Commit 524991fe authored by Yu Cheng's avatar Yu Cheng Committed by GitHub
Browse files

[Bugfix] Bugfix of pass order for hopper (#117)

parent 62843b88
......@@ -203,14 +203,14 @@ def lower(
mod = tl.transform.ThreadPartialSync("shared.dyn")(mod)
mod = tir.transform.InferFragment()(mod)
mod = tir.transform.LowerThreadAllreduce()(mod)
mod = tl.transform.LowerHopperIntrin()(mod)
mod = tl.transform.ThreadSync("shared")(mod)
mod = tl.transform.ThreadSync("shared.dyn")(mod)
mod = tir.transform.InjectPTXAsyncCopy()(mod)
mod = tl.transform.AnnotateDeviceRegions()(mod)
mod = tir.transform.SplitHostDevice()(mod)
mod = tir.transform.MergeSharedMemoryAllocations()(mod)
mod = tl.transform.ThreadSync("shared")(mod)
mod = tl.transform.ThreadSync("shared.dyn")(mod)
mod = tl.transform.LowerHopperIntrin()(mod)
mod = tir.transform.InjectPTXAsyncCopy()(mod)
mod = tl.transform.MakePackedAPI()(mod)
mod = tir.transform.LowerDeviceKernelLaunch()(mod)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment