rustc_codegen_ssa/back/
metadata.rs

1//! Reading of the rustc metadata for rlibs and dylibs
2
3use std::borrow::Cow;
4use std::fs::File;
5use std::io::Write;
6use std::path::Path;
7
8use itertools::Itertools;
9use object::write::{self, StandardSegment, Symbol, SymbolSection};
10use object::{
11    Architecture, BinaryFormat, Endianness, FileFlags, Object, ObjectSection, ObjectSymbol,
12    SectionFlags, SectionKind, SymbolFlags, SymbolKind, SymbolScope, elf, pe, xcoff,
13};
14use rustc_abi::Endian;
15use rustc_data_structures::memmap::Mmap;
16use rustc_data_structures::owned_slice::{OwnedSlice, try_slice_owned};
17use rustc_metadata::EncodedMetadata;
18use rustc_metadata::creader::MetadataLoader;
19use rustc_metadata::fs::METADATA_FILENAME;
20use rustc_middle::bug;
21use rustc_session::Session;
22use rustc_span::sym;
23use rustc_target::spec::{Abi, Os, RelocModel, Target, ef_avr_arch};
24use tracing::debug;
25
26use super::apple;
27
28/// The default metadata loader. This is used by cg_llvm and cg_clif.
29///
30/// # Metadata location
31///
32/// <dl>
33/// <dt>rlib</dt>
34/// <dd>The metadata can be found in the `lib.rmeta` file inside of the ar archive.</dd>
35/// <dt>dylib</dt>
36/// <dd>The metadata can be found in the `.rustc` section of the shared library.</dd>
37/// </dl>
38#[derive(Debug)]
39pub(crate) struct DefaultMetadataLoader;
40
41static AIX_METADATA_SYMBOL_NAME: &'static str = "__aix_rust_metadata";
42
43fn load_metadata_with(
44    path: &Path,
45    f: impl for<'a> FnOnce(&'a [u8]) -> Result<&'a [u8], String>,
46) -> Result<OwnedSlice, String> {
47    let file =
48        File::open(path).map_err(|e| format!("failed to open file '{}': {}", path.display(), e))?;
49
50    unsafe { Mmap::map(file) }
51        .map_err(|e| format!("failed to mmap file '{}': {}", path.display(), e))
52        .and_then(|mmap| try_slice_owned(mmap, |mmap| f(mmap)))
53}
54
55impl MetadataLoader for DefaultMetadataLoader {
56    fn get_rlib_metadata(&self, target: &Target, path: &Path) -> Result<OwnedSlice, String> {
57        debug!("getting rlib metadata for {}", path.display());
58        load_metadata_with(path, |data| {
59            let archive = object::read::archive::ArchiveFile::parse(&*data)
60                .map_err(|e| format!("failed to parse rlib '{}': {}", path.display(), e))?;
61
62            for entry_result in archive.members() {
63                let entry = entry_result
64                    .map_err(|e| format!("failed to parse rlib '{}': {}", path.display(), e))?;
65                if entry.name() == METADATA_FILENAME.as_bytes() {
66                    let data = entry
67                        .data(data)
68                        .map_err(|e| format!("failed to parse rlib '{}': {}", path.display(), e))?;
69                    if target.is_like_aix {
70                        return get_metadata_xcoff(path, data);
71                    } else {
72                        return search_for_section(path, data, ".rmeta");
73                    }
74                }
75            }
76
77            Err(format!("metadata not found in rlib '{}'", path.display()))
78        })
79    }
80
81    fn get_dylib_metadata(&self, target: &Target, path: &Path) -> Result<OwnedSlice, String> {
82        debug!("getting dylib metadata for {}", path.display());
83        if target.is_like_aix {
84            load_metadata_with(path, |data| {
85                let archive = object::read::archive::ArchiveFile::parse(&*data).map_err(|e| {
86                    format!("failed to parse aix dylib '{}': {}", path.display(), e)
87                })?;
88
89                // FIXME: rewrite in terms of `#![feature(exact_length_collection)]`. See: #149266
90                match Itertools::exactly_one(archive.members()) {
91                    Ok(lib) => {
92                        let lib = lib.map_err(|e| {
93                            format!("failed to parse aix dylib '{}': {}", path.display(), e)
94                        })?;
95                        let data = lib.data(data).map_err(|e| {
96                            format!("failed to parse aix dylib '{}': {}", path.display(), e)
97                        })?;
98                        get_metadata_xcoff(path, data)
99                    }
100                    Err(e) => Err(format!("failed to parse aix dylib '{}': {}", path.display(), e)),
101                }
102            })
103        } else {
104            load_metadata_with(path, |data| search_for_section(path, data, ".rustc"))
105        }
106    }
107}
108
109pub(super) fn search_for_section<'a>(
110    path: &Path,
111    bytes: &'a [u8],
112    section: &str,
113) -> Result<&'a [u8], String> {
114    let Ok(file) = object::File::parse(bytes) else {
115        // The parse above could fail for odd reasons like corruption, but for
116        // now we just interpret it as this target doesn't support metadata
117        // emission in object files so the entire byte slice itself is probably
118        // a metadata file. Ideally though if necessary we could at least check
119        // the prefix of bytes to see if it's an actual metadata object and if
120        // not forward the error along here.
