Peter Boyle
4e9df9e93c
GPU patches
2019-05-18 17:43:11 +01:00
Peter Boyle
9fe68857a9
Runs multiGPU with coalesced access on tesseract
2019-05-18 17:42:41 +01:00
Peter Boyle
37336c9e0c
Allow compress to be either vector or scalar types
2019-05-18 17:41:13 +01:00
Peter Boyle
6c4da3bbc7
Stencil now runs with coalesced accesses
2019-05-18 17:40:35 +01:00
Peter Boyle
a584b16c4a
Adding a non-blocking kernel launch
2019-05-18 17:39:54 +01:00
fionnoh
dbd7f3f0fc
Added variables that were missing from wall source setup
2019-05-17 19:10:09 +01:00
fionnoh
d14512ee03
Exposed a coulomb/landau enum to the gauge fixing module
2019-05-17 19:01:52 +01:00
Peter Boyle
48b1c806ed
Coulomb gauge added as an option
2019-05-17 17:36:32 +01:00
Felix Erben
8ce7ebdca3
fixed contraction issue
2019-05-17 10:52:55 +01:00
Felix Erben
435653490e
fixed contraction issue
2019-05-17 10:50:15 +01:00
Michael Marshall
10a052d695
3 issues preventing compilation under clang. Marked these with FELIX_ISSUE and made minimal change to make compile (as fix not obvious)
2019-05-17 09:59:01 +01:00
Felix Erben
acd5a01b65
some work on baryons
2019-05-16 15:11:50 +01:00
0a8b6724ef
Merge pull request #209 from fionnoh/develop
...
Added gauge transform option to eigpack IO
2019-05-15 18:09:44 +02:00
fionnoh
ce102ac550
More logging, timing, and 4d/5d logic for eigpack gauge transforms
2019-05-15 14:31:25 +01:00
fionnoh
94accec311
Added gauge transform option to eigpack IO
2019-05-15 13:35:47 +01:00
gfilaci
1a82533d22
fix inner product with thrust reduction
2019-05-14 15:35:54 +01:00
Michael Marshall
ec7d96ce3b
Merge branch 'develop' into feature/distil
...
* develop:
Hadron WeakEye and A2ALoop bug fixes, and WWVVContraction bug fix
DiskVector: fix of memory bug triggering segfault when the cache is accessed following a certain pattern
MFermion::GaugeProp fix for 4d fields
2019-05-14 13:10:40 +01:00
gfilaci
e3c56fd9b3
CayleyZeroCounters before benchmark loop
2019-05-13 15:52:00 +01:00
gfilaci
955cc7790f
MooeeInvDag offloaded to GPU
2019-05-13 14:25:29 +01:00
gfilaci
1179123ac2
MooeeInv offloaded to GPU
2019-05-13 12:37:12 +01:00
d8512b03f8
Merge pull request #195 from nils-asmussen/fix_GaugeProp_4d
...
MFermion::GaugeProp fix for 4d fields
2019-05-12 21:31:18 +02:00
d90cf9d022
Merge pull request #207 from fionnoh/develop
...
Weak Hamiltonian and contraction bug fixes
2019-05-12 21:30:20 +02:00
79e930ba12
Hadrons: Lepton Propagator for kl2, sign swap for antiperiodic boundary
2019-05-10 12:46:18 +01:00
gfilaci
22e35c9ddd
M5Ddag offloaded to GPU
2019-05-10 12:23:39 +01:00
gfilaci
698b45e163
remove unused typedef
2019-05-09 11:19:39 +01:00
gfilaci
f1744b3f01
M5D offloaded to GPU
2019-05-09 11:17:55 +01:00
gfilaci
2b3c22f03d
bandwidth dependent on grid default precision
2019-05-08 12:01:11 +01:00
gfilaci
8423a05940
duplicate CayleyFermion5D for gpu
2019-05-08 11:51:37 +01:00
fionnoh
2acd8ece65
Hadron WeakEye and A2ALoop bug fixes, and WWVVContraction bug fix
2019-05-08 10:57:36 +01:00
fionnoh
b638509c61
Merge branch 'develop' of https://github.com/paboyle/Grid into develop
2019-05-08 10:51:04 +01:00
Michael Marshall
c16916cc45
Multiple local slice fixes
2019-05-06 10:35:42 +01:00
Michael Marshall
a865caf0d2
Forgot a const in IndexName only version of NamedTensor constructor
2019-05-03 22:17:25 +01:00
Michael Marshall
9ae4d369f3
Use the definition of the Perambulator Index names given in Hadrons::MDistil
2019-05-03 22:00:50 +01:00
edeb590818
DiskVector: fix of memory bug triggering segfault when the cache is accessed following a certain pattern
2019-05-03 17:09:47 +01:00
Michael Marshall
ec24a1f828
Fixed 2 bugs in LapEvec: 1) InsertLocalSlice 2) ensure convergence assertion stops entire machine
2019-05-03 16:03:56 +01:00
Michael Marshall
0efe63f6fa
3D smearing fix
2019-05-02 19:37:59 +01:00
Michael Marshall
b7ead6c16a
Fixed bug: iff stout smearing disabled then gauge field uninitialised
2019-05-02 18:20:49 +01:00
gfilaci
d9438627d9
M5D benchmark without vector copy overhead
2019-05-02 11:10:57 +01:00
gfilaci
b23305dbe2
fix M5D flop count
2019-05-02 11:08:21 +01:00
gfilaci
d3b5c02e2d
measure M5D bandwidth and fix M5D flop count
2019-05-02 11:02:39 +01:00
gfilaci
8b6541fb60
Fix gpu MultRealPart and MaddRealPart bug
2019-05-02 10:58:17 +01:00
gfilaci
6da9aa9971
replace std::vector with Vector in benchmark
2019-05-02 10:56:22 +01:00
gfilaci
44e0360b97
replace std::vector with Vector
2019-05-02 10:55:36 +01:00
gfilaci
9003c4a07c
allocator copy constructor (to be fixed)
2019-05-02 10:53:37 +01:00
gfilaci
b52fa38f8c
seed initialisation of RNG5
2019-05-02 10:36:09 +01:00
gfilaci
3f1c4d8789
fix comment hash
2019-05-02 10:24:36 +01:00
Michael Marshall
62692b68b9
I'd forgotten that Intel '17 doesn't like auto var{value}; syntax
2019-05-01 20:45:16 +01:00
Michael Marshall
311c35a15c
Looking for fixes for Intel '17 compiler errors. std::cout << complex number ?
2019-05-01 18:22:08 +01:00
Michael Marshall
a3fe57f430
NamedTensor writes to tag NamedTensor by default (not filename) - so still usable in case user renames file.
...
Also tweaked tensor index name checking (which is used to ensure tensor is correct type)
2019-05-01 18:11:37 +01:00
Michael Marshall
8dc0587621
Post Michael / Felix review. Ready for Peter / Antonin review
2019-05-01 13:04:51 +01:00
