[Enhancement][Pipeline] More precise copy code block detection in pipeline (#384)

* Update legalize_safe_memory_access.cc

* Add cache path handling and file locking in Cython adapter

- Introduced a new cache path based on the code hash for the Cython JIT adapter, enhancing cache management.
- Added a lock file mechanism to ensure safe access during cache operations, improving concurrency handling.
- These changes aim to optimize the compilation process and prevent race conditions during library loading.

* lint fix

* refactor

* refactor

* Add GlobalCopyPatternDetector to identify global memory copy patterns

- Introduced a new class, GlobalCopyPatternDetector, to detect specific memory copy patterns in statements.
- Enhanced the PipelinePlanner to utilize this detector for determining copy stages based on global and local memory scopes.
- Improved code clarity and maintainability by encapsulating detection logic within the new class.

* Refactor copy stage detection logic in pipeline planning

- Simplified the determination of copy stages by directly assigning the result of GlobalCopyPatternDetector to pinfo.copy_stage.
- Removed redundant checks for read and write scopes, enhancing code clarity and maintainability.

* lint fix

src/transform/pipeline_planning.cc

@@ -24,6 +24,7 @@
 
 #include <tvm/arith/analyzer.h>
 #include <tvm/tir/analysis.h>
+#include <tvm/tir/builtin.h>
 #include <tvm/tir/stmt_functor.h>
 #include <tvm/tir/transform.h>
 
@@ -54,6 +55,68 @@ bool MayConflict(Region region1, Region region2) {
   return true;
 }
 
+/*!
+ * \brief Detect if a statement follows the global memory copy pattern:
+ *        1. Contains exactly one buffer store operation
+ *        2. Source buffer must be in global memory scope
+ *        3. Destination buffer must be in local or shared memory scope
+ */
+class GlobalCopyPatternDetector : public StmtExprVisitor {
+public:
+  static bool Detect(const Stmt &stmt) {
+    GlobalCopyPatternDetector detector;
+    detector.VisitStmt(stmt);
+    return detector.is_global_copy_pattern_;
+  }
+
+private:
+  void VisitStmt_(const BufferStoreNode *op) final {
+    Buffer store_buffer = op->buffer;
+    is_global_read_ = false;
+    this->VisitExpr(op->value);
+    if (is_global_read_ && (store_buffer.scope() == "shared" ||
+                            store_buffer.scope() == "shared.dyn" ||
+                            store_buffer.scope() == "local")) {
+      is_global_copy_pattern_ = true;
+    }
+    is_global_read_ = false;
+  }
+
+  void VisitExpr_(const BufferLoadNode *op) final {
+    if (op->buffer.scope() == "global") {
+      is_global_read_ = true;
+    }
+  }
+
+  void VisitExpr_(const CallNode *op) final {
+    auto args = op->args;
+    if (op->op.same_as(tir::builtin::if_then_else())) {
+      // Simplify nested if_then_else
+      // if (cond) { if (inner_cond) { inner_then_expr } else { inner_else_expr
+      // } } else { else_expr }
+      // => if (cond && inner_cond) { inner_then_expr } else { else_expr }
+      const PrimExpr &cond = op->args[0];
+      const PrimExpr &then_expr = op->args[1];
+      const PrimExpr &else_expr = op->args[2];
+      this->VisitExpr(then_expr);
+      this->VisitExpr(else_expr);
+    }
+  }
+
+  void VisitStmt_(const IfThenElseNode *op) final {
+    // Skip condition
+    this->VisitStmt(op->then_case);
+    if (op->else_case.defined()) {
+      this->VisitStmt(op->else_case.value());
+    }
+  }
+
+private:
+  bool is_global_read_ = false;
+  bool under_buffer_store_ = false;
+  bool is_global_copy_pattern_ = false;
+};
+
 class PipelinePlanner : public StmtExprMutator {
 public:
   static Stmt Substitute(const PrimFunc &f) {
@@ -89,20 +152,7 @@ private:
     pinfo.reads = std::move(access[0]);
     pinfo.writes = std::move(access[1]);
     pinfo.original_order = idx;
-
-    // copy stage should only have one reads and one writes
-    bool write_to_shared_or_local = false;
-    bool read_from_global = false;
-    for (auto region : pinfo.reads)
-      if (region->buffer.scope() == "global")
-        read_from_global = true;
-    for (auto region : pinfo.writes)
-      if (region->buffer.scope() == "shared" ||
-          region->buffer.scope() == "shared.dyn" ||
-          region->buffer.scope() == "local")
-        write_to_shared_or_local = true;
-
-    pinfo.copy_stage = write_to_shared_or_local && read_from_global;
+    pinfo.copy_stage = GlobalCopyPatternDetector::Detect(stmt);
 
