Update on "[reland][inductor] make thread order consistent with loop …

…order"


This PR relands #106827 which get reverted because of causing compilation error for some ads model.

Yanbo provide a repro in one of the 14k model ( `pytest ./generated/test_KaiyangZhou_deep_person_reid.py -k test_044`). This is also the model I used to confirm the fix and come up with a unit test. In this model, we call `tritoin_heuristics.triton_config` with size_hints [2048, 2]. Previously this would result in a trition config with XBLOCK=2048 and YBLOCK=2 . But since we change the mapping between size_hints and XYZ dimension, we now generate a triton config with XBLOCK=2 and YBLOCK=2048.  This fails compilation since we set max YBLOCK to be 1024.

My fix is to make sure we never generate a triton config that exceeds the maximum block size.


[ghstack-poisoned]

test/inductor/test_triton_heuristics.py

@@ -4,15 +4,16 @@
 import unittest
 
 try:
-    import triton
+    import triton  # noqa: F401
 except ImportError:
     if __name__ == "__main__":
         sys.exit(0)
     raise unittest.SkipTest("requires triton")
 
 from torch._dynamo.test_case import run_tests, TestCase
+from torch._inductor import config
 from torch._inductor.triton_heuristics import triton_config
-from  torch._inductor import config
+
 
 class TestTritonHeuristics(TestCase):
     def test_triton_config(self):
@@ -26,5 +27,6 @@ def test_triton_config(self):
                 continue
             self.assertTrue(cfg.kwargs[key] <= config.triton.max_block[label])
 
+
 if __name__ == "__main__":
     run_tests()