121        return Ok(bytes);
122    };
123    file.section_by_name(section)
124        .ok_or_else(|| format!("no `{}` section in '{}'", section, path.display()))?
125        .data()
126        .map_err(|e| format!("failed to read {} section in '{}': {}", section, path.display(), e))
127}
128
129fn add_gnu_property_note(
130    file: &mut write::Object<'static>,
131    architecture: Architecture,
132    binary_format: BinaryFormat,
133    endianness: Endianness,
134) {
135    // check bti protection
136    if binary_format != BinaryFormat::Elf
137        || !matches!(architecture, Architecture::X86_64 | Architecture::Aarch64)
138    {
139        return;
140    }
141
142    let section = file.add_section(
143        file.segment_name(StandardSegment::Data).to_vec(),
144        b".note.gnu.property".to_vec(),
145        SectionKind::Note,
146    );
147    let mut data: Vec<u8> = Vec::new();
148    let n_namsz: u32 = 4; // Size of the n_name field
149    let n_descsz: u32 = 16; // Size of the n_desc field
150    let n_type: u32 = object::elf::NT_GNU_PROPERTY_TYPE_0; // Type of note descriptor
151    let header_values = [n_namsz, n_descsz, n_type];
152    header_values.iter().for_each(|v| {
153        data.extend_from_slice(&match endianness {
154            Endianness::Little => v.to_le_bytes(),
155            Endianness::Big => v.to_be_bytes(),
156        })
157    });
158    data.extend_from_slice(b"GNU\0"); // Owner of the program property note
159    let pr_type: u32 = match architecture {
160        Architecture::X86_64 => object::elf::GNU_PROPERTY_X86_FEATURE_1_AND,
161        Architecture::Aarch64 => object::elf::GNU_PROPERTY_AARCH64_FEATURE_1_AND,
162        _ => unreachable!(),
163    };
164    let pr_datasz: u32 = 4; //size of the pr_data field
165    let pr_data: u32 = 3; //program property descriptor
166    let pr_padding: u32 = 0;
167    let property_values = [pr_type, pr_datasz, pr_data, pr_padding];
168    property_values.iter().for_each(|v| {
169        data.extend_from_slice(&match endianness {
170            Endianness::Little => v.to_le_bytes(),
171            Endianness::Big => v.to_be_bytes(),
172        })
173    });
174    file.append_section_data(section, &data, 8);
175}
176
177pub(super) fn get_metadata_xcoff<'a>(path: &Path, data: &'a [u8]) -> Result<&'a [u8], String> {
178    let Ok(file) = object::File::parse(data) else {
179        return Ok(data);
180    };
181    let info_data = search_for_section(path, data, ".info")?;
182    if let Some(metadata_symbol) =
183        file.symbols().find(|sym| sym.name() == Ok(AIX_METADATA_SYMBOL_NAME))
184    {
185        let offset = metadata_symbol.address() as usize;
186        // The offset specifies the location of rustc metadata in the .info section of XCOFF.
187        // Each string stored in .info section of XCOFF is preceded by a 4-byte length field.
188        if offset < 4 {
189            return Err(format!("Invalid metadata symbol offset: {offset}"));
190        }
191        // XCOFF format uses big-endian byte order.
