d1: sweep hd=64,128,256

tests/unit/test_d1_sweep.py

@@ -0,0 +1,37 @@
+"""D1: Quick test at hd=128 to narrow down the breakage."""
+import torch, math
+import cutlass.cute as cute
+import cutlass.torch as ct
+import cuda.bindings.driver as cuda
+from dsv4.kernels.attention.fmha import FmhaKernel
+
+for hd in [64, 128, 256]:
+    torch.manual_seed(42)
+    n = 128; m = 128
+    q = torch.randn(m, hd, 1, dtype=torch.bfloat16, device='cuda')
+    k = torch.randn(n, hd, 1, dtype=torch.bfloat16, device='cuda')
+    v = torch.randn(n, hd, dtype=torch.bfloat16, device='cuda')
+    c = torch.zeros(m, hd, 1, dtype=torch.bfloat16, device='cuda')
+
+    qf = q[:,:,0].float(); kf = k[:,:,0].float()
+    scale = 1.0 / math.sqrt(hd)
+    attn = qf @ kf.T * scale; attn = torch.softmax(attn, dim=-1)
+    ref = attn @ v.float()
+
+    # For hd>256, we'd need N-tiling, but 128 is fine as single tile
+    v_kernel = v.unsqueeze(-1)
+    mQ = ct.from_dlpack(q).mark_layout_dynamic(leading_dim=ct.get_leading_dim(q))
+    mK = ct.from_dlpack(k).mark_layout_dynamic(leading_dim=ct.get_leading_dim(k))
+    mV = ct.from_dlpack(v_kernel).mark_layout_dynamic(leading_dim=ct.get_leading_dim(v_kernel))
+    mC = ct.from_dlpack(c).mark_layout_dynamic(leading_dim=ct.get_leading_dim(c))
+    stream = cuda.CUstream(torch.cuda.current_stream().cuda_stream)
+
+    kernel = FmhaKernel(head_dim=hd, s_k=n)
+    print(f'hd={hd}: Compiling...', flush=True)
+    compiled = cute.compile(kernel, mQ, mK, mV, mC, stream)
+    compiled(mQ, mK, mV, mC, stream)
+    torch.cuda.synchronize()
+
+    out = c[:,:,0].float()
+    cos = torch.nn.functional.cosine_similarity(out.flatten().unsqueeze(0), ref.flatten().unsqueeze(0)).item()
+    print(f'hd={hd}: cos {cos:.6f}  {"PASS" if cos >= 0.97 else "FAIL"}')