Antonin Portelli portelli
0 Followers · 0 Following
  • Joined on 2017-10-25
Repositories
6
 Projects Packages Public Activity Starred Repositories
1
portelli synced commits to develop at portelli/Grid from mirror 2023-11-30 23:31:44 +00:00
b77a9b8947 SDDC compiles starting
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-11-29 14:51:44 +00:00
2c54be651c Further updates
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-11-28 22:31:44 +00:00
e859a199df Reduce volume to interior for coarse stencil -- worth up to 4x gain
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-11-28 14:21:44 +00:00
0a3682ad0b MultiRHS work
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-11-24 20:31:44 +00:00
59abaeb5cd Time stamp
portelli synced commits to refs/pull/379/merge at portelli/Grid from mirror 2023-11-24 12:21:43 +00:00
b391af4489 Merge 32e6d58356 into 7d077fe493
7d077fe493 Frontier compiel
Compare 2 commits »
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-11-24 04:11:44 +00:00
3e448435d3 Restrict to interior
a294bc3c5b Relax constraints for multiRHS
b302ad3d49 multiRHS test in place, passes Yay!
82fc4b1e94 Finalise
b4f1740380 Finalise message
Compare 7 commits »
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-11-16 00:11:44 +00:00
09946cf1ba Improved, works on 48^3 moving to multiRHS optimisations
f4fa95e7cb Use 5.3.0
100e29e35e Allow expression as argument to norm2
4cbe471a83 devVector
8bece1f861 Faster to transpose the matrix and apply with column major order
Compare 12 commits »
portelli synced commits to develop at portelli/Grid from mirror 2023-11-09 21:11:43 +00:00
7d077fe493 Frontier compiel
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-10-24 22:21:45 +01:00
4efa042f50 C++17 change
c7cb37e970 c++17 accepted
d34b207eab Avoid HIP warnings
0e6fa6f6b8 DOn't need the Cshift for the period optimisation
38b87de53f This works around a stacksize limit on AMD GPU
Compare 9 commits »
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-10-21 04:31:44 +01:00
351795ac3a Better messaging
9c9c42d0df Tests on frontier with real speed up . 3.5x on 16^3 at mq=0.01
b6ad1bafc7 Normal memory SendToRecvFrom asynchronous for use in general stencil
a5ca40f446 Better verbose -- track CPU GPU motion under --log Memory, others go to
9ab54c5565 Overlap comms & data copy/buffer assembly in Ghost zone exchange
Compare 18 commits »
portelli synced commits to refs/pull/379/merge at portelli/Grid from mirror 2023-10-18 03:01:44 +01:00
dc87a9de71 Merge 32e6d58356 into 51051df62c
51051df62c 3GeV run setup
33097681b9 FTHMC compiled and merged to develop
07e4900218 FTHMC commit
36ab567d67 FTHMC 3 Gev
Compare 11 commits »
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-10-17 02:31:44 +01:00
2207309f8a Spack rules
portelli synced commits to develop at portelli/Grid from mirror 2023-10-17 02:31:44 +01:00
51051df62c 3GeV run setup
portelli synced commits to develop at portelli/Grid from mirror 2023-10-14 01:01:43 +01:00
33097681b9 FTHMC compiled and merged to develop
07e4900218 FTHMC commit
36ab567d67 FTHMC 3 Gev
e19171523b FTHMC Status at lattice conference commit
9626a2c7c0 Asynch handling
Compare 9 commits »
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-10-07 05:41:44 +01:00
2111e7ab5f Run at physical mass
d29abfdcaf Transfer code to Frontier now
a751c42cc5 Checkpoint restore the setup
6a3bc9865e Verbose change
4d5f7e4377 Verbose change
Compare 8 commits »
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-10-06 21:31:44 +01:00
b58fd80379 I/O for coarse op and reorganise multigrid headers
7f6e0f57d0 No IO in file
cae27678d8 gpermute
48ff655bad Slightly less verbose
2525ad4623 Slight clean up
Compare 7 commits »
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-10-06 05:11:45 +01:00
3bc2da5321 Merge branch 'feature/scidac-wp1' of https://github.com/paboyle/Grid into feature/scidac-wp1
2d710d6bfd Optimised parameters for 16^3
6532b7f32b Eliminate older inefficient coarsening implementation
7b41b92d99 Only need to bad non-local dimensions
dd557af84b ADEF1 and ADEF2 2 level CG
Compare 10 commits »
portelli synced commits to refs/pull/379/merge at portelli/Grid from mirror 2023-10-05 21:01:44 +01:00
bb5e03f4bf Merge 32e6d58356 into 6d0c2de399
6d0c2de399 Deprecate teh PVC directory and make a PVC-OEM generic PVC target with
7786ea9921 Bug fix in script
d93eac7b1c Performance regressed and is OK in icpx 2023.2
afc316f501 Rename headers
Compare 171 commits »
portelli synced commits to feature/scidac-wp1 at portelli/Grid from mirror 2023-10-03 20:01:45 +01:00
737d3ffb98 ADEF1 and 1 hop projection