Antonin Portelli

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-21 18:09:36 +01:00

4d6931620b Gather based distributed dense inverse

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-20 17:39:37 +01:00

d62fc91996 Better ROCTX tracing

3ed05dbbb1 Verbose trasing

97c9178785 Ready for faster 3 level solve with dense coarse and BLAS on SRHS !

2fde4f3e50 Trace hooks

Compare 4 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-20 09:29:35 +01:00

b31547c903 no compile fix

2d56323012 No compile on Frontier

63fefb8ea5 new tests for lex lattice

Compare 3 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-20 01:19:35 +01:00

e4ec5d0009 FOr pvdagm preconditioners if they work

a5e4ce3e48 Better comment

2900ce33b5 New files, including v2 multiRHS coarse op

ea5bf89955 Failed experiment in the style of Saul Cohens non-red black CG approach to mgrid.

109b77ea75 Drop D2 field types

Compare 5 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-15 23:19:37 +01:00

0ac72cb6a3 Don't use host as it sucks.

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-15 15:09:36 +01:00

c4b089cf15 More timing updates

85e1bbf4bc Added gather benchmark to cross refer to the 'lazy' analogue of summing a vector with many zeroes in regions not owned by THIS rank

032f111c8b Instrument as not faster

Compare 3 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-15 06:59:37 +01:00

e38528888a Submission scripts for distributed schur coarse grid inverse

7c9a6d354e Updates for thrust

3ec90803ac no compil patches

c22473f15d First test of distributed schur recursive inverse

55c064de28 Batch blas test extension

Compare 5 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-14 06:29:36 +01:00

02d0301c9f Scaling perf

a6cdf20c18 IO tests

ad9a413892 Update

1fddd2c29b Updates to IO regression example for MPICH issue

2f75067569 Some I/O simple benchmarks

Compare 5 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-13 22:19:36 +01:00

6e8a00f215 Simplest starting point

702773e5fb BatchedBlas.h updates. May cause some churn. Trying to avoid scalar pointer copies to device.

4dfbd850ff Buf fix on DenseCoarseMatrix.h, coefficient caching optimisation on BatchGEMM

Compare 3 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-12 22:03:29 +01:00

b039e659af 4D pseudo fermion

d16d44dda0 Prepare for reduced Ls during evolution forces using 4D pseudo fermion.

1c19389ba6 Two flavour boson term

02fdff674c Test the PVdagM left prec pseudofermion formulation

fd8b6a23a6 PvDagM or other left prec precon two flavour ratio

Compare 5 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-12 05:43:30 +01:00

df908ee872 3 level dense

84715ff4b9 Default to fast path

5792195073 Check in codes used at time of Lattice conference for PVdagM multigrid

fb5662a449 Setup on Frontier

6b2ad3db80 Inverse improvements. Need to make this properly recursive

Compare 5 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-08-11 21:33:30 +01:00

1f55c13bc4 Test script for Frontier on I/O

d68b111d06 Switch on the faster IO be default

41f5a02204 IO updates for AllToAllV aggregation of contiguous chunks.

Compare 3 commits »

portelli synced commits to refs/heads/KS_shifted at portelli/Grid from mirror 2026-07-13 21:58:18 +01:00

01c3189d17 Reverted restart to inline in KrylovSchur

f1249685aa Refactording KrylovSchur to have separate operator() for non-shift and shift

722d68b0d9 Updating KrylovSchur example to include True Block KS

c1284e080a Adding TrueHarmonicBlockKrylovSchur

d2189d59e0 Adding parity flip and gamma 5 option for Block KS starting vector

Compare 5 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-06-26 21:43:30 +01:00

7647576863 Sourceme's for frontier

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-06-18 01:43:43 +01:00

f11ba18df2 Remove accelerator_inline on CPU only code

cf8587e401 Update booster compiule

7dd35ef749 Make disable accelerator aware mpi compile for CUDA

41e570ddce Annoying old CPU perfmon code should be removed or deprecated as not

a452131b50 Print improvement

Compare 5 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-06-17 01:23:12 +01:00

4e49ca55ab Merge branch 'develop' of https://github.com/paboyle/Grid into develop

c3f4474401 Adding Mattia's memory leak test

Compare 2 commits »

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-06-11 22:53:11 +01:00

3d3eff86f3 Modify move assignment operator to be noexcept

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-06-11 14:43:11 +01:00

fc9f154ac1 Modify Lattice move constructor to be noexcept

portelli synced commits to refs/heads/develop at portelli/Grid from mirror 2026-06-04 11:14:38 +01:00

4aa0bca4dc Change sum operation to use gpucub mistake in PR from Chris

portelli synced commits to refs/heads/feature/staggered-hdcg at portelli/Grid from mirror 2026-05-29 21:44:17 +01:00

94d8ee4268 More info

3fac61fc55 benchmarks: add DhopEO benchmarks to Benchmark_staggered and Benchmark_staggeredF

42cd9eda71 Some improvements that should have been there if in synch with develop,

34d8d003a8 staggered-hdcg: smoother shift tuning, CG baseline, Lanczos diagnostics

905651deaa Test_staggered_hdcg: fix GridParallelRNG and Lanczos grid bugs

Compare 14 commits »