NVIDIA/cutlass
1
0
Fork 0
mirror of https://github.com/NVIDIA/cutlass.git synced 2026-05-12 09:15:56 +00:00
Code Issues Packages Projects Releases Wiki Activity

update doc

Browse Source
This commit is contained in:
Fung Xie
2025-11-25 21:26:00 -08:00
parent f71892b824
commit 8fc9bc5dda
1 changed files with 0 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
media/docs/pythonDSL/cute_dsl_general/compile_with_tvm_ffi.rst
View File

@@ -351,18 +351,6 @@ The TVM FFI function supports the following |DSL|-specific types as arguments:
- Python tuple of corresponding call-time types.
Note on Stride Order
~~~~~~~~~~~~~~~~~~~~
Note that cute's convention is to write the stride order for dimensions from left to right,
where a lower order number means higher priority.
For shape ``(2, 3, 4)`` and stride order ``(0, 1, 2)``, the stride is ``(1, 2, 6)``. This is typically also known
as column major order. If you want to create a fake tensor with compact row major order,
you should explicitly pass in ``stride_order=reversed(range(len(shape)))``
to ``make_fake_compact_tensor``. Alternatively, you can always precisely control the
stride via the ``stride`` argument in the ``make_fake_tensor`` API.
Error handling
--------------