-
### Overview ###
This issue is a tracker for the next steps of enabling the data tiling fusion improvements, which are currently behind the `--iree-dispatch-creation-experimental-data-tiling` flag. O…
-
This issue happens with asan enabled:
```
wget https://storage.googleapis.com/iree-model-artifacts/ssd_mobilenet_v2_dynamic_1.0_float.tflite
iree-import-tflite ssd_mobilenet_v2_dynamic_1.0_float.…
-
After https://github.com/openxla/iree/pull/14938, reduce_window tests are constantly failing. Filing this issue to track. Note that there are known underminstic failures on macOS when running tests ht…
-
1. Following is the Conv2d pytorch module.
```
import torch
import torch.nn as nn
class op_conv2d(nn.Module):
def __init__(self):
super().__init__()
self.layers = nn.Seque…
-
## How I think about buffer allocation in data-tiling
1. The default path is now materializing encodings at very early stage (i.e., GlobalOpt), while we want to build the late materialization path.…
-
`input.mlir`
```mlir
module {
func.func @pipeline_dispatch_0_depthwise_conv_2d_nhwc_hwc_1x1082x1920x1x1x3(%extracted_slice_1: tensor, %extracted_slice_2: tensor ) -> tensor{
%arg7 = tensor.empty…
-
### What happened?
iree-compile: /_path_to_/iree/third_party/llvm-project/mlir/include/mlir/IR/UseDefLists.h:198: mlir::IRObjectWithUseList::~IRObjectWithUseList() [OperandType = mlir::OpOperand]: As…
-
I'm trying to get more programs running through WebGPU on the [`webgpu` branch](https://github.com/openxla/iree/tree/webgpu/) with the application code at [`experimental/web/sample_webgpu`](https://gi…
-
Hi,
I tried to bring my own llvm using [iree/build_tools/llvm/byo_llvm.sh](https://github.com/openxla/iree/blob/main/build_tools/llvm/byo_llvm.sh), and completed build (install) without any errors.…
-
Currently with a max padding size of 16 for allocations, we are discarding all tiles greater than 16 in any dimension, so these tiles are simply unused, and any ukernels designed for those sizes are a…