Han-Chung Wang

Thanks for the note, it is helpful! > We can't have the IR be in an inconsistent state where the host and device don't agree on the encodings. I see,...

> so your update code instead of isa TensorSizeOfOp would be hasTrait TensorPhaseOp then walk the op change any type attrs SG! I'm using tensor.sizeof op for prototype now; will...

I made some progress and got stuck in specialization. The issues are mostly about how we gather affinities, clone dispatches and update encodings, especially for multi-device concept. I was going...

I have a [prototype](https://github.com/iree-org/iree/pull/18738/commits/b0944b965aeeb3059c20098a97fdfbca3f78bfab) that addresses the dup config issue. One of the challenges is that the attribute is not mutable, so we can not update some field once we...

I wrote an example about running one matmul on device_a and the same matmul on the device_b; it gives me the [multi-device IR](https://gist.github.com/hanhanW/e84dcfb59506cdd698fb3845147a58f5) that we want to solve in SpecializeEncoding...

Note: here is the example input that I used in the prototype. What the IR does is 1. Compute matmul result on device_a 2. Transfer lhs/rhs tensors to device_b 3....

I have a [second take](https://github.com/iree-org/iree/pull/18738/commits/e992c932a45a9812c7c1af5908b1c736704ee139) for dup config issue w/o HAL attribute changes. It is still creating an additional level of wrapping but it is scoped within the Codegen directory....

Although I haven't finished the update of cloned executable part, but it looks like I'm doing something wrong. So I posted the update here and I'm looking for feedback. So...

Adding the status update to the issue: https://github.com/iree-org/iree/pull/18738 this PR has the prototype, and you can find the design doc at https://hackmd.io/@hwPnnvLBTB-JGVMeh-bCEA/Sy9nvDhb1e We had a great brainstorm yesterday, and here...

I have a [prototype](https://github.com/iree-org/iree/pull/18738/commits/6a8f2cc95b65fb155275c68d7ded5aba8c98adcf) for encoding information compression. It still carry the whole config (as an intermediate step) when we populate the attributes from HALAffinityAnalysisDialectInterface implementation. The main difference is...