Rollup merge of #121464 - alexcrichton:fix-wasm64, r=wesleywiser

rustc: Fix wasm64 metadata object files

It looks like LLD will detect object files being either 32 or 64-bit depending on any memory present. LLD will additionally reject 32-bit objects during a 64-bit link. Previously metadata objects did not have any memories in them which led LLD to conclude they were 32-bit objects which broke 64-bit targets for wasm.

This commit fixes this by ensuring that for 64-bit targets there's a memory object present to get LLD to detect it's a 64-bit target. Additionally this commit moves away from a hand-crafted wasm encoder to the `wasm-encoder` crate on crates.io as the complexity grows for the generated object file.

Closes #121460

Cargo.lock

@@ -3646,6 +3646,7 @@ dependencies = [
  "thin-vec",
  "thorin-dwp",
  "tracing",
+ "wasm-encoder",
  "windows",
 ]
 
@@ -6142,6 +6143,15 @@ version = "0.2.91"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "4f186bd2dcf04330886ce82d6f33dd75a7bfcf69ecf5763b89fcde53b6ac9838"
 
+[[package]]
+name = "wasm-encoder"
+version = "0.200.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b9e3fb0c8fbddd78aa6095b850dfeedbc7506cf5f81e633f69cf8f2333ab84b9"
+dependencies = [
+ "leb128",
+]
+
 [[package]]
 name = "wasmparser"
 version = "0.118.2"

compiler/rustc_codegen_ssa/Cargo.toml

@@ -38,6 +38,7 @@ tempfile = "3.2"
 thin-vec = "0.2.12"
 thorin-dwp = "0.7"
 tracing = "0.1"
+wasm-encoder = "0.200.0"
 # tidy-alphabetical-end
 
 [target.'cfg(unix)'.dependencies]

compiler/rustc_codegen_ssa/src/back/link.rs

@@ -315,8 +315,11 @@ fn link_rlib<'a>(
 
     let trailing_metadata = match flavor {
         RlibFlavor::Normal => {
-            let (metadata, metadata_position) =
-                create_wrapper_file(sess, b".rmeta".to_vec(), codegen_results.metadata.raw_data());
+            let (metadata, metadata_position) = create_wrapper_file(
+                sess,
+                ".rmeta".to_string(),
+                codegen_results.metadata.raw_data(),
+            );
             let metadata = emit_wrapper_file(sess, &metadata, tmpdir, METADATA_FILENAME);
             match metadata_position {
                 MetadataPosition::First => {
@@ -384,7 +387,7 @@ fn link_rlib<'a>(
             let path = find_native_static_library(filename.as_str(), true, &lib_search_paths, sess);
             let src = read(path)
                 .map_err(|e| sess.dcx().emit_fatal(errors::ReadFileError { message: e }))?;
-            let (data, _) = create_wrapper_file(sess, b".bundled_lib".to_vec(), &src);
+            let (data, _) = create_wrapper_file(sess, ".bundled_lib".to_string(), &src);
             let wrapper_file = emit_wrapper_file(sess, &data, tmpdir, filename.as_str());
             packed_bundled_libs.push(wrapper_file);
         } else {

compiler/rustc_codegen_ssa/src/back/metadata.rs

@@ -1,5 +1,6 @@
 //! Reading of the rustc metadata for rlibs and dylibs
 
+use std::borrow::Cow;
 use std::fs::File;
 use std::io::Write;
 use std::path::Path;
@@ -15,7 +16,6 @@ use rustc_data_structures::owned_slice::{try_slice_owned, OwnedSlice};
 use rustc_metadata::creader::MetadataLoader;
 use rustc_metadata::fs::METADATA_FILENAME;
 use rustc_metadata::EncodedMetadata;
-use rustc_serialize::leb128;
 use rustc_session::Session;
 use rustc_span::sym;
 use rustc_target::abi::Endian;
@@ -434,12 +434,15 @@ pub enum MetadataPosition {
 ///   automatically removed from the final output.
 pub fn create_wrapper_file(
     sess: &Session,
-    section_name: Vec<u8>,
+    section_name: String,
     data: &[u8],
 ) -> (Vec<u8>, MetadataPosition) {
     let Some(mut file) = create_object_file(sess) else {
         if sess.target.is_like_wasm {
-            return (create_metadata_file_for_wasm(data, &section_name), MetadataPosition::First);
+            return (
+                create_metadata_file_for_wasm(sess, data, &section_name),
+                MetadataPosition::First,
+            );
         }
 
