This PR;
- Fixes syntax errors like `.type values: dict[str,
Callable[[list[Any]], Any]]` to `:type values: dict[str,
Callable[[list[Any]], Any]]`,
- Fixes typos,
- Fixes formatting like `k ++` to ` k++`,
- Increases consistency (e.g. by transforming the minority `cd dir/` to
the majority `cd dir`).
This addition allows explanatory messages upon assertion failures:
```python
@triton.jit
def my_single_block_kernel(
matrix_extent: tl.constexpr,
block_size: tl.constexpr, # must be >= extent (single block)
matrix: Tensor,
...
):
tl.static_assert(matrix_extent <= block_size,
f"`matrix_extent` should not be more than the block size ({block_size}), but is {matrix_extent}")
```
Yielding, when called incorrectly:
```
AssertionError: `matrix_extent` should not be more than the block size (32), but is 57
```
This PR applies a minor patch that removes unnecessary masks in
`_dsd_kernel()`.
### Details
`offs_bn` is defined as follows and not updated after that.
```py
offs_bn = pid_m * TILE_N + tl.arange(0, TILE_N)
offs_bn = tl.max_contiguous(tl.multiple_of(offs_bn % DS0, TILE_N), TILE_N)
```
Because `offs_bn = offs_bn % DS0`, this mask is always `True`.
```py
b = tl.load(pb, mask=offs_bn[None, :] < DS0)
```
This PR removes this mask (as well as explicit `mask=True`).
Fixed `JITFunction.__init__` to mark args as constexpr only when the
annotation is actually `tl.constexpr`, rather than treating any
annotated arg as constexpr.
When the user set the `LLVM_SYSPATH` to use custom build llvm, it will
throw the error because there is no version.txt under the custom build
one.
This PR skips the version check If the `LLVM_SYSPATH` is set.
---------
Co-authored-by: Philippe Tillet <phil@openai.com>
This is cherry-picked from #1305
If you call a `JITFunction` twice in the same kernel, first with `int32`
then with `uint32`, the second call will treat the unsigned value as
signed. This passes through MLIR without error because MLIR uses the
same types for both, but different operation calls will be generated so
you may silently get the wrong result.
- Significant simplification of the optimizer pipeline. Right mma
version is now set directly after the coalescing pass. DotOperand layout
no longer hold a state to `isRow` argument, and instead query it from
their parent
- Moved a bunch of things from TritonGPUToLLVM/DotOpHelpers to
TritonGPUAttrDefs. All MMAv1 state is now queried from attributes.
- logic for getELemsPerThread is no longer duplicated in TypeConverter
`_triton.runtime.num_sm`, `_triton.runtime.clock_rate`,
`_triton.runtime.cc` seem no longer exist.
use the corresponding methods from `get_max_tensorcore_tflops` in the
same file.
* Cleaned up pipeline pass. Now works when there are element-wise ops
between the load and the dot
* Made `splat` compatible with varibales that have DotOperandLayout
* Moves rematerialization utils to separate Transforms/Utility.cpp file.
* Frontend:
- `int` kernel arguments are always signed
- Loop induction variable is now determine by integer promotion on
lb/ub/step
* Optimizer:
- Added new ExtractSliceOp that enforces 32-bit offsets
* Backend:
- Use 64-bit indices when lowering functions and control flow
- Removed `idx_val` macro and replaced it with `i32_val`
- Cleaned up comments
- Added new ArithToIndex pass to make sure operations on indices are
done with the `index` dialect, that gets converted to LLVM separately
using a 64-bit target
Per issue https://github.com/openai/triton/issues/1228. I believe we are
potentially exposed when a Triton executor (Pytorch for example) links
in two or more `triton_.so` shared objects and each has a stub for
`_launch`.
This fix ensures the `_launch` function is tied locally to the calling
`__triton_launcher` and can't be misused by another library.
Python 3.10 changes where packages are installed by default, causing
problems with Ubuntu into `/local`. See
[this](https://lists.debian.org/debian-python/2022/03/msg00039.html) and
[this](https://bugs.launchpad.net/ubuntu/+source/python3.10/+bug/1967920).
Triton seems to break when using 3.10 as it looks for the headers, but
the headers are not in `/local`, e.g. they are at
`/usr/include/python3.X` and not `/usr/local/include/python3.X`
Not 100% sure what's going on here since it's deep in python / pip, but
I think this should fix it. Otherwise, you have to hack around it in
dockerfiles, e.g. `ENV DEB_PYTHON_INSTALL_LAYOUT=deb`, which breaks
things with the release of pip that went.
---------
Co-authored-by: Keren Zhou <kerenzhou@openai.com>
Fix issue https://github.com/openai/triton/issues/244
Check `end` is greater than `start`.
Check if the range can fit in `int32`.
Check the number of elements less than or equal to
`TRITON_MAX_TENSOR_NUMEL = 131072`.
---------
Co-authored-by: Philippe Tillet <phil@openai.com>
