| 2247a93f | 10-Feb-2024 |
Rezgar Shakeri <42816410+rezgarshakeri@users.noreply.github.com> |
CeedMatrixPseudoinverse utility function (#1251)
* Added CeedMatrixPseudoinverse utility and use it CeedBasisGetCollocatedGrad
* Used pseudoinverse utility in CeedBasisApplyAtPoints
* Update d
CeedMatrixPseudoinverse utility function (#1251)
* Added CeedMatrixPseudoinverse utility and use it CeedBasisGetCollocatedGrad
* Used pseudoinverse utility in CeedBasisApplyAtPoints
* Update description of CeedMatrixPseudoinverse function
Co-authored-by: Jeremy L Thompson <jeremy@jeremylt.org>
* Used pseudoinverse utility in CeedBasisCreateProjectionMatrices
* Updated inline comments
* Added inline comment
* Round matrix-matrix product to zero in CeedBasisCreateProjectionMatrices
* increased tolerance in ex1-volume/src/main.rs
---------
Co-authored-by: Jeremy L Thompson <jeremy@jeremylt.org>
show more ...
|
| 1b7492f8 | 22-Jan-2024 |
Sebastian Grimberg <sjg@amazon.com> |
Fix a few incorrect JIT header includes |
| cf8cbdd6 | 22-Jan-2024 |
Sebastian Grimberg <sjg@amazon.com> |
Refactor GPU-backend element restriction source to only compile the required kernels, and do so lazily at first apply |
| cbfe683a | 20-Jan-2024 |
Sebastian Grimberg <sjg@amazon.com> |
Lazily compile GPU diagonal/point-block diagonal assembly kernels as needed
Also fixes CEED_SIZE -> USE_CEEDSIZE definition. |
| 7aa91133 | 08-Nov-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Add deterministic element restriction transpose GPU kernels for oriented and curl-oriented restrictions |
| 004e4986 | 16-Aug-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Operator full assembly and diagonal assembly for cuda-ref and hip-ref backends for H(div) and H(curl) elements |
| dce49693 | 15-Aug-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Initial commit for cuda-ref and hip-ref backend support for oriented and curl-oriented element restrictions
For now, the element restriction transpose for oriented or curl-oriented cases is always n
Initial commit for cuda-ref and hip-ref backend support for oriented and curl-oriented element restrictions
For now, the element restriction transpose for oriented or curl-oriented cases is always non-deterministic.
show more ...
|
| a36217cb | 15-Dec-2023 |
Jeremy L Thompson <jeremy@jeremylt.org> |
interface - remove legacy CEED_BASIS_COLLOCATED |
| b569baaf | 03-Jan-2024 |
Sebastian Grimberg <sebastiangrimb@gmail.com> |
Define CeedPragmaSIMD for Clang (#1432)
* Define CeedPragmaSIMD for Clang
* Change clang loop vectorization pragma |
| 6a8ec7bd | 20-Dec-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Fix comment |
| 897d4737 | 20-Dec-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Fix GCC warnings for casts between incompatible function types |
| 360501f4 | 20-Dec-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Silence compiler warnings for -Wstrict-prototypes |
| bafebce1 | 15-Sep-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Consistency fix for a few missed CeedElemRestriction rstr variables in the public API |
| ca94c3dd | 05-Dec-2023 |
Jeremy L Thompson <jeremy@jeremylt.org> |
Docstring consistency (#1409)
* doc - docstring consistency
* doc - update for bad linking of type names
* doc - update for bad linking of type names |
| 78cbaa95 | 30-Nov-2023 |
Sebastian Grimberg <sebastiangrimb@gmail.com> |
Merge pull request #1300 from CEED/sjg/hcurl-hdiv-basis-cuda-hip
H(div) and H(curl) basis support for `cuda-ref` and `hip-ref` backends |
| 48acf710 | 30-Nov-2023 |
Jeremy L Thompson <jeremy@jeremylt.org> |
OperatorAtPoints (#1390)
* op - add AtPoints interface
* op - reorder AtPoints QFunction for consistency
* op - add CEED_EVAL_WEIGHT for operator AtPoints
* doc - add CeedOperatorCreateAtPo
OperatorAtPoints (#1390)
* op - add AtPoints interface
* op - reorder AtPoints QFunction for consistency
* op - add CEED_EVAL_WEIGHT for operator AtPoints
* doc - add CeedOperatorCreateAtPoints to release notes
show more ...
|
| 47b00e8f | 17-Nov-2023 |
Jeremy L Thompson <jeremy@jeremylt.org> |
ceed - set main to non-release |
| 8d780e65 | 17-Nov-2023 |
Jeremy L Thompson <jeremy@jeremylt.org> |
Merge pull request #1396 from CEED/jeremy/context-bool
Add Boolan context field register/set/get/restore |
| 23dbfd29 | 17-Nov-2023 |
Jeremy L Thompson <jeremy@jeremylt.org> |
ctx - explicit int width in ctx fields |
| d075f50b | 11-Aug-2023 |
Sebastian Grimberg <sjg@amazon.com> |
H(div) and H(curl) basis support for cuda-ref and hip-ref backends |
| 127b3ffe | 22-Sep-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Resolve some missed header file function declarations inconsistencies |
| 93f4dbf1 | 07-Sep-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Remove legacy CeedChk and CeedChkBackend (should use CeedCall, CeedCallBackend instead) |
| 5b6ec284 | 15-Nov-2023 |
Jeremy L Thompson <jeremy@jeremylt.org> |
ctx - add Boolan field register/set/get/restore |
| 86ad04cc | 06-Nov-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Minor cleanup to magma non-tensor kernel |
| 1a0eda08 | 01-Nov-2023 |
Sebastian Grimberg <sjg@amazon.com> |
Fix regression but from shared-memory non-tensor basis transpose |