192        let len = u32::from_be_bytes(info_data[(offset - 4)..offset].try_into().unwrap()) as usize;
193        if offset + len > (info_data.len() as usize) {
194            return Err(format!(
195                "Metadata at offset {offset} with size {len} is beyond .info section"
196            ));
197        }
198        Ok(&info_data[offset..(offset + len)])
199    } else {
200        Err(format!("Unable to find symbol {AIX_METADATA_SYMBOL_NAME}"))
201    }
202}
203
204pub(crate) fn create_object_file(sess: &Session) -> Option<write::Object<'static>> {
205    let endianness = match sess.target.options.endian {
206        Endian::Little => Endianness::Little,
207        Endian::Big => Endianness::Big,
208    };
209    let Some((architecture, sub_architecture)) =
210        sess.target.object_architecture(&sess.unstable_target_features)
211    else {
212        return None;
213    };
214    let binary_format = sess.target.binary_format.to_object();
215
216    let mut file = write::Object::new(binary_format, architecture, endianness);
217    file.set_sub_architecture(sub_architecture);
218    if sess.target.is_like_darwin {
219        if macho_is_arm64e(&sess.target) {
220            file.set_macho_cpu_subtype(object::macho::CPU_SUBTYPE_ARM64E);
221        }
222
223        file.set_macho_build_version(macho_object_build_version_for_target(sess))
224    }
225    if binary_format == BinaryFormat::Coff {
226        // Disable the default mangler to avoid mangling the special "@feat.00" symbol name.
227        let original_mangling = file.mangling();
228        file.set_mangling(object::write::Mangling::None);
229
230        let mut feature = 0;
231
232        if file.architecture() == object::Architecture::I386 {
233            // When linking with /SAFESEH on x86, lld requires that all linker inputs be marked as
234            // safe exception handling compatible. Metadata files masquerade as regular COFF
235            // objects and are treated as linker inputs, despite containing no actual code. Thus,
236            // they still need to be marked as safe exception handling compatible. See #96498.
237            // Reference: https://docs.microsoft.com/en-us/windows/win32/debug/pe-format
238            feature |= 1;
239        }
240
241        file.add_symbol(object::write::Symbol {
242            name: "@feat.00".into(),
243            value: feature,
244            size: 0,
245            kind: object::SymbolKind::Data,
246            scope: object::SymbolScope::Compilation,
247            weak: false,
248            section: object::write::SymbolSection::Absolute,
249            flags: object::SymbolFlags::None,
250        });
251
252        file.set_mangling(original_mangling);
253    }
254    let e_flags = elf_e_flags(architecture, sess);
255    // adapted from LLVM's `MCELFObjectTargetWriter::getOSABI`
256    let os_abi = elf_os_abi(sess);
257    let abi_version = 0;
258    add_gnu_property_note(&mut file, architecture, binary_format, endianness);
259    file.flags = FileFlags::Elf { os_abi, abi_version, e_flags };
260    Some(file)
261}
262
263pub(super) fn elf_os_abi(sess: &Session) -> u8 {
264    match sess.target.options.os {
265        Os::Hermit => elf::ELFOSABI_STANDALONE,
266        Os::FreeBsd => elf::ELFOSABI_FREEBSD,
267        Os::Solaris => elf::ELFOSABI_SOLARIS,
268        _ => elf::ELFOSABI_NONE,
269    }
270}
271
272pub(super) fn elf_e_flags(architecture: Architecture, sess: &Session) -> u32 {
273    match architecture {
274        Architecture::Mips | Architecture::Mips64 | Architecture::Mips64_N32 => {
275            // "N32" indicates an "ILP32" data model on a 64-bit MIPS CPU
276            // like SPARC's "v8+", x86_64's "x32", or the watchOS "arm64_32".