     return std::move(pinfo);
   }

testing/python/kernel/test_tilelang_kernel_deepseek_nsa.py

+# ruff: noqa
+from tilelang import tvm as tvm
+import tilelang.testing
+import tilelang.language as T
+import torch
+from typing import Optional, Union
+from einops import rearrange, repeat
+
+tilelang.testing.set_random_seed(42)
+
+
+def naive_nsa_ref(q: torch.Tensor,
+                  k: torch.Tensor,
+                  v: torch.Tensor,
+                  g_slc: torch.Tensor,
+                  g_swa: torch.Tensor,
+                  block_indices: torch.LongTensor,
+                  block_counts: Optional[Union[torch.LongTensor, int]] = None,
+                  block_size: int = 64,
+                  window_size: int = 0,
+                  scale: Optional[float] = None,
+                  cu_seqlens: Optional[torch.LongTensor] = None,
+                  head_first: bool = False) -> torch.Tensor:
+
+    if scale is None:
+        scale = k.shape[-1]**-0.5
+    if cu_seqlens is not None:
+        assert q.shape[0] == 1, "batch size must be 1 when cu_seqlens are provided"
+        if head_first:
+            raise RuntimeError(
+                "Sequences with variable lengths are not supported for head-first mode")
+    if head_first:
+        q, k, v, block_indices = map(lambda x: rearrange(x, 'b h t d -> b t h d'),
+                                     (q, k, v, block_indices))
+        g_slc, g_swa = map(lambda x: rearrange(x, 'b h t -> b t h'), (g_slc, g_swa))
+        if isinstance(block_counts, torch.Tensor):
+            block_counts = rearrange(block_counts, 'b h t -> b t h')
+
+    dtype = q.dtype
+    G = q.shape[2] // k.shape[2]
+    BS = block_size
+    S = block_indices.shape[-1]
+    k, v, block_indices = (repeat(x, 'b t h d -> b t (h g) d', g=G) for x in (k, v, block_indices))
+    if isinstance(block_counts, torch.Tensor):
+        block_counts = repeat(block_counts, 'b t h -> b t (h g)', g=G)
+    c = torch.arange(S).repeat_interleave(BS).unsqueeze(1).expand(-1, q.shape[2]).to(q.device)
+    q, k, v = map(lambda x: x.float(), (q, k, v))
+
+    o_slc = torch.zeros_like(v)
+    o_swa = torch.zeros_like(v) if window_size > 0 else None
+    varlen = True
+    if cu_seqlens is None:
+        varlen = False
+        B, T = q.shape[:2]
+        cu_seqlens = torch.cat(
+            [block_indices.new_tensor(range(0, B * T, T)),
+             block_indices.new_tensor([B * T])])
+
+    for i in range(len(cu_seqlens) - 1):
+        if not varlen:
+            q_b, k_b, v_b, g_slc_b, g_swa_b, i_b = q[i], k[i], v[i], g_slc[i], g_swa[
+                i], block_indices[i]
+            if isinstance(block_counts, torch.Tensor):
+                s_b = block_counts[i]
+            else:
+                s_b = block_counts
+        else:
+            T = cu_seqlens[i + 1] - cu_seqlens[i]
+            q_b, k_b, v_b, g_slc_b, g_swa_b, i_b = map(
+                lambda x: x[0][cu_seqlens[i]:cu_seqlens[i + 1]],
+                (q, k, v, g_slc, g_swa, block_indices))
+            if isinstance(block_counts, torch.Tensor):
+                s_b = block_counts[0][cu_seqlens[i]:cu_seqlens[i + 1]]
+            else:
+                s_b = block_counts
+
+        i_b = i_b.unsqueeze(-1) * BS + i_b.new_tensor(range(BS))
+        # [T, S*BS, HQ]
+        i_b = i_b.view(T, block_indices.shape[2], -1).transpose(1, 2)
+        for i_q in range(T):
+            # [HQ, D]
+            q_i = q_b[i_q] * scale
+            # [HQ]
+            g_slc_i = g_slc_b[i_q]
+            # [HQ]
+            g_swa_i = g_swa_b[i_q]
+            # [S*BS, HQ]
+            i_i = i_b[i_q]
+            # [HQ]
+            if isinstance(block_counts, torch.Tensor):
+                s_i = s_b[i_q]
+            else:
+                s_i = s_b
+            # [S*BS, HQ, -1]
+            k_i_slc, v_i_slc = map(
+                lambda x: x.gather(
+                    0,
+                    i_i.clamp(0, T - 1).unsqueeze(-1).expand(*i_i.shape, x.shape[-1])), (k_b, v_b))
