spectrometerHBH opened a new pull request, #19785: URL: https://github.com/apache/tvm/pull/19785
Forward-port of a batch of tirx framework changes onto current `main`. Squashed into a single commit on top of latest upstream. Included changes: - FA4 framework support — env-driven ptxas reg-level, scheduler num_ctas - clusterlaunchcontrol device intrinsics + CLC tile scheduler - framework support for nvfp4 tcgen05 GEMM - scope-level operands for warp/wg/cta register elementwise - diagnose duplicate PrimFunc global symbols (LLVM codegen) - default CUDA compilation to NVRTC - tirx test/bench tooling updates Tests under `tests/python/tirx` pass locally on sm_100a (B200). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
