Code

C++17 parallelism #273

Code

Open

jeffhammond wants to merge 182 commits intoLLNL:developLLNL/RAJAPerf:developfrom jeffhammond:new-stdparjeffhammond/RAJAPerf:new-stdparCopy head branch name to clipboard

+8,127-42

Commits

Commits on Jul 7, 2022

• starting over with StdPar because git submodules are trash

  jeffhammond

  committed
  2543ff8

  Copy full SHA for 2543ff8

• starting over with StdPar because git submodules are trash

  jeffhammond

  committed
  a435877

  Copy full SHA for a435877

• starting over with StdPar because git submodules are trash

  jeffhammond

  committed
  9cf1e4b

  Copy full SHA for 9cf1e4b

Commits on Jul 8, 2022

• fix MEM***

  jeffhammond

  committed
  6bcc445

  Copy full SHA for 6bcc445

• fix MEM*** more

  jeffhammond

  committed
  bb16c99

  Copy full SHA for bb16c99

• add exec and alg to stdpar helper header

  jeffhammond

  committed
  7f784d1

  Copy full SHA for 7f784d1

• more stdpar

  jeffhammond

  committed
  41d9cd2

  Copy full SHA for 41d9cd2

• add SCAN and NODAL

  jeffhammond

  committed
  710a084

  Copy full SHA for 710a084

• add more stuff

  jeffhammond

  committed
  87e0111

  Copy full SHA for 87e0111

• index list

  jeffhammond

  committed
  4a674d3

  Copy full SHA for 4a674d3

• reduce sum

  jeffhammond

  committed
  48152bc

  Copy full SHA for 48152bc

• cout fix

  jeffhammond

  committed
  f0ad8c2

  Copy full SHA for f0ad8c2

• cout fix

  jeffhammond

  committed
  d55a2d5

  Copy full SHA for d55a2d5

• cleanup

  jeffhammond

  committed
  64a9768

  Copy full SHA for 64a9768

• README

  jeffhammond

  committed
  1cda0fd

  Copy full SHA for 1cda0fd

• fix unroll pragma

  jeffhammond

  committed
  a1470e1

  Copy full SHA for a1470e1

• GCC

  jeffhammond

  committed
  263821a

  Copy full SHA for 263821a

• remove RAJA_StdPar

  jeffhammond

  committed
  bf344cf

  Copy full SHA for bf344cf

• fix the issue with running

  jeffhammond

  committed
  8ce9c41

  Copy full SHA for 8ce9c41

• NVC note

  jeffhammond

  committed
  27f8007

  Copy full SHA for 27f8007

• nuke StdPar SORTPAIRS because it needs work

  jeffhammond

  committed
  e4f1604

  Copy full SHA for e4f1604

• notes on NVC++ multicore issues

  jeffhammond

  committed
  2e3bb2d

  Copy full SHA for 2e3bb2d

• NVC++ GPU fails here

  jeffhammond

  committed
  a85a5c6

  Copy full SHA for a85a5c6

• more errata

  jeffhammond

  committed
  fe8f916

  Copy full SHA for fe8f916

• all the erratum

  jeffhammond

  committed
  f56c87a

  Copy full SHA for f56c87a

• pointer to atomic for GPU

  jeffhammond

  committed
  ccbd4f0

  Copy full SHA for ccbd4f0

• update PI_ATOMIC

  jeffhammond

  committed
  51cbb35

  Copy full SHA for 51cbb35

• fix this one - nested way faster on GPU

  jeffhammond

  committed
  790d74e

  Copy full SHA for 790d74e

• fix this one - nested way faster on GPU

  jeffhammond

  committed
  2cc38e7

  Copy full SHA for 2cc38e7

• remove RAJA_StdPar

  jeffhammond

  committed
  1a77c06

  Copy full SHA for 1a77c06

Commits on Jul 12, 2022

• disable Lambda_StdPar; use par not par_unseq

  jeffhammond

  committed
  c986bf9

  Copy full SHA for c986bf9

• enable par_unseq even though it is slower

  jeffhammond

  committed
  563c969

  Copy full SHA for 563c969

• disable Lambda_StdPar

  jeffhammond

  committed
  ea4bccb

  Copy full SHA for ea4bccb

• move collapse choice here

  jeffhammond

  committed
  2d54a4b

  Copy full SHA for 2d54a4b

• partially implement SORTPAIRS with StdPar (GPU issues)

  jeffhammond

  committed
  d9a94c2

  Copy full SHA for d9a94c2

• partially implement SORTPAIRS with StdPar (GPU issues)

  jeffhammond

  committed
  6f1fbbd

  Copy full SHA for 6f1fbbd

• bring USE_STDPAR_COLLAPSE into common header

  jeffhammond

  committed
  3f39c8b

  Copy full SHA for 3f39c8b

• add StdPar impl using std::reduce

  jeffhammond

  committed
  6548d9c

  Copy full SHA for 6548d9c

• add comment why GPU disabled

  jeffhammond

  committed
  33bb44e

  Copy full SHA for 33bb44e

• add SCAN StdPar but wrong on GPU???

