-
As originally conceived, a "DIRECT" solve_type is supposed to result in the matrix M being inverted directly, while a "NORMEQ" solve inverts M^† M. From the NEWS file:
- Added 'solve_type' to …
-
Trying to fix issue 76. Compiles fine now, but when I call the cpuGaugeField constructor using QUDA_REFERENCE_FIELD_CREATE + QUDA_QDP_GAUGE_ORDER, I get a seg fault. The same constructor with QUDA_REF…
-
When we are strong scaling, one of the performance limiters is continually rebinding textures. We should work out a mechanism to cache texture binding so that within the course of a solve (or any oth…
-
This is an issue to track the integration of QUDA with the BQCD library. BQCD is written in Fortran and so requires that QUDA be extended to support being called from Fortran.
-
Steve Gottlieb has been running benchmark jobs on Keeneland lately, and he got in touch this week to say that the multi-mass solver was exhibiting some funny behaviour. At the end of the inversions, …
-
... I don't think (e.g., unitarize_force_quda.cu, unitarize_links_quda.cu, etc.). These should be included in a similar way to the fatlink code.
-
The tests directory has a unitarize_link_test.cpp file which is not compiled unless specifically targeted from the command line. Moreover it fails to compile:
mclark@nvsocal1:~/github/quda/tests$ ma…
-
We're having issues with the single-precision multi-mass staggered inverter. We specifiy a target residual and the inverter claims to have converged before the residual printed in Quda (cumulative and…
-
Hi, all.
I am trying to run dslash_test for the current (gb2b5915) multi-GPU Wilson dslash and getting the error
assert failed destNode != QMP_comm_get_node_number(comm).
From what I can tell, ala …
-
At the moment, as I understand it, the QUDA inverter routines do not return residuals. So currently, if you want to pass the inverter residual back to the client program, you call the inverter routine…