(tvm) branch main updated: [TIRx] Use canonical PTX async script API in s_tir test (#19739)

[tlopex]

This is an automated email from the ASF dual-hosted git repository.

tlopex pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git


The following commit(s) were added to refs/heads/main by this push:
     new 768d38bf07 [TIRx] Use canonical PTX async script API in s_tir test 
(#19739)
768d38bf07 is described below

commit 768d38bf07c0ff7623304e916ed5727a8b122c05
Author: Shushi Hong <820958...@qq.com>
AuthorDate: Fri Jun 12 00:38:01 2026 -0400

    [TIRx] Use canonical PTX async script API in s_tir test (#19739)
    
    Update `test_multiplication_nodes_are_inlined` to use the current TIRX
    PTX async script namespace:
    
    - `T.ptx.cp_async.commit_group()`
    - `T.ptx.cp_async.wait_group(0)`
    
    The test still used the older top-level names `T.ptx_commit_group()` and
    `T.ptx_wait_group(0)`, which are not exposed by the current
    `tvm.tirx.script` namespace. This caused parsing to fail before
    `InjectPTXAsyncCopy` could be tested.
    
    This keeps the test aligned with the rest of the TIRX PTX async tests
    and with the TVMScript printer output, without adding extra legacy
    aliases to the public script namespace.
---
 .../s_tir/transform/test_s_tir_transform_inject_ptx_async_copy.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git 
a/tests/python/s_tir/transform/test_s_tir_transform_inject_ptx_async_copy.py 
b/tests/python/s_tir/transform/test_s_tir_transform_inject_ptx_async_copy.py
index 9875114f66..e54b50e41a 100644
--- a/tests/python/s_tir/transform/test_s_tir_transform_inject_ptx_async_copy.py
+++ b/tests/python/s_tir/transform/test_s_tir_transform_inject_ptx_async_copy.py
@@ -926,8 +926,8 @@ def test_multiplication_nodes_are_inlined():
                 A_shared[T.Ramp(tx * T.int64(128) + cse_v1 * T.int64(8), 
T.int64(1), 8)] = (
                     A_flattened[T.Ramp(tx * T.int64(128) + cse_v1 * 
T.int64(8), T.int64(1), 8)]
                 )
-            T.ptx_commit_group()
-            T.ptx_wait_group(0)
+            T.ptx.cp_async.commit_group()
+            T.ptx.cp_async.wait_group(0)
 
     @I.ir_module(s_tir=True)
     class Expected:
@@ -946,8 +946,8 @@ def test_multiplication_nodes_are_inlined():
                     tx * T.int64(128) + cse_v1 * T.int64(8),
                     16,
                 )
-            T.ptx_commit_group()
-            T.ptx_wait_group(0)
+            T.ptx.cp_async.commit_group()
+            T.ptx.cp_async.wait_group(0)
 
     After = tvm.s_tir.transform.InjectPTXAsyncCopy()(Before)
     tvm.ir.assert_structural_equal(After, Expected)