Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Implement reference types in compiler-llvm. #2154

Merged
merged 4 commits into from
Mar 1, 2021
+626 −109
Merged

Implement reference types in compiler-llvm. #2154

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
b14afa1
Implement all reference types instructions in compiler-llvm.
nlewycky Mar 1, 2021
7d9dd4d
Correct the type of vmctx on these functions.
nlewycky Mar 1, 2021
3c6168d
Split funcref and externref into separate LLVM types, add anyref_ty.
nlewycky Mar 1, 2021
40f40bb
RefNull may return any reference type.
nlewycky Mar 1, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 8 additions & 7 deletions lib/cli/src/store.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -212,18 +212,19 @@ impl CompilerOptions {
// Converts a kind into a filename, that we will use to dump
// the contents of the IR object file to.
fn types_to_signature(types: &[Type]) -> String {
types.iter().map(|ty| {
match ty {
types
.iter()
.map(|ty| match ty {
Type::I32 => "i".to_string(),
Type::I64 => "I".to_string(),
Type::F32 => "f".to_string(),
Type::F64 => "F".to_string(),
Type::V128 => "v".to_string(),
_ => {
unimplemented!("Function type not yet supported for generated signatures in debugging");
}
}
}).collect::<Vec<_>>().join("")
Type::ExternRef => "e".to_string(),
Type::FuncRef => "r".to_string(),
})
.collect::<Vec<_>>()
.join("")
}
// Converts a kind into a filename, that we will use to dump
// the contents of the IR object file to.
Expand Down
2 changes: 1 addition & 1 deletion lib/compiler-cranelift/src/func_environ.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -575,7 +575,7 @@ impl<'module_environment> FuncEnvironment<'module_environment> {
(
sig,
local_memory_index.index(),
VMBuiltinFunctionIndex::get_local_memory_copy_index(),
VMBuiltinFunctionIndex::get_memory_copy_index(),
)
} else {
(
Expand Down
12 changes: 6 additions & 6 deletions lib/compiler-llvm/src/abi/aarch64_systemv.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,8 +140,7 @@ impl Abi for Aarch64SystemV {
Type::I32 | Type::F32 => 32,
Type::I64 | Type::F64 => 64,
Type::V128 => 128,
Type::ExternRef => unimplemented!("externref in the llvm backend"),
Type::FuncRef => unimplemented!("funcref in the llvm backend"),
Type::ExternRef | Type::FuncRef => 64, /* pointer */
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We may be able to use something in VMOffsets for this to help prevent breakage. I don't think it'll change any time soon, but something to think about

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The presence of VMOffsets is optional in this function. The issue is that we don't have a VMContext when producing trampolines, so we don't have a VMOffset for it either.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, it's fine. Really not a big deal because we do this thing in other places in compiler-llvm where we don't use data from VMOffsets so things may get out of sync. I think it's fine to not worry about it for now and we can update compiler-llvm later if we decide that that's a high enough priority thing to do.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Really, a lot of VMOffsets stuff is static. In this case I don't know if we actually need a VMOffsets but, yeah, generally speaking getting the size of things doesn't require anything other than knowing the size of a pointer on the target architecture. Maybe that means it needs to be split into two pieces

})
.collect::<Vec<i32>>();
match sig_returns_bitwidths.as_slice() {
Expand Down Expand Up @@ -285,8 +284,10 @@ impl Abi for Aarch64SystemV {
assert!(value.get_type() == intrinsics.i128_ty.as_basic_type_enum());
value
}
Type::ExternRef => unimplemented!("externref in the llvm backend"),
Type::FuncRef => unimplemented!("funcref in the llvm backend"),
Type::ExternRef | Type::FuncRef => {
assert!(value.get_type() == intrinsics.funcref_ty.as_basic_type_enum());
value
}
}
};

