Carleton DeTar comments

Results 7 comments of


                                            Carleton DeTar

error during autotuning of blocksolver

We need to know whether the multiRHS feature will improve our ECP FOM benchmarks. If we see a significant improvement, we will certainly consider incorporating it into production running on...

Feature/milc gaugefix ovr definition

Hi Evan, My mistake. I had forgotten about quda_milc_interface.h. I'll try that. Best, Carleton On 7/15/22 9:38 AM, Evan Weinberg wrote: > > Ah, here's the issue, the function you...

Staggered heavy-quark residual fails to regulate HISQ CG convergence properly with an odd checkerboard source

Thank you for helping with this, Evan. I should say first that I am building QUDA with Jim Simone's branch, "not_a_feature_rather_a_hope/staggered_correlator_gk", but he has been merging develop into it. In...

Staggered heavy-quark residual fails to regulate HISQ CG convergence properly with an odd checkerboard source

Thanks, Evan. Do you have access to Frontier? Or should I port the reproducer to Summit? MILC calls QUDA specifying the "parity" of the solve. For a strictly odd parity...

Staggered heavy-quark residual fails to regulate HISQ CG convergence properly with an odd checkerboard source

All of the MILC color vector field are full. On 5/3/23 9:03 AM, Evan Weinberg wrote: > > One question---is the host source in MILC single parity or the length...

Staggered heavy-quark residual fails to regulate HISQ CG convergence properly with an odd checkerboard source

Hi Evan, I sent a bunch of files in my last message, but I just discovered that our MIMEdefang probably deleted the shell script attachments. So here they are with...

Staggered heavy-quark residual fails to regulate HISQ CG convergence properly with an odd checkerboard source

Hi Evan, Any progress? Thanks, Carleton On 5/16/23 9:03 AM, Evan Weinberg wrote: > > I've reproduced the behavior on a 144^3 configuration, for both an odd > /and/ even...