<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><meta http-equiv=Content-Type content="text/html; charset=utf-8"><meta name=Generator content="Microsoft Word 15 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman",serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
span.EmailStyle17
{mso-style-type:personal-reply;
font-family:"Calibri",sans-serif;
color:#1F497D;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri",sans-serif;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]--></head><body lang=EN-US link=blue vlink=purple><div class=WordSection1><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D'>Daniel, I see correctness failures with the previous patch in AArch64 spec2006/h264ref and an internal test code we have.<o:p></o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D'>So I think your patch should only have the code fix for the delegation issue.<o:p></o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D'>I will try the new patch you attached and let you know the results.<o:p></o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D'><o:p> </o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D'>Thanks,<o:p></o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D'>Ana.<o:p></o:p></span></p><p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D'><o:p> </o:p></span></p><p class=MsoNormal><b><span style='font-size:11.0pt;font-family:"Calibri",sans-serif'>From:</span></b><span style='font-size:11.0pt;font-family:"Calibri",sans-serif'> llvmdev-bounces@cs.uiuc.edu [mailto:llvmdev-bounces@cs.uiuc.edu] <b>On Behalf Of </b>Daniel Berlin<br><b>Sent:</b> Tuesday, January 20, 2015 11:49 AM<br><b>To:</b> Hal Finkel<br><b>Cc:</b> Jiangning Liu; George Burgess IV; LLVM Developers Mailing List<br><b>Subject:</b> Re: [LLVMdev] question about enabling cfl-aa and collecting a57 numbers<o:p></o:p></span></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>So, I can make all these testcases work, but it's a little tricky (it involves tracking some things, like GEP byte range, and then checking bases and using getObjectSize, much like BasicAA does).<o:p></o:p></p><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>Because i really don't want to put that much "not well tested" code in a bugfix, and honestly, i'm not sure we will catch any cases here that BasicAA does not, i've attached a change to XFAIL these testcases, and updated the code to return MayAlias.<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>I will build and test a patch to get these back to PartialAlias, but this patch will at least get us to not be "giving wrong answers". I will also see if we catch anything with it that BasicAA does not, because if we don't, it doesn't seem worth it to me.<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal><span style='font-size:10.0pt'>Conservative new patch attached.</span><o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>(Note that i still updated the testcases, because we will *never* be able to legally return PartialAlias as they were written)<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><div><div><p class=MsoNormal><o:p> </o:p></p><div><p class=MsoNormal>On Sun Jan 18 2015 at 2:12:47 PM Daniel Berlin <<a href="mailto:dberlin@dberlin.org" target="_blank">dberlin@dberlin.org</a>> wrote:<o:p></o:p></p><blockquote style='border:none;border-left:solid #CCCCCC 1.0pt;padding:0in 0in 0in 6.0pt;margin-left:4.8pt;margin-right:0in'><div><p class=MsoNormal>On Sat Jan 17 2015 at 3:15:27 PM Hal Finkel <<a href="mailto:hfinkel@anl.gov" target="_blank">hfinkel@anl.gov</a>> wrote:<o:p></o:p></p><blockquote style='border:none;border-left:solid #CCCCCC 1.0pt;padding:0in 0in 0in 6.0pt;margin-left:4.8pt;margin-right:0in'><p class=MsoNormal style='margin-bottom:12.0pt'>----- Original Message -----<br>> From: "Daniel Berlin" <<a href="mailto:dberlin@dberlin.org" target="_blank">dberlin@dberlin.org</a>><br>> To: "Hal Finkel" <<a href="mailto:hfinkel@anl.gov" target="_blank">hfinkel@anl.gov</a>><br>> Cc: "Jiangning Liu" <<a href="mailto:Jiangning.Liu@arm.com" target="_blank">Jiangning.Liu@arm.com</a>>, "George Burgess IV" <<a href="mailto:george.burgess.iv@gmail.com" target="_blank">george.burgess.iv@gmail.com</a>>, "LLVM Developers<br>> Mailing List" <<a href="mailto:llvmdev@cs.uiuc.edu" target="_blank">llvmdev@cs.uiuc.edu</a>>, "Nick Lewycky" <<a href="mailto:nlewycky@google.com" target="_blank">nlewycky@google.com</a>><br>> Sent: Saturday, January 17, 2015 1:08:10 PM<br>> Subject: Re: [LLVMdev] question about enabling cfl-aa and collecting a57 numbers<br>><br>><br>><br>><br>> On Sat Jan 17 2015 at 12:03:33 AM Hal Finkel < <a href="mailto:hfinkel@anl.gov" target="_blank">hfinkel@anl.gov</a> ><br>> wrote:<br>><br>><br>> Hi Danny,<br>><br>> // Add TypeBasedAliasAnalysis before BasicAliasAnalysis so that<br>> // BasicAliasAnalysis wins if they disagree. This is intended to help<br>> // support "obvious" type-punning idioms.<br>> - if (UseCFLAA)<br>> - addPass( createCFLAliasAnalysisPass());<br>> addPass( createTypeBasedAliasAnalysisPa ss());<br>> addPass( createScopedNoAliasAAPass());<br>> + if (UseCFLAA)<br>> + addPass( createCFLAliasAnalysisPass());<br>> addPass( createBasicAliasAnalysisPass() );<br>><br>> Do we really want to change the order here? I had originally placed<br>> it after the metadata-based passes thinking that the compile-time<br>> would be better (guessing that the metadata queries would be faster<br>> than the CFL queries, so if the metadata could quickly return a<br>> NoAlias, then we'd cut out unecessary CFL queries). Perhaps this is<br>> an irrelevant effect, but we should have some documented rationale.