Problem with compiling OpenBLAS to work with R

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

Problem with compiling OpenBLAS to work with R

Erin Hodgess-2
Hello!

I'm not sure if this is the right place to post this, so apologies
in advance if I'm not in the right list.

I downloaded the OpenBLAS and am following Avraham Adler's great
instructions.  However, when I run make, things go well to a certain point,
and then go bad:

make
[snip]

touch cygopenblas_haswellp-r0.3.5.a
make -j 1 -C test all
make[1]: Entering directory
'/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
gfortran -O2 -Wall -frecursive -m64 -mavx2   -o sblat1 sblat1.o
../cygopenblas_haswellp-r0.3.5.a  -L/usr/lib/gcc/x86_64-pc-msys/7.3.0
-L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib/../lib
-L/usr/lib/../lib
-L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib
-L/usr/lib/w32api  -lmsys-2.0
D:/msys64/usr/lib/../lib/libpthread.a(t-d001088.o):fake:(.text+0x2):
undefined reference to `__imp_pthread_mutex_destroy'
D:/msys64/usr/lib/../lib/libpthread.a(t-d001090.o):fake:(.text+0x2):
undefined reference to `__imp_pthread_mutex_init'
D:/msys64/usr/lib/../lib/libpthread.a(t-d001091.o):fake:(.text+0x2):
undefined reference to `__imp_pthread_mutex_lock'
D:/msys64/usr/lib/../lib/libpthread.a(t-d001094.o):fake:(.text+0x2):
undefined reference to `__imp_pthread_mutex_trylock'
D:/msys64/usr/lib/../lib/libpthread.a(t-d001095.o):fake:(.text+0x2):
undefined reference to `__imp_pthread_mutex_unlock'
collect2.exe: error: ld returned 1 exit status
make[1]: *** [Makefile:134: sblat1] Error 1
make[1]: Leaving directory
'/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
make: *** [Makefile:124: tests] Error 2


I think it has something to do with the threads/pthreads but am not sure
how to fix it.  Any suggestions much appreciated.

Thanks,
Sincerely,
Erin

Erin Hodgess, PhD
mailto: [hidden email]

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Reply | Threaded
Open this post in threaded view
|

Re: Problem with compiling OpenBLAS to work with R

Kenny Bell-2
This person has had apparent success - you could follow what they did or
just download their product (with appropriate caution downloading a random
.exe).

https://github.com/thequackdaddy/R-OpenBLAS

On Thu, Feb 28, 2019 at 6:28 AM Erin Hodgess <[hidden email]>
wrote:

> Hello!
>
> I'm not sure if this is the right place to post this, so apologies
> in advance if I'm not in the right list.
>
> I downloaded the OpenBLAS and am following Avraham Adler's great
> instructions.  However, when I run make, things go well to a certain point,
> and then go bad:
>
> make
> [snip]
>
> touch cygopenblas_haswellp-r0.3.5.a
> make -j 1 -C test all
> make[1]: Entering directory
> '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
> gfortran -O2 -Wall -frecursive -m64 -mavx2   -o sblat1 sblat1.o
> ../cygopenblas_haswellp-r0.3.5.a  -L/usr/lib/gcc/x86_64-pc-msys/7.3.0
> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib/../lib
> -L/usr/lib/../lib
> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib
> -L/usr/lib/w32api  -lmsys-2.0
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001088.o):fake:(.text+0x2):
> undefined reference to `__imp_pthread_mutex_destroy'
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001090.o):fake:(.text+0x2):
> undefined reference to `__imp_pthread_mutex_init'
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001091.o):fake:(.text+0x2):
> undefined reference to `__imp_pthread_mutex_lock'
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001094.o):fake:(.text+0x2):
> undefined reference to `__imp_pthread_mutex_trylock'
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001095.o):fake:(.text+0x2):
> undefined reference to `__imp_pthread_mutex_unlock'
> collect2.exe: error: ld returned 1 exit status
> make[1]: *** [Makefile:134: sblat1] Error 1
> make[1]: Leaving directory
> '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
> make: *** [Makefile:124: tests] Error 2
>
>
> I think it has something to do with the threads/pthreads but am not sure
> how to fix it.  Any suggestions much appreciated.
>
> Thanks,
> Sincerely,
> Erin
>
> Erin Hodgess, PhD
> mailto: [hidden email]
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Reply | Threaded
Open this post in threaded view
|

Re: Problem with compiling OpenBLAS to work with R

Avraham Adler
I believe that repo just follows the directions on my blog. Without seeing
Dr. Hodges’s code, my initial concern is the many references to Cygwin. My
method specifically does not use Cygwin but MSYS2 and Mingw64/Rtools35.
That will likely change to solely Rtools40 once R3.6 is released due to the
Msys system being built in to it.

