Peter Boyle
497e7c1c40
Duplicate code
2020-12-02 17:55:30 -08:00
Peter Boyle
9dce101586
Merge branch 'feature/gpt' of https://github.com/lehner/Grid into lehner-feature/gpt
2020-11-24 06:10:16 -05:00
Peter Boyle
97e264d0ff
Christoph's changes
2020-11-23 15:46:11 +00:00
Peter Boyle
683a5e5bf5
Stencil use host vector for integera table on enable-shared=no and mirror it on device
2020-11-23 15:39:51 +00:00
Christoph Lehner
c61ea72949
Merge pull request #19 from paboyle/develop
...
Sync
2020-11-20 17:31:13 +01:00
Peter Boyle
612e468889
Configurable ALLOC_ALIGN and ALLOC_CACHE
2020-11-20 16:48:28 +01:00
Christoph Lehner
4ea8d128c2
Merge pull request #18 from paboyle/develop
...
Sync
2020-11-20 15:36:50 +01:00
Peter Boyle
f1c7480e3c
Warning remove
2020-11-19 19:23:03 +01:00
Peter Boyle
b5e7945dd9
Option for host or device Cshift implementation
2020-11-13 01:38:54 +01:00
Christoph Lehner
32ff766dbd
fix evict scheme, slab alloc
2020-09-13 14:02:53 -04:00
Peter Boyle
d3ce60713d
UVM, Device and Lattice/aligned allocators
2020-09-03 15:44:13 -04:00
Christoph Lehner
e33878e0de
Trigger re-run of CI
2020-08-06 11:50:24 +02:00
Christoph Lehner
6365a89ba3
create separate InitMessage for MemoryManager that can be called after communicator setup
2020-07-30 07:25:05 -04:00
Christoph Lehner
197612bc7a
fast cpu basisRotate and other small cleanups
2020-07-30 07:08:54 -04:00
Peter Boyle
093d1ee21b
Force initial values
2020-06-24 08:54:49 -04:00
Peter Boyle
2ff1fa6fad
UVM used shared for CPU alloccations andd ddont migrate
2020-06-23 22:14:56 -04:00
Peter Boyle
c48da35921
Memory Vector UVM and Lattice alignedAllocator separate
2020-06-22 20:21:53 -04:00
Peter Boyle
b949cf6b12
PeekLocal needs a view to keep thread safe.
...
ALLOCATION_CACHEE reenable
2020-06-19 17:13:27 -04:00
Christoph Lehner
3dccd7aa2c
Catch edge case in SharedMemoryMPI::GetShmDims; Change default units to consistent MB in init args; Want last element not past last element in MemoryManagerCache.cc
2020-06-14 13:26:01 -04:00
Christoph Lehner
b5e87e8d97
summit compile fixes
2020-06-12 18:16:12 -04:00
Christoph Lehner
5f5807d60a
cleanup
2020-06-12 14:48:23 -04:00
Christoph Lehner
7974acff54
merged sycl to feature-gpt
2020-06-12 06:49:38 -04:00
Peter Boyle
cdf0a04fc5
Merge branch 'develop' into sycl
2020-06-09 04:00:12 -04:00
Peter Boyle
616d3dd737
CCommpile updates
2020-06-08 18:57:41 -04:00
Peter Boyle
8b066baca8
Implement transient mechanism
2020-06-08 18:28:53 -04:00
Peter Boyle
87e5d2f4b7
Merge branch 'sycl' of https://www.github.com/paboyle/Grid into sycl
2020-06-05 17:32:21 -07:00
Peter Boyle
d720f10758
Liink error fix
2020-06-05 17:29:20 -07:00
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
f39c2a240b
Priintinig and device memory size detection
2020-06-04 14:58:03 -04:00
Peter Boyle
0d95805cde
Print improvement
2020-06-03 22:50:32 -04:00
Peter Boyle
6bf7f839ff
Better printing and logging
2020-06-03 09:28:57 -04:00
Peter Boyle
e3147881a9
Cache scheme
2020-06-03 09:23:48 -04:00
Peter Boyle
32237895bd
Reorg memory manager for O(1) hash table
2020-06-03 09:09:52 -04:00
Peter Boyle
c5c2dbc0ce
Optional CUDA info
2020-06-02 14:21:49 -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
Christoph Lehner
8358ee38c4
pull develop
2020-05-19 08:56:18 -04:00
Christoph Lehner
a7635fd5ba
summit mem
2020-05-18 17:52:26 -04:00
Peter Boyle
ebb60330c9
Automatic data motion options beginning
2020-05-17 16:34:25 -04:00
Peter Boyle
07c0c02f8c
Speed up Cshift
2020-05-11 17:02:01 -04:00
Peter Boyle
bbbee5660d
First compiile on HiP
2020-05-10 05:28:09 -04:00
Peter Boyle
ea08f193e7
Allocator cache spliit into large/small pools
2020-05-10 05:24:26 -04:00
Peter Boyle
2bb2c68e15
Separate pools for small and large allocations cache
2020-05-09 22:57:21 -04:00
Peter Boyle
efe5bc6a3c
Split allocator cache into two pools of different sizes
2020-05-09 22:27:56 -04:00
Peter Boyle
f8b8e00090
Systematise the accelerator primitives and locate to Grid/threads/Accelerator.h / Accelerator.cc
...
Aim to reduce the amount of cuda and other code variations floating around all over the place.
Will move GpuInit iinto Accelerator.cc from Init.cc
Need to worry about SharedMemoryMPI.cc and the Peer2Peer windows
2020-05-08 06:23:55 -07:00
Peter Boyle
28a1fcaaff
First compile against SYCL
2020-05-05 11:13:27 -07:00
Peter Boyle
afc7426f39
Much bigger pointer cache in case of Nvidia due to cost of setting up UVM allocations
2020-01-27 12:41:16 -05:00
Peter Boyle
339ea10cc7
First touch only on CPU code
2019-06-15 07:45:43 +01:00
gfilaci
9003c4a07c
allocator copy constructor (to be fixed)
2019-05-02 10:53:37 +01:00
Peter Boyle
b57a4d32aa
Merge branch 'develop' into feature/gpu-port
2018-12-13 05:11:34 +00:00
fb7d021b9d
Hadrons: moving Hadrons to root directory, build system improvements
2018-08-28 15:00:40 +01:00