[QE-users] information about PGI compiler v20+ download

Thu Nov 5 14:09:54 CET 2020

Dear Louis and Pierto,

Thanks for your effort in this... Infact, I tried with cuda 10.2 and with
the version offered by Peirto:
https://gitlab.com/QEF/q-e-gpu/-/archive/hotfix/q-e-gpu-hotfix.tar.bz2

$ ./install/configure CC=pgcc F77=pgf90 FC=pgf90 F90=pgf90 MPIF90=mpif90
--with-cuda=/apps/compilers/pgi/2020/Linux_x86_64/20.9/cuda/
--with-cuda-runtime=10.2 --with-cuda-cc=70 --enable-openmp
--with-scalapack=no
LIBS=-L/apps/compilers/pgi/2020/Linux_x86_64/20.9/cuda//lib64
-L/apps/compilers/pgi/2020/Linux_x86_64/20.9/cuda//lib64/stubs/

Both compilation and make pw are successfully installed. But now it seems
that when I try to run the job, the binary doesn't get loaded on the gpu
processor, so no output/report file is generated.
After scratching my head for a long time, eventually, I came down to 6.1
version, which is now running smoothly, no problem...

But I will try on the 6.6a1 and will let you know soon.

Thanks!
Sitangshu

On Mon, Nov 2, 2020 at 12:22 PM Sitangshu Bhattacharya <
sitangshu at iiita.ac.in> wrote:

