Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.

cblas_sgemv_batch_strided threading problem

MicPac
Beginner
311 Views

Hi,

with intel oneApi 2023.0:

 

I try to force cblas_sgemv_batch_strided to run on ONE thread.

I set :

mkl_set_num_threads_local( 1 )

or 
mkl_domain_set_num_threads(1, MKL_DOMAIN_BLAS)

or 

mkl_domain_set_num_threads(1, MKL_DOMAIN_ALL);

 

or both

but I can see a lot of tbb thread in all cases

 

I can see one thread only if I set MKL_THREADING sequential before my cmake find_package

but it's not possible in my use case

 

Is it a bug of that gemv function ?

 

With my best regards,

Michel

0 Kudos
1 Solution
SofeaAzrin_A_Intel
210 Views

Based on your previous comments, you are using the openMP related routine.

mkl_set_num_threads_local( 1 )

or mkl_domain_set_num_threads(1, MKL_DOMAIN_BLAS)


Thus, you need to link with openMP but not with TBB. For further details and understanding, you can refer to our user guide or get started documents.


View solution in original post

0 Kudos
2 Replies
SofeaAzrin_A_Intel
211 Views

Based on your previous comments, you are using the openMP related routine.

mkl_set_num_threads_local( 1 )

or mkl_domain_set_num_threads(1, MKL_DOMAIN_BLAS)


Thus, you need to link with openMP but not with TBB. For further details and understanding, you can refer to our user guide or get started documents.


0 Kudos
SofeaAzrin_A_Intel
169 Views

Hi,


We recommend you to look in TBB runtime design. The function that you mentioned is not possible to do.


Thanks.



0 Kudos
Reply