+            # [S*BS, HQ]
+            attn_slc = torch.einsum('h d, n h d -> n h', q_i, k_i_slc).masked_fill(
+                torch.logical_or(i_i < 0, i_i > i_q) |
+                (c >= s_i if block_counts is not None else False), float('-inf')).softmax(0)
+            if not varlen:
+                o_slc[i, i_q] = torch.einsum('n h, n h v -> h v', attn_slc,
+                                             v_i_slc) * g_slc_i.unsqueeze(-1)
+            else:
+                o_slc[0][cu_seqlens[i] + i_q] = torch.einsum('n h, n h v -> h v', attn_slc,
+                                                             v_i_slc) * g_slc_i.unsqueeze(-1)
+            if window_size > 0:
+                k_i_swa, v_i_swa = map(lambda x: x[max(0, i_q - window_size + 1):i_q + 1],
+                                       (k_b, v_b))
+                attn_swa = torch.einsum('h d, n h d -> n h', q_i, k_i_swa).softmax(0)
+                if not varlen:
+                    o_swa[i, i_q] = torch.einsum('n h, n h v -> h v', attn_swa,
+                                                 v_i_swa) * g_swa_i.unsqueeze(-1)
+                else:
+                    o_swa[0][cu_seqlens[i] + i_q] = torch.einsum('n h, n h v -> h v', attn_swa,
+                                                                 v_i_swa) * g_swa_i.unsqueeze(-1)
+
+    if head_first:
+        o_slc = rearrange(o_slc, 'b t h d -> b h t d')
+        o_swa = rearrange(o_swa, 'b t h d -> b h t d')
+
+    return o_slc.to(dtype) + o_swa.to(dtype) if o_swa is not None else o_slc.to(dtype)
+
+
+def native_sparse_attention(batch,
+                            heads,
+                            seq_len,
+                            dim,
+                            is_causal,
+                            scale=None,
+                            block_size=64,
+                            groups=16,
+                            selected_blocks=16,
+                            num_stages=0,
+                            threads=32):
+    if scale is None:
+        scale = (1.0 / dim)**0.5 * 1.44269504  # log2(e)
+    else:
+        scale = scale * 1.44269504  # log2(e)
+
+    head_kv = heads // groups
+    q_shape = [batch, seq_len, heads, dim]
+    kv_shape = [batch, seq_len, head_kv, dim]
+    block_indices_shape = [batch, seq_len, head_kv, selected_blocks]
+    block_indices_dtype = "int32"
+    dtype = "float16"
+    accum_dtype = "float"
+    block_S = block_size
+    block_T = min(128, tilelang.math.next_power_of_2(dim))
+
+    NK = tilelang.cdiv(dim, block_T)
+    NV = tilelang.cdiv(dim, block_T)
+    assert NK == 1, "The key dimension can not be larger than 256"
+
+    S = selected_blocks
+    G = groups
+    BS = block_S
+    BK = BV = block_T
+
+    @T.prim_func
+    def native_sparse_attention(
+            Q: T.Tensor(q_shape, dtype),
+            K: T.Tensor(kv_shape, dtype),
+            V: T.Tensor(kv_shape, dtype),
+            BlockIndices: T.Tensor(block_indices_shape, block_indices_dtype),
+            Output: T.Tensor(q_shape, dtype),
+    ):
+        with T.Kernel(seq_len, NV, batch * head_kv, threads=threads) as (bx, by, bz):
+            Q_shared = T.alloc_shared([G, BK], dtype)
+            K_shared = T.alloc_shared([BS, BK], dtype)
+            V_shared = T.alloc_shared([BS, BV], dtype)
+            O_shared = T.alloc_shared([G, BV], dtype)
+
+            acc_s = T.alloc_fragment([G, BS], accum_dtype)
+            acc_s_cast = T.alloc_fragment([G, BS], dtype)
+            acc_o = T.alloc_fragment([G, BV], accum_dtype)
+            scores_max = T.alloc_fragment([G], accum_dtype)
+            scores_max_prev = T.alloc_fragment([G], accum_dtype)
+            scores_scale = T.alloc_fragment([G], accum_dtype)
+            scores_sum = T.alloc_fragment([G], accum_dtype)
+            logsum = T.alloc_fragment([G], accum_dtype)
+
+            i_t, i_v, i_bh = bx, by, bz
+            i_b, i_h = i_bh // head_kv, i_bh % head_kv
