Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

{numlib}[GCC/9.2.0] OpenBLAS v0.3.8 #9852

Merged

Conversation

boegel
Copy link
Member

@boegel boegel commented Feb 10, 2020

(created using eb --new-pr)

@boegel boegel added the update label Feb 10, 2020
@boegel
Copy link
Member Author

boegel commented Feb 10, 2020

Test report by @boegel
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
generoso - Linux centos linux 7.6.1810, Intel(R) Xeon(R) CPU E5-2660 v4 @ 2.00GHz, Python 3.6.8
See https://gist.github.com/9115aebdd6b362ec3d693138811773df for a full test report.

@lexming
Copy link
Contributor

lexming commented Feb 10, 2020

Test report by @lexming
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
login2.cerberus.os - Linux centos linux 7.7.1908, Intel(R) Xeon(R) Gold 6126 CPU @ 2.60GHz, Python 2.7.5
See https://gist.github.com/bd720531fa5dfee6a1e234fcae63037e for a full test report.

@branfosj
Copy link
Member

Test report by @branfosj
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
bear-pg0305u05a.bear.cluster - Linux RHEL 7.6, 8335-GTX, Python 2.7.5
See https://gist.github.com/0c96fd40e369d814f4bb33a102e069ff for a full test report.

@branfosj
Copy link
Member

Test report by @branfosj
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
bear-pg0211u03a.bear.cluster - Linux centos linux 7.7.1908, Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz, Python 2.7.5
See https://gist.github.com/f26c6ca29dba97794704d919ca66c43a for a full test report.

@boegel boegel added this to the next release (4.1.2?) milestone Feb 10, 2020
@verdurin
Copy link
Member

Test report by @verdurin
SUCCESS
Build succeeded for 0 out of 0 (2 easyconfigs in this PR)
epyc000.hpc.in.bmrc.ox.ac.uk - Linux centos linux 7.6.1810, AMD EPYC 7502 32-Core Processor, Python 2.7.5
See https://gist.github.com/ca907fed4504c1e8620af407342ac284 for a full test report.

@verdurin
Copy link
Member

Test report by @verdurin
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
epyc000.hpc.in.bmrc.ox.ac.uk - Linux centos linux 7.6.1810, AMD EPYC 7502 32-Core Processor, Python 2.7.5
See https://gist.github.com/a830b1a57261862a1459e8521cba36ce for a full test report.

@Micket
Copy link
Contributor

Micket commented Feb 10, 2020

Test report by @Micket
SUCCESS
Build succeeded for 9 out of 9 (2 easyconfigs in this PR)
hebbe-c1 - Linux centos linux 7.7.1908, Intel Core Processor (Haswell, no TSX), Python 2.7.5
See https://gist.github.com/2efc7c837e9f6c7f05d2bc5b1d134612 for a full test report.

@boegel
Copy link
Member Author

boegel commented Feb 10, 2020

Based on the test reports, I think this is good to go.

@migueldiascosta Can you look into another test report from you, and then review/merge?

@verdurin
Copy link
Member

Test report by @verdurin
SUCCESS
Build succeeded for 9 out of 9 (2 easyconfigs in this PR)
rescomp2.hpc.in.bmrc.ox.ac.uk - Linux centos linux 7.6.1810, Intel(R) Xeon(R) Gold 6126 CPU @ 2.60GHz, Python 2.7.5
See https://gist.github.com/79477ac27a2ef406f2d6105aed2ef470 for a full test report.

@Micket
Copy link
Contributor

Micket commented Feb 10, 2020

Test report by @Micket
SUCCESS
Build succeeded for 9 out of 9 (2 easyconfigs in this PR)
vera-c1 - Linux centos linux 7.7.1908, Intel Xeon Processor (Skylake), Python 2.7.5
See https://gist.github.com/6d2300d1e8ae680605de587a517b80af for a full test report.

@orbsmiv
Copy link
Contributor

orbsmiv commented Feb 10, 2020

Test report by @orbsmiv
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
bear-pg0305u05a.bear.cluster - Linux RHEL 7.6, 8335-GTX, Python 2.7.5
See https://gist.github.com/dbb0d4b2f68b23521cd2b1cba8cc592c for a full test report.

@zao
Copy link
Contributor

zao commented Feb 10, 2020

Test report by @zao
SUCCESS
Build succeeded for 9 out of 9 (2 easyconfigs in this PR)
freja - Linux ubuntu 18.04, Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz, Python 2.7.17
See https://gist.github.com/c4d66b4965b518ba75374d94e01a5241 for a full test report.

@branfosj
Copy link
Member

Test report by @branfosj
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
bb2c1a09.bb2.cluster - Linux centos linux 7.7.1908, Intel(R) Xeon(R) CPU E5-2660 0 @ 2.20GHz, Python 2.7.5
See https://gist.github.com/ed97a3137ef0c34daaf382df96555c75 for a full test report.

@akesandgren
Copy link
Contributor

Test report by @akesandgren
SUCCESS
Build succeeded for 9 out of 9 (2 easyconfigs in this PR)
b-an03.hpc2n.umu.se - Linux ubuntu 16.04, Intel(R) Xeon(R) CPU E5-2690 v4 @ 2.60GHz, Python 2.7.12
See https://gist.github.com/a0b0c9153d3654ac29711ce831069883 for a full test report.

