[ConstraintGraph] Disallow subtype constraint contractions #11855
Conversation
xedin
force-pushed
the
disallow-subtype-contractions
branch
from
2f1330e to
ac311a0
Compare
|
@swift-ci please smoke test |
| $0.b = 0 | ||
| } | ||
|
|
||
| print(rdar34333874) |
| } | ||
|
|
||
| let rdar34333874 = foo(C()) { | ||
| $0.a = 42 |
There was a problem hiding this comment.
Is this really testing the same thing as the reported bug? This actually is mutating the input whereas the bug report wasn't.
There was a problem hiding this comment.
Example from the rdar does modify fields of $0
There was a problem hiding this comment.
The actual original code did mutate input inside the closure. I'd suggest testing both cases to be on the safe side.
There was a problem hiding this comment.
Sure, sounds good, I'll add one which doesn't mutate the fields.
| // RUN: rm -rf %t | ||
| // RUN: mkdir -p %t | ||
| // RUN: %target-build-swift %s -o %t/a.out | ||
| // RUN: %target-run %t/a.out |
There was a problem hiding this comment.
Drive-by comment: This should be %target-run-simple-swift, which gets run in more situations.
There was a problem hiding this comment.
Thanks, @jrose-apple ! I didn't know that existed...
xedin
force-pushed
the
disallow-subtype-contractions
branch
from
ac311a0 to
f7960bc
Compare
|
@swift-ci please smoke test |
| return this | ||
| } | ||
|
|
||
| let rdar34333874 = foo(C()) { |
There was a problem hiding this comment.
Hmm, apparently I didn't submit my comment about the test. Are you sure this is testing the same thing as the original report? This appears to actually be mutating something whereas the original report wasn't.
Presently subtype constraint is considered a candidate for contraction via `shouldContractEdge` when left-hand side of the subtype constraint is an inout type with type variable inside. Although it's intended to be used only in combination with BindParam constraint assigned to the same variable, that is actually never checked and contraction of subtype constraint like that is invalid. Resolves: rdar://problem/34333874
xedin
force-pushed
the
disallow-subtype-contractions
branch
from
f7960bc to
0a178fd
Compare
|
I've fixed a problem in |
|
@swift-ci please smoke test |
|
@swift-ci please test source compatibility |
Presently subtype constraint is considered a candidate for contraction
via
shouldContractEdgewhen left-hand side of the subtype constraintis an inout type with type variable inside. Although it's intended to
be used only in combination with BindParam constraint assigned to the
same variable, that is actually never checked and contraction of subtype
constraint like that is invalid.
Resolves: rdar://problem/34333874