Skip to content

Minimal GPU example which preserves the element routine #1291

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.

Sign up for GitHub

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

Open
termi-official wants to merge 45 commits into
base: master
Choose a base branch
from
Open

Minimal GPU example which preserves the element routine #1291

Show file tree
Hide file tree
Changes from 23 commits
Commits
Show all changes
45 commits
Select commit Hold shift + click to select a range
a7ba0c1
Minimal assembly example
termi-official Feb 27, 2026
489155a
Add minimal constraint handler
termi-official Feb 27, 2026
f3af4b9
Add actual subdofhandler
termi-official Feb 27, 2026
1311ef4
Add some brief docs and group functions. Also rename some stuff to be…
termi-official Feb 27, 2026
a92356b
Fix some more subtle Int issues and move most of the stuff into an ex…
termi-official Feb 27, 2026
2ce407b
Move heat assembly to tests
termi-official Feb 27, 2026
6442941
Runic
termi-official Feb 27, 2026
e0e1e84
Missing file.
termi-official Feb 27, 2026
b12d500
Remove StaticArrays dep
termi-official Feb 27, 2026
6826478
Fix trailing T
termi-official Feb 27, 2026
e687709
Remove self-qualification
termi-official Feb 27, 2026
f00dbb8
Typo
termi-official Feb 27, 2026
ff8b59a
Remove dynamic dispatch from error message generation
termi-official Feb 27, 2026
dc9954c
Define proper reinit for GPUCellCache
termi-official Feb 27, 2026
60a23c8
Make SOA transformation explicit.
termi-official Feb 28, 2026
ae09f6b
Try to trick CodeCov by using KA CPU.
termi-official Feb 28, 2026
12879fb
Runic
termi-official Feb 28, 2026
cf135fb
Partial recovery of detJ error
termi-official Feb 28, 2026
545a317
Revert to global memory usage and make sure the tests work.
termi-official Mar 3, 2026
04ff942
Fix and comment how-to for naive GPU assembly. For now this how-to is…
termi-official Mar 4, 2026
8467898
Remove old materialize fun
termi-official Mar 4, 2026
f715779
Make runic happy
termi-official Mar 4, 2026
560f8b5
Hide the SOA access behind a nicer user-interface
termi-official Mar 4, 2026
2535789
Fix KernelAbstractions example with Valentin's help
termi-official Mar 4, 2026
8571d45
Use global constant for simplicity in the example.
termi-official Mar 4, 2026
e60c93d
Remove todo from how-to
termi-official Mar 4, 2026
4fe3556
Rename GPU -> Device for most data structures
termi-official Mar 4, 2026
fdc1085
Split up extension into KA part and CUDA part. Also restore CPU error…
termi-official Mar 4, 2026
9df89bf
Runic
termi-official Mar 4, 2026
32eb310
Revert type stability issue and extend interfaces as suggested in the…
termi-official Mar 4, 2026
2b3ff9c
Docstring
termi-official Mar 4, 2026
651600e
Revert setdiff change as there is no measurable gain
termi-official Mar 4, 2026
530bdd2
Runic
termi-official Mar 4, 2026
ac3a916
Sneak GPU assembly into docs
termi-official Mar 4, 2026
e2af28c
Runic
termi-official Mar 4, 2026
179c993
Missing file.
termi-official Mar 4, 2026
280f35d
Tweak example
termi-official Mar 4, 2026
4e2d0db
Device assembler supertype
termi-official Apr 9, 2026
ff09269
More interface coverage
termi-official Apr 9, 2026
93de529
Fix CUDA compat
termi-official Apr 16, 2026
73d0ee2
Hoist the cache query out of the inner loop in the how-to
termi-official Apr 16, 2026
6f0b4e7
Dispatch only on adapt_structure
termi-official Apr 16, 2026
92d97f6
Merge master
termi-official Apr 16, 2026
bd60eab
[skip ci] Runic
termi-official Apr 16, 2026
f3459a2
Add device-compatible dof_range
termi-official Apr 17, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions Project.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,12 +16,16 @@ Tensors = "48a634ad-e948-5137-8d70-aa71f2a747f4"
WriteVTK = "64499a7a-5c06-52f2-abe2-ccb03c286192"

[weakdeps]
Adapt = "79e6a3ab-5dfb-504d-930d-738a2a938a0e"
BlockArrays = "8e7c35d0-a365-5155-bbbb-fb81a777f24e"
CUDA = "052768ef-5323-5732-b1bb-66c8b64840ba"
Metis = "2679e427-3c69-5b7f-982b-ece356f1e94b"
KernelAbstractions = "63c18a36-062a-441e-b654-da1e3ab1ce7c"
SparseMatricesCSR = "a0a7dd2c-ebf4-11e9-1f05-cf50bc540ca1"

[extensions]
FerriteBlockArrays = "BlockArrays"
FerriteCudaExt = ["Adapt", "CUDA", "KernelAbstractions"]
FerriteMetis = "Metis"
FerriteSparseMatrixCSR = "SparseMatricesCSR"

Expand Down
Loading
Loading