Peter Boyle
|
8b23a1546a
|
Force compile temporarily
|
2024-07-22 15:24:56 -04:00 |
|
Peter Boyle
|
a901e4e369
|
Regressed performance for paper
|
2024-07-22 15:24:04 -04:00 |
|
Peter Boyle
|
804d9367d4
|
Regressed performance
|
2024-07-22 15:23:25 -04:00 |
|
Peter Boyle
|
12b8be7cb9
|
Best so far on 96^3 350 Evecs converged on 4^4 block
|
2024-06-18 16:31:37 -04:00 |
|
Peter Boyle
|
d299c86633
|
Std::asin,acos
|
2024-06-11 16:41:23 -04:00 |
|
Peter Boyle
|
6ce52092e8
|
Merge branch 'develop' of https://github.com/paboyle/Grid into develop
|
2024-06-11 15:16:58 -04:00 |
|
Peter Boyle
|
b5926c1d21
|
Broadcast time info
|
2024-06-11 15:16:25 -04:00 |
|
Peter Boyle
|
9563238e9b
|
Force initial to identity
|
2024-06-11 17:51:58 +00:00 |
|
Peter Boyle
|
fb9b1d76ca
|
Merge branch 'develop' of https://github.com/paboyle/Grid into develop
|
2024-06-11 16:48:16 +00:00 |
|
Peter Boyle
|
1739146599
|
Property to initialise reduction
|
2024-06-11 16:47:35 +00:00 |
|
Peter Boyle
|
ed20b39ab3
|
Log files from Frontier benchmark
|
2024-06-11 11:16:20 -04:00 |
|
Peter Boyle
|
284fc05f15
|
Protect vs. missing LIME libarary
|
2024-06-11 11:08:00 -04:00 |
|
Peter Boyle
|
07a07b6fa3
|
Merge branch 'develop' of https://github.com/paboyle/Grid into develop
|
2024-06-10 15:09:25 -04:00 |
|
Peter Boyle
|
dc80b08969
|
96^3 test
|
2024-06-10 15:07:29 -04:00 |
|
Peter Boyle
|
a49a161f8d
|
SYCL update to use buffer on reduction variable
|
2024-06-08 16:05:18 +00:00 |
|
Peter Boyle
|
a6479ca50f
|
Shuhei's ComputeWilsonFlow main programme
|
2024-06-05 15:51:11 -04:00 |
|
Peter Boyle
|
0e607a55e7
|
Updated for 8^4 test
|
2024-05-26 20:53:05 +00:00 |
|
Peter Boyle
|
c4b9f71357
|
CPU compile ordering is important
|
2024-05-21 02:22:32 +01:00 |
|
Peter Boyle
|
394e506aea
|
Compile options for tursa update
|
2024-05-21 02:10:04 +01:00 |
|
Peter Boyle
|
e19b26341b
|
Tursa configure update
|
2024-05-21 01:14:27 +01:00 |
|
Peter Boyle
|
cfe1b13225
|
Back out zero change
|
2024-05-21 01:14:08 +01:00 |
|
Peter Boyle
|
890c5ea1cd
|
Warning disable
|
2024-05-20 20:08:31 +01:00 |
|
Peter Boyle
|
a87378d3b6
|
Update
|
2024-05-20 20:08:31 +01:00 |
|
Peter Boyle
|
9a1ad6a5eb
|
Merge branch 'develop' of https://github.com/paboyle/Grid into develop
|
2024-05-17 11:33:46 -04:00 |
|
Peter Boyle
|
a90eafad24
|
Merge branch 'feature/scidac-wp1' into develop
|
2024-05-17 11:32:00 -04:00 |
|
Peter Boyle
|
ad14a82742
|
Working aas good as possible on 48^3 in double
|
2024-05-16 10:55:45 -04:00 |
|
Peter Boyle
|
14e9d8ed9f
|
CG improvements for smoother
|
2024-05-16 10:55:18 -04:00 |
|
Peter Boyle
|
0ac85fa70b
|
Serialisation removal
|
2024-05-16 10:49:04 -04:00 |
|
Peter Boyle
|
c371de42b9
|
Some site tools for sitewise autocorr
|
2024-05-16 10:48:23 -04:00 |
|
Peter Boyle
|
ccf147d6c1
|
Select the compiler that gives better performance on sunspot
|
2024-05-07 18:45:56 +00:00 |
|
Peter Boyle
|
7aa12b446f
|
New config command for sunspot
|
2024-05-07 18:45:40 +00:00 |
|
Peter Boyle
|
c293228102
|
layout control
|
2024-05-07 18:45:21 +00:00 |
|
Peter Boyle
|
5c4c9f721a
|
Remove pbs file and replace with bench1 and bench2 for 1 and 2 nodes