277            let is_32bit = architecture == Architecture::Mips;
278            let mut e_flags = match sess.target.options.cpu.as_ref() {
279                "mips1" if is_32bit => elf::EF_MIPS_ARCH_1,
280                "mips2" if is_32bit => elf::EF_MIPS_ARCH_2,
281                "mips3" => elf::EF_MIPS_ARCH_3,
282                "mips4" => elf::EF_MIPS_ARCH_4,
283                "mips5" => elf::EF_MIPS_ARCH_5,
284                "mips32r2" if is_32bit => elf::EF_MIPS_ARCH_32R2,
285                "mips32r6" if is_32bit => elf::EF_MIPS_ARCH_32R6,
286                "mips64r2" if !is_32bit => elf::EF_MIPS_ARCH_64R2,
287                "mips64r6" if !is_32bit => elf::EF_MIPS_ARCH_64R6,
288                s if s.starts_with("mips32") && !is_32bit => {
289                    sess.dcx().fatal(format!("invalid CPU `{}` for 64-bit MIPS target", s))
290                }
291                s if s.starts_with("mips64") && is_32bit => {
292                    sess.dcx().fatal(format!("invalid CPU `{}` for 32-bit MIPS target", s))
293                }
294                _ if is_32bit => elf::EF_MIPS_ARCH_32R2,
295                _ => elf::EF_MIPS_ARCH_64R2,
296            };
297
298            // If the ABI is explicitly given, use it, or default to O32 on 32-bit MIPS,
299            // which is the only "true" 32-bit option that LLVM supports.
300            match sess.target.options.llvm_abiname.as_ref() {
301                "o32" if is_32bit => e_flags |= elf::EF_MIPS_ABI_O32,
302                "n32" if !is_32bit => e_flags |= elf::EF_MIPS_ABI2,
303                "n64" if !is_32bit => {}
304                "" if is_32bit => e_flags |= elf::EF_MIPS_ABI_O32,
305                "" => sess.dcx().fatal("LLVM ABI must be specified for 64-bit MIPS targets"),
306                s if is_32bit => {
307                    sess.dcx().fatal(format!("invalid LLVM ABI `{}` for 32-bit MIPS target", s))
308                }
309                s => sess.dcx().fatal(format!("invalid LLVM ABI `{}` for 64-bit MIPS target", s)),
310            };
311
312            if sess.target.options.relocation_model != RelocModel::Static {
313                // PIC means position-independent code. CPIC means "calls PIC".
314                // CPIC was mutually exclusive with PIC according to
315                // the SVR4 MIPS ABI https://refspecs.linuxfoundation.org/elf/mipsabi.pdf
316                // and should have only appeared on static objects with dynamically calls.
317                // At some point someone (GCC?) decided to set CPIC even for PIC.
318                // Nowadays various things expect both set on the same object file
319                // and may even error if you mix CPIC and non-CPIC object files,
320                // despite that being the entire point of the CPIC ABI extension!
321                // As we are in Rome, we do as the Romans do.
322                e_flags |= elf::EF_MIPS_PIC | elf::EF_MIPS_CPIC;
323            }
324            if sess.target.options.cpu.contains("r6") {
325                e_flags |= elf::EF_MIPS_NAN2008;
326            }
327            e_flags
328        }
329        Architecture::Riscv32 | Architecture::Riscv64 => {
330            // Source: https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/079772828bd10933d34121117a222b4cc0ee2200/riscv-elf.adoc
331            let mut e_flags: u32 = 0x0;
332
333            // Check if compression is enabled
334            // `unstable_target_features` is used here because "zca" is gated behind riscv_target_feature.
335            if sess.unstable_target_features.contains(&sym::zca) {
336                e_flags |= elf::EF_RISCV_RVC;
337            }
338
339            // Check if RVTSO is enabled
340            // `unstable_target_features` is used here because "ztso" is gated behind riscv_target_feature.
341            if sess.unstable_target_features.contains(&sym::ztso) {
342                e_flags |= elf::EF_RISCV_TSO;
343            }
344
345            // Set the appropriate flag based on ABI
346            // This needs to match LLVM `RISCVELFStreamer.cpp`
347            match &*sess.target.llvm_abiname {
348                "ilp32" | "lp64" => (),
349                "ilp32f" | "lp64f" => e_flags |= elf::EF_RISCV_FLOAT_ABI_SINGLE,
350                "ilp32d" | "lp64d" => e_flags |= elf::EF_RISCV_FLOAT_ABI_DOUBLE,
351                // Note that the `lp64e` is still unstable as it's not (yet) part of the ELF psABI.
