paboyle
c54d87a472
Aurora compile fix for new compiler
2025-11-06 18:17:33 +00:00
Peter Boyle
fe0db53842
FFT offload to GPU and MUCH faster comms.
...
40x speed up on Frontier
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
paboyle
41f344bbd3
Merge with Christoph GPT checksum debug
2025-07-15 03:06:09 +00:00
paboyle
a77cd50b2f
Update comms logging in Cshift
2025-07-11 14:36:10 +00:00
paboyle
5afcbcf0f3
Cshift uses flight recorder
2025-06-27 06:08:54 +00:00
paboyle
25ab9325e7
Use hostVector but remove construct resize
2025-03-11 15:02:32 +00:00
paboyle
19f9378b98
Should work on Aurora nowb
2025-03-11 13:50:43 +00:00
paboyle
4c3dd82d84
CSHIFT with bounce throuhgh Host memory on MPI packets
2025-02-12 19:09:53 +00:00
Peter Boyle
ce64271048
Remove the copying version
2024-10-18 13:56:24 -04:00
paboyle
e637fbacae
Verbose remove
2024-09-23 09:42:43 +00:00
paboyle
066544281f
Deprecate UVM
2024-09-17 13:34:27 +00:00
Peter Boyle
c9c4576237
Improved frontier cshift
2023-10-13 18:21:56 +03:00
Makis Kappas
7d62f1d6d2
Populate the Cshift_table in the GPU
...
Cshift is allocated in Unified memory and used
in the LambdaApply kernels but also populated
from the host. This creates a lot of Unified HtoD
and DtoH mem operations and has a negative effect
in performance. With this commit we populate the
Cshift table in the device with the
populate_Cshift_table() kernel.
2023-01-11 21:26:25 +00:00
Peter Boyle
a7fb25adf6
Make Cshift fields static to avoid repeated reallocaate overhead
2021-03-29 21:44:14 +02:00
Peter Boyle
6a429ee6d3
2d loop hits Nvidia 16bit limit on large local vols
2021-03-09 04:31:10 +01:00
Peter Boyle
7535566f54
Option for bounce through the SHM buffer
2020-11-12 22:54:27 +01:00
Peter Boyle
50b808ab33
Configure option between host and device
2020-11-12 22:28:12 +01:00
Peter Boyle
6dbd117aa5
Merge branch 'develop' of https://github.com/paboyle/Grid into develop
2020-09-03 20:30:49 -04:00
Peter Boyle
85b1c5df39
A never hit case that is not 100% confident is asserted for safety
2020-09-03 15:48:16 -04:00
Peter Boyle
5791021dcd
Speed up Cshift more with coalesced
2020-09-01 15:28:15 -04:00
Christoph Lehner
69db4816f7
fix variable capture in Scatter_plane_merge on accelerators
2020-08-06 16:57:16 +02:00
Peter Boyle
cdf0a04fc5
Merge branch 'develop' into sycl
2020-06-09 04:00:12 -04:00
Peter Boyle
1a4c8c3387
Global edit with change to View usage. autoView() creates a wrapper object that closes the view when scope closes.
2020-06-05 18:52:35 -04:00
Peter Boyle
7860a50f70
Make view specify where and drive data motion - first cut.
...
This is a compile tiime option --enable-unified=yes/no
2020-05-21 16:13:16 -04:00
Peter Boyle
cb7ee37562
Close expressions in arg to cshift
2020-05-12 17:56:40 -04:00
Peter Boyle
07c0c02f8c
Speed up Cshift
2020-05-11 17:02:01 -04:00
Peter Boyle
8c31c065b5
Keep the Vector fixed to protect it from realloc
2020-05-11 17:00:30 -04:00
Peter Boyle
68541606ab
Thread loop changes. Soon try these with accelerator loops and benchmark
2019-06-15 07:46:42 +01:00
Peter Boyle
4a96c067ae
Remove warnings from NVCC
2019-01-01 13:43:09 +00:00
Peter Boyle
b57a4d32aa
Merge branch 'develop' into feature/gpu-port
2018-12-13 05:11:34 +00:00
portelli
fb7d021b9d
Hadrons: moving Hadrons to root directory, build system improvements
2018-08-28 15:00:40 +01:00
paboyle
Peter Boyle
Makis Kappas
Peter Boyle
Christoph Lehner
portelli