         // Targets using this branch don't have support implemented here yet or
@@ -452,7 +455,7 @@ pub fn create_wrapper_file(
     } else {
         file.add_section(
             file.segment_name(StandardSegment::Debug).to_vec(),
-            section_name,
+            section_name.into_bytes(),
             SectionKind::Debug,
         )
     };
@@ -524,7 +527,7 @@ pub fn create_compressed_metadata_file(
 
     let Some(mut file) = create_object_file(sess) else {
         if sess.target.is_like_wasm {
-            return create_metadata_file_for_wasm(&packed_metadata, b".rustc");
+            return create_metadata_file_for_wasm(sess, &packed_metadata, ".rustc");
         }
         return packed_metadata.to_vec();
     };
@@ -624,51 +627,41 @@ pub fn create_compressed_metadata_file_for_xcoff(
 /// `data`.
 ///
 /// NB: the `object` crate does not yet have support for writing the wasm
-/// object file format. The format is simple enough that for now an extra crate
-/// from crates.io (such as `wasm-encoder`). The file format is:
+/// object file format. In lieu of that the `wasm-encoder` crate is used to
+/// build a wasm file by hand.
 ///
-/// * 4-byte header "\0asm"
-/// * 4-byte version number - 1u32 in little-endian format
-/// * concatenated sections, which for this object is always "custom sections"
-///
-/// Custom sections are then defined by:
-/// * 1-byte section identifier - 0 for a custom section
-/// * leb-encoded section length (size of the contents beneath this bullet)
-///   * leb-encoded custom section name length
-///   * custom section name
-///   * section contents
-///
-/// One custom section, `linking`, is added here in accordance with
+/// The wasm object file format is defined at
 /// <https://github.com/WebAssembly/tool-conventions/blob/main/Linking.md>
-/// which is required to inform LLD that this is an object file but it should
-/// otherwise basically ignore it if it otherwise looks at it. The linking
-/// section currently is defined by a single version byte (2) and then further
-/// sections, but we have no more sections, so it's just the byte "2".
+/// and mainly consists of a `linking` custom section. In this case the custom
+/// section there is empty except for a version marker indicating what format
+/// it's in.
 ///
-/// The next custom section is the one we're interested in.
-pub fn create_metadata_file_for_wasm(data: &[u8], section_name: &[u8]) -> Vec<u8> {
-    let mut bytes = b"\0asm\x01\0\0\0".to_vec();
+/// The main purpose of this is to contain a custom section with `section_name`,
+/// which is then appended after `linking`.
+///
+/// As a further detail the object needs to have a 64-bit memory if `wasm64` is
+/// the target or otherwise it's interpreted as a 32-bit object which is
+/// incompatible with 64-bit ones.
+pub fn create_metadata_file_for_wasm(sess: &Session, data: &[u8], section_name: &str) -> Vec<u8> {
+    assert!(sess.target.is_like_wasm);
+    let mut module = wasm_encoder::Module::new();
+    let mut imports = wasm_encoder::ImportSection::new();
 
-    let mut append_custom_section = |section_name: &[u8], data: &[u8]| {
-        let mut section_name_len = [0; leb128::max_leb128_len::<usize>()];
-        let off = leb128::write_usize_leb128(&mut section_name_len, section_name.len());
-        let section_name_len = &section_name_len[..off];
-
-        let mut section_len = [0; leb128::max_leb128_len::<usize>()];
-        let off = leb128::write_usize_leb128(
-            &mut section_len,
-            data.len() + section_name_len.len() + section_name.len(),
+    if sess.target.pointer_width == 64 {
+        imports.import(
+            "env",
+            "__linear_memory",
+            wasm_encoder::MemoryType { minimum: 0, maximum: None, memory64: true, shared: false },
         );
-        let section_len = &section_len[..off];
+    }
 
-        bytes.push(0u8);
-        bytes.extend_from_slice(section_len);
-        bytes.extend_from_slice(section_name_len);
-        bytes.extend_from_slice(section_name);
-        bytes.extend_from_slice(data);
-    };
-
-    append_custom_section(b"linking", &[2]);
-    append_custom_section(section_name, data);
-    bytes
+    if imports.len() > 0 {
+        module.section(&imports);
+    }
+    module.section(&wasm_encoder::CustomSection {
+        name: "linking".into(),
+        data: Cow::Borrowed(&[2]),
+    });
+    module.section(&wasm_encoder::CustomSection { name: section_name.into(), data: data.into() });
+    module.finish()
 }

src/tools/tidy/src/deps.rs

@@ -92,6 +92,7 @@ const EXCEPTIONS: ExceptionList = &[
     ("ryu", "Apache-2.0 OR BSL-1.0"), // BSL is not acceptble, but we use it under Apache-2.0                       // cargo/... (because of serde)
     ("self_cell", "Apache-2.0"),                             // rustc (fluent translations)
     ("snap", "BSD-3-Clause"),                                // rustc
+    ("wasm-encoder", "Apache-2.0 WITH LLVM-exception"),      // rustc
     ("wasmparser", "Apache-2.0 WITH LLVM-exception"),        // rustc
     // tidy-alphabetical-end
 ];
@@ -267,6 +268,7 @@ const PERMITTED_RUSTC_DEPENDENCIES: &[&str] = &[
     "jemalloc-sys",
     "jobserver",
     "lazy_static",
+    "leb128",
     "libc",
     "libloading",
     "linux-raw-sys",
@@ -380,6 +382,7 @@ const PERMITTED_RUSTC_DEPENDENCIES: &[&str] = &[
     "valuable",
     "version_check",
     "wasi",
+    "wasm-encoder",
     "wasmparser",
     "winapi",
     "winapi-i686-pc-windows-gnu",