There may be some library conflicts between Cygwin and msys2/mingw64. If
possible, my suggestion would be uninstall everything and then just install
msys2 (and add in make after you to the first msys update) and rtools35.
Then there should be no conflicting libraries.

Thanks,

Avi

On Thu, Feb 28, 2019 at 12:11 AM Kenny Bell <[hidden email]> wrote:

> This person has had apparent success - you could follow what they did or
> just download their product (with appropriate caution downloading a random
> .exe).
>
> https://github.com/thequackdaddy/R-OpenBLAS
>
> On Thu, Feb 28, 2019 at 6:28 AM Erin Hodgess <[hidden email]>
> wrote:
>
> > Hello!
> >
> > I'm not sure if this is the right place to post this, so apologies
> > in advance if I'm not in the right list.
> >
> > I downloaded the OpenBLAS and am following Avraham Adler's great
> > instructions.  However, when I run make, things go well to a certain
> point,
> > and then go bad:
> >
> > make
> > [snip]
> >
> > touch cygopenblas_haswellp-r0.3.5.a
> > make -j 1 -C test all
> > make[1]: Entering directory
> > '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
> > gfortran -O2 -Wall -frecursive -m64 -mavx2   -o sblat1 sblat1.o
> > ../cygopenblas_haswellp-r0.3.5.a  -L/usr/lib/gcc/x86_64-pc-msys/7.3.0
> > -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib/../lib
> > -L/usr/lib/../lib
> > -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib
> > -L/usr/lib/w32api  -lmsys-2.0
> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001088.o):fake:(.text+0x2):
> > undefined reference to `__imp_pthread_mutex_destroy'
> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001090.o):fake:(.text+0x2):
> > undefined reference to `__imp_pthread_mutex_init'
> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001091.o):fake:(.text+0x2):
> > undefined reference to `__imp_pthread_mutex_lock'
> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001094.o):fake:(.text+0x2):
> > undefined reference to `__imp_pthread_mutex_trylock'
> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001095.o):fake:(.text+0x2):
> > undefined reference to `__imp_pthread_mutex_unlock'
> > collect2.exe: error: ld returned 1 exit status
> > make[1]: *** [Makefile:134: sblat1] Error 1
> > make[1]: Leaving directory
> > '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
> > make: *** [Makefile:124: tests] Error 2
> >
> >
> > I think it has something to do with the threads/pthreads but am not sure
> > how to fix it.  Any suggestions much appreciated.
> >
> > Thanks,
> > Sincerely,
> > Erin
> >
> > Erin Hodgess, PhD
> > mailto: [hidden email]
> >
> >         [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > [hidden email] mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-devel
> >
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
--
Sent from Gmail Mobile

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Reply | Threaded
Open this post in threaded view
|

Re: Problem with compiling OpenBLAS to work with R

Erin Hodgess-2
The Cygwin is the problem.

I am redoing and will let you know how it goes.

Thanks for your help.

Sincerely,
Erin

On Wed, Feb 27, 2019 at 10:22 PM Avraham Adler <[hidden email]>
wrote:

