<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
06.12.2019 17:14, David Blaikie пишет:<br>
<blockquote type="cite"
cite="mid:CAENS6EvELzCTVqXvt0HiXpe9oaV12c+Vg84V3p63LP0ssot-Qg@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">Yep, I wasn't aware of that particular wording, but
that would make this feature a non-conforming extension. Perhaps
it should be off by default - but it's still super useful to
have around/I wouldn't be OK with removing it.<br>
</div>
</blockquote>
<p>I tried to add check into the DWARFVerifier that skeleton unit
has zero children but it did not pass because of this. <br>
</p>
<p>So yes, perhaps it would be good to set it off by default. <br>
</p>
<blockquote type="cite"
cite="mid:CAENS6EvELzCTVqXvt0HiXpe9oaV12c+Vg84V3p63LP0ssot-Qg@mail.gmail.com">
<div dir="ltr"><br>
In the future something like <a
href="http://wiki.dwarfstd.org/index.php?title=TwoLevelLineTables"
moz-do-not-send="true">http://wiki.dwarfstd.org/index.php?title=TwoLevelLineTables</a> might
be able to replace this functionality & at a lower cost in
object/executable size, but I'm not sure what state that is in
and whether it does cover all the use cases.</div>
</blockquote>
<p>interestingly. it looks like an option.<br>
</p>
<p>Thank you, Alexey.<br>
</p>
<blockquote type="cite"
cite="mid:CAENS6EvELzCTVqXvt0HiXpe9oaV12c+Vg84V3p63LP0ssot-Qg@mail.gmail.com"><br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Fri, Dec 6, 2019 at 7:46 AM
Alexey Lapshin via llvm-dev <<a
href="mailto:llvm-dev@lists.llvm.org" moz-do-not-send="true">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">Hi
DebugInfo folks,<br>
<br>
I have a question on using fsplit-dwarf-inlining option:<br>
<br>
"-fsplit-dwarf-inlining, -fno-split-dwarf-inlining<br>
<br>
Provide minimal debug info in the object/executable to
facilitate online <br>
symbolication/stack traces in the absence of .dwo/.dwp files
when using <br>
Split DWARF"<br>
<br>
i.e. it puts some debug info into compilation unit from First
<br>
partition(with Skeleton unit).<br>
<br>
At the same time, there is following requirement for split
DWARF :<br>
<br>
"When generating a split DWARF object file (see Section 7.3.2
on page <br>
187), the compilation unit in the .debug_info section is a
"skeleton" <br>
compilation unit with the tag DW_TAG_skeleton_unit<br>
<br>
A skeleton compilation unit has no children."<br>
<br>
Is the behavior of fsplit-dwarf-inlining option correct, then?<br>
<br>
<br>
_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org" target="_blank"
moz-do-not-send="true">llvm-dev@lists.llvm.org</a><br>
<a
href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev"
rel="noreferrer" target="_blank" moz-do-not-send="true">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
</blockquote>
</div>
</blockquote>
</body>
</html>