Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

reland: add morespecific rule for Type{Union{}} #49470

Merged
merged 4 commits into from
Apr 24, 2023
Merged

Conversation

vtjnash
Copy link
Member

@vtjnash vtjnash commented Apr 24, 2023

Reland #49349 after adding additional annotations for sagc checker. Apparently there was a merge conflict in the abilities of the checker to complain about this code.

vtjnash added 4 commits April 23, 2023 10:35
Removes some overly strict `@test_throws MethodError` for calls with
`Union{}` or `Any`, in preparation for making those errors more precise.
Make Type{Union{}} in method definitions always the most specific type
(normally these would end up being ambiguous). This ensures we do not
invalidate them, nor need to consider ambiguities that might arise from
intersections with them.
Based on the new morespecific rule for Union{} and method definitions of
the specific form `f(..., Type{Union{}}, Vararg)`. If a method
definition exists with that specific form, the intersection visitor will
ignore all intersections that have that as their only result, saving
significant effort when working with lookups involving `Type{<:T}`
(which usually ended up mostly ambiguous anyways).

Fixes: #33780

This pattern turns out to have still to been making package loading
slow. We could keep adding methods following the ambiguity pattern
#46000 for the couple specific
functions that need it (constructor, eltype, IteratorEltype,
IteratorSize, and maybe a couple others) so the internals can detect
those and optimize functions that have that method pair. But it seems
somewhat odd, convoluted, and non-obvious behavior there. Instead, this
breaks all ambiguities in which Union{} is present explicitly in favor
of the method with Union{}. This means that when computing method
matches, as soon as we see one method definition with Union{}, we can
record that the method is the only possible match for that slot.

This, in essence, permits creating a rule for dispatch that a TypeVar
lower bound must be strictly a supertype of Union{}, but this creates it
at the function level, instead of expecting the user to add it to every
TypeVar they use to define methods.

This also lets us improve the error message for these cases (generally
they should error to avoid polluting the inference result), since we can
be assured this method will be called, and not result in an ambiguous
MethodError instead!

Reverts the functional change of #46000
Since T cannot be Union{} here, the prior optimization would not get
detected post facto, but a priori this cannot be inhabited for
T=Union{}, so we can exclude it immediately. This does not happen during
inference, but shows up during edge validation somewhat often.
@giordano giordano added compiler:latency Compiler latency re-land This relands a PR that was previously merged but was later reverted. labels Apr 24, 2023
@vtjnash vtjnash merged commit cb1cc27 into master Apr 24, 2023
@vtjnash vtjnash deleted the jn/morespecific-bottom branch April 24, 2023 22:45
vtjnash added a commit that referenced this pull request Oct 30, 2023
With #49470, these can all be dispatched to the same method now,
avoiding unnecessary code duplication for this case.
vtjnash added a commit that referenced this pull request Nov 2, 2023
With #49470, these can all be dispatched to the same method now,
avoiding unnecessary code duplication for this case.
vtjnash added a commit that referenced this pull request Nov 2, 2023
With #49470, these can all be dispatched to the same method now,
avoiding unnecessary code duplication for this case.

This partly reverts f2dcc44, but our
edges against methods should be better than the ones against
MethodError these days.
vtjnash added a commit that referenced this pull request Dec 8, 2023
With #49470, these can all be dispatched to the same method now,
avoiding unnecessary code duplication for this case.

This partly reverts f2dcc44, but our
edges against methods should be better than the ones against
MethodError these days.
vtjnash added a commit that referenced this pull request Dec 11, 2023
With #49470, these can all be dispatched to the same method now,
avoiding unnecessary code duplication for this case.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
compiler:latency Compiler latency re-land This relands a PR that was previously merged but was later reverted.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants