<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Hi Paul,<br>
<br>
thanks for your comments.<br>
<blockquote type="cite"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">This is PR37240 (<a href="https://bugs.llvm.org/show_bug.cgi?id=37240">https://bugs.llvm.org/show_bug.cgi?id=37240</a>). I suspect this problem affects
all targets; your patch D68076 would address it only for AArch64. Although I would suggest you do some careful measurements to determine the runtime performance effect, to decide whether this is acceptable.</span></blockquote>
Yes, in principle the problem that instruction scheduling is dependent on the presence of cfi instruction should affect more targets than AArch64. However, this does not imply that all of these targets produce inconsistent assembly depending on debug information.<br>
<br>
<blockquote type="cite"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">The more complete approach in your steps 2 + 3 would solve this for all targets, assuming the solution did not have to be very target-specific. This would
benefit the entire community.</span></blockquote>
At least 2. would require a lot of target dependent changes because the insertion of cfi instructions would have to be moved from target specific frame lowering into an (probably again target specific) insertion pass.<br>
<br>
David<br>
<br>
<div class="moz-cite-prefix">On 26/09/2019 13:55, <a class="moz-txt-link-abbreviated" href="mailto:paul.robinson@sony.com">
paul.robinson@sony.com</a> wrote:<br>
</div>
<blockquote type="cite" cite="mid:E3B07FDB86BFF041819DC057DEED8FEA01420BDEF8@USCULXMSG13.am.sony.com">
<meta name="Generator" content="Microsoft Word 14 (filtered
medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 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;}
p
{mso-style-priority:99;
margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman","serif";}
span.EmailStyle18
{mso-style-type:personal-reply;
font-family:"Calibri","sans-serif";
color:#1F497D;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@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]-->
<div class="WordSection1">
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Hi David,<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">This is PR37240 (<a href="https://bugs.llvm.org/show_bug.cgi?id=37240" moz-do-not-send="true">https://bugs.llvm.org/show_bug.cgi?id=37240</a>). I suspect this
problem affects all targets; your patch D68076 would address it only for AArch64. Although I would suggest you do some careful measurements to determine the runtime performance effect, to decide whether this is acceptable.<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">The more complete approach in your steps 2 + 3 would solve this for all targets, assuming the solution did not have to be very target-specific. This would
benefit the entire community.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">--paulr<o:p></o:p></span></p>
<p class="MsoNormal"><a name="_MailEndCompose" moz-do-not-send="true"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></a></p>
<div style="border:none;border-left:solid blue 1.5pt;padding:0in
0in 0in 4.0pt">
<div>
<div style="border:none;border-top:solid #B5C4DF
1.0pt;padding:3.0pt 0in 0in 0in">
<p class="MsoNormal"><b><span style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">From:</span></b><span style="font-size:10.0pt;font-family:"Tahoma","sans-serif""> llvm-dev [<a class="moz-txt-link-freetext" href="mailto:llvm-dev-bounces@lists.llvm.org">mailto:llvm-dev-bounces@lists.llvm.org</a>]
<b>On Behalf Of </b>David Tellenbach via llvm-dev<br>
<b>Sent:</b> Thursday, September 26, 2019 5:57 AM<br>
<b>To:</b> llvm-dev<br>
<b>Cc:</b> nd<br>
<b>Subject:</b> [llvm-dev] [AArch64] Generated assembly differs depending on whether debug information is generated or not<o:p></o:p></span></p>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">Hi,</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">we at Arm have noticed that assembly can differ when compiling for AArch64</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">depending on whether debug information is generated or not.
</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">The issue is reproducible for the following small example compiled with `-O1`</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">for `aarch64-arm-linux-gnu`:</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> a() {</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> b(a);</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> for (;;)</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> c("", b);</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> }</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">The reason for the difference is that AArch64 frame lowering emits CFI</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">instructions if debug information is enabled but not if not. CFI instructions</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">act as scheduling boundaries during instruction scheduling and therefore lead to</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">differing scheduling regions and an overall different instruction scheduling.</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">We see several ways to fix the issue and would welcome comments on this:</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> 1. Enabling unwind tables by default for AArch64: By enabling unwind tables</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> by default CFI instructions will be inserted in both, debug and non-debug</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> mode. This should lead to smaller scheduling regions and probably to less</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> scheduling potential.</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> However, I've measured the average size of scheduling regions for randomly</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> generated programs with and without default unwind tables and found an</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> average difference of 0.5 to 1 instruction. Other architectures such as x86</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> do exactly this and therefore don't face the issue.</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> The following patch on Phabricator introduces the said change:</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">
<a class="moz-txt-link-freetext" href="https://reviews.llvm.org/D68076">https://reviews.llvm.org/D68076</a></span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> 2. Postpone insertion of CFI instructions until after instruction scheduling.</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> This would require a new pass running after instruction scheduling that
</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> inserts CFI instructions if needed. The downside I see is increased</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> compile-time and probably some code duplication with frame lowering.
</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> 3. Change instruction scheduling such that CFI instructions get tied together</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> with relevant instructions in such a way that they get scheduled together.</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> If this could work it would probably the cleanest solution.</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">To summarize:</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">1. would make scheduling in the non-debug case behave like in the</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">debug case and therefore probably cost some scheduling potential. However, it</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">would be by far the most easy to implement. 2. + 3. would probably lead to</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">better scheduling but seem to be more complex to implement.</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black">Comments and additional ideas are welcome.</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Courier
New";color:black"> David</span><span style="font-family:"Calibri","sans-serif";color:black"><o:p></o:p></span></p>
</div>
<p class="MsoNormal"><span style="font-family:"Calibri","sans-serif";color:black"><o:p> </o:p></span></p>
</div>
</div>
</div>
</blockquote>
<br>
</body>
</html>