Expand Down Expand Up @@ -322,8 +323,7 @@ impl Abi for Aarch64SystemV {
Type::I32 | Type::F32 => 32,
Type::I64 | Type::F64 => 64,
Type::V128 => 128,
Type::ExternRef => unimplemented!("externref in the llvm backend"),
Type::FuncRef => unimplemented!("funcref in the llvm backend"),
Type::ExternRef | Type::FuncRef => 64, /* pointer */
})
.collect::<Vec<i32>>();

Expand Down
40 changes: 8 additions & 32 deletions lib/compiler-llvm/src/abi/x86_64_systemv.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,8 +92,7 @@ impl Abi for X86_64SystemV {
Type::I32 | Type::F32 => 32,
Type::I64 | Type::F64 => 64,
Type::V128 => 128,
Type::ExternRef => unimplemented!("externref in the llvm backend"),
Type::FuncRef => unimplemented!("funcref in the llvm backend"),
Type::ExternRef | Type::FuncRef => 64, /* pointer */
})
.collect::<Vec<i32>>();

Expand Down Expand Up @@ -316,26 +315,7 @@ impl Abi for X86_64SystemV {
);
builder.build_bitcast(value, intrinsics.f32_ty, "")
}
Type::I64 => {
assert!(
value.get_type() == intrinsics.i64_ty.as_basic_type_enum()
|| value.get_type() == intrinsics.f64_ty.as_basic_type_enum()
);
builder.build_bitcast(value, intrinsics.i64_ty, "")
}
Type::F64 => {
assert!(
value.get_type() == intrinsics.i64_ty.as_basic_type_enum()
|| value.get_type() == intrinsics.f64_ty.as_basic_type_enum()
);
builder.build_bitcast(value, intrinsics.f64_ty, "")
}
Type::V128 => {
assert!(value.get_type() == intrinsics.i128_ty.as_basic_type_enum());
value
}
Type::ExternRef => unimplemented!("externref in the llvm backend"),
Type::FuncRef => unimplemented!("funcref in the llvm backend"),
_ => panic!("should only be called to repack 32-bit values"),
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just some other bug you noticed?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

On x86-64 this function is only called to cast 32-bit values, and should only be called to cast 32-bit values. I think this was copied out of the aarch64 code. I found this because I was auditing for places where ExternRef/FuncRef weren't handled and then investigated.

}
};

Expand Down Expand Up @@ -365,8 +345,7 @@ impl Abi for X86_64SystemV {
Type::I32 | Type::F32 => 32,
Type::I64 | Type::F64 => 64,
Type::V128 => 128,
Type::ExternRef => unimplemented!("externref in the llvm backend"),
Type::FuncRef => unimplemented!("funcref in the llvm backend"),
Type::ExternRef | Type::FuncRef => 64, /* pointer */
})
.collect::<Vec<i32>>();

Expand Down Expand Up @@ -475,15 +454,12 @@ impl Abi for X86_64SystemV {
.results()
.iter()
.map(|ty| match ty {
Type::I32 | Type::F32 => Ok(32),
Type::I64 | Type::F64 => Ok(64),
Type::V128 => Ok(128),
ty => Err(CompileError::Codegen(format!(
"is_sret: unimplemented wasmer_types type {:?}",
ty
))),
Type::I32 | Type::F32 => 32,
Type::I64 | Type::F64 => 64,
Type::V128 => 128,
Type::ExternRef | Type::FuncRef => 64, /* pointer */
})
.collect::<Result<Vec<i32>, _>>()?;
.collect::<Vec<i32>>();