352                "ilp32e" | "lp64e" => e_flags |= elf::EF_RISCV_RVE,
353                _ => bug!("unknown RISC-V ABI name"),
354            }
355
356            e_flags
357        }
358        Architecture::LoongArch32 | Architecture::LoongArch64 => {
359            // Source: https://github.com/loongson/la-abi-specs/blob/release/laelf.adoc#e_flags-identifies-abi-type-and-version
360            let mut e_flags: u32 = elf::EF_LARCH_OBJABI_V1;
361
362            // Set the appropriate flag based on ABI
363            // This needs to match LLVM `LoongArchELFStreamer.cpp`
364            match &*sess.target.llvm_abiname {
365                "ilp32s" | "lp64s" => e_flags |= elf::EF_LARCH_ABI_SOFT_FLOAT,
366                "ilp32f" | "lp64f" => e_flags |= elf::EF_LARCH_ABI_SINGLE_FLOAT,
367                "ilp32d" | "lp64d" => e_flags |= elf::EF_LARCH_ABI_DOUBLE_FLOAT,
368                _ => bug!("unknown LoongArch ABI name"),
369            }
370
371            e_flags
372        }
373        Architecture::Avr => {
374            // Resolve the ISA revision and set
375            // the appropriate EF_AVR_ARCH flag.
376            if let Some(ref cpu) = sess.opts.cg.target_cpu {
377                ef_avr_arch(cpu)
378            } else {
379                bug!("AVR CPU not explicitly specified")
380            }
381        }
382        Architecture::Csky => {
383            if matches!(sess.target.options.abi, Abi::AbiV2) {
384                elf::EF_CSKY_ABIV2
385            } else {
386                elf::EF_CSKY_ABIV1
387            }
388        }
389        Architecture::PowerPc64 => {
390            const EF_PPC64_ABI_UNKNOWN: u32 = 0;
391            const EF_PPC64_ABI_ELF_V1: u32 = 1;
392            const EF_PPC64_ABI_ELF_V2: u32 = 2;
393
394            match sess.target.options.llvm_abiname.as_ref() {
395                // If the flags do not correctly indicate the ABI,
396                // linkers such as ld.lld assume that the ppc64 object files are always ELFv2
397                // which leads to broken binaries if ELFv1 is used for the object files.
398                "elfv1" => EF_PPC64_ABI_ELF_V1,
399                "elfv2" => EF_PPC64_ABI_ELF_V2,
400                "" if sess.target.options.binary_format.to_object() == BinaryFormat::Elf => {
401                    bug!("No ABI specified for this PPC64 ELF target");
402                }
403                // Fall back
404                _ => EF_PPC64_ABI_UNKNOWN,
405            }
406        }
407        _ => 0,
408    }
409}
410
411/// Mach-O files contain information about:
412/// - The platform/OS they were built for (macOS/watchOS/Mac Catalyst/iOS simulator etc).
413/// - The minimum OS version / deployment target.
414/// - The version of the SDK they were targetting.
415///
416/// In the past, this was accomplished using the LC_VERSION_MIN_MACOSX, LC_VERSION_MIN_IPHONEOS,
417/// LC_VERSION_MIN_TVOS or LC_VERSION_MIN_WATCHOS load commands, which each contain information
418/// about the deployment target and SDK version, and implicitly, by their presence, which OS they
419/// target. Simulator targets were determined if the architecture was x86_64, but there was e.g. a
420/// LC_VERSION_MIN_IPHONEOS present.
421///
422/// This is of course brittle and limited, so modern tooling emit the LC_BUILD_VERSION load
423/// command (which contains all three pieces of information in one) when the deployment target is
424/// high enough, or the target is something that wouldn't be encodable with the old load commands
425/// (such as Mac Catalyst, or Aarch64 iOS simulator).
426///
427/// Since Xcode 15, Apple's LD apparently requires object files to use this load command, so this
428/// returns the `MachOBuildVersion` for the target to do so.