  jeffhammond

  committed
  0cd37a3

  Copy full SHA for 0cd37a3

• start working on INDEXLIST StdPar

  jeffhammond

  committed
  4f8e7ab

  Copy full SHA for 4f8e7ab

• implement DAXPY_ATOMIC StdPar using a variety of atomics, because C++20 atomic_ref is not widely available

  jeffhammond

  committed
  c7c292b

  Copy full SHA for c7c292b

• use std:: not RAJA min/max

  jeffhammond

  committed
  d4f47a5

  Copy full SHA for d4f47a5

• implement REDUCE_STRUCT Base_StdPar

  jeffhammond

  committed
  16c2099

  Copy full SHA for 16c2099

• remove warning

  jeffhammond

  committed
  7ece3bf

  Copy full SHA for 7ece3bf

• FIRST_MIN Lambda_StdPar unimplemented

  jeffhammond

  committed
  7a58180

  Copy full SHA for 7a58180

• s/RAJA_MAX/std::max/g

  jeffhammond

  committed
  5fa7743

  Copy full SHA for 5fa7743

• CPU StdPar

  jeffhammond

  committed
  7ea72bc

  Copy full SHA for 7ea72bc

• Merge branch 'new-stdpar' of https://github.com/jeffhammond/RAJAPerf into new-stdpar

