Allow serializing all cranelift-module data structures (#6172)

* Remove ModuleCompiledFunction

The same information can be retrieved using

ctx.compiled_code().unwrap().code_info().total_size

In addition for Module implementations that don't immediately compile the
given function there is no correct value that can be returned.

* Don't give anonymous functions and data objects an internal name

This internal name can conflict if a module is serialized and then
deserialized into another module. It also wasn't used by any of the
Module implementations anyway.

* Allow serializing all cranelift-module data structures

This allows a Module implementation to serialize it's internal state and
deserialize it in another compilation session. For example to implement
LTO or to load the module into cranelift-interpreter.

* Use expect
This commit is contained in:
bjorn3
2023-04-21 14:39:15 +02:00
committed by GitHub
parent 8078404a15
commit 91d1d246cd
8 changed files with 359 additions and 100 deletions

View File

@@ -1,17 +1,14 @@
//! Defines `ObjectModule`.
use anyhow::anyhow;
use cranelift_codegen::binemit::{Addend, CodeOffset, Reloc};
use cranelift_codegen::entity::SecondaryMap;
use cranelift_codegen::isa::{OwnedTargetIsa, TargetIsa};
use cranelift_codegen::{self, ir, MachReloc};
use cranelift_codegen::{
binemit::{Addend, CodeOffset, Reloc},
CodegenError,
};
use cranelift_control::ControlPlane;
use cranelift_module::{
DataDescription, DataId, FuncId, Init, Linkage, Module, ModuleCompiledFunction,
ModuleDeclarations, ModuleError, ModuleExtName, ModuleReloc, ModuleResult,
DataDescription, DataId, FuncId, Init, Linkage, Module, ModuleDeclarations, ModuleError,
ModuleExtName, ModuleReloc, ModuleResult,
};
use log::info;
use object::write::{
@@ -21,7 +18,6 @@ use object::{
RelocationEncoding, RelocationKind, SectionKind, SymbolFlags, SymbolKind, SymbolScope,
};
use std::collections::HashMap;
use std::convert::TryInto;
use std::mem;
use target_lexicon::PointerWidth;
@@ -135,8 +131,6 @@ pub struct ObjectModule {
libcall_names: Box<dyn Fn(ir::LibCall) -> String + Send + Sync>,
known_symbols: HashMap<ir::KnownSymbol, SymbolId>,
per_function_section: bool,
anon_func_number: u64,
anon_data_number: u64,
}
impl ObjectModule {
@@ -156,8 +150,6 @@ impl ObjectModule {
libcall_names: builder.libcall_names,
known_symbols: HashMap::new(),
per_function_section: builder.per_function_section,
anon_func_number: 0,
anon_data_number: 0,
}
}
}
@@ -219,16 +211,15 @@ impl Module for ObjectModule {
}
fn declare_anonymous_function(&mut self, signature: &ir::Signature) -> ModuleResult<FuncId> {
// Symbols starting with .L are completely omitted from the symbol table after linking.
// Using hexadecimal instead of decimal for slightly smaller symbol names and often slightly
// faster linking.
let name = format!(".Lfn{:x}", self.anon_func_number);
self.anon_func_number += 1;
let id = self.declarations.declare_anonymous_function(signature)?;
let symbol_id = self.object.add_symbol(Symbol {
name: name.as_bytes().to_vec(),
name: self
.declarations
.get_function_decl(id)
.linkage_name(id)
.into_owned()
.into_bytes(),
value: 0,
size: 0,
kind: SymbolKind::Text,
@@ -287,12 +278,6 @@ impl Module for ObjectModule {
}
fn declare_anonymous_data(&mut self, writable: bool, tls: bool) -> ModuleResult<DataId> {
// Symbols starting with .L are completely omitted from the symbol table after linking.
// Using hexadecimal instead of decimal for slightly smaller symbol names and often slightly
// faster linking.
let name = format!(".Ldata{:x}", self.anon_data_number);
self.anon_data_number += 1;
let id = self.declarations.declare_anonymous_data(writable, tls)?;
let kind = if tls {
@@ -302,7 +287,12 @@ impl Module for ObjectModule {
};
let symbol_id = self.object.add_symbol(Symbol {
name: name.as_bytes().to_vec(),
name: self
.declarations
.get_data_decl(id)
.linkage_name(id)
.into_owned()
.into_bytes(),
value: 0,
size: 0,
kind,
@@ -321,7 +311,7 @@ impl Module for ObjectModule {
func_id: FuncId,
ctx: &mut cranelift_codegen::Context,
ctrl_plane: &mut ControlPlane,
) -> ModuleResult<ModuleCompiledFunction> {
) -> ModuleResult<()> {
info!("defining function {}: {}", func_id, ctx.func.display());
let mut code: Vec<u8> = Vec::new();
@@ -344,21 +334,20 @@ impl Module for ObjectModule {
alignment: u64,
bytes: &[u8],
relocs: &[MachReloc],
) -> ModuleResult<ModuleCompiledFunction> {
) -> ModuleResult<()> {
info!("defining function {} with bytes", func_id);
let total_size: u32 = match bytes.len().try_into() {
Ok(total_size) => total_size,
_ => Err(CodegenError::CodeTooLarge)?,
};
let decl = self.declarations.get_function_decl(func_id);
if !decl.linkage.is_definable() {
return Err(ModuleError::InvalidImportDefinition(decl.name.clone()));
return Err(ModuleError::InvalidImportDefinition(
decl.linkage_name(func_id).into_owned(),
));
}
let &mut (symbol, ref mut defined) = self.functions[func_id].as_mut().unwrap();
if *defined {
return Err(ModuleError::DuplicateDefinition(decl.name.clone()));
return Err(ModuleError::DuplicateDefinition(
decl.linkage_name(func_id).into_owned(),
));
}
*defined = true;
@@ -391,18 +380,22 @@ impl Module for ObjectModule {
});
}
Ok(ModuleCompiledFunction { size: total_size })
Ok(())
}
fn define_data(&mut self, data_id: DataId, data: &DataDescription) -> ModuleResult<()> {
let decl = self.declarations.get_data_decl(data_id);
if !decl.linkage.is_definable() {
return Err(ModuleError::InvalidImportDefinition(decl.name.clone()));
return Err(ModuleError::InvalidImportDefinition(
decl.linkage_name(data_id).into_owned(),
));
}
let &mut (symbol, ref mut defined) = self.data_objects[data_id].as_mut().unwrap();
if *defined {
return Err(ModuleError::DuplicateDefinition(decl.name.clone()));
return Err(ModuleError::DuplicateDefinition(
decl.linkage_name(data_id).into_owned(),
));
}
*defined = true;