| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
| |
When looking for all values written to a field, we can ignore
values that are loaded from that same field, i.e., are copied from
something already present there. Such operations never introduce
new values.
This helps by a small but non-zero amount on j2cl.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
ConstantFieldPropagation (#4064)
Previously we tracked them in the same way. That means that we did the same
when seeing if either a struct.new or a struct.set can write to the memory that
is read by a struct.get, where the rule is that if either type is a subtype of the
other then they might. But with struct.new we know the precise type, which
means we can do better.
Specifically, if we see a new of type B, then only a get of a supertype of B can
possibly read that data: it is not possible for our struct of type B to appear in
a location that requires a subtype of B. Conceptually:
A = type struct
B = type extends A
C = type extends B
x = struct.new<B>
struct.get<A>(y) // x might appear here, as it can be assigned to a
// variable y of a supertype
struct.get<C>(y) // x cannot appear here
This allows more devirtualization. It is a followup for #4052 that implements
a TODO from there.
The diff without whitespace is simpler.
|
|
|
A field in a struct is constant if we can see that in the entire program we
only ever write the same constant value to it. For example, imagine a
vtable type that we construct with the same funcrefs each time, then (if
we have no struct.sets, or if we did, and they had the same value), we
could replace a get with that constant value, since it cannot be anything
else:
(struct.new $T (i32.const 10) (rtt))
..no other conflicting values..
(struct.get $T 0) => (i32.const 10)
If the value is a function reference, then this may allow other passes
to turn what was a call_ref into a direct call and perhaps also get
inlined, effectively a form of devirtualization.
This only works in nominal typing, as we need to know the supertype
of each type. (It could work in theory in structural, but we'd need to do
hard work to find all the possible supertypes, and it would also
become far less effective.)
This deletes a trivial test for running -O on GC content. We have
many more tests for GC at this point, so that test is not needed, and
this PR also optimizes the code into something trivial and
uninteresting anyhow.
|
