v4.3.2 update. (#2839)

This commit is contained in:
Junkai-Wu
2025-12-04 10:14:32 -05:00
committed by GitHub
parent 52ae719eda
commit bc680c7f67
19 changed files with 169 additions and 301 deletions
@@ -130,6 +130,9 @@ After serialization, compiled MLIR bytecode is stored in file.
The cache directory is ``/tmp/{current_user}/cutlass_python_cache``.
The cache loads from files into memory during |DSL| initialization and saves back to files when the process exits.
Note that for efficiency, the default cache directory is located in a temporary folder. However, this location is not persistent, it may be cleared by the system (for example, during a reboot or disk space cleanup).
If you wish to preserve the cache across sessions, set the ``CUTE_DSL_CACHE_DIR`` environment variable to point to a persistent directory.
The following environment variables control file caching:
.. code-block:: bash
@@ -140,6 +143,9 @@ The following environment variables control file caching:
# Maximum number of cache files allowed, defaults to 1000.
export CUTE_DSL_FILE_CACHING_CAPACITY=1000
# Cache directory, defaults to /tmp/{current_user}/cutlass_python_cache.
export CUTE_DSL_CACHE_DIR=/home/user/local_cutlass_python_cache/dense_gemm_cache/
Limitations
~~~~~~~~~~~~~~~~~~~~~