate redundant alias checks while vectorizing epilog loop.</span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class="">Redundant alias checks are expensive, we like to reuse the results of already computed alias checks.</span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class="">With metadata we can limit the width of epilog loop, but not sure about reusing alias check result.</span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class="">Any thoughts on rerunning vectorizer with reusing the alias check result ?</span><o:p class=""></o:p></div></div></div></blockquote><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 7.5pt; font-family: Helvetica, sans-serif;" class=""><br class=""><span style="background-color: white; background-position: initial initial; background-repeat: initial initial;" class="">One way of looking at this is: Reusing the alias-check result is really just a conditional propagation problem; if we don't already have an optimization that can combine these after the fact, then we should.</span></span><o:p class=""></o:p></div></div></blockquote><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">+Danny<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Isn’t Extended SSA supposed to help with this?<o:p class=""></o:p></div></div></div></div></blockquote><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Yes, it will solve this with no issue already. GVN probably does already too.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">even if if you have<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">if (a == b)<o:p class=""></o:p></div></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">if (a == c)<o:p class=""></o:p></div></div></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> if (a == d)<o:p class=""></o:p></div></div></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> if (a == e)<o:p class=""></o:p></div></div></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> if (a == g)<o:p class=""></o:p></div></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">and we can prove a ... g equivalent, newgvn will eliminate them all and set all the branches true.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">If you need a simpler clean up pass, we could run it on sub-graphs.<o:p class=""></o:p></div></div></div></div></div></div></blockquote><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Yes we probably don’t want to run a full GVN after the “loop-scheduling” passes.<o:p class=""></o:p></div></div></div></blockquote><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><br class="">FWIW, we could, just without the memory-dependence analysis enabled (i.e. set the NoLoads constructor parameter to true). GVN is pretty fast in that mode.<o:p class=""></o:p></div></div></div></blockquote><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">OK. Another data point is that I’ve seen cases in the past where the alias checks required for the loop passes could enable GVN to remove redundant loads/stores. Currently we can only pick these up with LTO when GVN is rerun.<o:p class=""></o:p></div></div></div></div></blockquote><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">This is just GVN brokenness, newgvn should not have this problem.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">If it does, i'd love to see it.<o:p class=""></o:p></div></div></div></div></div></blockquote><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><br class="">I thought that the problem is that we just don't run GVN after that point in the pipeline.<o:p class=""></o:p></div></div></div></div></div></blockquote><div class=""><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Yeah, that is the problem but I think Danny misunderstood what I was trying to say.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">This was a datapoint to possibly rerun GVN with memory-awareness.<o:p class=""></o:p></div></div><p class="MsoNormal" style="margin: 0in 0in 12pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;"><o:p class=""> </o:p></p><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div class=""><p class="MsoNormal" style="margin: 0in 0in 12pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;"><br class=""> -Hal<o:p class=""></o:p></p><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div class=""><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">(I'm working on the last few parts of turning it on by default, but it requires a new getModRefInfo interface to be able to get the last few testcases)<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div></div></div></div></blockquote><p class="MsoNormal" style="margin: 0in 0in 12pt 1.5in; font-size: 12pt; font-family: 'Times New Roman', serif;"><o:p class=""> </o:p></p><pre style="margin: 0in 0in 0.0001pt 1.5in; font-size: 10pt; font-family: 'Courier New';" class="">-- <o:p class=""></o:p></pre><pre style="margin: 0in 0in 0.0001pt 1.5in; font-size: 10pt; font-family: 'Courier New';" class="">Hal Finkel<o:p class=""></o:p></pre><pre style="margin: 0in 0in 0.0001pt 1.5in; font-size: 10pt; font-family: 'Courier New';" class="">Lead, Compiler Technology and Programming Languages<o:p class=""></o:p></pre><pre style="margin: 0in 0in 0.0001pt 1.5in; font-size: 10pt; font-family: 'Courier New';" class="">Leadership Computing Facility<o:p class=""></o:p></pre><pre style="margin: 0in 0in 0.0001pt 1.5in; font-size: 10pt; font-family: 'Courier New';" class="">Argonne National Laboratory</pre></div></div></blockquote></div></div></div></div></div></blockquote></div></div></div></div></blockquote></div><br class=""></body></html>