infinite loop in C code when trying to add a method #54516
Labels
domain:types and dispatch
Types, subtyping and method dispatch
kind:bug
Indicates an unexpected problem or unintended behavior
kind:regression
Regression in behavior compared to a previous version
Milestone
Comments
nsajko
added
kind:bug
|
When trying to precompile LaurentPolynomials.jl: |
|
Continuing from above, after second interrupt: |
|
Em that loop might be infinite if innervals have circular bounds. julia> S = Tuple{Val{S1} where {S1<:T1}, Union{Int,T1}} where {T1}
Tuple{Val{S1} where S1<:T1, Union{Int64, T1}} where T1
julia> T = Tuple{Union{Int,T2}, Val{S2} where {S2<:T2}} where {T2}
Tuple{Union{Int64, T2}, Val{S2} where S2<:T2} where T2
julia> typeintersect(S, T) |> Base.has_free_typevars
true
julia> versioninfo()
Julia Version 1.10.3
Commit 0b4590a550 (2024-04-30 10:59 UTC)
Build Info:
Official https://julialang.org/ release
Platform Info:
OS: Windows (x86_64-w64-mingw32)
CPU: 16 × 12th Gen Intel(R) Core(TM) i7-12650H
WORD_SIZE: 64
LIBM: libopenlibm
LLVM: libLLVM-15.0.7 (ORCJIT, alderlake)
Threads: 1 default, 0 interactive, 1 GC (on 16 virtual cores) |
|
Add the milestone as #53675 has been backported. |
This was referenced May 22, 2024
Merged
vtjnash
pushed a commit
that referenced
this issue
May 28, 2024
…4545) The infinite loop encountered in #54516 has been traced back to a circular bound during `finish_unionall`. As we insert innervar more eagerly now, the direct `jl_has_typevar` could not find all circularity. To address this, `has_typevar_via_flatten_env` is added to perform thorough check. Although there is some code duplication with `reachable_var`, it could be improved in future refactoring. #54516 also highlighted another free var escaping regression since v1.10. This regression is not solely the result of incomplete checks, it is also caused by the missing final substitution of `vb`'s bound, which has now been corrected. At last, this PR adds an assertion of sorting complexity, which should facilitate the detection of similar issues by PkgEval. close #54516
KristofferC
pushed a commit
that referenced
this issue
May 29, 2024
…4545) The infinite loop encountered in #54516 has been traced back to a circular bound during `finish_unionall`. As we insert innervar more eagerly now, the direct `jl_has_typevar` could not find all circularity. To address this, `has_typevar_via_flatten_env` is added to perform thorough check. Although there is some code duplication with `reachable_var`, it could be improved in future refactoring. #54516 also highlighted another free var escaping regression since v1.10. This regression is not solely the result of incomplete checks, it is also caused by the missing final substitution of `vb`'s bound, which has now been corrected. At last, this PR adds an assertion of sorting complexity, which should facilitate the detection of similar issues by PkgEval. close #54516 (cherry picked from commit 92dfdca)
|
This problem seems to have been backported on 1.11 beta2. |
|
Yes, backporting #54545 will fix that. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
MWE:
FTR this makes it impossible to compile LaurentPolynomials.jl.
Version info:
The text was updated successfully, but these errors were encountered: