KA kernels for 3D P4est #2590
Draft
KA kernels for 3D P4est #2590
Conversation
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
In order to eventually support GPU computation we need
to use Adapt.jl to allow GPU backend packages to swap
out host-array types like `CuArray` with device-side types
like `CuDeviceArray`.
Additionally this will allow us to change the element type
of a simulation by using `adapt(Array{Float32}`.
Co-authored-by: Lars Christmann <[email protected]>
Co-authored-by: Benedict Geihe <[email protected]>
In order to eventually support GPU computation we need
to use Adapt.jl to allow GPU backend packages to swap
out host-array types like `CuArray` with device-side types
like `CuDeviceArray`.
Additionally this will allow us to change the element type
of a simulation by using `adapt(Array{Float32}`.
Co-authored-by: Lars Christmann <[email protected]>
Co-authored-by: Benedict Geihe <[email protected]>
vchuravy
reviewed
Oct 2, 2025
vchuravy
reviewed
Oct 2, 2025
vchuravy
reviewed
Oct 2, 2025
vchuravy
reviewed
Oct 2, 2025
| max_lambda1 + max_lambda2 + max_lambda3) | ||
| end | ||
| # TODO GPU dt on CPU? (time integration happens on CPU) | ||
| max_scaled_speed = max(nextfloat(zero(t)), maximum(max_scaled_speeds)) |
Member
There was a problem hiding this comment.
Can we write this as a mapreduce? E.g. something like:
AcceleratedKernels.mapreduce(max, 1:num_elements, backend) do element
max_scaled_speed_element(u, meshT, equations, dg,
contravariant_vectors,
inverse_jacobian,
element)
end
AcceleratedKernels does use Polyester for it's mapreduce on the CPU IIRC (grrml) so we could test the performance against @batch, another alternative is tmapreduce from OhMyThreads.
commit f4bbcd9 Author: Daniel Doehring <[email protected]> Date: Sun Oct 5 08:33:38 2025 +0200 Comment `temperature` and /3 (#2594) --------- Co-authored-by: Hendrik Ranocha <[email protected]> commit 68c0c71 Author: Daniel Doehring <[email protected]> Date: Fri Oct 3 15:06:55 2025 +0200 Second-Order Finite Volume Integral in 1D (#2022) * Pick up where Gregor left * preliminary example * more limiters * comments * fmt * continue * comments * print some more info * Add unit tests * add comment * Remove some alternative limiter implementations. * move, comments, fmt * Use second order timestepping * debug superbee * prim2cons 1D Adv * test * fmt, typo * typos * some more tests * fmt * Update src/solvers/dgsem_tree/finite_volume_O2.jl * Update test/test_unit.jl * Update src/solvers/dgsem_tree/dg_1d.jl * fmt * add different recontruction mode * Update src/solvers/dgsem_tree/finite_volume_O2.jl Co-authored-by: Andrés Rueda-Ramírez <[email protected]> * test + fmt * comments * correct way cells dim * increase coverage * revisit * continue * fmt * shorten * extra test * comment "inverse_weights" * change files * test vals * Update test/test_tree_1d_euler.jl * Update examples/structured_1d_dgsem/elixir_euler_source_terms_nonperiodic_fvO2.jl * Update examples/tree_1d_dgsem/elixir_euler_convergence_pure_fvO2.jl * Update test/test_tree_1d_euler.jl * fix * test compact print * comment * relabel * comments * comments * comemnts * commenbts * rm * test * rename * docstrings * comments * Apply suggestions from code review Co-authored-by: Joshua Lampert <[email protected]> * fmt * fmt * mv * fix * Apply suggestions from code review Co-authored-by: Joshua Lampert <[email protected]> --------- Co-authored-by: Andrés Rueda-Ramírez <[email protected]> Co-authored-by: Joshua Lampert <[email protected]> commit 96c7aef Author: Daniel Doehring <[email protected]> Date: Thu Oct 2 18:19:13 2025 +0200 Bundle identical `rhs!` (#2552) * Bundle identical `rhs!` * fix 1d * comment * bring back --------- Co-authored-by: Hendrik Ranocha <[email protected]> commit 5c97803 Author: Daniel Doehring <[email protected]> Date: Thu Oct 2 15:39:55 2025 +0200 Use variable name `have_nonconservative_terms` (#2592) * Use variable name `have_nonconservative_terms` * fix * cons fmt --------- Co-authored-by: Benedict <[email protected]> Co-authored-by: Hendrik Ranocha <[email protected]> commit 2688623 Author: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Date: Thu Oct 2 08:10:10 2025 +0200 Bump crate-ci/typos from 1.35.7 to 1.37.1 (#2593) * Bump crate-ci/typos from 1.35.7 to 1.37.1 Bumps [crate-ci/typos](https://github.com/crate-ci/typos) from 1.35.7 to 1.37.1. - [Release notes](https://github.com/crate-ci/typos/releases) - [Changelog](https://github.com/crate-ci/typos/blob/master/CHANGELOG.md) - [Commits](crate-ci/typos@v1.35.7...v1.37.1) --- updated-dependencies: - dependency-name: crate-ci/typos dependency-version: 1.37.1 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <[email protected]> * fix typos --------- Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Joshua Lampert <[email protected]>
Co-authored-by: Valentin Churavy <[email protected]>
…rixi.jl into feature-gpu-offloading
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
vchuravy
reviewed
Dec 3, 2025
| index_range, interface) | ||
| end | ||
|
|
||
| function calc_interface_flux_interface!(surface_flux_values, |
Member
There was a problem hiding this comment.
Check with @device_code but you likely must manually annotate functions like this or the callsite above with @inline
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.
Add KernelAbstractions.jl kernels for basic rhs! functions for 3D P4estMesh implementation.