> Hi Paulo and Louis,
> Thanks for your advice, I got the config done with this!:
>
> module load codes/cuda/10.2
> module load compilers/pgi/2020
> export CUDADIR=/apps/compilers/pgi/2020/Linux_x86_64/20.9/cuda/
> ./configure CC=pgcc F77=pgf90 FC=pgf90 F90=pgf90 MPIF90=mpif90
> --with-cuda=$CUDADIR --with-cuda-runtime=10.2 --with-cuda-cc=70
> --enable-openmp --with-scalapack=no LIBS="-L$CUDADIR/lib64/
> -L$CUDADIR/lib64/stubs/"
>
> However, while doing the make pw, I am getting the following error:
>
> NVFORTRAN-S-0519-More than one device-resident object in assignment
> (device_util_tmp.f90: 3924)
> NVFORTRAN-S-0519-More than one device-resident object in assignment
> (device_util_tmp.f90: 3932)
> NVFORTRAN-S-0519-More than one device-resident object in assignment
> (device_util_tmp.f90: 3932)
> NVFORTRAN-S-0519-More than one device-resident object in assignment
> (device_util_tmp.f90: 3932)
> NVFORTRAN-S-0519-More than one device-resident object in assignment
> (device_util_tmp.f90: 3942)
> NVFORTRAN-S-0519-More than one device-resident object in assignment
> (device_util_tmp.f90: 3950)
>   0 inform,   0 warnings,  14 severes, 0 fatal for sp_dev_mat_upd_dmd_c2d
> make[1]: *** [device_util.o] Error 2
> make[1]: Leaving directory
> `/data_hpc/home/sitanshu/softwares/q-e-gpu-qe-gpu-6.6a2/UtilXlib'
> make: *** [libutil] Error 1
>
> Please suggest...
>
> With regards,
> Sitangshu
>
>
>
>
>
>
> On Thu, Oct 29, 2020 at 11:13 AM Sitangshu Bhattacharya <
> sitangshu at iiita.ac.in> wrote:
>
>> Dear Louis,
>>
>> Thanks for your suggestion. I made the changes as you suggested, but some
>> other compiling error in line 92 is constantly appearing.
>> I am attaching the config.log file for you to check please...
>>
>> Regards,
>> Sitangshu
>>
>> On Tue, Oct 20, 2020 at 9:06 PM Sitangshu Bhattacharya <
>> sitangshu at iiita.ac.in> wrote:
>>
>>> Dear Louis, many thanks for your help, I am attaching the log file...
>>>
>>> Regards
>>> Sitangshu
>>>
>>> On Tue, Oct 20, 2020 at 8:33 PM Sitangshu Bhattacharya <
>>> sitangshu at iiita.ac.in> wrote:
>>>
>>>> Hi Peirto, I tried to run the QE config file after installing the pgi
>>>> compiler (20.9). It seems that pgifortran issue is now solved, but stuck
>>>> with libcuda:
>>>> Please let me know...
>>>> Thanks,
>>>> Sitangshu
>>>>
>>>> [root at agni q-e-gpu-qe-gpu-6.6a2]# ./configure CC=pgcc F77=pgf90 FC=pgf90 F90=pgf90 MPIF90=mpif90 --with-cuda=/usr/local/cuda/bin --with-cuda-runtime=9.2 --with-cuda-cc=70 --enable-openmp
>>>> checking build system type... x86_64-pc-linux-gnu
>>>> checking ARCH... x86_64
>>>> checking setting AR... ... ar
>>>> checking setting ARFLAGS... ... ruv
>>>> checking whether the Fortran compiler works... yes
>>>> checking for Fortran compiler default output file name... a.out
>>>> checking for suffix of executables...
>>>> checking whether we are cross compiling... no
>>>> checking for suffix of object files... o
>>>> checking whether we are using the GNU Fortran compiler... no
>>>> checking whether pgf90 accepts -g... yes
>>>> configure: WARNING: F90 value is set to be consistent with value of MPIF90
>>>> checking for mpif90... mpif90
>>>> checking whether we are using the GNU Fortran compiler... no
>>>> checking whether mpif90 accepts -g... yes
>>>> checking version of mpif90... nvfortran 20.9-0
>>>> checking for Fortran flag to compile .f90 files... none
>>>> setting F90... nvfortran
>>>> setting MPIF90... mpif90
>>>> checking whether we are using the GNU C compiler... yes
>>>> checking whether pgcc accepts -g... yes
>>>> checking for pgcc option to accept ISO C89... none needed
>>>> setting CC... pgcc
>>>> setting CFLAGS... -fast -Mpreprocess
>>>> checking whether we are using the GNU Fortran 77 compiler... no
>>>> checking whether pgf90 accepts -g... yes
>>>> setting F77... pgf90
>>>> using F90... nvfortran
>>>> setting FFLAGS... -O1
>>>> setting F90FLAGS... $(FFLAGS)
>>>> setting FFLAGS_NOOPT... -O0
>>>> setting CPP... cpp
>>>> setting CPPFLAGS... -P -traditional -Uvector
>>>> setting LD... mpif90
>>>> setting LDFLAGS...
>>>> checking for Fortran flag to compile .f90 files... (cached) none
>>>> checking whether Fortran compiler accepts -Mcuda... yes
>>>> checking for nvcc... /apps/compilers/pgi/2020/Linux_x86_64/20.9/compilers/bin/nvcc
>>>> checking whether nvcc works... no
>>>> configure: WARNING: CUDA compiler has problems.
>>>> checking for cuInit in -lcuda... no
