Compressed sections are a general kind of section. You can compress literally any kind of section. All relocations are in terms of the decompressed section. We can't use ELFSectionWriter on say symbol tables because the actual data there isn't finalized until finalization. We don't have to concern ourselves with how compression and modification interact for debug sections because we don't modify those.<br><br><div class="gmail_quote"><div dir="ltr">On Wed, Aug 1, 2018, 2:59 AM James Henderson via Phabricator <<a href="mailto:reviews@reviews.llvm.org">reviews@reviews.llvm.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">jhenderson added a comment.<br>
<br>
I think @jakehehrlich's proposal more or less matches my thoughts. But I'm wondering why we need the OriginalData field? Do we have anything else that can modify the contents of a debug section? If not, why do we need a different field here? I may be missing something obviously...<br>
<br>
On the note of weird diffs, are you developing on Windows, and if so, did you remember to strip line endings?<br>
<br>
<br>
Repository:<br>
rL LLVM<br>
<br>
<a href="https://reviews.llvm.org/D49678" rel="noreferrer" target="_blank">https://reviews.llvm.org/D49678</a><br>
<br>
<br>
<br>
</blockquote></div>