|
2024-05-07 18:44:49 +00:00 |
|
Peter Boyle
|
057f86c1de
|
2 queues works ok in performance
|
2024-05-07 18:42:50 +00:00 |
|
Peter Boyle
|
cd52e3cbc2
|
Jobs on subspot
|
2024-05-07 18:38:15 +00:00 |
|
Peter Boyle
|
24602e1259
|
Accidental synchronise
|
2024-05-07 17:28:38 +00:00 |
|
Peter Boyle
|
5c3ace7c3e
|
Merge branch 'develop' into feature/scidac-wp1
|
2024-04-30 05:26:06 -04:00 |
|
Peter Boyle
|
aa148455b7
|
Updated todo list
|
2024-04-30 05:24:39 -04:00 |
|
Peter Boyle
|
98cf247f33
|
prepare to switch to mixed precision
|
2024-04-30 05:23:45 -04:00 |
|
Peter Boyle
|
0cf16522d1
|
Refine with HDCG choice
|
2024-04-30 05:22:14 -04:00 |
|
Peter Boyle
|
7b7c75f9e5
|
Setup
|
2024-04-30 05:21:02 -04:00 |
|
Peter Boyle
|
aefd255a3c
|
Verbose
|
2024-04-30 05:20:41 -04:00 |
|
Peter Boyle
|
1c5aa939fd
|
Subspace setup changes
|
2024-04-30 05:19:09 -04:00 |
|
Peter Boyle
|
3a0ff17be0
|
Verbose changes
|
2024-04-30 05:17:28 -04:00 |
|
Peter Boyle
|
47829ae5cc
|
Verbose changes
|
2024-04-30 05:16:46 -04:00 |
|
Peter Boyle
|
bfa7b69aff
|
Verbose changes
|
2024-04-16 15:42:46 -04:00 |
|
Peter Boyle
|
2aaa959b5f
|
Printing changes
|
2024-04-16 15:41:25 -04:00 |
|
Peter Boyle
|
ce2970b93a
|
Printing changes
|
2024-04-16 15:40:38 -04:00 |
|
Peter Boyle
|
7b76970d10
|
Verbose changes
|
2024-04-16 15:40:10 -04:00 |
|
Peter Boyle
|
9fd41882d2
|
Herm Op update
|
2024-04-16 15:39:27 -04:00 |
|
Peter Boyle
|
5147a42818
|
Updated hdcg
|
2024-04-05 01:05:57 -04:00 |
|
Peter Boyle
|
57552d8ca3
|
Assign from non-lattice made accelerator resident
|
2024-04-05 01:05:12 -04:00 |
|
Peter Boyle
|
13713b2a76
|
Much faster little dirac operator calculation
|
2024-04-05 01:04:40 -04:00 |
|
Peter Boyle
|
36a14e4ee3
|
Best setup and introduce an HDCG refine method
|
2024-04-05 01:03:33 -04:00 |
|
Peter Boyle
|
b4cc788b8c
|
First version used in mrhsHDCG
Need to consolidate files.
Plan: Make this version able to go virtual base, then absorb chulwoos
version when it is proven
|
2024-04-05 01:02:21 -04:00 |
|
Peter Boyle
|
0f0e7512f3
|
Keep MRHS in a different file
|
2024-04-05 00:59:53 -04:00 |
|
Peter Boyle
|
1196b1a161
|
Less verbose
|
2024-04-05 00:58:58 -04:00 |
|
Peter Boyle
|
2c8c3be9ee
|
Adef2Mrhs
|
2024-04-05 00:57:13 -04:00 |
|
Peter Boyle
|
5b79d51c22
|
Improvements
|
2024-04-01 14:18:40 -04:00 |
|
Peter Boyle
|
da890dc293
|
Verbose changes
|
2024-04-01 14:18:00 -04:00 |
|
Peter Boyle
|
93d0a1e73a
|
HISQ view call
|
2024-04-01 14:16:47 -04:00 |
|
Peter Boyle
|
f0a8c7d045
|
Playing with chebyshevs
|
2024-04-01 14:16:11 -04:00 |
|
Peter Boyle
|
db8793777c
|
Logging/verbose
|
2024-04-01 14:15:41 -04:00 |
|
Peter Boyle
|
c745484e65
|
9.5x speed up version
|
2024-04-01 14:14:30 -04:00 |
|
Peter Boyle
|
da59379612
|
Large reg file for double
|
2024-03-26 17:03:20 +00:00 |
|
Peter Boyle
|
3ef2a41518
|
ifdef guard ommitted
|
2024-03-26 14:50:32 +00:00 |
|
Peter Boyle
|
aa96f420c6
|
Acclerator ware MPI guard on the Unix domain sockets
|
2024-03-26 14:41:25 +00:00 |
|
Peter Boyle
|
49e9e4ed0e
|
Fences
|
2024-03-26 14:14:06 +00:00 |
|
Peter Boyle
|
f7b8163016