  jeffhammond

  committed
  c2147ee

  Copy full SHA for c2147ee

• disable atomic_ref

  jeffhammond

  committed
  013c0ad

  Copy full SHA for 013c0ad

• CPU info

  jeffhammond

  committed
  e99f8c3

  Copy full SHA for e99f8c3

• Merge remote-tracking branch 'origin/new-stdpar' into new-stdpar

  jeffhammond

  committed
  88baaeb

  Copy full SHA for 88baaeb

• fix no GPU StdPar in SCAN

  jeffhammond

  committed
  7213acc

  Copy full SHA for 7213acc

• Lambda_StdPar HEAT_3D added

  jeffhammond

  committed
  e1791c7

  Copy full SHA for e1791c7

Commits on Aug 2, 2022

• Merge branch 'LLNL:develop' into new-stdpar

  jeffhammond

  authored
  ffb45e3

  Copy full SHA for ffb45e3

Commits on Aug 31, 2022

• starting over with StdPar because git submodules are trash

  jeffhammond

  committed
  d06f0ce

  Copy full SHA for d06f0ce

• starting over with StdPar because git submodules are trash

  jeffhammond

  committed
  4b6d7c9

  Copy full SHA for 4b6d7c9

• starting over with StdPar because git submodules are trash

  jeffhammond

  committed
  baae1b2

  Copy full SHA for baae1b2

• fix MEM***

  jeffhammond

  committed
  0044ac9

  Copy full SHA for 0044ac9

• add exec and alg to stdpar helper header

  jeffhammond

  committed
  4a57467

  Copy full SHA for 4a57467

• more stdpar

  jeffhammond

  committed
  e0349b4

  Copy full SHA for e0349b4

• add SCAN and NODAL

  jeffhammond

  committed
  cbe5356

  Copy full SHA for cbe5356

• add more stuff

  jeffhammond

  committed
  43b9f8f

  Copy full SHA for 43b9f8f

• index list

  jeffhammond

  committed
  d7c7a68

  Copy full SHA for d7c7a68

• reduce sum

  jeffhammond

  committed
  854871b

  Copy full SHA for 854871b

• cout fix

  jeffhammond

  committed
  ab62562

  Copy full SHA for ab62562

• cout fix

  jeffhammond

  committed
  1f10cdc

  Copy full SHA for 1f10cdc

• cleanup

  jeffhammond

  committed
  c3d1dbe

  Copy full SHA for c3d1dbe

• README

  jeffhammond

  committed
  4c3519e

  Copy full SHA for 4c3519e

• fix unroll pragma

  jeffhammond

  committed
  8c34177

  Copy full SHA for 8c34177

• GCC

  jeffhammond

  committed
  a6d6bc4

  Copy full SHA for a6d6bc4

• remove RAJA_StdPar

  jeffhammond

  committed
  c9a6ec5

  Copy full SHA for c9a6ec5

• fix the issue with running

  jeffhammond

  committed
  1e3f624

  Copy full SHA for 1e3f624

• NVC note

  jeffhammond

  committed
  21885f7

  Copy full SHA for 21885f7

• nuke StdPar SORTPAIRS because it needs work

  jeffhammond

  committed
  6a26a5f

  Copy full SHA for 6a26a5f

• notes on NVC++ multicore issues

  jeffhammond

  committed
  de6a3d2

  Copy full SHA for de6a3d2

• NVC++ GPU fails here

  jeffhammond

  committed
  f07036b

  Copy full SHA for f07036b

• more errata

  jeffhammond

  committed
  00a107a

  Copy full SHA for 00a107a

• all the erratum

  jeffhammond

  committed
  92f8d12

  Copy full SHA for 92f8d12

• pointer to atomic for GPU

  jeffhammond

  committed
  645bb4a

  Copy full SHA for 645bb4a

• update PI_ATOMIC

  jeffhammond

  committed
  dda1d64

  Copy full SHA for dda1d64

• fix this one - nested way faster on GPU

  jeffhammond

  committed
  8bbcf1a

  Copy full SHA for 8bbcf1a

• fix this one - nested way faster on GPU

  jeffhammond

  committed
  88547f2

  Copy full SHA for 88547f2

• remove RAJA_StdPar

  jeffhammond

  committed
  65963b6

  Copy full SHA for 65963b6

• disable Lambda_StdPar; use par not par_unseq

  jeffhammond

  committed
  bdf03ba

  Copy full SHA for bdf03ba

• enable par_unseq even though it is slower

  jeffhammond

  committed
  eeb41c8

  Copy full SHA for eeb41c8

• disable Lambda_StdPar

  jeffhammond

  committed
  c9e981b

  Copy full SHA for c9e981b

• move collapse choice here

  jeffhammond

  committed
  9a07e35

  Copy full SHA for 9a07e35

• partially implement SORTPAIRS with StdPar (GPU issues)

  jeffhammond

  committed
  de7ab8b

  Copy full SHA for de7ab8b

• partially implement SORTPAIRS with StdPar (GPU issues)

  jeffhammond

  committed
  e4e06b0

  Copy full SHA for e4e06b0

• bring USE_STDPAR_COLLAPSE into common header

  jeffhammond

  committed
  e2fb576

  Copy full SHA for e2fb576

• add StdPar impl using std::reduce

  jeffhammond

  committed
  6bf1600

  Copy full SHA for 6bf1600

• add comment why GPU disabled

  jeffhammond

  committed
  cc5bc50

  Copy full SHA for cc5bc50

• add SCAN StdPar but wrong on GPU???

  jeffhammond

  committed
  ac4a8f1

  Copy full SHA for ac4a8f1

• start working on INDEXLIST StdPar

  jeffhammond

  committed
  252c98c

  Copy full SHA for 252c98c

• implement DAXPY_ATOMIC StdPar using a variety of atomics, because C++20 atomic_ref is not widely available

  jeffhammond

  committed
  dd4a5e0

  Copy full SHA for dd4a5e0

• use std:: not RAJA min/max

  jeffhammond

  committed
  2e4a4ef

  Copy full SHA for 2e4a4ef

• implement REDUCE_STRUCT Base_StdPar

  jeffhammond

  committed
  34b8324

  Copy full SHA for 34b8324

• remove warning

  jeffhammond

  committed
  de072c1

  Copy full SHA for de072c1

• FIRST_MIN Lambda_StdPar unimplemented

  jeffhammond

  committed
  7b05b04

  Copy full SHA for 7b05b04

• s/RAJA_MAX/std::max/g

  jeffhammond

  committed
  97e5348

  Copy full SHA for 97e5348

• CPU StdPar

  jeffhammond

  committed
  c98264d

  Copy full SHA for c98264d

• disable atomic_ref

  jeffhammond

  committed
  15ed8ac

  Copy full SHA for 15ed8ac

• CPU info

  jeffhammond

  committed
  0d75450

  Copy full SHA for 0d75450

• fix no GPU StdPar in SCAN

  jeffhammond

  committed
  e25ec99

  Copy full SHA for e25ec99

• Lambda_StdPar HEAT_3D added

  jeffhammond

  committed
  f842555

  Copy full SHA for f842555

• Merge branch 'new-stdpar' of https://github.com/jeffhammond/RAJAPerf into new-stdpar

