optimize MiniMax-Text-01 lightning_attn_decode triton (#2966)

benchmark/kernels/minmax-text-01-lighting_attention/benchmark_lighting_attention_decode.py

@@ -23,7 +23,10 @@ def _decode_kernel(
     h: tl.constexpr,
     n: tl.constexpr,
     d: tl.constexpr,
+    d_original: tl.constexpr,
     e: tl.constexpr,
+    e_original: tl.constexpr,
+    BLOCK_SIZE: tl.constexpr = 32,
 ):
     off_bh = tl.program_id(0)
     off_h = off_bh % h
@@ -39,21 +42,38 @@ def _decode_kernel(
     d_idx = tl.arange(0, d)
     e_idx = tl.arange(0, e)
 
-    q = tl.load(Q + qk_offset + d_idx)
+    # Create masks for original dimensions
-    k = tl.load(K + qk_offset + d_idx)
+    d_mask = d_idx < d_original
-    v = tl.load(V + v_offset + e_idx)
+    e_mask = e_idx < e_original
 
-    kv = tl.load(KV + kv_offset + d_idx[:, None] * e + e_idx[None, :])
+    # Load with masking
+    q = tl.load(Q + qk_offset + d_idx, mask=d_mask, other=0.0)
+    k = tl.load(K + qk_offset + d_idx, mask=d_mask, other=0.0)
+    v = tl.load(V + v_offset + e_idx, mask=e_mask, other=0.0)
 
+    # Load KV with 2D masking
+    kv = tl.load(
+        KV + kv_offset + d_idx[:, None] * e + e_idx[None, :],
+        mask=(d_mask[:, None] & e_mask[None, :]),
+        other=0.0,
+    )
+
+    # Compute outer product using element-wise operations
     k_v_prod = k[:, None] * v[None, :]
     kv = ratio * kv + k_v_prod
 
+    # Store KV with 2D masking
     tl.store(
-        KV + kv_offset + d_idx[:, None] * e + e_idx[None, :], kv.to(KV.dtype.element_ty)
+        KV + kv_offset + d_idx[:, None] * e + e_idx[None, :],
+        kv.to(KV.dtype.element_ty),
+        mask=(d_mask[:, None] & e_mask[None, :]),
     )
 
+    # Compute matrix-vector multiplication using element-wise operations and reduction
     o = tl.sum(q[:, None] * kv, axis=0)
-    tl.store(Out + o_offset + e_idx, o.to(Out.dtype.element_ty))
+
+    # Store output with masking
+    tl.store(Out + o_offset + e_idx, o.to(Out.dtype.element_ty), mask=e_mask)
 
 
 def lightning_attn_decode(q, k, v, kv, s):
@@ -62,26 +82,27 @@ def lightning_attn_decode(q, k, v, kv, s):
     e = v.shape[-1]
     assert n == 1, "Sequence length must be 1 in decode mode"
 
-    # Pad dimensions to power of 2
+    # Get padded dimensions (power of 2)
     d_padded = next_power_of_2(d)
     e_padded = next_power_of_2(e)
 
-    # Pad inputs
-    q_padded = F.pad(q, (0, d_padded - d))
-    k_padded = F.pad(k, (0, d_padded - d))
-    v_padded = F.pad(v, (0, e_padded - e))
-    kv_padded = F.pad(kv, (0, e_padded - e, 0, d_padded - d))
-
-    # Ensure inputs are contiguous
-    q_padded = q_padded.contiguous()
-    k_padded = k_padded.contiguous()
-    v_padded = v_padded.contiguous()
-    kv_padded = kv_padded.contiguous().to(torch.float32)
-    s = s.contiguous()
-
     # Create output tensor (padded)
     o_padded = torch.empty(b, h, n, e_padded, dtype=v.dtype, device=v.device)
 
+    # Create padded tensors without actually padding the data
+    q_padded = torch.empty(b, h, n, d_padded, dtype=q.dtype, device=q.device)
+    k_padded = torch.empty(b, h, n, d_padded, dtype=k.dtype, device=k.device)
+    v_padded = torch.empty(b, h, n, e_padded, dtype=v.dtype, device=v.device)
+    kv_padded = torch.empty(
+        b, h, d_padded, e_padded, dtype=torch.float32, device=kv.device
+    )
+
+    # Copy data to padded tensors
+    q_padded[..., :d] = q
+    k_padded[..., :d] = k
+    v_padded[..., :e] = v
+    kv_padded[..., :d, :e] = kv
+
     # Launch kernel
     grid = (b * h, 1)
     _decode_kernel[grid](
@@ -95,10 +116,12 @@ def lightning_attn_decode(q, k, v, kv, s):
         h=h,
         n=n,
         d=d_padded,
+        d_original=d,
         e=e_padded,
+        e_original=e,
     )
 
-    # Remove padding
+    # Get unpadded outputs
     o = o_padded[..., :e]
     kv_out = kv_padded[..., :d, :e]
 
@@ -351,6 +374,8 @@ def test_lightning_attention_implementations(model_params):
         msg="Lightning attention implementations produce different kv results",
     )
 
+    print("✅ Two implementations match")
+
 
 def _build_slope_tensor(n_attention_heads: int):
     def get_slopes(n):
@@ -375,7 +400,7 @@ def _build_slope_tensor(n_attention_heads: int):
 
 
 def get_benchmark():
-    batch_size_range = [2**i for i in range(0, 12)]  # max 2048
+    batch_size_range = [i for i in range(1, 33)]  # max 32
     seq_length_range = [1]  # decode mode sequence length is fixed to 1
     configs = list(itertools.product(batch_size_range, seq_length_range))