32654db366
Test_planned_fft: fix PlannedFFT template parameter to use ::vector_object
...
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-21 12:34:30 -04:00
cd340cfab3
tests: add Test_planned_fft exercising PlannedFFT<vobj>
...
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-21 12:34:30 -04:00
f32866b2ff
tests/fft: remove PlanDestroy calls (FFT handles plans per-call)
...
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-21 12:34:30 -04:00
e79adc9d31
FFT: cache plans per vobj type across calls
...
Plans are created lazily on the first FFT_dim call and reused for all
subsequent calls on the same FFT object. PlanCreate<vobj>() can be
called explicitly to pre-warm the cache. PlanDestroy() must be called
before switching to a different vobj type; the destructor cleans up any
live plans automatically.
Update Test_fft.cc and Test_fftf.cc to call PlanDestroy() between the
LatticeComplex and LatticeSpinMatrix sections that reuse the same FFT object.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-21 12:34:30 -04:00
fe0db53842
FFT offload to GPU and MUCH faster comms.
...
40x speed up on Frontier
2025-08-21 16:45:38 -04:00
76c0ada1e1
Benchmark for En Hung
2025-08-21 16:45:38 -04:00
paboyle
9e6a4a4737
Assertion updates to macros (mostly) with backtrace.
...
WIlson flow to include options for DBW2, Iwasaki, Symanzik.
View logging for data assurance
2025-08-07 15:48:38 +00:00
b91fc1b6b4
Merge branch 'feature/boosted' into feature/deprecate-uvm
...
Fixed boosted free field test
2024-10-28 16:53:09 -04:00
eafc150034
Test fft asserts
2024-10-23 16:46:26 -04:00
1e893af775
GPU happy
2024-10-23 14:52:15 -04:00
d9f430a575
Happy GPU
2024-10-23 14:51:16 -04:00
paboyle
066544281f
Deprecate UVM
2024-09-17 13:34:27 +00:00
paboyle
160969a758
UVM tester, doesn't turn up anything
2024-09-10 18:09:42 +00:00
b507fe209c
Added SpinColourMatrix case to sliceSum Test
2024-02-27 11:28:32 -05:00
15878f7613
sliceSumReduction_cub_large now also faster than CPU on Frontier
2024-02-16 13:55:21 -05:00
6f3455900e
Adding sliceSumReduction_cub_small/large since hipcub cannot deal with arb. large vobjs
2024-02-16 13:15:02 -05:00
b5659d106e
more test cases
2024-02-09 13:37:14 -05:00
9514035b87
refactor slicesum: slicesum uses GPU version by default now
2024-02-09 13:02:28 -05:00
ab2de131bd
work towards sliceSum for sycl backend
2024-02-06 13:24:45 -05:00
b8b9dc952d
Async memcpy's and cleanup
2024-02-01 17:55:35 -05:00
79a6ed32d8
Use accelerator_for2d and DeviceSegmentedRecude to avoid kernel launch latencies
2024-02-01 16:41:03 -05:00
caa5f97723
Add sliceSum gpu using cub/hipcub
2024-01-31 16:50:06 -05:00
c5f1420dea
Merge remote-tracking branch 'LupoA/develop' into LupoA-develop
2023-10-02 16:22:35 -04:00
b8a7004365
Partial fraction test
2023-08-14 15:17:03 -04:00
c797cbe737
deal with post-merge trauma
2023-06-16 14:20:37 +01:00
e09dfbf1c2
definetely the right merge upstream/develop
2023-06-16 14:19:46 +01:00
fe88a0c12f
cleaner twoindex class, cleaner tests
2023-05-26 16:55:30 +01:00
de8daa3824
group is SUn by default
2023-05-26 13:44:41 +01:00
5897b93dd4
debug tests, fix dimension
2023-05-22 13:42:21 +01:00
e8ad1fef53
Unify TwoIndex
2023-05-12 14:35:50 +01:00
9c8750f261
Merge branch 'develop' of https://github.com/paboyle/Grid into develop
2023-05-11 12:29:09 -04:00
91efd08179
Option for Qlat generator basis
2023-05-11 12:27:45 -04:00
7b3b7093fa
cleaning up requested by Ed
2023-05-02 12:50:57 +01:00
178376f24b
minor stylistic changes
2023-04-06 12:08:17 +01:00
866f48391a
Temporary fix for develop incorrect results
2023-03-30 17:10:13 -04:00
23298acb81
Merge pull request #424 from giltirn/feature/dirichlet-precchange
...
Precision change implementation
2023-03-22 23:04:52 -04:00
b5b759df73
Merge branch 'develop' into feature/dirichlet
2023-03-21 16:05:46 -04:00
d6ff644aab
Towards the day all tests compile
2023-03-14 10:43:25 +00:00
29586f6b5e
Deactivate some tests for Nc!=3
2023-03-13 08:17:14 +00:00
1db58a8acc
Precision change improvements
...
Added a new, much faster implementation of precision change that uses (optionally) a precomputed workspace containing pointer offsets that is device resident, such that all lattice copying occurs only on the device and no host<->device transfer is required, other than the pointer table. It also avoids the need to unpack and repack the fields using explicit lane copying. When this new precisionChange is called without a workspace, one will be computed on-the-fly; however it is still considerably faster than the original implementation.
In the special case of using double2 and when the Grids are the same, calls to the new precisionChange will automatically use precisionChangeFast, such that there is a single API call for all precision changes.
Reliable update and mixed-prec multishift have been modified to precompute precision change workspaces
Renamed the original precisionChange as precisionChangeOrig
Fixed incorrect pointer offset bug in copyLane
Added a test and a benchmark for precisionChange
Added a test for reliable update CG
2023-02-21 10:52:42 -05:00
4ca1bf7cca
Added gauge invariance test
2022-12-21 07:23:16 -05:00
ede02b6883
Memory manager debug Felix case
2022-12-20 05:10:23 -05:00
d8c29f5fcf
Updated FFT test for PETSc
2022-12-18 12:05:00 -05:00
281f8101fe
Matt FFT test
2022-12-17 20:35:33 -05:00
472ed2dd5c
Merge branch 'feature/dirichlet' of https://github.com/paboyle/Grid into feature/dirichlet
2022-12-17 20:17:09 -05:00
4f85672674
Simpler test for PETSc
2022-12-17 20:16:11 -05:00
505fa49983
Renamed SUn.h -> GaugeGroup.h
2022-11-30 17:09:48 +00:00
6e750ecb0e
Remove apparently forgotten file
2022-11-28 16:33:46 +00:00
e51eaedc56
Making tests compile
2022-11-15 22:58:30 -05:00
8208a6214f
Merge branch 'feature/dirichlet-gparity' into feature/dirichlet
2022-06-15 19:23:48 -04:00
Peter Boyle
paboyle
Dennis Bollweg
dbollweg
Alessandro Lupo
Julian Lenz
Peter Boyle
Christopher Kelly