  jeffhammond

  committed
  185d2c4

  Copy full SHA for 185d2c4

• fix correctness issues caused by emplace_back in parallel

  Show description for 7a96a08
  jeffhammond

  committed
  7a96a08

  Copy full SHA for 7a96a08

• deal with exception mess

  jeffhammond

  committed
  a38b190

  Copy full SHA for a38b190

• fix emplace_back and deal with exceptions

  jeffhammond

  committed
  7b4039c

  Copy full SHA for 7b4039c

• remove listed issues that have been fixed

  jeffhammond

  committed
  8f7e804

  Copy full SHA for 8f7e804

• no clue what to do with these

  jeffhammond

  committed
  d7716e5

  Copy full SHA for d7716e5

• code compiles but is wrong, like Base

  jeffhammond

  committed
  ea0feff

  Copy full SHA for ea0feff

• code compiles but is wrong, like Base

  jeffhammond

  committed
  aea0bca

  Copy full SHA for aea0bca

Commits on Oct 11, 2022

• change output

  jeffhammond

  committed
  69ce57c

  Copy full SHA for 69ce57c

• addd comment

  jeffhammond

  committed
  e4248d6

  Copy full SHA for e4248d6

• collapse in StdPar is really important for GPU

  jeffhammond

  committed
  db398c2

  Copy full SHA for db398c2

• collapse in StdPar is really important for GPU

  jeffhammond

  committed
  274ad60

  Copy full SHA for 274ad60

• fixed lambda collapse - need to do the base version and cleanup

  jeffhammond

  committed
  076abfb

  Copy full SHA for 076abfb

Commits on Oct 12, 2022

• fix atomic_ref

  jeffhammond

  committed
  a884cd1

  Copy full SHA for a884cd1

• move openacc.h header to stdpar common header

  jeffhammond

  committed
  1456962

  Copy full SHA for 1456962

• use for_each everywhere - correct with intel and nvhpc it seems

  jeffhammond

  committed
  00bd052

  Copy full SHA for 00bd052

• debug FDTD_2D with GCC

  jeffhammond

  committed
  cfa64d9

  Copy full SHA for cfa64d9

• debug the GCC problem more

  jeffhammond

  committed
  2aeb29b

  Copy full SHA for 2aeb29b

• add a note about a bad idea

  jeffhammond

  committed
  eef71d5

  Copy full SHA for eef71d5

• ugh atomics are such a pain right now

  jeffhammond

  committed
  788e2b4

  Copy full SHA for 788e2b4

• change atomics again

  jeffhammond

  committed
  1f10a86

  Copy full SHA for 1f10a86

• solve lambda capture t issue a diff way

  jeffhammond

  committed
  3c93637

  Copy full SHA for 3c93637

• switch to for_each_n in algorithm

  jeffhammond

  committed
  f64306d

  Copy full SHA for f64306d

• switch to for_each_n in stream

  jeffhammond

  committed
  50a661d

  Copy full SHA for 50a661d

• switch to for_each_n in basic

  jeffhammond

  committed
  3c2368b

  Copy full SHA for 3c2368b

• for_each_n

  jeffhammond

  committed
  0b67f20

  Copy full SHA for 0b67f20

• for_each_n

  jeffhammond

  committed
  c18465d

  Copy full SHA for c18465d

• fix collapsed case

  jeffhammond

  committed
  1aa331f

  Copy full SHA for 1aa331f

• for_each_n

  jeffhammond

  committed
  27e59b3

  Copy full SHA for 27e59b3

Commits on Oct 13, 2022

• for_each_n

  jeffhammond

  committed
  456e722

  Copy full SHA for 456e722

• for_each_n

  jeffhammond

  committed
  a31da74

  Copy full SHA for a31da74

• add failed experiment w xform red

  jeffhammond

  committed
  7fbae76

  Copy full SHA for 7fbae76

• for_each_n

  jeffhammond

  committed
  6853072

  Copy full SHA for 6853072

• for_each_n

  jeffhammond

  committed
  4a82e0e

  Copy full SHA for 4a82e0e

• for_each_n

  jeffhammond

  committed
  8207cb1

  Copy full SHA for 8207cb1

• for_each_n

  jeffhammond

  committed
  02d5466

  Copy full SHA for 02d5466

• for_each_n

  jeffhammond

  committed
  e0a0331

  Copy full SHA for e0a0331

• for_each_n

  jeffhammond

  committed
  7af9e3d

  Copy full SHA for 7af9e3d

• for_each_n

  jeffhammond

  committed
  f70e3c9

  Copy full SHA for f70e3c9

• fixed unused warnings