> I believe that repo just follows the directions on my blog. Without seeing
> Dr. Hodges’s code, my initial concern is the many references to Cygwin. My
> method specifically does not use Cygwin but MSYS2 and Mingw64/Rtools35.
> That will likely change to solely Rtools40 once R3.6 is released due to the
> Msys system being built in to it.
>
> There may be some library conflicts between Cygwin and msys2/mingw64. If
> possible, my suggestion would be uninstall everything and then just install
> msys2 (and add in make after you to the first msys update) and rtools35.
> Then there should be no conflicting libraries.
>
> Thanks,
>
> Avi
>
> On Thu, Feb 28, 2019 at 12:11 AM Kenny Bell <[hidden email]> wrote:
>
>> This person has had apparent success - you could follow what they did or
>> just download their product (with appropriate caution downloading a random
>> .exe).
>>
>> https://github.com/thequackdaddy/R-OpenBLAS
>>
>> On Thu, Feb 28, 2019 at 6:28 AM Erin Hodgess <[hidden email]>
>> wrote:
>>
>> > Hello!
>> >
>> > I'm not sure if this is the right place to post this, so apologies
>> > in advance if I'm not in the right list.
>> >
>> > I downloaded the OpenBLAS and am following Avraham Adler's great
>> > instructions.  However, when I run make, things go well to a certain
>> point,
>> > and then go bad:
>> >
>> > make
>> > [snip]
>> >
>> > touch cygopenblas_haswellp-r0.3.5.a
>> > make -j 1 -C test all
>> > make[1]: Entering directory
>> > '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
>> > gfortran -O2 -Wall -frecursive -m64 -mavx2   -o sblat1 sblat1.o
>> > ../cygopenblas_haswellp-r0.3.5.a  -L/usr/lib/gcc/x86_64-pc-msys/7.3.0
>> >
>> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib/../lib
>> > -L/usr/lib/../lib
>> > -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib
>> > -L/usr/lib/w32api  -lmsys-2.0
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001088.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_destroy'
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001090.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_init'
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001091.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_lock'
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001094.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_trylock'
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001095.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_unlock'
>> > collect2.exe: error: ld returned 1 exit status
>> > make[1]: *** [Makefile:134: sblat1] Error 1
>> > make[1]: Leaving directory
>> > '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
>> > make: *** [Makefile:124: tests] Error 2
>> >
>> >
>> > I think it has something to do with the threads/pthreads but am not sure
>> > how to fix it.  Any suggestions much appreciated.
>> >
>> > Thanks,
>> > Sincerely,
>> > Erin
>> >
>> > Erin Hodgess, PhD
>> > mailto: [hidden email]
>> >
>> >         [[alternative HTML version deleted]]
>> >
>> > ______________________________________________
>> > [hidden email] mailing list
>> > https://stat.ethz.ch/mailman/listinfo/r-devel
>> >
>>
>>         [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> [hidden email] mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-devel
>>
> --
> Sent from Gmail Mobile
>
--
Erin Hodgess, PhD
mailto: [hidden email]

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Reply | Threaded
Open this post in threaded view
|

Re: Problem with compiling OpenBLAS to work with R

Erin Hodgess-2
In reply to this post by Avraham Adler
Yay!  I re-installed everything and got through "Make distribution"!

