Repository navigation

Abstract CUDA hardcodes into configurable te_device_type / te_platform - #3113

Closed

lxd-cumt wants to merge 8 commits into

NVIDIA:release_v2.14from

lxd-cumt:cuda_patch

Closed

Abstract CUDA hardcodes into configurable te_device_type / te_platform#3113
lxd-cumt wants to merge 8 commits into
NVIDIA:release_v2.14from
lxd-cumt:cuda_patch

Commits on Jun 17, 2026
- fake: try to patch cuda hardcodes
  lxd-cumt
  committed
  cdefc94
- remove vendor related
  lxd-cumt
  committed
  e5e85a9
- support apply paatches
  lxd-cumt
  committed
  aff2e8a
- fix: use transformer_engine_plugin_fl interface and replace assert with explicit raise
  lxd-cumt
  committed
  561efb3
- fix: patch cuda hardcodes for plugin compatibility
  lxd-cumt
  committed
  8ae24c3
Commits on Jul 7, 2026
- fix: move torch-dependent globals to pytorch/__init__.py and remove redundant code
  lxd-cumt
  committed
  7529fba
- fix: use NVTE_PLUGIN env var instead of hardcoded plugin module name
  lxd-cumt
  committed
  b6f31a8
Commits on Aug 19, 2026
- Simplify plugin device/platform interface and fix minor issues
  lxd-cumt
  committed
  bccb702