>>>> configure: error: in `/data_hpc/home/sitanshu/softwares/q-e-gpu-qe-gpu-6.6a2':
>>>> configure: error: Couldn't find libcuda
>>>>
>>>>
>>>> On Sun, Oct 18, 2020 at 12:50 PM Sitangshu Bhattacharya <
>>>> sitangshu at iiita.ac.in> wrote:
>>>>
>>>>> Thank you Peitro, but the link only tells about nvhpc. I couldn't get
>>>>> the pgi compilers there...
>>>>>
>>>>> Regards,
>>>>> Sitangshu
>>>>>
>>>>> On Sun, Oct 18, 2020 at 12:14 PM Sitangshu Bhattacharya <
>>>>> sitangshu at iiita.ac.in> wrote:
>>>>>
>>>>>> Dear all,
>>>>>>
>>>>>> I am in search of a pgi compiler V20+ to run QE on a V100 GPU. I
>>>>>> looked in the nvidia and pgi websites, but it seems some sort of tie up is
>>>>>> needed between the guest and pgi organizations. As this may take some time,
>>>>>> may I know if there are any quick download links...?
>>>>>>
>>>>>>
>>>>>> Regards,
>>>>>> **********************************************
>>>>>> Sitangshu Bhattacharya (সিতাংশু ভট্টাচার্য), Ph.D
>>>>>> Assistant Professor,
>>>>>> Room No. 2221, CC-1,
>>>>>> Electronic Structure Theory Group,
>>>>>> Department of Electronics and Communication Engineering,
>>>>>> Indian Institute of Information Technology-Allahabad
>>>>>> Uttar Pradesh 211 012
>>>>>> India
>>>>>> Telephone: 91-532-2922000 Extn.: 2131
>>>>>> Web-page: http://profile.iiita.ac.in/sitangshu/
>>>>>> Institute: http://www.iiita.ac.in/
>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> **********************************************
>>>>> Sitangshu Bhattacharya (সিতাংশু ভট্টাচার্য), Ph.D
>>>>> Assistant Professor,
>>>>> Room No. 2221, CC-1,
>>>>> Electronic Structure Theory Group,
>>>>> Department of Electronics and Communication Engineering,
>>>>> Indian Institute of Information Technology-Allahabad
>>>>> Uttar Pradesh 211 012
>>>>> India
>>>>> Telephone: 91-532-2922000 Extn.: 2131
>>>>> Web-page: http://profile.iiita.ac.in/sitangshu/
>>>>> Institute: http://www.iiita.ac.in/
>>>>>
>>>>>
>>>>
>>>> --
>>>> **********************************************
>>>> Sitangshu Bhattacharya (সিতাংশু ভট্টাচার্য), Ph.D
>>>> Assistant Professor,
>>>> Room No. 2221, CC-1,
>>>> Electronic Structure Theory Group,
>>>> Department of Electronics and Communication Engineering,
>>>> Indian Institute of Information Technology-Allahabad
>>>> Uttar Pradesh 211 012
>>>> India
>>>> Telephone: 91-532-2922000 Extn.: 2131
>>>> Web-page: http://profile.iiita.ac.in/sitangshu/
>>>> Institute: http://www.iiita.ac.in/
>>>>
>>>>
>>>
>>> --
>>> **********************************************
>>> Sitangshu Bhattacharya (সিতাংশু ভট্টাচার্য), Ph.D
>>> Assistant Professor,
>>> Room No. 2221, CC-1,
>>> Electronic Structure Theory Group,
>>> Department of Electronics and Communication Engineering,
>>> Indian Institute of Information Technology-Allahabad
>>> Uttar Pradesh 211 012
>>> India
>>> Telephone: 91-532-2922000 Extn.: 2131
>>> Web-page: http://profile.iiita.ac.in/sitangshu/
>>> Institute: http://www.iiita.ac.in/
>>>
>>>
>>
>> --
>> **********************************************
>> Sitangshu Bhattacharya (সিতাংশু ভট্টাচার্য), Ph.D
>> Assistant Professor,
>> Room No. 2221, CC-1,
>> Electronic Structure Theory Group,
>> Department of Electronics and Communication Engineering,
>> Indian Institute of Information Technology-Allahabad
>> Uttar Pradesh 211 012
>> India
>> Telephone: 91-532-2922000 Extn.: 2131
>> Web-page: http://profile.iiita.ac.in/sitangshu/
>> Institute: http://www.iiita.ac.in/
>>
>>
>
> --
> **********************************************
> Sitangshu Bhattacharya (সিতাংশু ভট্টাচার্য), Ph.D
> Assistant Professor,
> Room No. 2221, CC-1,
> Electronic Structure Theory Group,
> Department of Electronics and Communication Engineering,
> Indian Institute of Information Technology-Allahabad
> Uttar Pradesh 211 012
> India
> Telephone: 91-532-2922000 Extn.: 2131
> Web-page: http://profile.iiita.ac.in/sitangshu/
> Institute: http://www.iiita.ac.in/
>
>

-- 
**********************************************
Sitangshu Bhattacharya (সিতাংশু ভট্টাচার্য), Ph.D
Assistant Professor,
Room No. 2221, CC-1,
Electronic Structure Theory Group,
Department of Electronics and Communication Engineering,
Indian Institute of Information Technology-Allahabad
Uttar Pradesh 211 012
India
Telephone: 91-532-2922000 Extn.: 2131
Web-page: http://profile.iiita.ac.in/sitangshu/
Institute: http://www.iiita.ac.in/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20201105/817daa5c/attachment.html>