ytyt

test/test_pallas_spmd.py

@@ -10,6 +10,7 @@
 import torch_xla.distributed.spmd as xs
 from torch_xla import runtime as xr
 from torch_xla._internal import tpu
+import torch.nn.functional as F
 
 if xr.device_type() == 'TPU':
   from torch_xla.experimental.custom_kernel import flash_attention
@@ -122,8 +123,10 @@ def test_flash_attention_spmd_data_parallel_with_segment_ids(self):
         torch_xla._XLAC._get_xla_sharding_spec(o),
         f"{{devices=[{n_devices},1,1,1]0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15}}")
 
-    attention_mask = kv_segment_ids.repeat_interleave(32, dim=0)
-    attention_mask = attention_mask.view(4, 32, -1, 128)
+    attention_mask = F.pad(kv_segment_ids, (0, 16256), value=0.0)
+    attention_mask = attention_mask.repeat_interleave(2, dim=0)
+    attention_mask = attention_mask.view(4, 2, 128, 128)
+    # attention_mask = torch.ones(4, 2, 128, 128).to("xla")
     # head_size = self.heads
     # current_length: int = attention_mask.shape[-1]
     # if current_length != target_length:
@@ -137,6 +140,17 @@ def test_flash_attention_spmd_data_parallel_with_segment_ids(self):
     # )
 
     expected_o = self._attention(q, k, v, attn_mask=attention_mask)
+    diff = (expected_o - o).abs()
+    # z = torch.allclose(o.cpu(), expected_o.cpu(), atol=1e-05)
+
+    true_count = torch.sum(diff < 0.00001).item()
+    false_count = diff.numel() - true_count
+
+    print(f"Number of True: {true_count}")
+    print(f"Number of False: {false_count}")
+
+    print(diff.max().cpu())
+    print(diff.min().cpu())
     self.assertTrue(torch.allclose(o.cpu(), expected_o.cpu(), atol=1e-05))
     jax.config.update('jax_default_matmul_precision', "default")