Thank you, Barry, Satish, and Matt, for all your answers. Per your feedbacks, 
we will pursue using MKL as the provider of blas/lapack for PETSc. If we 
continue to have issues, I will contact you via petsc-maint.

I hope you are all doing well also.

Very best,

Ernesto.



Schlumberger-Private
From: Barry Smith <[email protected]>
Sent: Wednesday, March 16, 2022 11:15 AM
To: Ernesto Prudencio <[email protected]>
Cc: PETSc users list <[email protected]>
Subject: [Ext] Re: [petsc-users] Two simple questions on building


  Hi Ernesto, I hope you are doing well.

  I agree with Satish. It would be best to resolve the issues with the pure MKL 
approach. Any "hack" that got the libraries to work mixing fblaslapack and MKL 
would be fragile and untrustworthy.

  Feel free to email to [email protected]<mailto:[email protected]> 
the configure.log make.log and failure information in the pure MKL approach so 
we can take a look at it. Since you doing this in a controlled environment 
presumably we can even reproduce the problem with enough information on your 
build process and track down the underlying cause.

  Barry


On Mar 16, 2022, at 1:03 AM, Ernesto Prudencio via petsc-users 
<[email protected]<mailto:[email protected]>> wrote:

Hi.

I have an application that uses MKL for some convolution operations. Such MKL 
functionality uses, I suppose, BLAS/LAPACK underneath.

This same application of mine also uses PETSc for other purposes. I can supply 
blas and lapack to PETSc in two ways:

  1.  Using the configuration 
option--with-blaslapack-lib="-L${MKL_DIR}/lib/intel64 -lfile1 -lfile2 ... ".  
For reasons related to compilation environments + docker images + cloud, I am 
having issues with this option (a) _after_ PETSc builds successfully (both make 
and make install work fine).
  2.  Using the configuration option --download-fblaslapack=yes. This options 
works fine for the purpose of generating my application executable.

If I use option (b), I understand that I will have two different blas/lapack 
codes available during the execution of my application: one from MKL, the other 
being the one that PETSc downloads during its configuration.

Question 1) Do you foresee any potential run time issue with option (b)?

Question 2) In the case PETSc, is there any problem if run "make" and "make 
install" without specifying PETSC_ARCH?

Thank you in advance,

Ernesto.

Schlumberger-Private

Reply via email to