Code

Update CI matrix to use NVKS nodes.#3572

Code

Merged

alliepiper merged 9 commits intoNVIDIA:mainNVIDIA/cccl:mainfrom alliepiper:nvks_runnersalliepiper/cccl:nvks_runnersCopy head branch name to clipboardJan 30, 2025

+90-61

Commits

Commits on Jan 28, 2025

• Update CI matrix to use NVKS nodes.

  Show description for b8e1549
  alliepiper

  committed
  b8e1549

  Copy full SHA for b8e1549

• Update windows CI scripts to accept -arch.

  alliepiper

  committed
  50864b9

  Copy full SHA for 50864b9

Commits on Jan 29, 2025

• Move all non-Catch2 device algo tests to lid0/lid1.

  Show description for 571e729
  alliepiper

  committed
  571e729

  Copy full SHA for 571e729

• Move libcudacxx builds to t4 temporarily.

  Show description for 6434ef5
  alliepiper

  committed
  6434ef5

  Copy full SHA for 6434ef5

• libcudacxx test fail on t4, too. Moving back to v100.

  alliepiper

  committed
  bf68a88

  Copy full SHA for bf68a88

• TEMP override matrix to experiment with libcu++ failures.

  alliepiper

  committed
  2977b2e

  Copy full SHA for 2977b2e

• Remove sm: 'gpu' from most CI jobs.

  Show description for 9f70b22
  alliepiper

  committed
  9f70b22

  Copy full SHA for 9f70b22

• Move t4 jobs to v100 while runners team investigates issues.

  alliepiper

  committed
  ee22996

  Copy full SHA for ee22996

• Switch to all rtx queues:

  Show description for ea4d40e
  alliepiper

  committed
  ea4d40e

  Copy full SHA for ea4d40e