I was playing with multi-threading in Julia master and noticed that my code currently gives wrong results when more than one thread is enabled.
Julia 0.5:
julia> versioninfo()
Julia Version 0.5.0
Commit 3c9d753 (2016-09-19 18:14 UTC)
Platform Info:
System: Linux (x86_64-pc-linux-gnu)
CPU: Intel(R) Core(TM) i7-4700MQ CPU @ 2.40GHz
WORD_SIZE: 64
BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Haswell)
LAPACK: libopenblas64_
LIBM: libopenlibm
LLVM: libLLVM-3.7.1 (ORCJIT, haswell)
julia> Threads.nthreads()
4
julia> function foo()
p = Vector{Float64}(20)
Threads.@threads for i in eachindex(p)
p[i] = i
end
return p
end
foo (generic function with 1 method)
julia> foo()
20-element Array{Float64,1}:
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
18.0
19.0
20.0
Julia master:
julia> versioninfo()
Julia Version 0.6.0-dev.1924
Commit 2356ebb3f (2017-01-05 06:38 UTC)
Platform Info:
OS: Linux (x86_64-linux-gnu)
CPU: Intel(R) Core(TM) i7-4700MQ CPU @ 2.40GHz
WORD_SIZE: 64
BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Haswell)
LAPACK: libopenblas64_
LIBM: libopenlibm
LLVM: libLLVM-3.9.1 (ORCJIT, haswell)
julia> Threads.nthreads()
4
julia> function foo()
p = Vector{Float64}(20)
Threads.@threads for i in eachindex(p)
p[i] = i
end
return p
end
foo (generic function with 1 method)
julia> foo()
20-element Array{Float64,1}:
1.0
2.0
3.0
4.0
5.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Also assigning p[i] = Threads.threadid(), like in the example of the manual, gives wrong results.
Am I doing something wrong? Is the different behavior to be expected? I checked out the manual, but didn’t find any difference in the multi-threading section.