Update wasm to llvm 21

Author: hamishmack

overlays/nix-prefetch-git-minimal.nix

@@ -17,9 +17,12 @@ final: prev: {
   # This can reduce closure size of nix-tools:
   #  * Eliminates dependency on python3 (70MB)
   #  * Allows sharing with `fetchgit` as it also uses `gitMinimal` (50MB)
-  inherit (final.callPackages (final.path + "/pkgs/tools/package-management/nix-prefetch-scripts") {
-    git = final.gitMinimal;
-  }) nix-prefetch-git;
+  inherit (
+    let f = import (final.path + "/pkgs/tools/package-management/nix-prefetch-scripts");
+    in
+      if (builtins.functionArgs f) ? git
+        then final.callPackages f { git = final.gitMinimal; }
+        else prev) nix-prefetch-git;
 
   # fetchgit use `buildPackages.gitMinimal` and on nixpkgs 21.11
   # and earlier that causes problems when cross compiling.

overlays/patches/wasm/llvm/gnu-install-dirs.patch

@@ -1,8 +1,8 @@
 diff --git a/CMakeLists.txt b/CMakeLists.txt
-index c9ff3696e22d..bd96aab5e237 100644
+index 73c4fc14f031..42284703a52c 100644
 --- a/CMakeLists.txt
 +++ b/CMakeLists.txt
-@@ -1133,9 +1133,9 @@ if (NOT TENSORFLOW_AOT_PATH STREQUAL "")
+@@ -1153,9 +1153,9 @@ if (NOT TENSORFLOW_AOT_PATH STREQUAL "")
    add_subdirectory(${TENSORFLOW_AOT_PATH}/xla_aot_runtime_src
      ${CMAKE_ARCHIVE_OUTPUT_DIRECTORY}/tf_runtime)
    install(TARGETS tf_xla_runtime EXPORT LLVMExports
@@ -15,7 +15,7 @@ index c9ff3696e22d..bd96aab5e237 100644
    # Once we add more modules, we should handle this more automatically.
    if (DEFINED LLVM_OVERRIDE_MODEL_HEADER_INLINERSIZEMODEL)
 diff --git a/cmake/modules/AddLLVM.cmake b/cmake/modules/AddLLVM.cmake
-index baf47677b247..81954240a9bf 100644
+index 52bf7ca0906a..ba2f4be17f08 100644
 --- a/cmake/modules/AddLLVM.cmake
 +++ b/cmake/modules/AddLLVM.cmake
 @@ -974,8 +974,8 @@ macro(add_llvm_library name)
@@ -29,17 +29,17 @@ index baf47677b247..81954240a9bf 100644
                RUNTIME DESTINATION "${CMAKE_INSTALL_BINDIR}" COMPONENT ${name})
 
        if (NOT LLVM_ENABLE_IDE)
-@@ -2240,7 +2240,7 @@ function(llvm_install_library_symlink name dest type)
+@@ -2258,7 +2258,7 @@ function(llvm_install_library_symlink name dest type)
      set(LLVM_LINK_OR_COPY copy)
    endif()
 