Ok(!matches!(
func_sig_returns_bitwidths.as_slice(),
Expand Down
56 changes: 53 additions & 3 deletions lib/compiler-llvm/src/object_file.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -54,15 +54,65 @@ where
{
// TODO: use perfect hash function?
let mut libcalls = HashMap::new();
libcalls.insert("wasmer_raise_trap".to_string(), LibCall::RaiseTrap);
libcalls.insert("truncf".to_string(), LibCall::TruncF32);
libcalls.insert("trunc".to_string(), LibCall::TruncF64);
libcalls.insert("ceilf".to_string(), LibCall::CeilF32);
libcalls.insert("ceil".to_string(), LibCall::CeilF64);
libcalls.insert("floorf".to_string(), LibCall::FloorF32);
libcalls.insert("floor".to_string(), LibCall::FloorF64);
libcalls.insert("nearbyintf".to_string(), LibCall::NearestF32);
libcalls.insert("nearbyint".to_string(), LibCall::NearestF64);
libcalls.insert("truncf".to_string(), LibCall::TruncF32);
libcalls.insert("trunc".to_string(), LibCall::TruncF64);
libcalls.insert("wasmer_f32_ceil".to_string(), LibCall::CeilF32);
libcalls.insert("wasmer_f64_ceil".to_string(), LibCall::CeilF64);
libcalls.insert("wasmer_f32_floor".to_string(), LibCall::FloorF32);
libcalls.insert("wasmer_f64_floor".to_string(), LibCall::FloorF64);
libcalls.insert("wasmer_f32_nearest".to_string(), LibCall::NearestF32);
libcalls.insert("wasmer_f64_nearest".to_string(), LibCall::NearestF64);
libcalls.insert("wasmer_f32_trunc".to_string(), LibCall::TruncF32);
libcalls.insert("wasmer_f64_trunc".to_string(), LibCall::TruncF64);
Comment on lines +65 to +72
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Are we using these?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nope!

libcalls.insert("wasmer_memory32_size".to_string(), LibCall::Memory32Size);
libcalls.insert(
"wasmer_imported_memory32_size".to_string(),
LibCall::ImportedMemory32Size,
);
libcalls.insert("wasmer_table_copy".to_string(), LibCall::TableCopy);
libcalls.insert("wasmer_table_init".to_string(), LibCall::TableInit);
libcalls.insert("wasmer_table_fill".to_string(), LibCall::TableFill);
libcalls.insert("wasmer_table_size".to_string(), LibCall::TableSize);
libcalls.insert(
"wasmer_imported_table_size".to_string(),
LibCall::ImportedTableSize,
);
libcalls.insert("wasmer_table_get".to_string(), LibCall::TableGet);
libcalls.insert(
"wasmer_imported_table_get".to_string(),
LibCall::ImportedTableGet,
);
libcalls.insert("wasmer_table_set".to_string(), LibCall::TableSet);
libcalls.insert(
"wasmer_imported_table_set".to_string(),
LibCall::ImportedTableSet,
);
libcalls.insert("wasmer_table_grow".to_string(), LibCall::TableGrow);
libcalls.insert(
"wasmer_imported_table_grow".to_string(),
LibCall::ImportedTableGrow,
);
libcalls.insert("wasmer_func_ref".to_string(), LibCall::FuncRef);
libcalls.insert("wasmer_elem_drop".to_string(), LibCall::ElemDrop);
libcalls.insert("wasmer_memory32_copy".to_string(), LibCall::Memory32Copy);
libcalls.insert(
"wasmer_imported_memory32_copy".to_string(),
LibCall::ImportedMemory32Copy,
);
libcalls.insert("wasmer_memory32_fill".to_string(), LibCall::Memory32Fill);
libcalls.insert(
"wasmer_imported_memory32_fill".to_string(),
LibCall::ImportedMemory32Fill,
);
libcalls.insert("wasmer_memory32_init".to_string(), LibCall::Memory32Init);
libcalls.insert("wasmer_data_drop".to_string(), LibCall::DataDrop);
libcalls.insert("wasmer_raise_trap".to_string(), LibCall::RaiseTrap);
libcalls.insert("wasmer_probestack".to_string(), LibCall::Probestack);

let elf = goblin::elf::Elf::parse(&contents).map_err(map_goblin_err)?;
Expand Down
Loading