R devel - Oct 2023 - About FlexiBLAS in the R-admin docs

On 10/11/23 23:12, George Ostrouchov wrote:
> I agree that the "R Installation and Administration" guide
statement about FlexiBLAS needs an update. Standard R in CentOS Stream 9, which
is now downstream from Fedora, gives the following result for Matrix products in
sessionInfo():
>
>> sessionInfo()
> R version 4.3.1 (2023-06-16)
> Platform: x86_64-redhat-linux-gnu (64-bit)
> Running under: CentOS Stream 9
>
> Matrix products: default
> BLAS/LAPACK: FlexiBLAS OPENBLAS-OPENMP;  LAPACK version 3.9.0
> ?
>
> So OpenBLAS compiled with OpenMP is the default. FlexiBLAS (via the
flexiblas package) is my favorite way to manage core use in multithreaded BLAS
computations. It is also the default in the EasyBuild foss compiler toolchain
for HPC clusters: https://docs.easybuild.io/common-toolchains/
The output of session info is based on that flexiblas is used and on 
what flexiblas tells R is the backend it uses. R does not attempt to 
check that optimized LAPACK functions from the backend really end up 
called via flexiblas, and I don't think it could be realistically checked.

But I've checked one case manually in Fedora 38 using Linux perf tool. 
The following code:

S <- toeplitz((10:1)/10)
repeat { R <- rWishart(10, 20, S) }

uses dpotrf from LAPACK, which is optimized in OpenBLAS and ATLAS and 
the corresponding optimized implementations really appeared on the 
sampling profile for me from the backend libraries.

The comment from R Admin has been removed now and if anyone runs into 
the problem (that an optimized LAPACK function is not called from a 
backend that provides it), it would be best to report it with sufficient 
detail to flexiblas.

Best
Tomas
> What would be a good way to suggest or construct an update?
>
> Best regards,
> George
>
>
>> Date: Wed, 27 Sep 2023 15:14:43 +0200
>> From: =?UTF-8?Q?I=C3=B1aki_Ucar?= <iucar at fedoraproject.org>
>> To: "r-devel at r-project.org" <R-devel at
r-project.org>
>> Subject: [Rd] About FlexiBLAS in the R-admin docs
>> Message-ID:
>> 	<CALEXWq2x9gygnxt6S906BkQHd139CP783bTDQC4wmeM-uKuW3g at
mail.gmail.com>
>> Content-Type: text/plain; charset="utf-8"
>>
>> Hi,
>>
>> Not sure if this is the right place for this. The "R Installation
and
>> Administration" guide states:
>>
>>> Apparently undocumented: FlexiBLAS on Fedora provides a complete
LAPACK, but not the enhanced routines from ATLAS or OpenBLAS.
>> I'm not sure what this means. FlexiBLAS does provide 100% of BLAS
and
>> LAPACK, and if the active backend (say, OpenBLAS) implements an
>> enhanced LAPACK routine, then the call is redirected to the backend.
>> If the user switches to another backend and that routine is not
>> available there, then the original LAPACK routine is dispatched
>> instead.
>>
>> Best,
>> -- 
>> I?aki ?car
>>
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel

On Tue, 24 Oct 2023 at 12:53, Tomas Kalibera <tomas.kalibera at gmail.com>
wrote:
>
> The output of session info is based on that flexiblas is used and on
> what flexiblas tells R is the backend it uses. R does not attempt to
> check that optimized LAPACK functions from the backend really end up
> called via flexiblas, and I don't think it could be realistically
checked.
>
> But I've checked one case manually in Fedora 38 using Linux perf tool.
> The following code:
>
> S <- toeplitz((10:1)/10)
> repeat { R <- rWishart(10, 20, S) }
>
> uses dpotrf from LAPACK, which is optimized in OpenBLAS and ATLAS and
> the corresponding optimized implementations really appeared on the
> sampling profile for me from the backend libraries.
>
> The comment from R Admin has been removed now and if anyone runs into
> the problem (that an optimized LAPACK function is not called from a
> backend that provides it), it would be best to report it with sufficient
> detail to flexiblas.
Thanks, Tomas. Yes, if an expected redirection to an optimized
function does not happen, that would be a bug in FlexiBLAS and should
be reported upstream.

Related to this, a small detail... I noticed that the R configure
script reports "BLAS(FlexiBlas)" correctly as an external library, but
then "LAPACK(generic)". This should be FlexiBLAS too instead of
"generic". It doesn't make any difference, since the LAPACK
symbols in
FlexiBLAS are called anyway, but it's misleading.

Best,
-- 
I?aki ?car