 -  set(output_dir lib${LLVM_LIBDIR_SUFFIX})
 +  set(output_dir ${CMAKE_INSTALL_FULL_LIBDIR}${LLVM_LIBDIR_SUFFIX})
-   if(WIN32 AND "${type}" STREQUAL "SHARED")
+   if((WIN32 OR CYGWIN) AND "${type}" STREQUAL "SHARED")
      set(output_dir "${CMAKE_INSTALL_BINDIR}")
    endif()
-@@ -2516,15 +2516,37 @@ function(llvm_setup_rpath name)
-
+@@ -2534,15 +2534,37 @@ function(llvm_setup_rpath name)
+ 
    if (APPLE)
      set(_install_name_dir INSTALL_NAME_DIR "@rpath")
 -    set(_install_rpath "@loader_path/../lib${LLVM_LIBDIR_SUFFIX}" ${extra_libdir})
@@ -79,19 +79,6 @@ index baf47677b247..81954240a9bf 100644
      if(${CMAKE_SYSTEM_NAME} MATCHES "(FreeBSD|DragonFly)")
        set_property(TARGET ${name} APPEND_STRING PROPERTY
                     LINK_FLAGS " -Wl,-z,origin ")
-@@ -2539,9 +2561,9 @@ function(llvm_setup_rpath name)
-   endif()
-
-   # Enable BUILD_WITH_INSTALL_RPATH unless CMAKE_BUILD_RPATH is set.
--  if("${CMAKE_BUILD_RPATH}" STREQUAL "")
--    set_property(TARGET ${name} PROPERTY BUILD_WITH_INSTALL_RPATH ON)
--  endif()
-+  #if("${CMAKE_BUILD_RPATH}" STREQUAL "")
-+  #  set_property(TARGET ${name} PROPERTY BUILD_WITH_INSTALL_RPATH ON)
-+  #endif()
-
-   set_target_properties(${name} PROPERTIES
-                         INSTALL_RPATH "${_install_rpath}"
 diff --git a/cmake/modules/AddOCaml.cmake b/cmake/modules/AddOCaml.cmake
 index 2d9116b08a52..2dd7cad4ec66 100644
 --- a/cmake/modules/AddOCaml.cmake
@@ -122,7 +109,7 @@ index ef4cfa3acdb5..7478e157a7c2 100644
    "${LLVM_CONFIG_LIBRARY_DIR}"
    # FIXME: Should there be other entries here?
 diff --git a/tools/llvm-config/BuildVariables.inc.in b/tools/llvm-config/BuildVariables.inc.in
-index 370005cd8d7d..7e790bc52111 100644
+index e4e1d449bf4d..3aab6ea7bf8b 100644
 --- a/tools/llvm-config/BuildVariables.inc.in
 +++ b/tools/llvm-config/BuildVariables.inc.in
 @@ -23,6 +23,7 @@
@@ -134,10 +121,10 @@ index 370005cd8d7d..7e790bc52111 100644
  #define LLVM_INSTALL_PACKAGE_DIR "@LLVM_INSTALL_PACKAGE_DIR@"
  #define LLVM_TARGETS_BUILT "@LLVM_TARGETS_BUILT@"
 diff --git a/tools/llvm-config/llvm-config.cpp b/tools/llvm-config/llvm-config.cpp
-index d5b76b1bb6c1..1dbdb2a8f10d 100644
+index 49df8fdcb7f7..c7cb05b82821 100644
 --- a/tools/llvm-config/llvm-config.cpp
 +++ b/tools/llvm-config/llvm-config.cpp
-@@ -366,7 +366,11 @@ int main(int argc, char **argv) {
+@@ -365,7 +365,11 @@ int main(int argc, char **argv) {
        sys::fs::make_absolute(ActivePrefix, Path);
        ActiveBinDir = std::string(Path);
      }

overlays/patches/wasm/llvm/haskell-wasm-llvm-project.patch

@@ -1,5 +1,5 @@
 diff --git a/clang/lib/Driver/ToolChains/WebAssembly.cpp b/clang/lib/Driver/ToolChains/WebAssembly.cpp
-index bd25fd1a8933..b906bff5d5c7 100644
+index 5054868b5ff4..e4fcd949ff71 100644
 --- a/clang/lib/Driver/ToolChains/WebAssembly.cpp
 +++ b/clang/lib/Driver/ToolChains/WebAssembly.cpp
 @@ -6,6 +6,8 @@
@@ -9,17 +9,17 @@ index bd25fd1a8933..b906bff5d5c7 100644
 +#include <sstream>
 +
  #include "WebAssembly.h"
- #include "CommonArgs.h"
  #include "Gnu.h"
-@@ -19,6 +21,7 @@
+ #include "clang/Config/config.h"
+@@ -17,6 +19,7 @@
  #include "llvm/Option/ArgList.h"
  #include "llvm/Support/FileSystem.h"
  #include "llvm/Support/Path.h"
 +#include "llvm/Support/Process.h"
  #include "llvm/Support/VirtualFileSystem.h"
 
  using namespace clang::driver;
-@@ -168,21 +171,12 @@ void wasm::Linker::ConstructJob(Compilation &C, const JobAction &JA,
+@@ -179,21 +182,12 @@ void wasm::Linker::ConstructJob(Compilation &C, const JobAction &JA,
    CmdArgs.push_back("-o");
    CmdArgs.push_back(Output.getFilename());
 
@@ -44,7 +44,7 @@ index bd25fd1a8933..b906bff5d5c7 100644
    }
 
    if (!WasmOptPath.empty()) {
-@@ -193,29 +187,27 @@ void wasm::Linker::ConstructJob(Compilation &C, const JobAction &JA,
+@@ -204,29 +198,27 @@ void wasm::Linker::ConstructJob(Compilation &C, const JobAction &JA,
                                           ResponseFileSupport::AtFileCurCP(),
                                           Linker, CmdArgs, Inputs, Output));
 