I have one more question, please:  I am running the make check-all.  I have
an error at reg-1d.  It stops the process.  However, the mean difference
(as per the file) is 2.0e-12.  I'm ok with that.  How do I bypass this,
please?

Thanks,
Erin

Erin Hodgess, PhD
mailto: [hidden email]


On Wed, Feb 27, 2019 at 10:22 PM Avraham Adler <[hidden email]>
wrote:

> I believe that repo just follows the directions on my blog. Without seeing
> Dr. Hodges’s code, my initial concern is the many references to Cygwin. My
> method specifically does not use Cygwin but MSYS2 and Mingw64/Rtools35.
> That will likely change to solely Rtools40 once R3.6 is released due to the
> Msys system being built in to it.
>
> There may be some library conflicts between Cygwin and msys2/mingw64. If
> possible, my suggestion would be uninstall everything and then just install
> msys2 (and add in make after you to the first msys update) and rtools35.
> Then there should be no conflicting libraries.
>
> Thanks,
>
> Avi
>
> On Thu, Feb 28, 2019 at 12:11 AM Kenny Bell <[hidden email]> wrote:
>
>> This person has had apparent success - you could follow what they did or
>> just download their product (with appropriate caution downloading a random
>> .exe).
>>
>> https://github.com/thequackdaddy/R-OpenBLAS
>>
>> On Thu, Feb 28, 2019 at 6:28 AM Erin Hodgess <[hidden email]>
>> wrote:
>>
>> > Hello!
>> >
>> > I'm not sure if this is the right place to post this, so apologies
>> > in advance if I'm not in the right list.
>> >
>> > I downloaded the OpenBLAS and am following Avraham Adler's great
>> > instructions.  However, when I run make, things go well to a certain
>> point,
>> > and then go bad:
>> >
>> > make
>> > [snip]
>> >
>> > touch cygopenblas_haswellp-r0.3.5.a
>> > make -j 1 -C test all
>> > make[1]: Entering directory
>> > '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
>> > gfortran -O2 -Wall -frecursive -m64 -mavx2   -o sblat1 sblat1.o
>> > ../cygopenblas_haswellp-r0.3.5.a  -L/usr/lib/gcc/x86_64-pc-msys/7.3.0
>> >
>> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib/../lib
>> > -L/usr/lib/../lib
>> > -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib
>> > -L/usr/lib/w32api  -lmsys-2.0
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001088.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_destroy'
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001090.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_init'
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001091.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_lock'
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001094.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_trylock'
>> > D:/msys64/usr/lib/../lib/libpthread.a(t-d001095.o):fake:(.text+0x2):
>> > undefined reference to `__imp_pthread_mutex_unlock'
>> > collect2.exe: error: ld returned 1 exit status
>> > make[1]: *** [Makefile:134: sblat1] Error 1
>> > make[1]: Leaving directory
>> > '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test'
>> > make: *** [Makefile:124: tests] Error 2
>> >
>> >
>> > I think it has something to do with the threads/pthreads but am not sure
>> > how to fix it.  Any suggestions much appreciated.
>> >
>> > Thanks,
>> > Sincerely,
>> > Erin
>> >
>> > Erin Hodgess, PhD
>> > mailto: [hidden email]
>> >
>> >         [[alternative HTML version deleted]]
>> >
>> > ______________________________________________
>> > [hidden email] mailing list
>> > https://stat.ethz.ch/mailman/listinfo/r-devel
>> >
>>
>>         [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> [hidden email] mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-devel
>>
> --
> Sent from Gmail Mobile
>

