Code

Cudastf#794

Code

Draft

sidelnik wants to merge 41 commits intoNVIDIA:mainNVIDIA/MatX:mainfrom

sidelnik:cudastfsidelnik/MatX:cudastfCopy head branch name to clipboard

Commits

Commits on Nov 4, 2024

• Update build config to pull CUDASTF

  sidelnik

  committed
  d6dc01d

  Copy full SHA for d6dc01d

• remove const expr

  sidelnik

  committed
  245b20f

  Copy full SHA for 245b20f

Commits on Nov 5, 2024

• Updates to get basic cudastf functionality working with matx

  sidelnik

  committed
  9b35ec8

  Copy full SHA for 9b35ec8

• move to void_interface

  sidelnik

  committed
  7d298d4

  Copy full SHA for 7d298d4

• add stf executor

  sidelnik

  committed
  154b3f9

  Copy full SHA for 154b3f9

• support for cgsolve operator and a few examples

  sidelnik

  committed
  c8ef988

  Copy full SHA for c8ef988

Commits on Dec 3, 2024

• make the sync() that is part of stfexecutor call ctx.task_fence()

  sidelnik

  committed
  52b18c9

  Copy full SHA for 52b18c9

• fix typo

  sidelnik

  committed
  d726b10

  Copy full SHA for d726b10

Commits on Dec 17, 2024

• Added test case

  sidelnik

  committed
  5e7576c

  Copy full SHA for 5e7576c

• Fixes to the sync

  sidelnik

  committed
  1373699

  Copy full SHA for 1373699

• add support for cgsolve

  sidelnik

  committed
  92e7204

  Copy full SHA for 92e7204

• update to the simple radar code

  sidelnik

  committed
  a608f3f

  Copy full SHA for a608f3f

• minor typo fix

  sidelnik

  committed
  b062577

  Copy full SHA for b062577

Commits on Dec 19, 2024

• update version of stf

  sidelnik

  committed
  bbf9abc

  Copy full SHA for bbf9abc

• cleanup constexpr case for stfexecutor

  sidelnik

  committed
  3e831ea

  Copy full SHA for 3e831ea

• cleanup constexpr case for stfexecutor

  sidelnik

  committed
  702fe79

  Copy full SHA for 702fe79

• add conditional support for cudagraph to the stf executor

  sidelnik

  committed
  5bfe21e

  Copy full SHA for 5bfe21e

Commits on Jan 9, 2025

• update to latest cudastf

  sidelnik

  committed
  f407256

  Copy full SHA for f407256

• switch to use logical token

  sidelnik

  committed
  221599c

  Copy full SHA for 221599c

• update parameters for radar code

  sidelnik

  committed
  7a5bb6c

  Copy full SHA for 7a5bb6c

• update to radar code to work with command line args

  sidelnik

  committed
  0c2432f

  Copy full SHA for 0c2432f

• cleanup to support different executor

  sidelnik

  committed
  3ae267b

  Copy full SHA for 3ae267b

Commits on Jan 24, 2025

• cleanup radar code to emit stf and cuda versions

  sidelnik

  committed
  6a75794

  Copy full SHA for 6a75794

• test script that runs simple radar with different input sizes. outputs a .csv file of results

  sidelnik

  committed
  f1facca

  Copy full SHA for f1facca

• enable cuda graphs as a command line argument enableGraphs

  sidelnik

  committed
  0199e75

  Copy full SHA for 0199e75

Commits on Jan 27, 2025

• add support for the random/randomOp generator

  sidelnik

  committed
  39b16f4

  Copy full SHA for 39b16f4

• get the basic spectrogram code working with stf

  sidelnik

  committed
  9b7c4b0

  Copy full SHA for 9b7c4b0

• get spectrogram cudagraph code working with stf

  sidelnik

  committed
  f9e09f1

  Copy full SHA for f9e09f1

Commits on Feb 10, 2025

• add assert in the case stream capture is turned on if creating a plan

  sidelnik

  committed
  6c9a791

  Copy full SHA for 6c9a791

Commits on Mar 19, 2025

• Merge branch 'cudastf' into cudastf_latest

  sidelnik

  authored
  a1efd1c

  Copy full SHA for a1efd1c

• Merge pull request #2 from sidelnik/cudastf_latest

  Show description for 6437eab
  sidelnik

  authored
  6437eab

  Copy full SHA for 6437eab

Commits on Mar 24, 2025

• Apps using matx with stf should get these flags

  caugonnet

  committed
  bbb9aae

  Copy full SHA for bbb9aae

• fix constructor

  caugonnet

  committed
  e13c9b6

  Copy full SHA for e13c9b6

Commits on Apr 21, 2025

• fix typo/bug

  sidelnik

  committed
  7244399

  Copy full SHA for 7244399

• update to example code to fix compile error

  sidelnik

  committed
  66f6850

  Copy full SHA for 66f6850

• update to example code to fix compile error

  sidelnik

  committed
  89e2a43

  Copy full SHA for 89e2a43

• update test script for radar code

  sidelnik

  committed
  973886b

  Copy full SHA for 973886b

• temp fix to the allocator dtor

  sidelnik

  committed
  92885e7

  Copy full SHA for 92885e7

• remove warning to work with latest stf

  sidelnik

  committed
  8607840

  Copy full SHA for 8607840

• replace logical token with token

  sidelnik

  committed
  14e0985

  Copy full SHA for 14e0985

• update version to use cccl from main

  sidelnik

  committed
  92e04d5

  Copy full SHA for 92e04d5