  jeffhammond

  committed
  45fc93b

  Copy full SHA for 45fc93b

• disable this, since it requires SLM

  jeffhammond

  committed
  8074f41

  Copy full SHA for 8074f41

• remove lambda version - will not bother with this

  jeffhammond

  committed
  eb3ecee

  Copy full SHA for eb3ecee

• remove since not implementing this

  jeffhammond

  committed
  c1b4034

  Copy full SHA for c1b4034

• remove since not implementing this

  jeffhammond

  committed
  9893a3a

  Copy full SHA for 9893a3a

• need empty impl after all

  jeffhammond

  committed
  c59a972

  Copy full SHA for c59a972

• need empty impl after all

  jeffhammond

  committed
  83ccc76

  Copy full SHA for 83ccc76

• enable INDEXLIST_3LOOP StdPar

  jeffhammond

  committed
  1f85e58

  Copy full SHA for 1f85e58

• fix for_each_n

  jeffhammond

  committed
  1499974

  Copy full SHA for 1499974

• INDEXLIST_3LOOP validated manually; Intel busted

  jeffhammond

  committed
  6a91b3d

  Copy full SHA for 6a91b3d

• GCC broke too

  jeffhammond

  committed
  b406723

  Copy full SHA for b406723

Commits on Oct 26, 2022

• suppress misspelled

  jeffhammond

  committed
  e51cd8c

  Copy full SHA for e51cd8c

• cleanup workaround for __throw_bad_array_new_length

  jeffhammond

  committed
  bc5976b

  Copy full SHA for bc5976b

Commits on Oct 27, 2022

• merge develop

  jeffhammond

  committed
  72c1065

  Copy full SHA for 72c1065

Commits on Oct 31, 2022

• rewrite to use transform not for_each_n

  jeffhammond

  committed
  85f65ff

  Copy full SHA for 85f65ff

• rewrite to use transform not for_each_n

  jeffhammond

  committed
  00e8690

  Copy full SHA for 00e8690

Commits on Nov 3, 2022

• explicit seq because cannot find the implicit one for who knows why

  jeffhammond

  committed
  9d8d3e2

  Copy full SHA for 9d8d3e2

• Merge branch 'new-stdpar' of https://github.com/jeffhammond/RAJAPerf into new-stdpar

  jeffhammond

  committed
  eb81d25

  Copy full SHA for eb81d25

• use transform instead

  jeffhammond

  committed
  00a5498

  Copy full SHA for 00a5498

• use transform instead

  jeffhammond

  committed
  25ca55d

  Copy full SHA for 25ca55d

Commits on Mar 28, 2023

• dunno

  jeffhammond

  committed
  c9f0295

  Copy full SHA for c9f0295

• fix all merge problems

  jeffhammond

  committed
  f54afcd

  Copy full SHA for f54afcd

• fix bad merge

  jeffhammond

  committed
  c86144c

  Copy full SHA for c86144c

• add markdown

  jeffhammond

  committed
  834b519

  Copy full SHA for 834b519

• this was WIP - restore serial fallback

  jeffhammond

  committed
  74c9fda

  Copy full SHA for 74c9fda

• give up on std::atomic<double> for now

  jeffhammond

  committed
  16b9f09

  Copy full SHA for 16b9f09

Commits on Mar 31, 2023

• Merge branch 'develop' into new-stdpar

  jeffhammond

  committed
  323d780

  Copy full SHA for 323d780

• add RAJA_ENABLE_STDPAR guard

  jeffhammond

  committed
  043bb5a

  Copy full SHA for 043bb5a

• add RAJA_ENABLE_STDPAR guard

  jeffhammond

  committed
  4c40ab0

  Copy full SHA for 4c40ab0

• add RAJA_ENABLE_STDPAR guard

  jeffhammond

  committed
  75a6eb6

  Copy full SHA for 75a6eb6

• add RAJA_ENABLE_STDPAR guard

  jeffhammond

  committed
  6770f11

  Copy full SHA for 6770f11

• add RAJA_ENABLE_STDPAR guard

  jeffhammond

  committed
  8402b34

  Copy full SHA for 8402b34

• N namespace oops

  jeffhammond

  committed
  d4db911

  Copy full SHA for d4db911

• add RAJA_ENABLE_STDPAR guard

  jeffhammond

  committed
  75b16b9

  Copy full SHA for 75b16b9

• end namespace oops

  jeffhammond

  committed
  23d04fe

  Copy full SHA for 23d04fe

• change guard name to BUILD_STDPAR

  jeffhammond

  committed
  bbe72dd

  Copy full SHA for bbe72dd

• Update ADD-StdPar.cpp

  Show description for 38bf9bd
  jeffhammond

  authored
  38bf9bd

  Copy full SHA for 38bf9bd