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Reply | Threaded
Open this post in threaded view
|

Re: Problem with compiling OpenBLAS to work with R

Martin Maechler
>>>>> Erin Hodgess
>>>>>     on Fri, 1 Mar 2019 12:30:35 -0700 writes:

    > Yay!  I re-installed everything and got through "Make
    > distribution"!  I have one more question, please: I am
    > running the make check-all.  I have an error at reg-1d.
    > It stops the process.  However, the mean difference (as
    > per the file) is 2.0e-12.  I'm ok with that.  How do I
    > bypass this, please?

If you'd be really precise in your report (use correct file
names; show the exact error message + context lines of the *.Rout.fail
file) I could help you.

Note (by the way) that this may be yet another example showing
that OpenBLAS may be faster but less accurate than the
(plain+ minor tweaks) version of BLAS that ships with R.

Martin

    > Thanks, Erin

    > Erin Hodgess, PhD mailto: [hidden email]


    > On Wed, Feb 27, 2019 at 10:22 PM Avraham Adler
    > <[hidden email]> wrote:

    >> I believe that repo just follows the directions on my
    >> blog. Without seeing Dr. Hodges’s code, my initial
    >> concern is the many references to Cygwin. My method
    >> specifically does not use Cygwin but MSYS2 and
    >> Mingw64/Rtools35.  That will likely change to solely
    >> Rtools40 once R3.6 is released due to the Msys system
    >> being built in to it.
    >>
    >> There may be some library conflicts between Cygwin and
    >> msys2/mingw64. If possible, my suggestion would be
    >> uninstall everything and then just install msys2 (and add
    >> in make after you to the first msys update) and rtools35.
    >> Then there should be no conflicting libraries.
    >>
    >> Thanks,
    >>
    >> Avi
    >>
    >> On Thu, Feb 28, 2019 at 12:11 AM Kenny Bell
    >> <[hidden email]> wrote:
    >>
    >>> This person has had apparent success - you could follow
    >>> what they did or just download their product (with
    >>> appropriate caution downloading a random .exe).
    >>>
    >>> https://github.com/thequackdaddy/R-OpenBLAS
    >>>
    >>> On Thu, Feb 28, 2019 at 6:28 AM Erin Hodgess
    >>> <[hidden email]> wrote:
    >>>
    >>> > Hello!
    >>> >
    >>> > I'm not sure if this is the right place to post this,
    >>> so apologies > in advance if I'm not in the right list.
    >>> >
    >>> > I downloaded the OpenBLAS and am following Avraham
    >>> Adler's great > instructions.  However, when I run make,
    >>> things go well to a certain point, > and then go bad:
    >>> >
    >>> > make > [snip]
    >>> >
    >>> > touch cygopenblas_haswellp-r0.3.5.a > make -j 1 -C
    >>> test all > make[1]: Entering directory >
    >>> '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test' >
    >>> gfortran -O2 -Wall -frecursive -m64 -mavx2 -o sblat1
    >>> sblat1.o > ../cygopenblas_haswellp-r0.3.5.a
    >>> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0
    >>> >
    >>> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib/../lib
    >>> > -L/usr/lib/../lib >
    >>> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib
    >>> > -L/usr/lib/w32api -lmsys-2.0 >
    >>> D:/msys64/usr/lib/../lib/libpthread.a(t-d001088.o):fake:(.text+0x2):
    >>> > undefined reference to `__imp_pthread_mutex_destroy' >
    >>> D:/msys64/usr/lib/../lib/libpthread.a(t-d001090.o):fake:(.text+0x2):
    >>> > undefined reference to `__imp_pthread_mutex_init' >
    >>> D:/msys64/usr/lib/../lib/libpthread.a(t-d001091.o):fake:(.text+0x2):
    >>> > undefined reference to `__imp_pthread_mutex_lock' >
    >>> D:/msys64/usr/lib/../lib/libpthread.a(t-d001094.o):fake:(.text+0x2):
    >>> > undefined reference to `__imp_pthread_mutex_trylock' >
    >>> D:/msys64/usr/lib/../lib/libpthread.a(t-d001095.o):fake:(.text+0x2):
    >>> > undefined reference to `__imp_pthread_mutex_unlock' >
    >>> collect2.exe: error: ld returned 1 exit status >
    >>> make[1]: *** [Makefile:134: sblat1] Error 1 > make[1]:
    >>> Leaving directory >
    >>> '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test' >
    >>> make: *** [Makefile:124: tests] Error 2
    >>> >
    >>> >
    >>> > I think it has something to do with the
    >>> threads/pthreads but am not sure > how to fix it.  Any
    >>> suggestions much appreciated.
    >>> >
    >>> > Thanks, > Sincerely, > Erin
    >>> >
    >>> > Erin Hodgess, PhD > mailto: [hidden email]
    >>> >
    >>> > [[alternative HTML version deleted]]
    >>> >
    >>> > ______________________________________________ >
    >>> [hidden email] mailing list >
    >>> https://stat.ethz.ch/mailman/listinfo/r-devel
    >>> >
    >>>
    >>> [[alternative HTML version deleted]]
    >>>
    >>> ______________________________________________
    >>> [hidden email] mailing list
    >>> https://stat.ethz.ch/mailman/listinfo/r-devel
    >>>
    >> --
    >> Sent from Gmail Mobile
    >>

    > [[alternative HTML version deleted]]

    > ______________________________________________
    > [hidden email] mailing list
    > https://stat.ethz.ch/mailman/listinfo/r-devel

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Reply | Threaded
Open this post in threaded view
|

Re: Problem with compiling OpenBLAS to work with R

Erin Hodgess-2
I actually figured it out.  I went back in any changed the mean difference
from 2-e12 to 5e-12.

Thanks,
Erin

Erin Hodgess, PhD
mailto: [hidden email]


On Mon, Mar 4, 2019 at 2:10 AM Martin Maechler <[hidden email]>
wrote:

