[TorchToLinalg]Lower torch.gcd to linalg and scf

[bratislavSyrmia]

Add verify() method to check if tensors are of
integer type. Also check if tensors are of same shape, or if the second tensor is a single element tensor.

Add e2e tests. Put them into onnx and stablehlo
xfailed sets.

[bratislavSyrmia]

force push: added math::cttz instead of counting trailing zeros manually

[vivekkhandelwal1]

Hi @bratislavSyrmia, I would like you to explore a non-loop path for this lowering, since these kind of lowerings usually causes issues in the downstream pipeline especially the code-generation part.

[IanWood1]

Hi @bratislavSyrmia, I would like you to explore a non-loop path for this lowering, since these kind of lowerings usually causes issues in the downstream pipeline especially the code-generation part.

@vivekkhandelwal1 out of curiosity, do you have an algorithm/solution in mind? The best I could think of was to use linalg.generic's library_call attr and define a GCD function but maybe that runs into the same problem.

[vivekkhandelwal1]

Hi @bratislavSyrmia, I would like you to explore a non-loop path for this lowering, since these kind of lowerings usually causes issues in the downstream pipeline especially the code-generation part.

@vivekkhandelwal1 out of curiosity, do you have an algorithm/solution in mind? The best I could think of was to use linalg.generic's library_call attr and define a GCD function but maybe that runs into the same problem.

Hi @IanWood1, I did not spend time on thinking about it that's why I asked @bratislavSyrmia to explore the possibility of any such solution.

But if there exists a solution based on linalg.generic then it would still be a better approach then the current one.

[bratislavSyrmia]

Hi @bratislavSyrmia, I would like you to explore a non-loop path for this lowering, since these kind of lowerings usually causes issues in the downstream pipeline especially the code-generation part.

@vivekkhandelwal1 out of curiosity, do you have an algorithm/solution in mind? The best I could think of was to use linalg.generic's library_call attr and define a GCD function but maybe that runs into the same problem.

Hi @IanWood1, I did not spend time on thinking about it that's why I asked @bratislavSyrmia to explore the possibility of any such solution.

But if there exists a solution based on linalg.generic then it would still be a better approach then the current one.

I have thought about it but I have no idea how I would find the greatest common divisor between two numbers without using loops