-
Notifications
You must be signed in to change notification settings - Fork 576
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
Tpetra: device version of sort_and_merge_crs #11751
Tpetra: device version of sort_and_merge_crs #11751
Conversation
Status Flag 'Pre-Test Inspection' - Auto Inspected - Inspection Is Not Necessary for this Pull Request. |
Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects: Pull Request Auto Testing STARTING (click to expand)Build InformationTest Name: Trilinos_PR_gcc-8.3.0
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-serial
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-debug
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_clang-11.0.1
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_python3
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_cuda-11.4.2-uvm-off
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_intel-2021.3
Jenkins Parameters
Using Repos:
Pull Request Author: lucbv |
Status Flag 'Pull Request AutoTester' - Jenkins Testing: 1 or more Jobs FAILED Note: Testing will normally be attempted again in approx. 2 Hrs 30 Mins. If a change to the PR source branch occurs, the testing will be attempted again on next available autotester run. Pull Request Auto Testing has FAILED (click to expand)Build InformationTest Name: Trilinos_PR_gcc-8.3.0
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-serial
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-debug
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_clang-11.0.1
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_python3
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_cuda-11.4.2-uvm-off
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_intel-2021.3
Jenkins Parameters
|
44f4301
to
6a77a08
Compare
@@ -712,6 +719,19 @@ sortAndMergeCrsEntries (const Teuchos::ArrayView<size_t> &CRS_rowptr, | |||
return sortAndMergeCrsEntries(CRS_rowptr, CRS_colind, CRS_vals); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why sort and merge dummy values here? There is a sort_and_merge_graph
which doesn't have values as input/output.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I am not quite sure, but this can be updated eventually, at this point it seems to be a host only view based function.
Not too sure who wanted it that way, we can make an issue about that and fix it after this PR gets merged.
6a77a08
to
b2a78ae
Compare
Status Flag 'Pull Request AutoTester' - User Requested Retest - Label AT: RETEST will be reset after testing. |
Status Flag 'Pre-Test Inspection' - Auto Inspected - Inspection is Not Necessary for this Pull Request. |
Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects: Pull Request Auto Testing STARTING (click to expand)Build InformationTest Name: Trilinos_PR_gcc-8.3.0
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-serial
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-debug
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_clang-11.0.1
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_python3
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_cuda-11.4.2-uvm-off
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_intel-2021.3
Jenkins Parameters
Using Repos:
Pull Request Author: lucbv |
Status Flag 'Pull Request AutoTester' - Jenkins Testing: 1 or more Jobs FAILED Note: Testing will normally be attempted again in approx. 2 Hrs 30 Mins. If a change to the PR source branch occurs, the testing will be attempted again on next available autotester run. Pull Request Auto Testing has FAILED (click to expand)Build InformationTest Name: Trilinos_PR_gcc-8.3.0
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-serial
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-debug
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_clang-11.0.1
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_python3
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_cuda-11.4.2-uvm-off
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_intel-2021.3
Jenkins Parameters
|
@sebrowne No cdash results show up when I click the link above. |
Excellent catch, the new autotester is sending a link for testing.sandia.gov/cdash instead of trilinos-cdash.sandia.gov I'll ask SEMS to fix that, must have missed that in the autotester configurations. |
b4e1645
to
c6d5791
Compare
In transfer and fill complete, the sort of crs matrix data is currently done on host using ArrayRCP, we are moving this call to device and storing data in View. Data is subsequently copied and moved back to host and ArrayRCP while other routines are being moved on device as well until we can skip the use of ArrayRCP all together.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
lgtm
fixing typo in comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
thanks!
Status Flag 'Pull Request AutoTester' - User Requested Retest - Label AT: RETEST will be reset after testing. |
Status Flag 'Pre-Test Inspection' - Auto Inspected - Inspection is Not Necessary for this Pull Request. |
Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects: Pull Request Auto Testing STARTING (click to expand)Build InformationTest Name: Trilinos_PR_gcc-8.3.0
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-serial
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-debug
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_clang-11.0.1
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_python3
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_cuda-11.4.2-uvm-off
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_intel-2021.3
Jenkins Parameters
Using Repos:
Pull Request Author: lucbv |
Status Flag 'Pull Request AutoTester' - User Requested Retest - Label AT: RETEST will be reset after testing. |
Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects: Pull Request Auto Testing STARTING (click to expand)Build InformationTest Name: Trilinos_PR_gcc-8.3.0
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-serial
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-debug
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_clang-11.0.1
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_python3
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_cuda-11.4.2-uvm-off
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_intel-2021.3
Jenkins Parameters
Using Repos:
Pull Request Author: lucbv |
Status Flag 'Pull Request AutoTester' - Jenkins Testing: all Jobs PASSED Pull Request Auto Testing has PASSED (click to expand)Build InformationTest Name: Trilinos_PR_gcc-8.3.0
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-serial
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_gcc-8.3.0-debug
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_clang-11.0.1
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_python3
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_cuda-11.4.2-uvm-off
Jenkins Parameters
Build InformationTest Name: Trilinos_PR_intel-2021.3
Jenkins Parameters
|
Status Flag 'Pre-Merge Inspection' - SUCCESS: The last commit to this Pull Request has been INSPECTED AND APPROVED by [ jhux2 ]! |
Status Flag 'Pull Request AutoTester' - Pull Request will be Automerged |
Merge on Pull Request# 11751: IS A SUCCESS - Pull Request successfully merged |
After the refactor done in PR trilinos#11751, these new changes remove the copy to host after sorting the CRS matrix and instead just call the setAllValues overload that accepts Kokkos Views. Should improve performance a bit and this provides a step toward completing issue trilinos#11689.
After the refactor done in PR trilinos#11751, these new changes remove the copy to host after sorting the CRS matrix and instead just call the setAllValues overload that accepts Kokkos Views. Should improve performance a bit and this provides a step toward completing issue trilinos#11689.
This should increase the speed of MueLu by a nice amount on GPU platforms : )
@trilinos/tpetra
Motivation
There is no reason to not perform the sort and merge of csr entries on device.
This code should increase the performance of Tpetra and MueLu on GPU architectures.
Related Issues
Stakeholder Feedback
Testing
Tested on Crusher