Thread per point - #10
Draft
MTCam wants to merge 3 commits into
Draft
Conversation
Imported from Theseus-parallelism, branch project/03-parallelism. Source range: 9cd39a4552c8ea558ef672fbe1fd8c8d594b5783..e8a1f82 Original development history: - Extract inviscid interior-face point kernel - Add point-parallel Euler interior faces - Extract Navier-Stokes interior-face point kernels - Add point-parallel Navier-Stokes interior faces - Add point-parallel gradient volume assembly - Add point-parallel viscous volume assembly - Add point-parallel Euler volume assembly - Add point-parallel Navier-Stokes volume assembly - Merge branch 'main' into project/03-parallelism-sync
Contributor
There was a problem hiding this comment.
Pull request overview
Adds optional point-parallel volume and interior-face RHS assembly, alongside timestep performance reporting.
Changes:
- Adds per-point DGSEM kernels for Euler and Navier–Stokes assembly.
- Adds CMake switches and resizes wave-speed caches for point-parallel execution.
- Adds aggregate timestep timing and reporting.
Reviewed changes
Copilot reviewed 8 out of 8 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
CMakeLists.txt |
Adds point-parallel build options and adjusts timer configuration. |
include/DGSEMIntegrator.hpp |
Implements point-level volume, face, viscous, and gradient kernels. |
include/EulerOperator_impl.hpp |
Integrates optional point-parallel Euler assembly. |
include/NSOperator_impl.hpp |
Integrates optional point-parallel Navier–Stokes assembly. |
include/dgsem_cache.hpp |
Documents expanded wave-speed storage. |
include/dgsem_cache_utilities.hpp |
Sizes and validates point-level wave-speed caches. |
include/timer.hpp |
Adds timestep statistics collection and reporting. |
src/Simulation.cpp |
Integrates timestep timing and updates runtime output. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| const double mean_mean = | ||
| mean_sum * inv_nranks; | ||
|
|
||
| os << std::fixed << std::setprecision(3); |
Comment on lines
+25
to
+28
| option(ENABLE_POINT_PARALLEL_INTERIOR_FACES | ||
| "Use one device thread per interior-face point." OFF) | ||
| option(ENABLE_POINT_PARALLEL_VOLUME | ||
| "Use one device thread per element volume point where supported." OFF) |
Comment on lines
+13
to
+17
| #include <algorithm> | ||
| #include <chrono> | ||
| #include <cstdint> | ||
| #include <iomanip> | ||
| #include <iostream> |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This change set adds on thread per point in RHS assembly. The default is one thread per element for volume, one thread per interior face, and one thread per boundary point. If enabled, this change set can optionally change volume to one thread per volume point (duplicating some computation on each thread for two-point fluxing) - and/or optionally change to one thread per interior face point.