T0b1/wasmtime
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Support explicit endianness in Cranelift IR MemFlags

Browse Source 
WebAssembly memory operations are by definition little-endian even on
big-endian target platforms.  However, other memory accesses will require
native target endianness (e.g. to access parts of the VMContext that is
also accessed by VM native code).  This means on big-endian targets,
the code generator will have to handle both little- and big-endian
memory accesses.  However, there is currently no way to encode that
distinction into the Cranelift IR that describes memory accesses.

This patch provides such a way by adding an (optional) explicit
endianness marker to an instance of MemFlags.  Since each Cranelift IR
instruction that describes memory accesses already has an instance of
MemFlags attached, this can now be used to provide endianness
information.

Note that by default, memory accesses will continue to use the native
target ISA endianness.  To override this to specify an explicit
endianness, a MemFlags value that was built using the set_endianness
routine must be used.  This patch does so for accesses that implement
WebAssembly memory operations.

This patch addresses issue #2124.
This commit is contained in:
Ulrich Weigand
2020-12-09 16:23:10 +01:00
parent 2cec20aa57
commit 467a1af83a
5 changed files with 83 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
cranelift/wasm/src/code_translator.rs
View File

@@ -2056,7 +2056,9 @@ fn prepare_load<FE: FuncEnvironment + ?Sized>(
// Note that we don't set `is_aligned` here, even if the load instruction's
// alignment immediate says it's aligned, because WebAssembly's immediate
// field is just a hint, while Cranelift's aligned flag needs a guarantee.
let flags = MemFlags::new();
// WebAssembly memory accesses are always little-endian.
let mut flags = MemFlags::new();
flags.set_endianness(ir::Endianness::Little);
Ok((flags, base, offset.into()))
}
@@ -2103,7 +2105,8 @@ fn translate_store<FE: FuncEnvironment + ?Sized>(
builder,
);
// See the comments in `prepare_load` about the flags.
let flags = MemFlags::new();
let mut flags = MemFlags::new();
flags.set_endianness(ir::Endianness::Little);
builder
.ins()
.Store(opcode, val_ty, flags, offset.into(), val, base);
@@ -2207,7 +2210,8 @@ fn translate_atomic_rmw<FE: FuncEnvironment + ?Sized>(
finalise_atomic_mem_addr(linear_mem_addr, memarg, access_ty, builder, state, environ)?;
// See the comments in `prepare_load` about the flags.
let flags = MemFlags::new();
let mut flags = MemFlags::new();
flags.set_endianness(ir::Endianness::Little);
let mut res = builder
.ins()
.atomic_rmw(access_ty, flags, op, final_effective_address, arg2);
@@ -2260,7 +2264,8 @@ fn translate_atomic_cas<FE: FuncEnvironment + ?Sized>(
finalise_atomic_mem_addr(linear_mem_addr, memarg, access_ty, builder, state, environ)?;
// See the comments in `prepare_load` about the flags.
let flags = MemFlags::new();
let mut flags = MemFlags::new();
flags.set_endianness(ir::Endianness::Little);
let mut res = builder
.ins()
.atomic_cas(flags, final_effective_address, expected, replacement);
@@ -2302,7 +2307,8 @@ fn translate_atomic_load<FE: FuncEnvironment + ?Sized>(
finalise_atomic_mem_addr(linear_mem_addr, memarg, access_ty, builder, state, environ)?;
// See the comments in `prepare_load` about the flags.
let flags = MemFlags::new();
let mut flags = MemFlags::new();
flags.set_endianness(ir::Endianness::Little);
let mut res = builder
.ins()
.atomic_load(access_ty, flags, final_effective_address);
@@ -2348,7 +2354,8 @@ fn translate_atomic_store<FE: FuncEnvironment + ?Sized>(
finalise_atomic_mem_addr(linear_mem_addr, memarg, access_ty, builder, state, environ)?;
// See the comments in `prepare_load` about the flags.
let flags = MemFlags::new();
let mut flags = MemFlags::new();
flags.set_endianness(ir::Endianness::Little);
builder
.ins()
.atomic_store(flags, data, final_effective_address);