@@ -95,10 +95,10 @@ index bd25fd1a8933..b906bff5d5c7 100644
  }
 
 diff --git a/compiler-rt/cmake/Modules/AddCompilerRT.cmake b/compiler-rt/cmake/Modules/AddCompilerRT.cmake
-index c3e734f72392..282a321fd70b 100644
+index fb2aee8e42ee..a5de2dca1bc9 100644
 --- a/compiler-rt/cmake/Modules/AddCompilerRT.cmake
 +++ b/compiler-rt/cmake/Modules/AddCompilerRT.cmake
-@@ -382,8 +382,8 @@ function(add_compiler_rt_runtime name type)
+@@ -390,8 +390,8 @@ function(add_compiler_rt_runtime name type)
        target_link_libraries(${libname} PRIVATE ${builtins_${libname}})
      endif()
      if(${type} STREQUAL "SHARED")
@@ -110,10 +110,10 @@ index c3e734f72392..282a321fd70b 100644
        if(WIN32 AND NOT CYGWIN AND NOT MINGW)
          set_target_properties(${libname} PROPERTIES IMPORT_PREFIX "")
 diff --git a/lld/wasm/InputChunks.cpp b/lld/wasm/InputChunks.cpp
-index 0e6c4e691be1..bfa438505d98 100644
+index 181221a77b10..c813e62e7b93 100644
 --- a/lld/wasm/InputChunks.cpp
 +++ b/lld/wasm/InputChunks.cpp
-@@ -361,12 +361,11 @@ uint64_t InputChunk::getVA(uint64_t offset) const {
+@@ -409,12 +409,11 @@ uint64_t InputChunk::getVA(uint64_t offset) const {
  // Generate code to apply relocations to the data section at runtime.
  // This is only called when generating shared libraries (PIC) where address are
  // not known at static link time.
@@ -127,7 +127,7 @@ index 0e6c4e691be1..bfa438505d98 100644
    unsigned opcode_ptr_const = is64 ? WASM_OPCODE_I64_CONST
                                     : WASM_OPCODE_I32_CONST;
    unsigned opcode_ptr_add = is64 ? WASM_OPCODE_I64_ADD
-@@ -385,6 +384,14 @@ bool InputChunk::generateRelocationCode(raw_ostream &os) const {
+@@ -433,6 +432,14 @@ bool InputChunk::generateRelocationCode(raw_ostream &os) const {
      if (!requiresRuntimeReloc)
        continue;
 
@@ -142,7 +142,7 @@ index 0e6c4e691be1..bfa438505d98 100644
      LLVM_DEBUG(dbgs() << "gen reloc: type=" << relocTypeToString(rel.Type)
                        << " addend=" << rel.Addend << " index=" << rel.Index
                        << " output offset=" << offset << "\n");
-@@ -439,9 +446,7 @@ bool InputChunk::generateRelocationCode(raw_ostream &os) const {
+@@ -487,9 +494,7 @@ bool InputChunk::generateRelocationCode(raw_ostream &os) const {
      writeU8(os, opcode_reloc_store, "I32_STORE");
      writeUleb128(os, 2, "align");
      writeUleb128(os, 0, "offset");
@@ -153,23 +153,23 @@ index 0e6c4e691be1..bfa438505d98 100644
 
  // Split WASM_SEG_FLAG_STRINGS section. Such a section is a sequence of
 diff --git a/lld/wasm/InputChunks.h b/lld/wasm/InputChunks.h
-index f545449e1246..d231382a5f5e 100644
+index 1fe78d76631f..2c721f5f92e8 100644
 --- a/lld/wasm/InputChunks.h
 +++ b/lld/wasm/InputChunks.h
-@@ -78,7 +78,7 @@ public:
- 
+@@ -79,7 +79,7 @@ public:
    size_t getNumRelocations() const { return relocations.size(); }
+   size_t getNumLiveRelocations() const;
    void writeRelocations(llvm::raw_ostream &os) const;
 -  bool generateRelocationCode(raw_ostream &os) const;
 +  void generateRelocationCode(std::vector<std::string> &funcs) const;
 
    bool isTLS() const { return flags & llvm::wasm::WASM_SEG_FLAG_TLS; }
    bool isRetained() const { return flags & llvm::wasm::WASM_SEG_FLAG_RETAIN; }
 diff --git a/lld/wasm/SyntheticSections.cpp b/lld/wasm/SyntheticSections.cpp
-index 0e2aa57e9048..e5df7d8c0be0 100644
+index e1192706ea91..1b97039b1f1f 100644
 --- a/lld/wasm/SyntheticSections.cpp
 +++ b/lld/wasm/SyntheticSections.cpp
-@@ -299,6 +299,8 @@ void FunctionSection::writeBody() {
+@@ -312,6 +312,8 @@ void FunctionSection::writeBody() {
  void FunctionSection::addFunction(InputFunction *func) {
    if (!func->live)
      return;
@@ -179,10 +179,10 @@ index 0e2aa57e9048..e5df7d8c0be0 100644
        out.importSec->getNumImportedFunctions() + inputFunctions.size();
    inputFunctions.emplace_back(func);
 diff --git a/lld/wasm/Writer.cpp b/lld/wasm/Writer.cpp
-index 2bf4b370a7db..19fca2616c7a 100644
+index b704677d36c9..ac9eab2b9dd5 100644
 --- a/lld/wasm/Writer.cpp
 +++ b/lld/wasm/Writer.cpp
-@@ -1452,20 +1452,21 @@ void Writer::createStartFunction() {
+@@ -1455,20 +1455,21 @@ void Writer::createStartFunction() {
  void Writer::createApplyDataRelocationsFunction() {
    LLVM_DEBUG(dbgs() << "createApplyDataRelocationsFunction\n");
    // First write the body's contents to a string.
@@ -213,7 +213,7 @@ index 2bf4b370a7db..19fca2616c7a 100644
      writeU8(os, WASM_OPCODE_END, "END");
    }
 
