-
### Describe the bug
Performing two back-to-back bmm calls on NVIDIA GPU triggers an internal assert. On triton 3.1.0:
```
python3: /project/lib/Conversion/TritonGPUToLLVM/ConvertLayoutOpToLLVM.cpp…
-
Platforms: linux
This test was disabled because it is failing in CI. See [recent examples](https://hud.pytorch.org/flakytest?name=test_comprehensive_bmm_cpu_float64&suite=TestInductorOpInfoCPU&limit=…
-
### What happened?
With `--iree-llvmcpu-enable-ukernels=all` or `--iree-llvmcpu-enable-ukernels=unpack`, the IREE will report the following message for tinyllama model in `LLVMCPUCheckIRBeforeLLVMCon…
-
- [x] Export playback errors from npaw to metabase
- [x] Create Dashboard(s)
- [x] Make sure they are finadable and available to the team
A dashboard that gives insight on general issues
e.g. referen…
-
Finish was uploaded yesterday eve
But it was not fully processed this morning, so, it looks like that th…
-
would it be possible to register operations such as na2d using `torch.library.custom_op`,
or otherwise ensure that they participate in operation dispatch?
torch's built in flop counter, FlopCoun…
-
Hi Mattjis / Pieter,
I have been taking a detailed look at the Archie code for handling specific RMs. We want to add an RM, and I'm seeing some code (not much) in places like `archie.utils` that appe…
-
Additional tasks (added by Ven on May 22):
* [x] Update last figure in readme to reflect addition of priors
* [x] Update NEWS
* [x] fix load of default_prior etc
* [x] Review developer's notes
…
-
### The feature, motivation and pitch
### Description
Please update the implementation of the `get_tflops` function in `benchmark/test_blas_perf.py` under the `benchmark` directory. The function shou…
-
[~/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/utils/math.py](https://github.com/isaac-sim/IsaacLab/blob/main/source/extensions/omni.isaac.lab/omni/isaac/lab/utils/math.py)
there are two…