<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=us-ascii">
<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:11.0pt;
font-family:"Calibri",sans-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.msonormal0, li.msonormal0, div.msonormal0
{mso-style-name:msonormal;
margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
span.EmailStyle19
{mso-style-type:personal-reply;
font-family:"Calibri",sans-serif;
color:windowtext;}
.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]-->
</head>
<body bgcolor="white" lang="EN-US" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal">As has been mentioned elsewhere, Sony generally fixes up references from debug info to stripped functions (of any length) using -1, because that’s a less-likely-to-be-real address than 0x0 or 0x1. (0x0 is a typical base address for shared
libraries, I’d think using it has the potential to mislead various consumers.) For .debug_ranges we use -2, because both a 0/0 pair and a -1/-1 pair have a reserved meaning in that section.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">If you’re looking only at zero-length functions, you can stop there; but I’m not sure why stopping there solves much of a real problem, as zero-length functions seem like a weird corner case. Linkers know how to strip dead functions (gc)
or deduplicate them (icf, COMDAT) and people do this all the time, in some cases (COMDAT) without explicitly asking for it, so non-zero-length functions seem like the much more interesting case. In that situation, -1 (or -2) seems like a much wiser choice
of blessed-as-not-real address, versus 0x0 or 0x1.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Stripping non-zero-length functions does mean you have to care about more sections. For example .debug_locs would want to be fixed up the same way as .debug_ranges, not because a debugger would care but so that dumpers would not run into
the 0/0 brick wall. We also fix up lengths in .debug_aranges to zero, although there might be history behind that tactic that I’m not aware of; it seems like it ought to be unnecessary, if consumers are aware of the special address(es).<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">--paulr<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></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 #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in">
<p class="MsoNormal"><b>From:</b> Alexey Lapshin <alapshin@accesssoftek.com> <br>
<b>Sent:</b> Thursday, May 28, 2020 9:03 AM<br>
<b>To:</b> Sriraman Tallam <tmsriram@google.com>; Wei Mi <wmi@google.com>; Robinson, Paul <paul.robinson@sony.com>; Adrian Prantl <aprantl@apple.com>; Jonas Devlieghere <jdevlieghere@apple.com>; Alexey Lapshin <a.v.lapshin@mail.ru>; Eric Christopher <echristo@gmail.com>;
Fangrui Song <maskray@google.com>; David Blaikie <dblaikie@gmail.com>; llvm-dev@lists.llvm.org<br>
<b>Subject:</b> Re: [llvm-dev] Range lists, zero-length functions, linker gc<o:p></o:p></p>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<p><span style="font-size:12.0pt;color:black">Hi David,<o:p></o:p></span></p>
<p><span style="font-size:12.0pt;color:black"><o:p> </o:p></span></p>
<p><span style="font-size:12.0pt;color:black">>So there have been several recent discussions about the issues around
<o:p></o:p></span></p>
<p><span style="font-size:12.0pt;color:black">>DWARF-agnostic linking and gc-sections, linkonce function definitions being
<o:p></o:p></span></p>
<p><span style="font-size:12.0pt;color:black">>dropped, etc - and just how much DWARF-awareness would be suitable
<o:p></o:p></span></p>
<p><span style="font-size:12.0pt;color:black">>in a linker to help with this situation.<o:p></o:p></span></p>
<div>
<div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121"><br>
> I'd like to discuss a narrower instance of this issue: Zero length gc'd/deduplicated functions.<br>
<br>
> LLVM seems to at least produce zero length functions in a few cases:<br>
> * non-void function without a return statement<br>
> * function definition containing only llvm_unreachable<br>
> (both of these trap at -O0, but at higher optimization levels even the trap <o:p>
</o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> instruction is removed & you get the full power UB of control flowing off
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> the end of the function into whatever other bytes are after that function)<br>
<br>
> So, for context, debug_ranges (this whole issue doesn't exist in DWARFv5, <o:p>
</o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> FWIW) is a list of address pairs, terminated by a pair of zeros.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> With function sections, or even just with normal C++ inline functions,
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> the CU will have a range entry for that function that consists of two relocations
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> - to the start and end of the function. Generally the start of the function is the
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> start of the section, and the end is "start of function + length of function (aka addend)".<br>
<br>
> Usually any relocation to the section would keep that section "alive" during linking -
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> but that would cause debug info to defeat linker GC and deduplication. So there's
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> special rules for how linkers handle these relocations in debug info to allow the
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> sections to be dropped - what do you write in the bytes that requested the relocation?<br>
<br>
> Binutils ld: Special cases only debug_ranges, resolving all relocations to dead
