[lld] [llvm] [llvm-objcopy][WebAssembly] Allow --strip-debug to operate on relocatable files. (PR #102978)

[Sam Clegg via llvm-commits]

https://github.com/sbc100 created https://github.com/llvm/llvm-project/pull/102978

This change is enough to allow `--strip-debug` to work on object files, without breaking the relocation information or symbol table.

A more complete version of this change would instead reconstruct the symbol table and relocation sections, but that is much larger change.

Fixes: #102002

>From 2a09d4c812efff1f7e8b12e8b28a003398aa15ac Mon Sep 17 00:00:00 2001
From: Sam Clegg <sbc at chromium.org>
Date: Mon, 12 Aug 2024 13:53:33 -0700
Subject: [PATCH] [llvm-objcopy][WebAssembly] Allow --strip-debug to operator
 on relocatable files.

This change is enough to allow `--strip-debug` to work on object files,
without breaking the relocation information or symbol table.

A more complete version of this change would instead reconstruct the
symbol table and relocation sections, but that is much larger change.

Fixes: #102002
---
 lld/wasm/Writer.cpp                   |  5 +++++
 llvm/lib/ObjCopy/wasm/WasmObjcopy.cpp |  2 +-
 llvm/lib/ObjCopy/wasm/WasmObject.cpp  | 19 +++++++++++++++++--
 llvm/lib/ObjCopy/wasm/WasmObject.h    |  1 +
 llvm/lib/ObjCopy/wasm/WasmReader.cpp  |  1 +
 5 files changed, 25 insertions(+), 3 deletions(-)

diff --git a/lld/wasm/Writer.cpp b/lld/wasm/Writer.cpp
index 6a66a29d2498c5..6beef81d39a381 100644
--- a/lld/wasm/Writer.cpp
+++ b/lld/wasm/Writer.cpp
@@ -138,6 +138,11 @@ void Writer::calculateCustomSections() {
       // Exclude COMDAT sections that are not selected for inclusion
       if (section->discarded)
         continue;
+      // Ignore empty custom sections.  In particular objcopy/strip will
+      // sometimes replace stripped sections with empty custom sections to
+      // avoid section re-numbering.
+      if (section->getSize() == 0)
+        continue;
       StringRef name = section->name;
       // These custom sections are known the linker and synthesized rather than
       // blindly copied.
diff --git a/llvm/lib/ObjCopy/wasm/WasmObjcopy.cpp b/llvm/lib/ObjCopy/wasm/WasmObjcopy.cpp
index 5bba1dea9adf0e..cf3d884bee3bd8 100644
--- a/llvm/lib/ObjCopy/wasm/WasmObjcopy.cpp
+++ b/llvm/lib/ObjCopy/wasm/WasmObjcopy.cpp
@@ -22,7 +22,7 @@ using namespace object;
 using SectionPred = std::function<bool(const Section &Sec)>;
 
 static bool isDebugSection(const Section &Sec) {
-  return Sec.Name.starts_with(".debug");
+  return Sec.Name.starts_with(".debug") || Sec.Name.starts_with("reloc..debug");
 }
 
 static bool isLinkerSection(const Section &Sec) {
diff --git a/llvm/lib/ObjCopy/wasm/WasmObject.cpp b/llvm/lib/ObjCopy/wasm/WasmObject.cpp
index 28a2de6e6e4f12..bee9c260645137 100644
--- a/llvm/lib/ObjCopy/wasm/WasmObject.cpp
+++ b/llvm/lib/ObjCopy/wasm/WasmObject.cpp
@@ -25,8 +25,23 @@ void Object::addSectionWithOwnedContents(
 }
 
 void Object::removeSections(function_ref<bool(const Section &)> ToRemove) {
-  // TODO: remove reloc sections for the removed section, handle symbols, etc.
-  llvm::erase_if(Sections, ToRemove);
+  if (isRelocatableObject) {
+    // For relocatable objects, avoid actually removing any sections,
+    // since that can invalidate the symbol table an relocation sections.
+    // TODO: Allow removal of sections by re-generating symbol table and
+    // relocation sections here instead.
+    for (auto &Sec : Sections) {
+      if (ToRemove(Sec)) {
+        //static std::vector<char> empty;
+        Sec.Name = ".objcopy.removed";
+        Sec.SectionType = wasm::WASM_SEC_CUSTOM;
+        Sec.Contents = {};
+        Sec.HeaderSecSizeEncodingLen = std::nullopt;
+      }
+    }
+  } else {
+    llvm::erase_if(Sections, ToRemove);
+  }
 }
 
 } // end namespace wasm
diff --git a/llvm/lib/ObjCopy/wasm/WasmObject.h b/llvm/lib/ObjCopy/wasm/WasmObject.h
index f860ec697e5684..374e523cd879b6 100644
--- a/llvm/lib/ObjCopy/wasm/WasmObject.h
+++ b/llvm/lib/ObjCopy/wasm/WasmObject.h
@@ -32,6 +32,7 @@ struct Object {
   llvm::wasm::WasmObjectHeader Header;
   // For now don't discriminate between kinds of sections.
   std::vector<Section> Sections;
+  bool isRelocatableObject = false;
 
   void addSectionWithOwnedContents(Section NewSection,
                                    std::unique_ptr<MemoryBuffer> &&Content);
diff --git a/llvm/lib/ObjCopy/wasm/WasmReader.cpp b/llvm/lib/ObjCopy/wasm/WasmReader.cpp
index 578e78955af3e4..420d17f9864323 100644
--- a/llvm/lib/ObjCopy/wasm/WasmReader.cpp
+++ b/llvm/lib/ObjCopy/wasm/WasmReader.cpp
@@ -18,6 +18,7 @@ using namespace llvm::wasm;
 Expected<std::unique_ptr<Object>> Reader::create() const {
   auto Obj = std::make_unique<Object>();
   Obj->Header = WasmObj.getHeader();
+  Obj->isRelocatableObject = WasmObj.isRelocatableObject();
   std::vector<Section> Sections;
   Obj->Sections.reserve(WasmObj.getNumSections());
   for (const SectionRef &Sec : WasmObj.sections()) {