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

machinst: fix the pinned reg hack;

Browse Source 
The pinned register hack didn't work because the GetPinnedReg is marked
as having side-effects, so that GVN wouldn't try to common it out.

This commit tweaks the function used during lowering to vcode, so that
the GetPinnedReg opcode is specially handled. It's a bit lame, but it
makes the hack work again.

Also, use_input needs to be a no-op for real registers.
This commit is contained in:
Benjamin Bouvier
2020-08-19 14:52:42 +02:00
parent efff43e769
commit ee76e01efc
2 changed files with 12 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
cranelift/codegen/src/inst_predicates.rs
View File

@@ -41,9 +41,11 @@ pub fn has_side_effect(func: &Function, inst: Inst) -> bool {
trivially_has_side_effects(opcode) || is_load_with_defined_trapping(opcode, data)
}
/// Does the given instruction have any side-effect as per [has_side_effect], or else is a load?
pub fn has_side_effect_or_load(func: &Function, inst: Inst) -> bool {
has_side_effect(func, inst) || func.dfg[inst].opcode().can_load()
/// Does the given instruction have any side-effect as per [has_side_effect], or else is a load,
/// but not the get_pinned_reg opcode?
pub fn has_side_effect_or_load_not_get_pinned_reg(func: &Function, inst: Inst) -> bool {
let op = func.dfg[inst].opcode();
op != Opcode::GetPinnedReg && (has_side_effect(func, inst) || op.can_load())
}
/// Is the given instruction a constant value (`iconst`, `fconst`, `bconst`) that can be