-
### What happened?
The program
```mlir
func.func @trace_args(%arg0: tensor) {
flow.tensor.trace "debug_sink_test" = [
%arg0: tensor
]
return
}
```
Fails to compile with assert fa…
-
This issue is a method of tracking/deconflicting efforts to change the Python API. If anyone has any PRs to change the Python API please let me know and we can record them here! We can also use this i…
-
I installed torch-mlir with following script in my own conda env, where 20240825.194 is the newest torch-mlir in the url [https://github.com/llvm/torch-mlir-release/releases/expanded_assets/dev-wheels…
-
### Description
Got `Page size must be divisible by sizeof(uint32_t) because buffers hold uint32_t values` error while running the compiled framework model in Bfloat16
### Reproduce
```bash
git chec…
-
During the testing of the expm1 op, I noticed major discrepancies between the results of the exp and expm1 ops ran on CPU and Tenstorrent silicon. This was the cause for widening the atol tolerance in…
-
```patch
diff --git a/mlir/lib/Driver/CompilerDriver.cpp b/mlir/lib/Driver/CompilerDriver.cpp
index d75406bf6..2114a7a78 100644
--- a/mlir/lib/Driver/CompilerDriver.cpp
+++ b/mlir/lib/Driver/Compi…
-
I am trying to create a python object fo mlir ModuleOp using the code below.
```
static py::object getMlirIrClass(const char* className) {
return py::module::import(MAKE_MLIR_PYTHON_QUALNAME("i…
-
Similar issue from https://github.com/llvm/clangir/issues/1065, we cannot use globals at all (even if we ignore the absence of `alias`) to start representing aliases. `llvm.mlir.global`s don't seem to…
-
Reproducer:
```f90
real :: r1
complex :: c1
!$omp atomic write
c1 = r1
end
```
Stack dump:
```console
#0 0x00007f6aacc5f727 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int)
#1 0x000…
-
For [this ](https://gist.github.com/nirvedhmeshram/344f11443b96fb9ff022fa283cc6cd8a) matmul like + elementwise IR, we go down the LLVMGPUSIMT pipeline, see dump [here](https://gist.github.com/nirvedhm…