tilelang.engine.phase

Functions

allow_warp_specialized([pass_ctx, target])
allow_tma_and_warp_specialized([pass_ctx, target])
allow_fence_proxy([target])
allow_vectorize([pass_ctx])
allow_global_thread_synchronization([pass_ctx])
should_enable_aggressive_merge([pass_ctx, target])
LowerAndLegalize(mod, target)
OptimizeForTarget(mod, target)

Module Contents

tilelang.engine.phase.allow_warp_specialized(pass_ctx=None, target=None)

Parameters:

• pass_ctx (Optional[tilelang.transform.PassContext])

• target (Optional[tvm.target.Target])

Return type:

bool

tilelang.engine.phase.allow_tma_and_warp_specialized(pass_ctx=None, target=None)

Parameters:

• pass_ctx (Optional[tilelang.transform.PassContext])

• target (Optional[tvm.target.Target])

Return type:

bool

tilelang.engine.phase.allow_fence_proxy(target=None)

Parameters:

target (Optional[tvm.target.Target])

Return type:

bool

tilelang.engine.phase.allow_vectorize(pass_ctx=None)

Parameters:

pass_ctx (Optional[tilelang.transform.PassContext])

Return type:

bool

tilelang.engine.phase.allow_global_thread_synchronization(pass_ctx=None)

Parameters:

pass_ctx (Optional[tilelang.transform.PassContext])

Return type:

bool

tilelang.engine.phase.should_enable_aggressive_merge(pass_ctx=None, target=None)

Parameters:

• pass_ctx (Optional[tilelang.transform.PassContext])

• target (Optional[tvm.target.Target])

Return type:

bool

tilelang.engine.phase.LowerAndLegalize(mod, target)

Parameters:

• mod (tvm.IRModule)

• target (tvm.target.Target)

Return type:

tvm.IRModule

tilelang.engine.phase.OptimizeForTarget(mod, target)

Parameters:

• mod (tvm.IRModule)

• target (tvm.target.Target)

Return type:

tvm.IRModule