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.

Sign up for GitHub

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

Jump to bottom

more inference fixes #16962

Closed
wants to merge 3 commits into from
Closed

more inference fixes #16962

wants to merge 3 commits into from

Conversation

vtjnash
Copy link
Member

@vtjnash vtjnash commented Jun 16, 2016

--compile=all is broken right now. This should fix most of the issues, although #16907 introduced a new segfault:

$ ./julia --compile=all -J usr/lib/julia/sys.dylib --output-o y.o --startup-file=no -e nothing
(lldb) r
Process 74790 launched: './usr/bin/julia' (x86_64)
found 26662 uncompiled methods for compile-all
WARNING: could not attach metadata for @simd loop.
 265 / 26662
signal (11): Segmentation fault: 11
while loading no file, in expression starting on line 0
_ZL13emit_functionP17_jl_lambda_info_tP20_jl_llvm_functions_t at /Users/jameson/julia/src/codegen.cpp:4008
_ZL11to_functionP17_jl_lambda_info_t at /Users/jameson/julia/src/codegen.cpp:824
_compile_all_deq at /Users/jameson/julia/src/gf.c:1358
julia_save at /Users/jameson/julia/src/init.c:733
main at /Users/jameson/julia/./usr/bin/julia (unknown line)
Allocations: 37935807 (Pool: 37934653; Big: 1154); GC: 54

vtjnash added 3 commits June 16, 2016 02:39
@vtjnash
correct precise_container_types to handle TypeConstructor correctly
d011b11
@vtjnash
inference: remove duplicated handling of sparam_vals
8f2a2cc
@vtjnash
fail jl_specializations_lookup more conservatively in the presence of…
f47322b 
… incorrect UnionAll types

previously, inserting a type and then looking up that exact type could fail,
since it would compute the UnionAll in the wrong position and thus not consider the types to be equivalent
JeffBezanson
JeffBezanson reviewed Jun 16, 2016
View reviewed changes
test/inference.jl
typealias NInt{N} Tuple{Vararg{Int, N}}
@test Base.eltype(NInt) === Int
fNInt(x::NInt) = (x...)
gNInt() = f(x)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Did you mean fNInt?

@JeffBezanson
Copy link
Member

lgtm

@vtjnash vtjnash closed this Jun 16, 2016
@vtjnash vtjnash deleted the jn/fix-delete_non_inlineable branch June 16, 2016 16:11
@vtjnash
Copy link
Member Author

vtjnash commented Jun 16, 2016

merged in 7ad5eee, 85cf5e6, 8b103cf (with fix to ensure it is actually being tested)

tkelman referenced this pull request Jun 16, 2016
@vtjnash
correct precise_container_types to handle TypeConstructor correctly
8b103cf
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@vtjnash @JeffBezanson