Scott Todd
Scott Todd
(I'm not sure how actionable this issue on its own will be, like with https://github.com/iree-org/iree/issues/11994, but I'm at least hoping to start some conversations and generate interesting reproducers / traces...
> Is there a way to add dispatch count to sdxl CI . @ScottTodd of things, that is most important for now. Yeah. We can adapt what's in https://github.com/iree-org/iree/blob/main/build_tools/benchmarks/collect_compilation_statistics.py, or...
https://github.com/iree-org/iree/issues/16250 maybe use this?
Still seeing these, and it's making updating XFAIL lists between local builds and CI builds slightly tricky. Should * Fix the crash * Enable asserts on pkgci
Tracy has been updated in IREE. Let's update the capstone version upstream and then update again?
Fixed by https://github.com/iree-org/iree/pull/19842
Do we have a workaround for this or any patches we could try? I'm also seeing unusably slow behavior after running `LLVMCPUVectorTransferLowering` on `open_llama_3b_v2_f16_gguf` from https://github.com/nod-ai/sharktank. Logs and IR here:...
> Perhaps you can try [llvm/torch-mlir#3277](https://github.com/llvm/torch-mlir/pull/3277) . It should fix the embedding lookup issue at torch level. That gets further, yeah :D. Might be enough to call this particular issue...
> @ScottTodd can you provide the mlir file? @pashu123 please help triage and provide possible solutions This is the input file I'm working with: https://sharkpublic.blob.core.windows.net/sharkpublic/scotttodd/issue_reports/open_llama_3b_v2_f16.mlir