> >>>>> Erin Hodgess
> >>>>>     on Fri, 1 Mar 2019 12:30:35 -0700 writes:
>
>     > Yay!  I re-installed everything and got through "Make
>     > distribution"!  I have one more question, please: I am
>     > running the make check-all.  I have an error at reg-1d.
>     > It stops the process.  However, the mean difference (as
>     > per the file) is 2.0e-12.  I'm ok with that.  How do I
>     > bypass this, please?
>
> If you'd be really precise in your report (use correct file
> names; show the exact error message + context lines of the *.Rout.fail
> file) I could help you.
>
> Note (by the way) that this may be yet another example showing
> that OpenBLAS may be faster but less accurate than the
> (plain+ minor tweaks) version of BLAS that ships with R.
>
> Martin
>
>     > Thanks, Erin
>
>     > Erin Hodgess, PhD mailto: [hidden email]
>
>
>     > On Wed, Feb 27, 2019 at 10:22 PM Avraham Adler
>     > <[hidden email]> wrote:
>
>     >> I believe that repo just follows the directions on my
>     >> blog. Without seeing Dr. Hodges’s code, my initial
>     >> concern is the many references to Cygwin. My method
>     >> specifically does not use Cygwin but MSYS2 and
>     >> Mingw64/Rtools35.  That will likely change to solely
>     >> Rtools40 once R3.6 is released due to the Msys system
>     >> being built in to it.
>     >>
>     >> There may be some library conflicts between Cygwin and
>     >> msys2/mingw64. If possible, my suggestion would be
>     >> uninstall everything and then just install msys2 (and add
>     >> in make after you to the first msys update) and rtools35.
>     >> Then there should be no conflicting libraries.
>     >>
>     >> Thanks,
>     >>
>     >> Avi
>     >>
>     >> On Thu, Feb 28, 2019 at 12:11 AM Kenny Bell
>     >> <[hidden email]> wrote:
>     >>
>     >>> This person has had apparent success - you could follow
>     >>> what they did or just download their product (with
>     >>> appropriate caution downloading a random .exe).
>     >>>
>     >>> https://github.com/thequackdaddy/R-OpenBLAS
>     >>>
>     >>> On Thu, Feb 28, 2019 at 6:28 AM Erin Hodgess
>     >>> <[hidden email]> wrote:
>     >>>
>     >>> > Hello!
>     >>> >
>     >>> > I'm not sure if this is the right place to post this,
>     >>> so apologies > in advance if I'm not in the right list.
>     >>> >
>     >>> > I downloaded the OpenBLAS and am following Avraham
>     >>> Adler's great > instructions.  However, when I run make,
>     >>> things go well to a certain point, > and then go bad:
>     >>> >
>     >>> > make > [snip]
>     >>> >
>     >>> > touch cygopenblas_haswellp-r0.3.5.a > make -j 1 -C
>     >>> test all > make[1]: Entering directory >
>     >>> '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test' >
>     >>> gfortran -O2 -Wall -frecursive -m64 -mavx2 -o sblat1
>     >>> sblat1.o > ../cygopenblas_haswellp-r0.3.5.a
>     >>> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0
>     >>> >
>     >>>
> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib/../lib
>     >>> > -L/usr/lib/../lib >
>     >>> -L/usr/lib/gcc/x86_64-pc-msys/7.3.0/../../../../x86_64-pc-msys/lib
>     >>> > -L/usr/lib/w32api -lmsys-2.0 >
>     >>>
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001088.o):fake:(.text+0x2):
>     >>> > undefined reference to `__imp_pthread_mutex_destroy' >
>     >>>
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001090.o):fake:(.text+0x2):
>     >>> > undefined reference to `__imp_pthread_mutex_init' >
>     >>>
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001091.o):fake:(.text+0x2):
>     >>> > undefined reference to `__imp_pthread_mutex_lock' >
>     >>>
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001094.o):fake:(.text+0x2):
>     >>> > undefined reference to `__imp_pthread_mutex_trylock' >
>     >>>
> D:/msys64/usr/lib/../lib/libpthread.a(t-d001095.o):fake:(.text+0x2):
>     >>> > undefined reference to `__imp_pthread_mutex_unlock' >
>     >>> collect2.exe: error: ld returned 1 exit status >
>     >>> make[1]: *** [Makefile:134: sblat1] Error 1 > make[1]:
>     >>> Leaving directory >
>     >>> '/home/erinm/OPB_HOME/xianyi-OpenBLAS-eebc189/test' >
>     >>> make: *** [Makefile:124: tests] Error 2
>     >>> >
>     >>> >
>     >>> > I think it has something to do with the
>     >>> threads/pthreads but am not sure > how to fix it.  Any
>     >>> suggestions much appreciated.
>     >>> >
>     >>> > Thanks, > Sincerely, > Erin
>     >>> >
>     >>> > Erin Hodgess, PhD > mailto: [hidden email]
>     >>> >
>     >>> > [[alternative HTML version deleted]]
>     >>> >
>     >>> > ______________________________________________ >
>     >>> [hidden email] mailing list >
>     >>> https://stat.ethz.ch/mailman/listinfo/r-devel
>     >>> >
>     >>>
>     >>> [[alternative HTML version deleted]]
>     >>>
>     >>> ______________________________________________
>     >>> [hidden email] mailing list
>     >>> https://stat.ethz.ch/mailman/listinfo/r-devel
>     >>>
>     >> --
>     >> Sent from Gmail Mobile
>     >>
>
>     >   [[alternative HTML version deleted]]
>
>     > ______________________________________________
>     > [hidden email] mailing list
>     > https://stat.ethz.ch/mailman/listinfo/r-devel
>

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel