<div dir="ltr"><div dir="ltr">Thank you all for replying!<br></div><div dir="ltr"><br></div><div>As Paul suggested, today, if clang doesn't attach source locations for 'artificial' marked functions, then inline pass will automatically attach call site location on the inlined instructions from callee.</div><div>Anyway, I filed <a href="https://github.com/llvm/llvm-project/issues/53472">an issue</a> for this.</div><div><br></div><div>Fangqing</div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, Jan 28, 2022 at 12:32 PM David Blaikie via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org">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"><div dir="ltr"><div dir="ltr">On Fri, Jan 28, 2022 at 12:07 PM Robinson, Paul <<a href="mailto:paul.robinson@sony.com" target="_blank">paul.robinson@sony.com</a>> wrote:<br></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div lang="EN-US">
<div>
<p class="MsoNormal">In order to get the described effect, I think we’d need to put DW_AT_artificial on the subprogram, and then not attach source locations to the generated code. That would have the effect of not allowing break/step inside, but the subprogram
name is still available. This is not the same as `nodebug`, where you lose the subprogram name too.<u></u><u></u></p>
<p class="MsoNormal">Shouldn’t be at all hard to get this to work.</p></div></div></blockquote><div><br>That doesn't seem to be the described or actual effect (testing GCC's debug info, even for __attribute__((artificial)) functions, the locations inside the function (inlined or non-inlined) are attributed to their original source location, not to the call site) - I believe the intent is that only effect of the attribute is to add the DW_AT_artificial attribute to the function (& that seems to be the implemented effect) - but if GCC is emitting some other debug info format (not sure what other formats it supports) that doesn't have an equivalent to DW_AT_artificial, it does the next best thing - which is to attribute the inlined source locations to the call site. (unclear what it does for non-inlined code in that case - though interestingly the attribute is only valid in GCC on functions that are marked "inline"... which seems like a weird limitation to me).<br><br>I believe Clang is implementing the same behavior as GCC, as far as DWARF is concerned at least. It appears to be a documentation issue to me.<br><br>- Dave<br> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div lang="EN-US"><div><p class="MsoNormal"><u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">Fangqing, would you be able to file an issue for this?<u></u><u></u></p>
<p class="MsoNormal">Thanks,<u></u><u></u></p>
<p class="MsoNormal">--paulr<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<div style="border-top:none;border-right:none;border-bottom:none;border-left:1.5pt solid blue;padding:0in 0in 0in 4pt">
<div>
<div style="border-right:none;border-bottom:none;border-left:none;border-top:1pt solid rgb(225,225,225);padding:3pt 0in 0in">
<p class="MsoNormal"><b>From:</b> llvm-dev <<a href="mailto:llvm-dev-bounces@lists.llvm.org" target="_blank">llvm-dev-bounces@lists.llvm.org</a>> <b>On Behalf Of
</b>David Blaikie via llvm-dev<br>
<b>Sent:</b> Friday, January 28, 2022 2:37 PM<br>
<b>To:</b> Reid Kleckner <<a href="mailto:rnk@google.com" target="_blank">rnk@google.com</a>><br>
<b>Cc:</b> llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>><br>
<b>Subject:</b> Re: [llvm-dev] about __attribute__((artificial)) and inline pass<u></u><u></u></p>
</div>
</div>
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<div>
<p class="MsoNormal">On Fri, Jan 28, 2022 at 11:26 AM Reid Kleckner <<a href="mailto:rnk@google.com" target="_blank">rnk@google.com</a>> wrote:<u></u><u></u></p>
</div>
<div>
<blockquote style="border-top:none;border-right:none;border-bottom:none;border-left:1pt solid rgb(204,204,204);padding:0in 0in 0in 6pt;margin-left:4.8pt;margin-right:0in">
<div>
<div>
<p class="MsoNormal">I thought `artificial` and `nodebug` were supposed to be synonyms, just different names invented by different communities.
<u></u><u></u></p>
</div>
</div>
</blockquote>
<div>
<p class="MsoNormal"><br>
Doesn't look like it, based on the GCC documentation and behavior - GCC puts the DW_AT_artificial attribute on the entity, but still fully describes it (DWARF remains the same except for that attribute). I think the second option in the GCC docs is for debug
info formats that don't have an equivalent of DW_AT_artificial, in which case it downgrades to the equivalent of "nodebug".<br>
<u></u><u></u></p>
</div>
<blockquote style="border-top:none;border-right:none;border-bottom:none;border-left:1pt solid rgb(204,204,204);padding:0in 0in 0in 6pt;margin-left:4.8pt;margin-right:0in">
<div>
<div>
<p class="MsoNormal">The linked patch doesn't implement the LLVM side of things, so I'm not sure when that got added.<u></u><u></u></p>
</div>
</div>
</blockquote>
<div>
<p class="MsoNormal"><br>
The patch added the artificial flag on the metadata which was already implemented/supported in LLVM (used for things like implicit constructors, etc, which don't have user-written code in them) and matches the GCC behavior - no extra work was needed on the
LLVM side.<br>
<u></u><u></u></p>
</div>
<blockquote style="border-top:none;border-right:none;border-bottom:none;border-left:1pt solid rgb(204,204,204);padding:0in 0in 0in 6pt;margin-left:4.8pt;margin-right:0in">
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<div>
<p class="MsoNormal">On Fri, Jan 28, 2022 at 11:03 AM David Blaikie <<a href="mailto:dblaikie@gmail.com" target="_blank">dblaikie@gmail.com</a>> wrote:<u></u><u></u></p>
</div>
<blockquote style="border-top:none;border-right:none;border-bottom:none;border-left:1pt solid rgb(204,204,204);padding:0in 0in 0in 6pt;margin-left:4.8pt;margin-right:0in">
<div>
<p class="MsoNormal">Hmm, seems I missed this feature (or at least don't remember) when it went in.<br>
<br>
Looks like the GCC spec might be more accurate to the current implementation than what Clang's documentation says:<br>
<br>
"<span style="font-size:13.5pt;font-family:Times,serif;color:black">artificial</span><u></u><u></u></p>
<p class="MsoNormal" style="margin-left:0.5in"><a name="m_-7780019829097547105_m_2324470475014252717_m_8334339238115289765_m_4238466721377605"></a><span style="font-size:13.5pt;font-family:Times,serif;color:black">This attribute is useful for small inline wrappers which if possible should appear during
debugging as a unit, depending on the debug info format it will either mean marking the function as artificial or using the caller location for all instructions within the inlined body."<u></u><u></u></span></p>
<p class="MsoNormal"><br>
& the "marking the function as artificial" is the thing that this patch implemented in Clang. This did not have the same effect as nodebug, and doesn't "use the caller location for all instructions within the inlined body".<br>
<br>
It's possible we could implement the "artificial" attribute as a an alias for "nodebug" which would also be conforming with GCC's spec, but would be less expressive than the current implementation (though would save debug info size in a way the current Clang
implementation does not).<br>
<br>
It looks like Clang's implementation is consistent with GCC's implementation for DWARF - "marking the function as artificial", not "using the caller location for instructions within the inlined body", but Clang's documentation is incorrect, since it documents
the latter and not the former.<u></u><u></u></p>
</div>
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<div>
<p class="MsoNormal">On Fri, Jan 28, 2022 at 9:44 AM Fangqing Du via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>> wrote:<u></u><u></u></p>
</div>
<blockquote style="border-top:none;border-right:none;border-bottom:none;border-left:1pt solid rgb(204,204,204);padding:0in 0in 0in 6pt;margin-left:4.8pt;margin-right:0in">
<div>
<div>
<p class="MsoNormal">Dear all,<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal"><u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal">In clang/llvm 7.0, `__attribute__((artificial))` can be identified by clang and attach `DIFlagArtificial` into function metadata in
<a href="https://urldefense.com/v3/__https:/reviews.llvm.org/D43259*change-jbvcnfxttyAz__;Iw!!JmoZiZGBv3RvKRSx!pEsMF8kDNPtKteuvhreMp6mZjOVtB-7XVCgfOi_ELf630AbjxHC2VDLB3BoX9IMFWg$" target="_blank">
this change.</a><u></u><u></u></p>
</div>
<div>
<p class="MsoNormal"><u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal">And according to the <a href="https://urldefense.com/v3/__https:/clang.llvm.org/docs/AttributeReference.html*artificial__;Iw!!JmoZiZGBv3RvKRSx!pEsMF8kDNPtKteuvhreMp6mZjOVtB-7XVCgfOi_ELf630AbjxHC2VDLB3BrUKhXpvw$" target="_blank">
specification</a> of `artificial`, if I understand correctly, when the function (which is marked with `artificial` attribute) is inlined, the inlined function instructions should be associated with the callsite line number, instead of the line of inlined callee.<u></u><u></u></p>
</div>
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<p class="MsoNormal">But from my observation, with and without this attribute, the inline pass behavior is not affected, and inlined instructions debug location still point to the callee instead of callsite.<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal">Here is my experiment: <a href="https://urldefense.com/v3/__https:/godbolt.org/z/oW8Td5d4M__;!!JmoZiZGBv3RvKRSx!pEsMF8kDNPtKteuvhreMp6mZjOVtB-7XVCgfOi_ELf630AbjxHC2VDLB3BqqabukgA$" target="_blank">
godbolt experiment link</a><u></u><u></u></p>
</div>
<div>
<p class="MsoNormal"><u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal">Is my understanding correct?<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal">Thanks,<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal">Fangqing<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal">Xilinx Inc.<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal"><u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal">== ------------ example -------------===<u></u><u></u></p>
</div>
<div>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 1 int bar(int);</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 2 </span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 3 inline int __attribute__((artificial)) __attribute__((always_inline)) foo(int x)</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 4 {</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 5 return bar(x + 1); </span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 6 }</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 7 </span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 8 void baz(void)</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 9 {</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 10 auto x = foo(1);</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 11 x = foo(2);</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
<p style="margin:0in;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal">
<span style="font-size:10pt;font-family:"Helvetica Neue",serif;color:black;background:rgb(204,204,204)"> 12 }</span><span style="font-size:10pt;font-family:"Helvetica Neue",serif"><u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal">When function 'foo' is inlined, the debug location of call instruction 'bar' still points to line #5, instead of line #10 and #11.<u></u><u></u></p>
</div>
</div>
<p class="MsoNormal">_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
<a href="https://urldefense.com/v3/__https:/lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev__;!!JmoZiZGBv3RvKRSx!pEsMF8kDNPtKteuvhreMp6mZjOVtB-7XVCgfOi_ELf630AbjxHC2VDLB3Bp86eErug$" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><u></u><u></u></p>
</blockquote>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote></div></div>
_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
<a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
</blockquote></div></div>