externref: implement stack map-based garbage collection

For host VM code, we use plain reference counting, where cloning increments
the reference count, and dropping decrements it. We can avoid many of the
on-stack increment/decrement operations that typically plague the
performance of reference counting via Rust's ownership and borrowing system.
Moving a `VMExternRef` avoids mutating its reference count, and borrowing it
either avoids the reference count increment or delays it until if/when the
`VMExternRef` is cloned.

When passing a `VMExternRef` into compiled Wasm code, we don't want to do
reference count mutations for every compiled `local.{get,set}`, nor for
every function call. Therefore, we use a variation of **deferred reference
counting**, where we only mutate reference counts when storing
`VMExternRef`s somewhere that outlives the activation: into a global or
table. Simultaneously, we over-approximate the set of `VMExternRef`s that
are inside Wasm function activations. Periodically, we walk the stack at GC
safe points, and use stack map information to precisely identify the set of
`VMExternRef`s inside Wasm activations. Then we take the difference between
this precise set and our over-approximation, and decrement the reference
count for each of the `VMExternRef`s that are in our over-approximation but
not in the precise set. Finally, the over-approximation is replaced with the
precise set.

The `VMExternRefActivationsTable` implements the over-approximized set of
`VMExternRef`s referenced by Wasm activations. Calling a Wasm function and
passing it a `VMExternRef` moves the `VMExternRef` into the table, and the
compiled Wasm function logically "borrows" the `VMExternRef` from the
table. Similarly, `global.get` and `table.get` operations clone the gotten
`VMExternRef` into the `VMExternRefActivationsTable` and then "borrow" the
reference out of the table.

When a `VMExternRef` is returned to host code from a Wasm function, the host
increments the reference count (because the reference is logically
"borrowed" from the `VMExternRefActivationsTable` and the reference count
from the table will be dropped at the next GC).

For more general information on deferred reference counting, see *An
Examination of Deferred Reference Counting and Cycle Detection* by Quinane:
https://openresearch-repository.anu.edu.au/bitstream/1885/42030/2/hon-thesis.pdf

cc #929

Fixes #1804
This commit is contained in:
Nick Fitzgerald
2020-06-03 09:21:34 -07:00
parent 357fb11f46
commit f30ce1fe97
32 changed files with 1415 additions and 235 deletions

View File

@@ -88,7 +88,8 @@
use crate::address_map::{FunctionAddressMap, InstructionAddressMap};
use crate::cache::{ModuleCacheDataTupleType, ModuleCacheEntry};
use crate::compilation::{
Compilation, CompileError, CompiledFunction, Relocation, RelocationTarget, TrapInformation,
Compilation, CompileError, CompiledFunction, Relocation, RelocationTarget, StackMapInformation,
TrapInformation,
};
use crate::func_environ::{get_func_name, FuncEnvironment};
use crate::{CacheConfig, FunctionBodyData, ModuleLocal, ModuleTranslation, Tunables};
@@ -204,6 +205,27 @@ impl binemit::TrapSink for TrapSink {
}
}
#[derive(Default)]
struct StackMapSink {
infos: Vec<StackMapInformation>,
}
impl binemit::StackmapSink for StackMapSink {
fn add_stackmap(&mut self, code_offset: binemit::CodeOffset, stack_map: binemit::Stackmap) {
self.infos.push(StackMapInformation {
code_offset,
stack_map,
});
}
}
impl StackMapSink {
fn finish(mut self) -> Vec<StackMapInformation> {
self.infos.sort_by_key(|info| info.code_offset);
self.infos
}
}
fn get_function_address_map<'data>(
context: &Context,
data: &FunctionBodyData<'data>,
@@ -294,6 +316,7 @@ fn compile(env: CompileEnv<'_>) -> Result<ModuleCacheDataTupleType, CompileError
let mut value_ranges = PrimaryMap::with_capacity(env.function_body_inputs.len());
let mut stack_slots = PrimaryMap::with_capacity(env.function_body_inputs.len());
let mut traps = PrimaryMap::with_capacity(env.function_body_inputs.len());
let mut stack_maps = PrimaryMap::with_capacity(env.function_body_inputs.len());
env.function_body_inputs
.into_iter()
@@ -354,14 +377,14 @@ fn compile(env: CompileEnv<'_>) -> Result<ModuleCacheDataTupleType, CompileError
let mut code_buf: Vec<u8> = Vec::new();
let mut reloc_sink = RelocSink::new(func_index);
let mut trap_sink = TrapSink::new();
let mut stackmap_sink = binemit::NullStackmapSink {};
let mut stack_map_sink = StackMapSink::default();
context
.compile_and_emit(
isa,
&mut code_buf,
&mut reloc_sink,
&mut trap_sink,
&mut stackmap_sink,
&mut stack_map_sink,
)
.map_err(|error| {
CompileError::Codegen(pretty_error(&context.func, Some(isa), error))
@@ -391,6 +414,7 @@ fn compile(env: CompileEnv<'_>) -> Result<ModuleCacheDataTupleType, CompileError
context.func.stack_slots,
trap_sink.traps,
unwind_info,
stack_map_sink.finish(),
))
})
.collect::<Result<Vec<_>, CompileError>>()?
@@ -405,6 +429,7 @@ fn compile(env: CompileEnv<'_>) -> Result<ModuleCacheDataTupleType, CompileError
sss,
function_traps,
unwind_info,
stack_map,
)| {
functions.push(CompiledFunction {
body: function,
@@ -416,6 +441,7 @@ fn compile(env: CompileEnv<'_>) -> Result<ModuleCacheDataTupleType, CompileError
value_ranges.push(ranges.unwrap_or_default());
stack_slots.push(sss);
traps.push(function_traps);
stack_maps.push(stack_map);
},
);
@@ -428,6 +454,7 @@ fn compile(env: CompileEnv<'_>) -> Result<ModuleCacheDataTupleType, CompileError
value_ranges,
stack_slots,
traps,
stack_maps,
))
}