<br>><br>><br>><br>> Yeah, this was a mistake (Chandler had suggested it was right<br>> earlier, but we were both wrong :P)<br>><br>><br>><br>> ; CHECK: PartialAlias: i16* %bigbase0, i8* %phi<br>> -define i8 @test0(i8* %base, i1 %x) {<br>> +define i8 @test0(i1 %x) {<br>> entry:<br>> + %base = alloca i8, align 4<br>> %baseplusone = getelementptr i8* %base, i64 1<br>> br i1 %x, label %red, label %green<br>> red:<br>> @@ -25,8 +26,9 @@ green:<br>> }<br>><br>> why should this return PartialAlias? %ohi does partially overlap, so<br>> this correct, but what happens when the overlap is partial or<br>> control dependent?<br>> So, after talking with some people offline, they convinced me in SSA<br>> form, the name would change in these situations, and the only<br>> situations you can get into trouble is with things "based on pointer<br>> arguments" (because you have no idea what their initial state is),<br>> or "globals" (because they are not in SSA form)<br>> I could not come up with a case otherwise<br><br>Okay; that part of the code could really use some more commentary. I'd really appreciate it if you should put these thoughts in written form that could be added as comments.<o:p></o:p></p></blockquote><div><p class=MsoNormal><o:p> </o:p></p></div></div><div><div><p class=MsoNormal>Will do<o:p></o:p></p></div></div><div><div><p class=MsoNormal> <o:p></o:p></p></div><blockquote style='border:none;border-left:solid #CCCCCC 1.0pt;padding:0in 0in 0in 6.0pt;margin-left:4.8pt;margin-right:0in'><p class=MsoNormal>> But i'm welcome to hear if you think this is wrong.<br>><br>> FWIW: I bootstrapped/tested the compiler with an assert that<br>> triggered if CFL-AA was going to return PartialAlias and BasicAA<br>> would have returned NoAlias, and it did not trigger with this<br>> change.<br>><br>><br>> (It would have triggered before this set of changes)<br>><br>> Not proof of course, but it at least tells me it's not "obviously<br>> wrong" :)<br>><br>><br><br>That's good :) -- but, not exactly what I was concerned about. Our general convention has been that PartialAlias is a "strong" result, like MustAlias, but implies that AA has proved that only a partial overlap will occur.<br><br>So, in this test case we get the right result:<br><br>; CHECK: PartialAlias: i16* %bigbase0, i8* %phi<br>define i8 @test0(i1 %x) {<br>entry:<br> %base = alloca i8, align 4<br> %baseplusone = getelementptr i8* %base, i64 1<br> br i1 %x, label %red, label %green<br>red:<br> br label %green<br>green:<br> %phi = phi i8* [ %baseplusone, %red ], [ %base, %entry ]<br> store i8 0, i8* %phi<br><br> %bigbase0 = bitcast i8* %base to i16*<br> store i16 -1, i16* %bigbase0<br><br> %loaded = load i8* %phi<br> ret i8 %loaded<br>}<br><br>because %phi will have a partial overlap with %bigbase0, the only uncertainty is whether the overlap is with the low byte or the high byte. But if I modify the test to be this:<br><br>define i8 @test0x(i1 %x) {<br>entry:<br> %base = alloca i8, align 4<br> %baseplustwo = getelementptr i8* %base, i64 2<br> br i1 %x, label %red, label %green<br>red:<br> br label %green<br>green:<br> %phi = phi i8* [ %baseplustwo, %red ], [ %base, %entry ]<br> store i8 0, i8* %phi<br><br> %bigbase0 = bitcast i8* %base to i16*<br> store i16 -1, i16* %bigbase0<br><br> %loaded = load i8* %phi<br> ret i8 %loaded<br>}<br><br>I still get this result:<br> PartialAlias: i16* %bigbase0, i8* %phi<o:p></o:p></p></blockquote><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal><span style='font-size:10.0pt'> </span><o:p></o:p></p></div><blockquote style='border:none;border-left:solid #CCCCCC 1.0pt;padding:0in 0in 0in 6.0pt;margin-left:4.8pt;margin-right:0in'><p class=MsoNormal><br>but now %phi might not overlap %bigbase0 at all (although, when it does, there is a partial overlap), so we should just return MayAlias (perhaps without delegation because this is a definitive result?).<o:p></o:p></p></blockquote><div><p class=MsoNormal><o:p> </o:p></p></div></div><div><div><p class=MsoNormal>Yeah, i have to do some size checking, let me see if we have the info and i'll update the patch.<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>Otherwise, my view is that we should always delegate MayAlias, because we have no idea what order the passes are in or what pass someone has inserted where :)<o:p></o:p></p></div><div><p class=MsoNormal><o:p> </o:p></p></div><div><p class=MsoNormal>(WIW: I believe the same about everything except MustAlias and NoAlias, but currently we don't delegate PartialAlias.<o:p></o:p></p></div><div><p class=MsoNormal>We claim PartialAlias is a definitive result, but it really isn't.<o:p></o:p></p></div><div><p class=MsoNormal>Right now we have TBAA that will give NoAlias results on things other passes claim are PartialAlias, and that result is correct. That's just in our default, we have no idea what other people do. Even if you ignore TBAA, plenty of other compilers have noalias/mustalias metadata that would have the same effect.<o:p></o:p></p></div></div></blockquote></div></div></div></div></div></body></html>