Issues
- 0
Schwarz preconditioned CG gives `nan`s in develop
#1535 opened by weinbe2 - 2
Generating null-vector using bicgstab broken
#1532 opened by chaoos - 3
Hang in mixed HISQ outer/naive staggered pseudofine MG
#1533 opened by weinbe2 - 1
Covariant derivative gives wrong result after applying an Wilson/Clover-Wilson matrix.
#1527 opened by SaltyChiang - 1
Out of bounds error when running multi-GPU/partitioned HISQ MG with long links dropped
#1512 opened by weinbe2 - 0
QUDA's device-side RNG needs to be updated to support global volumes > `2^31 - 1`
#1513 opened by weinbe2 - 1
Use of bool in quda.h
#1521 opened by V3-vvv - 0
Domain Decomposed BLAS support
#1519 opened by maddyscientist - 0
Batched Memcpy
#1514 opened by maddyscientist - 1
- 0
Setting gauge field scale for floating point types
#1508 opened by jcosborn - 0
- 0
- 4
This file doesn't run well
#1499 opened by fwinter - 2
NVSHEM run error
#1501 opened by nicolin - 2
- 2
Show progress report while tuning
#1498 opened by bjoo - 2
host omp parallel for loop variable
#1483 opened by jxy - 1
Storing and restoring the multigrid setup
#1482 opened by sunpho84 - 4
infinite hang when autotuning is disabled
#1477 opened by amwe210 - 5
build error on default branch - Building CUDA object lib/CMakeFiles/quda.dir/blas_quda.cu.o
#1476 opened by twobombs - 5
ApplyLaplace no longer supports nSpin=4
#1475 opened by aniketsen - 0
Fermion smearing interface cleanup
#1474 opened by maddyscientist - 0
`dslash_test` failing with `QUDA_REORDER_LOCATION=CPU`
#1466 opened by sbacchio - 1
Turn on domain wall code builds in ROCm CI
#1440 opened by dmcdougall - 1
Clean up hard-coded loop bounds
#1461 opened by dmcdougall - 4
Add robust signal captures to QUDA
#1449 opened by weinbe2 - 0
Cleanup/optimize staggered invert/eigensolver tests analogous to Wilson cleanup in #1441
#1453 opened by weinbe2 - 2
Constrain the tuning space for host <-> device MILCSiteOrder gauge field copies
#1451 opened by weinbe2 - 0
Integer overflow with staggered MG
#1448 opened by maddyscientist - 12
- 5
- 1
- 0
- 0
- 3
apparent "number of iterations regression" in double-half CG for (nd)tmclover
#1433 opened by kostrzewa - 4
Default tolerance values for fp64 and fp32.
#1425 opened by SaltyChiang - 0
Merge `hisq_stencil_test` and `hist_stencil_ctest` into a single file, following the pattern of `[staggered_]invert_test`
#1430 opened by weinbe2 - 0
Add Wilson functionality to contraction codes, ensure the covariant derivative code is properly tested.
#1423 opened by cpviolator - 0
- 3
mild failures in some DW ctests
#1410 opened by jcosborn - 1
Missing break in Gamma ?
#1406 opened by mathiaswagner - 2
CPS staggered phase in QUDA is different from CPS and Chroma's convention.
#1408 opened by SaltyChiang - 2
STRICT build of dslash_mdw_fused_* fails with sm_86
#1403 opened by jcosborn - 1
Sanity checking on the CPM.cmake download
#1399 opened by dmcdougall - 1
`staggered_dslash_test` asqtad verify fails with recon-9/13, partitioning enabled, and computing the fat/long gauge links
#1394 opened by weinbe2 - 6
- 0
Broad HQ residual cleanup
#1389 opened by weinbe2 - 1
Chroma build with QUDA and QDPXX
#1388 opened by Naresh-Eddula - 11
`site unroll not supported for nSpin = 2 nColor = 32` in coarse-grid-deflated MG
#1378 opened by kostrzewa