-@@ -1478,24 +1479,67 @@ void Writer::createApplyDataRelocationsFunction() {
+@@ -1481,24 +1482,67 @@ void Writer::createApplyDataRelocationsFunction() {
        make<SyntheticFunction>(nullSignature, "__wasm_apply_data_relocs"));
    def->markLive();
 
@@ -288,10 +288,10 @@ index 2bf4b370a7db..19fca2616c7a 100644
 
  // Similar to createApplyDataRelocationsFunction but generates relocation code
 diff --git a/llvm/cmake/modules/AddLLVM.cmake b/llvm/cmake/modules/AddLLVM.cmake
-index d3e9377c8d2f..50a34184919a 100644
+index 83772ed8d2b1..0a7360d9357a 100644
 --- a/llvm/cmake/modules/AddLLVM.cmake
 +++ b/llvm/cmake/modules/AddLLVM.cmake
-@@ -2524,8 +2524,7 @@ function(llvm_setup_rpath name)
+@@ -2547,8 +2547,7 @@ function(llvm_setup_rpath name)
      # FIXME: update this when there is better solution.
      set(_install_rpath "${LLVM_LIBRARY_OUTPUT_INTDIR}" "${CMAKE_INSTALL_PREFIX}/lib${LLVM_LIBDIR_SUFFIX}" ${extra_libdir})
    elseif(UNIX)
@@ -301,7 +301,7 @@ index d3e9377c8d2f..50a34184919a 100644
      if(${CMAKE_SYSTEM_NAME} MATCHES "(FreeBSD|DragonFly)")
        set_property(TARGET ${name} APPEND_STRING PROPERTY
                     LINK_FLAGS " -Wl,-z,origin ")
-@@ -2539,16 +2538,9 @@ function(llvm_setup_rpath name)
+@@ -2562,16 +2561,9 @@ function(llvm_setup_rpath name)
      return()
    endif()
 
@@ -320,3 +320,22 @@ index d3e9377c8d2f..50a34184919a 100644
    endif()
 
    set_target_properties(${name} PROPERTIES
+diff --git a/llvm/lib/MC/MCParser/WasmAsmParser.cpp b/llvm/lib/MC/MCParser/WasmAsmParser.cpp
+index 1f824b80bcd4..1025d3ee3026 100644
+--- a/llvm/lib/MC/MCParser/WasmAsmParser.cpp
++++ b/llvm/lib/MC/MCParser/WasmAsmParser.cpp
+@@ -228,7 +228,13 @@ public:
+     if (WasmSym->isFunction()) {
+       // Ignore .size directives for function symbols.  They get their size
+       // set automatically based on their content.
+-      Warning(Loc, ".size directive ignored for function symbols");
++      //
++      // Upstream LLVM treats this as a warning, we turn this into an
++      // error since it almost certainly signals severely malformed
++      // assembly due to miscompilation, and data/function symbol kind
++      // confusion is not always caught at link-time and might
++      // manifest as wasm runtime crashes :/
++      Error(Loc, ".size directive ignored for function symbols");
+     } else {
+       getStreamer().emitELFSize(Sym, Expr);
+     }

overlays/wasm.nix

@@ -1,11 +1,11 @@
 final: prev: prev.lib.optionalAttrs prev.stdenv.targetPlatform.isWasm {
-  llvmPackages = final.llvmPackages_20.override {
+  llvmPackages = final.llvmPackages_21.override {
     patchesFn = p: p // { "llvm/gnu-install-dirs.patch" = [{path = ./patches/wasm;}]; };
     monorepoSrc =
       final.stdenv.mkDerivation {
         pname = "llvm-source";
-        version = final.llvmPackages_20.llvm.version + "-haskell";
-        src = final.llvmPackages_20.llvm.monorepoSrc;
+        version = final.llvmPackages_21.llvm.version + "-haskell";
+        src = final.llvmPackages_21.llvm.monorepoSrc;
         patches = ./patches/wasm/llvm/haskell-wasm-llvm-project.patch;
         buildPhase = "true";
         installPhase = ''