Lei Wang
fded655836
[Runtime][Cache] Detect and repair corrupted cache entries ( #3074 )
...
Harden cache entries against partial writes
2026-08-24 17:15:45 +08:00
Lei Wang
e3bef1ceac
[Runtime][Cache] Remove separate cache temporary directory ( #3069 )
...
* [CUDA][Cache] Keep binary cache writes on one filesystem
* [Runtime][Cache] Remove separate cache temporary directory
* [Tests] Drop redundant autotune staging regression
* [Tests] Keep kernel cache cleanup focused
* [Tests] Keep CUDA cache cleanup focused
2026-08-23 21:33:47 +08:00
Chaofan Lin
fef9f76092
[Backend] Add unified backend resolution policy ( #2318 )
...
* Refactor backend module registration
* Define backends with explicit specs
* Require explicit backend for codegen
* Rename backend spec to backend module
* Keep host codegen on selected backend
* Introduce backend context for compilation
* Remove legacy backend compatibility APIs
* Restore backend API docstrings
2026-08-06 02:41:57 +08:00
Lei Wang
0d97af896e
[CUDA][Cache] Include compile options in binary cache key ( #2532 )
...
* Include CUDA compile options in binary cache key
* enhance
2026-07-07 00:09:22 +08:00
Lei Wang
46f3d1c924
[CUDA][JIT][Cache] Add cross-host CUDA binary cache ( #2459 )
...
* Add CUDA target code attr support
* Add CUDA target normalization support
Introduced a new function to normalize CUDA targets, ensuring that the target is correctly identified and transformed based on the detected architecture. Added a test to verify that the bare CUDA target uses the detected architecture accurately.
* Add cross-host CUDA binary cache
* Nest host kernel cache under version namespace
2026-06-26 15:56:04 +08:00
Lei Wang