"examples/sampling/vscode:/vscode.git/clone" did not exist on "bd74c44c39e037e645dfd3e7e580d4395b211a14"
API_Reference_Guide.rst 1.14 KB
Newer Older
pmaybank's avatar
pmaybank committed
1

2
*******************
pmaybank's avatar
pmaybank committed
3
API Reference Guide
4
*******************
pmaybank's avatar
pmaybank committed
5

6
=================
pmaybank's avatar
pmaybank committed
7
Introduction
8
=================
pmaybank's avatar
pmaybank committed
9
10
11
12
13
14
15
16
17
18

This document contains details of the APIs for the Composable Kernel (CK) library and introduces some of the key design
principles that are used to write new classes that extend CK functionality.

=================
Using CK API
=================

This section describes how to use the CK library API.

19
=================
pmaybank's avatar
pmaybank committed
20
CK Datatypes
21
22
23
24
=================

-----------------
DeviceMem
pmaybank's avatar
pmaybank committed
25
26
-----------------

27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
.. doxygenstruct:: DeviceMem

---------------------------
Kernels For Flashattention
---------------------------

The Flashattention algorithm is defined in :cite:t:`dao2022flashattention`.  This sections lists the classes that are
used in the CK GPU implementation of Flashattention.

**Gridwise classes**

.. doxygenstruct:: ck::GridwiseBatchedGemmSoftmaxGemm_Xdl_CShuffle

**Blockwise classes**

.. doxygenstruct:: ck::ThreadGroupTensorSliceTransfer_v4r1

.. doxygenstruct:: ck::BlockwiseGemmXdlops_v2

.. doxygenstruct:: ck::BlockwiseSoftmax

**Threadwise classes**

.. doxygenstruct:: ck::ThreadwiseTensorSliceTransfer_StaticToStatic

Sam Wu's avatar
Sam Wu committed
52
.. bibliography::