@terjekv
Copy link
Collaborator

terjekv commented Feb 10, 2020

Test report by @terjekv
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
ninhursaga.uio.no - Linux red hat enterprise linux 8.1, Intel(R) Xeon(R) Silver 4216 CPU @ 2.10GHz, Python 3.6.8
See https://gist.github.com/1e22d4e1d89f8b5387be2b24de303772 for a full test report.

@terjekv
Copy link
Collaborator

terjekv commented Feb 10, 2020

Builds fine, but is still affected by #9754

@migueldiascosta
Copy link
Member

Test report by @migueldiascosta
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
sms - Linux centos linux 7.6.1810, AMD EPYC 7601 32-Core Processor, Python 2.7.5
See https://gist.github.com/5bd2309a9ef83de5f638aaf8aa01b674 for a full test report.

Copy link
Member

@migueldiascosta migueldiascosta left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

@migueldiascosta
Copy link
Member

Going in, thanks @boegel!

@migueldiascosta migueldiascosta merged commit ccd1065 into easybuilders:develop Feb 11, 2020
@akesandgren
Copy link
Contributor

Test report by @akesandgren
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
b-cn1404.hpc2n.umu.se - Linux ubuntu 16.04, Intel(R) Xeon(R) Gold 6132 CPU @ 2.60GHz, Python 2.7.12
See https://gist.github.com/885410b91aedbe59ea45994cb4b34736 for a full test report.

@casparvl
Copy link
Contributor

As @lexming reported in the maintainers channel, there is a problem with this build on Skylake nodes.

The EasyBuild installation completes succesfully, however:

== 2020-02-11 14:53:07,807 run.py:589 INFO parse_log_for_error (some may be harmless) regExp (?<![(,-]|\w)(?:error|segmentation fault|failed)(?![(,-]|\.?\w) found:
 2049 |        printf("error code=%d,\trelease->address=%lx\n",errsv,release->address);
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DGEMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DGBMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DSYMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DSBMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DSPMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DGEMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DGBMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DSYMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DSBMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* DSPMV  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
 ******* cblas_dgemv  FAILED ON CALL NUMBER:
 ******* cblas_dgemv  FAILED ON CALL NUMBER:
 ******* FATAL ERROR - TESTS ABANDONED *******
== 2020-02-11 14:53:07,807 run.py:546 WARNING Found 25 errors in command output (output:  2049 |        printf("error code=%d,\trelease->address=%lx\n",errsv,release->address);
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DGEMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DGBMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSYMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSBMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSPMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DGEMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DGBMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSYMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSBMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSPMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* cblas_dgemv  FAILED ON CALL NUMBER:
         ******* cblas_dgemv  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - TESTS ABANDONED *******)
== 2020-02-11 14:53:07,807 run.py:546 WARNING Found 25 errors in command output (output:  2049 |        printf("error code=%d,\trelease->address=%lx\n",errsv,release->address);
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DGEMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DGBMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSYMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSBMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSPMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DGEMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DGBMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSYMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSBMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* DSPMV  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE *******
         ******* cblas_dgemv  FAILED ON CALL NUMBER:
         ******* cblas_dgemv  FAILED ON CALL NUMBER:
         ******* FATAL ERROR - TESTS ABANDONED *******)

This was on a Intel(R) Xeon(R) Silver 4110 CPU @ 2.10GHz optimizing for AVX-512. In other words: we cannot rely on this build in its current state...

Some action points...

  • Fix EasyBuild (probably: the OpenBLAS EasyBlock) to make sure the install doesn't complete succesfully for these cases...
  • Try and fix the build
  • Contact OpenBLAS devs

@akesandgren
Copy link
Contributor

I also get
FATAL ERROR - COMPUTED RESULT IS LESS THAN HALF ACCURATE
on Skylakes, which doesn't cause errors in EB

@akesandgren
Copy link
Contributor

See OpenMathLib/OpenBLAS#2408

@branfosj
Copy link
Member

I'm seeing the same erros on our CascadeLake nodes. (Sandybridge, Haswell, and Power pass the tests.)

@lexming
Copy link
Contributor

lexming commented Feb 11, 2020

Quick fix for OpenMathLib/OpenBLAS#2408 if you need to deploy this asap is to add toolchainopts = {'vectorize': False} to the easyconfig OpenBLAS-0.3.8-GCC-9.2.0.eb

@akesandgren
Copy link
Contributor

akesandgren commented Feb 11, 2020

@lexming can you make a PR for this? (vec: False...)

@lexming
Copy link
Contributor

lexming commented Feb 11, 2020

@akesandgren sure, but first we should fix the openblas easyblock to properly check such errors. Working on it.

@boegel
Copy link
Member Author

boegel commented Feb 12, 2020

should be fixed with #9867...

@Flamefire
Copy link
Contributor

Test report by @Flamefire
SUCCESS
Build succeeded for 9 out of 9 (2 easyconfigs in this PR)
taurusml2 - Linux RHEL 7.6, 8335-GTX, Python 2.7.5
See https://gist.github.com/ed0590e8e91c2f1c42e2696b17c75573 for a full test report.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.