Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[doc] Add simt functions in operators page #5333

Merged
merged 1 commit into from
Jul 11, 2022
Merged

[doc] Add simt functions in operators page #5333

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
49 changes: 49 additions & 0 deletions docs/lang/articles/basic/operator.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -323,3 +323,52 @@ a@a # @ denotes matrix multiplication
For now, determinant() and inverse() only works in Taichi-scope, and the
size of the matrix must be 1x1, 2x2, 3x3 or 4x4.
:::


## Supported SIMT intrinsics
For CUDA backend, Taichi now supports warp-level and block-level intrinsics that
are needed for writing high-performance SIMT kernels. You can use them in Taichi
similar to the [usage in CUDA kernels](https://developer.nvidia.com/blog/using-cuda-warp-level-primitives/). Currently, the following functions are supported:


| Operation | Mapped CUDA intrinsic |
| -------------------------- | ----------------------------- |
|`ti.simt.warp.all_nonzero` | `__all_sync` |
|`ti.simt.warp.any_nonzero` | `__any_sync` |
|`ti.simt.warp.unique` | `__uni_sync` |
|`ti.simt.warp.ballot` | `__ballot_sync` |
|`ti.simt.warp.shfl_sync_i32`| `__shfl_sync` |
|`ti.simt.warp.shfl_sync_f32`| `__shfl_sync` |
|`ti.simt.warp.shfl_up_i32` | `__shfl_up_sync` |
|`ti.simt.warp.shfl_up_f32` | `__shfl_up_sync` |
|`ti.simt.warp.shfl_down_i32`| `__shfl_down_sync`|
|`ti.simt.warp.shfl_down_f32`| `__shfl_down_sync`|
|`ti.simt.warp.shfl_xor_i32` | `__shfl_xor_sync` |
|`ti.simt.warp.match_any` | `__match_any_sync`|
|`ti.simt.warp.match_all` | `__match_all_sync`|
|`ti.simt.warp.active_mask` | `__activemask` |
|`ti.simt.warp.sync` | `__syncwarp` |

Please refer to our [API docs](https://docs.taichi.graphics/api/taichi/lang/simt/warp/#module-taichi.lang.simt.warp)
jim19930609 marked this conversation as resolved.
Show resolved Hide resolved
for more information on each function.

Here is an example to perform data exchange within a warp in Taichi:


```python
a = ti.field(dtype=ti.i32, shape=32)

@ti.kernel
def foo():
ti.loop_config(block_dim=32)
for i in range(32):
a[i] = ti.simt.warp.shfl_up_i32(ti.u32(0xFFFFFFFF), a[i], 1)

for i in range(32):
a[i] = i * i

foo()

for i in range(1, 32):
assert a[i] == (i - 1) * (i - 1)
```