429fn macho_object_build_version_for_target(sess: &Session) -> object::write::MachOBuildVersion {
430    /// The `object` crate demands "X.Y.Z encoded in nibbles as xxxx.yy.zz"
431    /// e.g. minOS 14.0 = 0x000E0000, or SDK 16.2 = 0x00100200
432    fn pack_version(apple::OSVersion { major, minor, patch }: apple::OSVersion) -> u32 {
433        let (major, minor, patch) = (major as u32, minor as u32, patch as u32);
434        (major << 16) | (minor << 8) | patch
435    }
436
437    let platform = apple::macho_platform(&sess.target);
438    let min_os = sess.apple_deployment_target();
439
440    let mut build_version = object::write::MachOBuildVersion::default();
441    build_version.platform = platform;
442    build_version.minos = pack_version(min_os);
443    // The version here does not _really_ matter, since it is only used at runtime, and we specify
444    // it when linking the final binary, so we will omit the version. This is also what LLVM does,
445    // and the tooling also allows this (and shows the SDK version as `n/a`). Finally, it is the
446    // semantically correct choice, as the SDK has not influenced the binary generated by rustc at
447    // this point in time.
448    build_version.sdk = 0;
449
450    build_version
451}
452
453/// Is Apple's CPU subtype `arm64e`s
454fn macho_is_arm64e(target: &Target) -> bool {
455    target.llvm_target.starts_with("arm64e")
456}
457
458pub(crate) enum MetadataPosition {
459    First,
460    Last,
461}
462
463/// For rlibs we "pack" rustc metadata into a dummy object file.
464///
465/// Historically it was needed because rustc linked rlibs as whole-archive in some cases.
466/// In that case linkers try to include all files located in an archive, so if metadata is stored
467/// in an archive then it needs to be of a form that the linker is able to process.
468/// Now it's not clear whether metadata still needs to be wrapped into an object file or not.
469///
470/// Note, though, that we don't actually want this metadata to show up in any
471/// final output of the compiler. Instead this is purely for rustc's own
472/// metadata tracking purposes.
473///
474/// With the above in mind, each "flavor" of object format gets special
475/// handling here depending on the target:
476///
477/// * MachO - macos-like targets will insert the metadata into a section that
478///   is sort of fake dwarf debug info. Inspecting the source of the macos
479///   linker this causes these sections to be skipped automatically because
480///   it's not in an allowlist of otherwise well known dwarf section names to
481///   go into the final artifact.
482///
483/// * WebAssembly - this uses wasm files themselves as the object file format
484///   so an empty file with no linking metadata but a single custom section is
485///   created holding our metadata.
486///
487/// * COFF - Windows-like targets create an object with a section that has
488///   the `IMAGE_SCN_LNK_REMOVE` flag set which ensures that if the linker
489///   ever sees the section it doesn't process it and it's removed.
490///
491/// * ELF - All other targets are similar to Windows in that there's a
492///   `SHF_EXCLUDE` flag we can set on sections in an object file to get
493///   automatically removed from the final output.
494pub(crate) fn create_wrapper_file(
495    sess: &Session,
496    section_name: String,
497    data: &[u8],
498) -> (Vec<u8>, MetadataPosition) {
499    let Some(mut file) = create_object_file(sess) else {
500        if sess.target.is_like_wasm {
501            return (
502                create_metadata_file_for_wasm(sess, data, &section_name),
503                MetadataPosition::First,
504            );
505        }
506
507        // Targets using this branch don't have support implemented here yet or
508        // they're not yet implemented in the `object` crate and will likely
509        // fill out this module over time.
510        return (data.to_vec(), MetadataPosition::Last);
511    };
512    let section = if file.format() == BinaryFormat::Xcoff {
513        file.add_section(Vec::new(), b".info".to_vec(), SectionKind::Debug)
514    } else {
515        file.add_section(
516            file.segment_name(StandardSegment::Debug).to_vec(),
517            section_name.into_bytes(),
518            SectionKind::Debug,
519        )
520    };
521    match file.format() {
522        BinaryFormat::Coff => {
523            file.section_mut(section).flags =
524                SectionFlags::Coff { characteristics: pe::IMAGE_SCN_LNK_REMOVE };
525        }
526        BinaryFormat::Elf => {
527            file.section_mut(section).flags =
528                SectionFlags::Elf { sh_flags: elf::SHF_EXCLUDE as u64 };
529        }
530        BinaryFormat::Xcoff => {
531            // AIX system linker may aborts if it meets a valid XCOFF file in archive with no .text, no .data and no .bss.
532            file.add_section(Vec::new(), b".text".to_vec(), SectionKind::Text);
533            file.section_mut(section).flags =
534                SectionFlags::Xcoff { s_flags: xcoff::STYP_INFO as u32 };
535            // Encode string stored in .info section of XCOFF.
536            // FIXME: The length of data here is not guaranteed to fit in a u32.
537            // We may have to split the data into multiple pieces in order to
538            // store in .info section.
539            let len: u32 = data.len().try_into().unwrap();
540            let offset = file.append_section_data(section, &len.to_be_bytes(), 1);
541            // Add a symbol referring to the data in .info section.
542            file.add_symbol(Symbol {
543                name: AIX_METADATA_SYMBOL_NAME.into(),
544                value: offset + 4,
545                size: 0,
546                kind: SymbolKind::Unknown,
547                scope: SymbolScope::Compilation,
548                weak: false,
549                section: SymbolSection::Section(section),
550                flags: SymbolFlags::Xcoff {
551                    n_sclass: xcoff::C_INFO,
552                    x_smtyp: xcoff::C_HIDEXT,
553                    x_smclas: xcoff::C_HIDEXT,
554                    containing_csect: None,
555                },
556            });
557        }
558        _ => {}
559    };
560    file.append_section_data(section, data, 1);
561    (file.write().unwrap(), MetadataPosition::First)
562}
563
564// Historical note:
565//
566// When using link.exe it was seen that the section name `.note.rustc`
567// was getting shortened to `.note.ru`, and according to the PE and COFF
568// specification:
569//
570// > Executable images do not use a string table and do not support
571// > section names longer than 8 characters
572//
573// https://docs.microsoft.com/en-us/windows/win32/debug/pe-format
574//
575// As a result, we choose a slightly shorter name! As to why
576// `.note.rustc` works on MinGW, see
577// https://github.com/llvm/llvm-project/blob/llvmorg-12.0.0/lld/COFF/Writer.cpp#L1190-L1197
578pub fn create_compressed_metadata_file(
579    sess: &Session,
580    metadata: &EncodedMetadata,
581    symbol_name: &str,
582) -> Vec<u8> {
583    let mut packed_metadata = rustc_metadata::METADATA_HEADER.to_vec();
584    packed_metadata.write_all(&(metadata.stub_or_full().len() as u64).to_le_bytes()).unwrap();
585    packed_metadata.extend(metadata.stub_or_full());
586
587    let Some(mut file) = create_object_file(sess) else {
588        if sess.target.is_like_wasm {
589            return create_metadata_file_for_wasm(sess, &packed_metadata, ".rustc");
590        }
591        return packed_metadata.to_vec();
592    };
593    if file.format() == BinaryFormat::Xcoff {
594        return create_compressed_metadata_file_for_xcoff(file, &packed_metadata, symbol_name);
595    }
596    let section = file.add_section(
597        file.segment_name(StandardSegment::Data).to_vec(),
598        b".rustc".to_vec(),
599        SectionKind::ReadOnlyData,
600    );
601    match file.format() {
602        BinaryFormat::Elf => {
603            // Explicitly set no flags to avoid SHF_ALLOC default for data section.
604            file.section_mut(section).flags = SectionFlags::Elf { sh_flags: 0 };
605        }
606        _ => {}
607    };
608    let offset = file.append_section_data(section, &packed_metadata, 1);
609
610    // For MachO and probably PE this is necessary to prevent the linker from throwing away the
611    // .rustc section. For ELF this isn't necessary, but it also doesn't harm.
612    file.add_symbol(Symbol {
613        name: symbol_name.as_bytes().to_vec(),
614        value: offset,
615        size: packed_metadata.len() as u64,
616        kind: SymbolKind::Data,
617        scope: SymbolScope::Dynamic,
618        weak: false,
619        section: SymbolSection::Section(section),
620        flags: SymbolFlags::None,
621    });
622
623    file.write().unwrap()
624}
625
626/// * Xcoff - On AIX, custom sections are merged into predefined sections,
627///   so custom .rustc section is not preserved during linking.
628///   For this reason, we store metadata in predefined .info section, and
629///   define a symbol to reference the metadata. To preserve metadata during
630///   linking on AIX, we have to
631///   1. Create an empty .text section, a empty .data section.
632///   2. Define an empty symbol named `symbol_name` inside .data section.
633///   3. Define an symbol named `AIX_METADATA_SYMBOL_NAME` referencing
634///      data inside .info section.
