-
- ttir.all_gather
- ttir.line_all_gather
- ttir.reduce_scatter
Related issue:
- #700
Follow this tutorial for adding a new op to the stack: https://docs.tenstorrent.com/tt-mlir/adding-an-op.html
-
### What happened?
For the given IR
```
module {
func.func @torch_jit( %arg1: !torch.vtensor, %arg3: !torch.vtensor) -> !torch.vtensor attributes {torch.onnx_meta.ir_version = 7 : si64, torch.onn…
-
## Problem
Compiling llama 405b without data tiling explicitly disabled causes a bunch of `tensor.extract_slice` dispatches. Here is the difference in dispatch count:
| Data-Tiling Opt| Dispatch Cou…
-
At some point mlir switched over from using attributes to properties, but still calls everything an Attribute in tablegen. As a result, there are a lot of places where we are using `attr_def` where we…
-
### What happened?
I compiled GPT and tried to run it using `iree-run-module`. It errored with the following message:
```
iree/runtime/src/iree/hal/drivers/cuda/memory_pools.c:236: RESOURCE_EXHAUSTE…
-
I know quantized model isn't supported yet. I would like to confirm this is the symptom due to Dequantize/Quantize operators?
The model I'm testing is ShuffleNet-v2-int8 from [here](https://github.…
-
After https://github.com/llvm/llvm-project/pull/93519, when building ArmSMEStubs.cpp on mingw with `LLVM_BUILD_LLVM_DYLIB=ON`, `mlir_arm_sme_abi_stubs_EXPORTS` is no longer defined as expected from th…
-
Reproduction instructions:
Save this reproducer to `repro.mlir`:
```mlir
module {
func.func @intfail() -> !torch.int {
%0 = torch.vtensor.literal(dense : tensor) : !torch.vtensor
…
-
# 编译器与中间表示: LLVM IR, SPIR-V, 以及 MLIR | Lei.Chat()
总体介绍编译器和中间表示 (LLVM IR, SPIR-V, and MLIR) 的发展历史和演进趋势
[https://www.lei.chat/zh/posts/compilers-and-irs-llvm-ir-spirv-and-mlir/](https://www.lei.chat/z…
-
@srush
Always appreciate your wonderful OSS educational contributions!
I'm relatively familiar with `CUDA` and `triton` but less so with machine learning compilers and am interested in getting …