Hi DNF,

Those instructions (if they help in all cases) only work if you build Julia 
yourself by cloning the Julia git repository. You have installed julia via 
homebrew.
Unfortunately that route doesn’t work for me:

```
rob$ /usr/local/bin/julia
Illegal instruction: 4
```

I expect that maybe this problem is due to older stuff left behind somewhere 
(/usr/local/Cellar?).

Just now I installed Julia 0.4.1 from julialang.org/downloads and then moved 
the Julia-0.4.1.app to /Applications (after double clicking the Julia disk). 
That gave me simd without any problems. Maybe you could try that route?

Regards,
Rob

———————————————————————————————————————————————————————————

On Nov 9, 2015, at 11:47 AM, DNF <oyv...@gmail.com> wrote:

Thank you very much for taking the time to look into this, Rob.

My understanding now is that this is something that has to do with the build 
process of Julia, or perhaps with what version of openblas is being used. Am I 
getting that correctly? Do you think this something I could address to the 
maintainer of homebrew-julia?

I see that you are running some 'make' commands. Is 'make' something I run 
after git pulling directly from the main julia git repository, or does it 
somehow work with homebrew as well?


> On Nov 8, 2015, at 2:51 PM, Rob J. Goedman <goed...@icloud.com> wrote:
> 
> On another, slightly older system, I noticed similar (approximately 
> identical) timings for the simd.jl test script using Julia 0.5:
> 
> julia> include("/Users/rob/Projects/Julia/Rob/Julia/simd.jl")
> 
> Julia Version 0.5.0-dev+720
> Commit 5920633* (2015-10-11 15:15 UTC)
> Platform Info:
>   System: Darwin (x86_64-apple-darwin15.0.0)
>   CPU: Intel(R) Core(TM) i7-3720QM CPU @ 2.60GHz
>   WORD_SIZE: 64
>   BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Sandybridge)
>   LAPACK: libopenblas64_
>   LIBM: libopenlibm
>   LLVM: libLLVM-3.3
> 
> First call to timeit(1000,1000):
> 
> GFlop        = 0.6092165323090373
> GFlop (SIMD) = 0.4607065672339039
> 
> Second call to timeit(1000,1000):
> 
> GFlop        = 0.5935117884795207
> GFlop (SIMD) = 0.42286883095163036
> 
> On that same system Julia 0.4 (installed from the Julia site) did show 
> improved Gflop numbers and about a 6x improvement with simd.
> 
> To see if that would help with Julia 0.5, I did (in the cloned julia 
> directory, in a terminal):
> 
> git pull https://github.com/JuliaLang/julia 
> <https://github.com/JuliaLang/julia> master
> make -j 4
> 
> Lots of compile messages/warnings, but in the end:
> 
> clang: error: linker command failed with exit code 1 (use -v to see 
> invocation)
> make[3]: *** [libopenblas64_p-r0.2.15.dylib] Error 1
> make[2]: *** [shared] Error 2
> *** Clean the OpenBLAS build with 'make -C deps clean-openblas'. Rebuild with 
> 'make OPENBLAS_USE_THREAD=0 if OpenBLAS had trouble linking libpthread.so, 
> and with 'make OPENBLAS_TARGET_ARCH=NEHALEM' if there were errors building 
> SandyBridge support. Both these options can also be used simultaneously. ***
> make[1]: *** [build/openblas/libopenblas64_.dylib] Error 1
> make: *** [julia-deps] Error 2
> 
> I tried:
> 
> brew update
> brew upgrade
> make -C deps clean-openblas
> make -j 4
> 
> and running the simd.jl script now shows:
> 
> julia> include("/Users/rob/Projects/Julia/Rob/Julia/simd.jl")
> 
> Julia Version 0.5.0-dev+1195
> Commit 68667a3* (2015-11-08 21:05 UTC)
> Platform Info:
>   System: Darwin (x86_64-apple-darwin15.0.0)
>   CPU: Intel(R) Core(TM) i7-3720QM CPU @ 2.60GHz
>   WORD_SIZE: 64
>   BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Sandybridge)
>   LAPACK: libopenblas64_
>   LIBM: libopenlibm
>   LLVM: libLLVM-3.3
> 
> First call to timeit(1000,1000):
> 
> GFlop        = 1.4006308441321973
> GFlop (SIMD) = 13.561988458747821
> 
> Second call to timeit(1000,1000):
> 
> GFlop        = 2.300048186009497
> GFlop (SIMD) = 12.84397777991844
> 
> 