635///   From XCOFF's view, (2) creates a csect entry in the symbol table, the
636///   symbol created by (3) is a info symbol for the preceding csect. Thus
637///   two symbols are preserved during linking and we can use the second symbol
638///   to reference the metadata.
639pub fn create_compressed_metadata_file_for_xcoff(
640    mut file: write::Object<'_>,
641    data: &[u8],
642    symbol_name: &str,
643) -> Vec<u8> {
644    assert!(file.format() == BinaryFormat::Xcoff);
645    // AIX system linker may aborts if it meets a valid XCOFF file in archive with no .text, no .data and no .bss.
646    file.add_section(Vec::new(), b".text".to_vec(), SectionKind::Text);
647    let data_section = file.add_section(Vec::new(), b".data".to_vec(), SectionKind::Data);
648    let section = file.add_section(Vec::new(), b".info".to_vec(), SectionKind::Debug);
649    file.add_file_symbol("lib.rmeta".into());
650    file.section_mut(section).flags = SectionFlags::Xcoff { s_flags: xcoff::STYP_INFO as u32 };
651    // Add a global symbol to data_section.
652    file.add_symbol(Symbol {
653        name: symbol_name.as_bytes().into(),
654        value: 0,
655        size: 0,
656        kind: SymbolKind::Data,
657        scope: SymbolScope::Dynamic,
658        weak: true,
659        section: SymbolSection::Section(data_section),
660        flags: SymbolFlags::None,
661    });
662    let len: u32 = data.len().try_into().unwrap();
663    let offset = file.append_section_data(section, &len.to_be_bytes(), 1);
664    // Add a symbol referring to the rustc metadata.
665    file.add_symbol(Symbol {
666        name: AIX_METADATA_SYMBOL_NAME.into(),
667        value: offset + 4, // The metadata is preceded by a 4-byte length field.
668        size: 0,
669        kind: SymbolKind::Unknown,
670        scope: SymbolScope::Dynamic,
671        weak: false,
672        section: SymbolSection::Section(section),
673        flags: SymbolFlags::Xcoff {
674            n_sclass: xcoff::C_INFO,
675            x_smtyp: xcoff::C_HIDEXT,
676            x_smclas: xcoff::C_HIDEXT,
677            containing_csect: None,
678        },
679    });
680    file.append_section_data(section, data, 1);
681    file.write().unwrap()
682}
683
684/// Creates a simple WebAssembly object file, which is itself a wasm module,
685/// that contains a custom section of the name `section_name` with contents
686/// `data`.
687///
688/// NB: the `object` crate does not yet have support for writing the wasm
689/// object file format. In lieu of that the `wasm-encoder` crate is used to
690/// build a wasm file by hand.
691///
692/// The wasm object file format is defined at
693/// <https://github.com/WebAssembly/tool-conventions/blob/main/Linking.md>
694/// and mainly consists of a `linking` custom section. In this case the custom
695/// section there is empty except for a version marker indicating what format
696/// it's in.
697///
698/// The main purpose of this is to contain a custom section with `section_name`,
699/// which is then appended after `linking`.
700///
701/// As a further detail the object needs to have a 64-bit memory if `wasm64` is
702/// the target or otherwise it's interpreted as a 32-bit object which is
703/// incompatible with 64-bit ones.
704pub fn create_metadata_file_for_wasm(sess: &Session, data: &[u8], section_name: &str) -> Vec<u8> {
705    assert!(sess.target.is_like_wasm);
706    let mut module = wasm_encoder::Module::new();
707    let mut imports = wasm_encoder::ImportSection::new();
708
709    if sess.target.pointer_width == 64 {
710        imports.import(
711            "env",
712            "__linear_memory",
713            wasm_encoder::MemoryType {
714                minimum: 0,
715                maximum: None,
716                memory64: true,
717                shared: false,
718                page_size_log2: None,
719            },
720        );
721    }
722
723    if imports.len() > 0 {
724        module.section(&imports);
725    }
726    module.section(&wasm_encoder::CustomSection {
727        name: "linking".into(),
728        data: Cow::Borrowed(&[2]),
729    });
730    module.section(&wasm_encoder::CustomSection { name: section_name.into(), data: data.into() });
731    module.finish()
732}
rustc_codegen_ssa/back/metadata.rs

rustc_codegen_ssa/back/
metadata.rs