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> code to 1. In other debug sections, these values are all resolved to zero.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> Gold and lld: Special cases all debug info sections - resolving all relocations
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> to "addend" (so begin usually goes to zero, end goes to "size of function")<br>
<br>
> These special rules are designed to ensure omitted/gc'd/deduplicated functions <o:p>
</o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> don't cause the range list to terminate prematurely (which would happen if begin/end
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> were both resolved to zero).<br>
<br>
>But with an empty function, gold and lld's strategy here fails to avoid terminating a
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">>range list by accident.<br>
<br>
> What should we do about it?<br>
<br>
> 1) Ensure no zero-length functions exist? (doesn't address backwards <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> compatibility/existing functions/other compilers)<br>
> 2) adopt the binutils approach to this (at least in debug_ranges - maybe in all
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> debug sections? (doing it in other sections could break )<br>
> 3) Revisit the discussion about using an even more 'blessed' value, <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> like int max-1? ( <a href="https://urldefense.com/v3/__https:/reviews.llvm.org/D59553__;!!JmoZiZGBv3RvKRSx!r2Jqc2yEgxrb2QcQEocDHJBizj0KUKE70_57b4_rsj1TN0qB8NpBvVKtY63HSqgMOg$" target="_blank">https://reviews.llvm.org/D59553</a> )<br>
<br>
> (I don't have links to all the recent threads about this discussion - I think D59553
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> might've spawned a separate broader discussion/non-review - oh, Alexey wrote a
<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> good summary with links to other discussions here:<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">> <a href="https://urldefense.com/v3/__http:/lists.llvm.org/pipermail/llvm-dev/2019-September/135068.html__;!!JmoZiZGBv3RvKRSx!r2Jqc2yEgxrb2QcQEocDHJBizj0KUKE70_57b4_rsj1TN0qB8NpBvVKtY638NIRu2g$">
http://lists.llvm.org/pipermail/llvm-dev/2019-September/135068.html</a> )<br>
<br>
> Thoughts?<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121">I think for the problem of "zero length functions and .debug_ranges" binutils approach looks good:<br>
<br>
>Special cases only debug_ranges, resolving all relocations to<br>
>dead code to 1. In other debug sections, these values are all resolved to<br>
>zero.<br>
<br>
But, this would not completely solve the problem from <a href="https://urldefense.com/v3/__https:/reviews.llvm.org/D59553__;!!JmoZiZGBv3RvKRSx!r2Jqc2yEgxrb2QcQEocDHJBizj0KUKE70_57b4_rsj1TN0qB8NpBvVKtY63HSqgMOg$">
https://reviews.llvm.org/D59553</a> - Overlapped address ranges. Binutils approach will solve the problem if the address range specified as start_address:end_address. While resolving relocations, it would replace such a range with 1:1.<br>
However, It would not work if address ranges were specified as start_address:length since the length is not relocated. This case could be additionally fixed by fast scan debug_info for High_PC defined as length and changing it to 1. Something which you suggested
here: <a href="https://urldefense.com/v3/__http:/lists.llvm.org/pipermail/llvm-dev/2020-May/141599.html__;!!JmoZiZGBv3RvKRSx!r2Jqc2yEgxrb2QcQEocDHJBizj0KUKE70_57b4_rsj1TN0qB8NpBvVKtY63PsubKJQ$">
http://lists.llvm.org/pipermail/llvm-dev/2020-May/141599.html</a>.<br>
<br>
So it looks like following solution could fix both problems and be relatively fast:
<br>
<br>
"Resolve all relocations from debug sections into dead code to 1. Parse debug sections and replace HighPc of an address range pointing to dead code and specified as length to 1".<br>
<br>
As the result all address ranges pointing into dead code would be marked as zero length.<br>
<br>
There still exist another problem: <br>
<br>
DWARF4: "A range list entry (but not a base address selection or end of list entry) whose beginning and<br>
ending addresses are equal has no effect because the size of the range covered by such an<br>
entry is zero."<br>
<br>
DWARF5: "A bounded range entry whose beginning and ending address offsets are equal<br>
(including zero) indicates an empty range and may be ignored."<br>
<br>
These rules allow us to ignore zero-length address ranges. I.e., some tool reading DWARF is permitted to ignore related DWARF entries. In that case, there could be ignored essential descriptions. That problem could happen with -flto=thin example
<a href="https://urldefense.com/v3/__https:/reviews.llvm.org/D54747*1503720__;Iw!!JmoZiZGBv3RvKRSx!r2Jqc2yEgxrb2QcQEocDHJBizj0KUKE70_57b4_rsj1TN0qB8NpBvVKtY637ju_eQw$">
https://reviews.llvm.org/D54747#1503720</a> . In this example, all type definitions except one were replaced with declarations by thinlto. The definition, which was left, is in a piece of debug info related to deleted code. According to zero-length rule, that
definition could be ignored, and finally, incomplete debug info could be used.<br>
<br>
So, it probably should be forbidden to generate debug_info, which could become incomplete after removing pieces related to zero length address ranges. Otherwise, creating zero-length address ranges could lead to incomplete debug info.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:12.0pt;color:#212121"><o:p> </o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span style="font-size:12.0pt;color:#212121">Thank you, Alexey.<br>
<br>
<br>
<o:p></o:p></span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>