> Not sure if this helps or is even the right way to remedy this.
> 
> Regards,
> Rob
> 
> 
> <simd.jl>
> 
> 
> 
>> On Nov 6, 2015, at 5:35 PM, Rob J. Goedman <goed...@icloud.com 
>> <mailto:goed...@icloud.com>> wrote:
>> 
>> Thanks Seth,
>> 
>> That's the end of my first attempt to figure out what’s happening here. Back 
>> to the drawing board!
>> 
>> Regards,
>> Rob
>> 
>>> On Nov 6, 2015, at 4:53 PM, Seth <catch...@bromberger.com 
>>> <mailto:catch...@bromberger.com>> wrote:
>>> 
>>> Hi Rob,
>>> 
>>> I built it (and openblas) myself (via git clone) since I'm testing out 
>>> Cxx.jl. Xcode is Version 7.1 (7B91b).
>>> 
>>> Seth.
>>> 
>>> 
>>> On Friday, November 6, 2015 at 3:54:04 PM UTC-8, Rob J Goedman wrote:
>>> Seth,
>>> 
>>> You must have built  Julia 0.4.1-pre yourself. Did you use brew?
>>> 
>>> It looks like you are on Yosemite and picked up a newer libLLVM. Which 
>>> Xcode are you using?
>>> In the Julia.rb formula there is a test ENV.compiler, could it be clang is 
>>> not being used? 
>>> 
>>> Rob
>>> 
>>>> On Nov 6, 2015, at 3:01 PM, Seth <catc...@bromberger.com <javascript:>> 
>>>> wrote:
>>>> 
>>>> For what it's worth, I'm getting
>>>> 
>>>> julia> timeit(1000,1000)
>>>> GFlop        = 2.3913033081289967
>>>> GFlop (SIMD) = 2.2694726426420293
>>>> 
>>>> 
>>>> julia> versioninfo()
>>>> Julia Version 0.4.1-pre+22
>>>> Commit 669222e* (2015-11-01 00:06 UTC)
>>>> Platform Info:
>>>>   System: Darwin (x86_64-apple-darwin14.5.0)
>>>>   CPU: Intel(R) Core(TM) i7-4870HQ CPU @ 2.50GHz
>>>>   WORD_SIZE: 64
>>>>   BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Haswell)
>>>>   LAPACK: libopenblas64_
>>>>   LIBM: libopenlibm
>>>>   LLVM: libLLVM-svn
>>>> 
>>>> so it doesn't look like I'm taking advantage of simd either. :(
>>>> 
>>>> On Friday, November 6, 2015 at 11:43:41 AM UTC-8, Rob J Goedman wrote:
>>>> Hi DNF,
>>>> 
>>>> In below versioninfo’s only libopenblas appears different. You installed 
>>>> using brew. The first thing I would try is to execute the steps under 
>>>> Common Issues listed on https://github.com/staticfloat/homebrew-julia 
>>>> <https://github.com/staticfloat/homebrew-julia>. A bit further down on 
>>>> that site there is also some additional openblas related info.
>>>> 
>>>> Rob
>>>> 
>>>>> On Nov 6, 2015, at 10:35 AM, DNF <oyv...@gmail.com <>> wrote:
>>>>> 
>>>>> Thanks for the feedback. It seems like this is not a problem for most.
>>>>> 
>>>>> If anyone has even the faintest clue where I could start looking for a 
>>>>> solution to this, I would be grateful. Perhaps there is some software I 
>>>>> could run that would detect hardware problems, or maybe I am missing 
>>>>> software dependencies of some kind? What could I even google for? All my 
>>>>> searches just seem to bring up general info about SIMD, nothing like what 
>>>>> I'm describing.
>>>> 
>>>> 
>>>> 
>>>> On Friday, November 6, 2015 at 12:15:47 AM UTC+1, DNF wrote:
>>>> I install using homebrew from here: 
>>>> https://github.com/staticfloat/homebrew-julia 
>>>> <https://github.com/staticfloat/homebrew-julia>
>>>> I have limited understanding of the process, but believe there is some 
>>>> compilation involved.
>>>> 
>>>> 
>>>> Julia Version 0.4.0
>>>> Commit 0ff703b* (2015-10-08 06:20 UTC)
>>>> Platform Info:
>>>>   System: Darwin (x86_64-apple-darwin13.4.0)
>>>>   CPU: Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz
>>>>   WORD_SIZE: 64
>>>>   BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Haswell)
>>>>   LAPACK: libopenblas64_
>>>>   LIBM: libopenlibm
>>>>   LLVM: libLLVM-3.3
>>>> 
>>>> Julia Version 0.4.0
>>>> Commit 0ff703b* (2015-10-08 06:20 UTC)
>>>> Platform Info:
>>>>   System: Darwin (x86_64-apple-darwin15.0.0)
>>>>   CPU: Intel(R) Core(TM) i7-4850HQ CPU @ 2.30GHz
>>>>   WORD_SIZE: 64
>>>>   BLAS: libopenblas (DYNAMIC_ARCH NO_AFFINITY Haswell)
>>>>   LAPACK: libopenblas
>>>>   LIBM: libopenlibm
>>>>   LLVM: libLLVM-3.3
>>>> 
>>> 
>> 
> 

Reply via email to