|
Deterministic MPI reduce options
|
2024-03-26 14:11:40 +00:00 |
|
Peter Boyle
|
93769eacd3
|
Updated configure for bounce through host
|
2024-03-26 14:10:24 +00:00 |
|
Peter Boyle
|
59b0cc11df
|
REduce the time in single
|
2024-03-26 00:42:40 +00:00 |
|
Peter Boyle
|
f32c275376
|
Updated config options for MPI not being aware of GPU
|
2024-03-26 00:42:00 +00:00 |
|
Peter Boyle
|
5404fc66ab
|
Merge needs a fence on SYCL
|
2024-03-26 00:38:41 +00:00 |
|
Peter Boyle
|
1f53458af8
|
Options to bounce through a host buffer if
--disable-accelerator-aware-mpi
|
2024-03-26 00:37:19 +00:00 |
|
Peter Boyle
|
434c3e7f1d
|
We have a choice of GET or PUT across NVlink
|
2024-03-25 14:32:44 +00:00 |
|
Peter Boyle
|
500b119f3d
|
Deterministic MPI
|
2024-03-22 15:55:23 +00:00 |
|
Peter Boyle
|
4b87259c1b
|
New config command for sunspot
|
2024-03-22 15:43:49 +00:00 |
|
Peter Boyle
|
503dec34ef
|
This appears working now on Sunspot
|
2024-03-22 15:43:30 +00:00 |
|
Peter Boyle
|
d1e9fe50d2
|
Xor csum for repro testing
|
2024-03-22 15:42:57 +00:00 |
|
Peter Boyle
|
d01e5fa838
|
Improved FlightRecorder
|
2024-03-22 15:42:32 +00:00 |
|
Peter Boyle
|
a477c25e8c
|
Sunspot repro tests
|
2024-03-22 15:42:11 +00:00 |
|
Peter Boyle
|
1bd20cd9e8
|
FlightRecorder
|
2024-03-22 15:40:01 +00:00 |
|
Peter Boyle
|
e49e95b037
|
Upgrade of the Britney test with flight recorder and fast xor checksum
|
2024-03-22 15:39:27 +00:00 |
|
Peter Boyle
|
6f59fed563
|
Flight recorder, resurrecting the "world famous" Britney test
|
2024-03-22 15:32:32 +00:00 |
|
Peter Boyle
|
60b7f6c99d
|
Flight recorder, resurrecting the "world famous" Britney test
|
2024-03-22 15:32:26 +00:00 |
|
Peter Boyle
|
b92dfcc8d3
|
Flight recorder, resurrecting the "world famous" Britney test
|
2024-03-22 15:30:27 +00:00 |
|
Peter Boyle
|
f6fd6dd053
|
Flight recorder, resurrecting the "world famous" Britney test
|
2024-03-22 15:30:01 +00:00 |
|
Peter Boyle
|
79ad567dd5
|
Merge branch 'develop' of https://github.com/paboyle/Grid into develop
|
2024-03-19 15:43:42 +00:00 |
|
Peter Boyle
|
fab1efb48c
|
More britney logging improvements
|
2024-03-19 14:36:21 +00:00 |
|
Peter Boyle
|
660eb76d93
|
FFTW from OneAPI
|
2024-03-19 14:28:33 +00:00 |
|
Peter Boyle
|
62e7bf024a
|
Updated flight logging for Britney test
|
2024-03-12 20:10:04 +00:00 |
|
Peter Boyle
|
95f3d69cf9
|
Extra hardware test hook
|
2024-03-12 20:09:37 +00:00 |
|
Peter Boyle
|
cc04dc42dc
|
Merge branch 'develop' into feature/scidac-wp1
|
2024-03-06 14:55:21 -05:00 |
|
Peter Boyle
|
070b61f08f
|
Simplifying the MultiRHS solver to make it do SRHS *and* MRHS
|
2024-03-06 14:04:33 -05:00 |
|
Peter Boyle
|
7e5bd46dd3
|
Booster update
|
2024-03-06 19:03:45 +01:00 |
|
Peter Boyle
|
228bbb9d81
|
Benchmark results
|
2024-03-06 19:03:35 +01:00 |
|
Peter Boyle
|
c805f86343
|
USQCD benchmark
|
2024-03-01 00:05:04 -05:00 |
|
Peter Boyle
|
04ca065281
|
Only one rank opens
|
2024-02-29 20:09:11 -05:00 |
|
Peter Boyle
|
88d8fa43d7
|
Benchmark development
|
2024-02-29 20:01:44 -05:00 |
|
Peter Boyle
|
3c49762875
|
Propagate in the blas routine
|
2024-02-29 15:33:06 -05:00 |
|