+
+            NS = S
+            T.copy(Q[i_b, i_t, i_h * G:(i_h + 1) * G, :], Q_shared)
+
+            T.fill(acc_o, 0)
+            T.fill(logsum, 0)
+            T.fill(scores_max, -T.infinity(accum_dtype))
+
+            for i in T.Pipelined(NS, num_stages=num_stages):
+                i_s = BlockIndices[i_b, i_t, i_h, i] * BS
+                if i_s <= i_t and i_s >= 0:
+                    # [BS, BK]
+                    T.copy(K[i_b, i_s:i_s + BS, i_h, :], K_shared)
+
+                    if is_causal:
+                        for i, j in T.Parallel(G, BS):
+                            acc_s[i, j] = T.if_then_else(i_t >= (i_s + j), 0,
+                                                         -T.infinity(acc_s.dtype))
+                    else:
+                        T.clear(acc_s)
+
+                    T.gemm(
+                        Q_shared,
+                        K_shared,
+                        acc_s,
+                        transpose_B=True,
+                        policy=T.GemmWarpPolicy.FullRow)
+
+                    # Softmax
+                    T.copy(scores_max, scores_max_prev)
+                    T.fill(scores_max, -T.infinity(accum_dtype))
+                    T.reduce_max(acc_s, scores_max, dim=1, clear=True)
+                    for i in T.Parallel(G):
+                        scores_scale[i] = T.exp2(scores_max_prev[i] * scale - scores_max[i] * scale)
+                    for i, j in T.Parallel(G, BS):
+                        acc_s[i, j] = T.exp2(acc_s[i, j] * scale - scores_max[i] * scale)
+                    T.reduce_sum(acc_s, scores_sum, dim=1)
+                    for i in T.Parallel(G):
+                        logsum[i] = logsum[i] * scores_scale[i] + scores_sum[i]
+                    T.copy(acc_s, acc_s_cast)
+
+                    # Rescale
+                    for i, j in T.Parallel(G, BV):
+                        acc_o[i, j] *= scores_scale[i]
+
+                    # V * softmax(Q * K)
+                    T.copy(V[i_b, i_s:i_s + BS, i_h, i_v * BV:(i_v + 1) * BV], V_shared)
+                    T.gemm(acc_s_cast, V_shared, acc_o, policy=T.GemmWarpPolicy.FullRow)
+
+            for i, j in T.Parallel(G, BV):
+                acc_o[i, j] /= logsum[i]
+            T.copy(acc_o, O_shared)
+            T.copy(O_shared, Output[i_b, i_t, i_h * G:(i_h + 1) * G, i_v * BV:(i_v + 1) * BV])
+
+    return native_sparse_attention
+
+
+def run_native_sparse_attention(batch,
+                                heads,
+                                seq_len,
+                                dim,
+                                is_causal,
+                                scale=None,
+                                block_size=64,
+                                groups=16,
+                                selected_blocks=16,
+                                num_stages=0,
+                                threads=32):
+    dtype = torch.float16
+    head_kv = heads // groups
+    program = native_sparse_attention(batch, heads, seq_len, dim, is_causal, scale, block_size,
+                                      groups, selected_blocks, num_stages, threads)
+    kernel = tilelang.compile(program, out_idx=-1)
+    Q = torch.randn((batch, seq_len, heads, dim), dtype=dtype).cuda()
+    K = torch.randn((batch, seq_len, head_kv, dim), dtype=dtype).cuda()
+    V = torch.randn((batch, seq_len, head_kv, dim), dtype=dtype).cuda()
+    g_slc = torch.ones((batch, seq_len, heads), dtype=dtype).cuda()
+    g_swa = torch.ones((batch, seq_len, heads), dtype=dtype).cuda()
+
+    block_indices = torch.full((batch, seq_len, head_kv, selected_blocks),
+                               seq_len,
+                               dtype=torch.long,
+                               device='cuda')
+    for b in range(batch):
+        for t in range(seq_len):
+            for h in range(head_kv):
+                i_i = torch.randperm(max(1, (t // block_size)))[:selected_blocks]
+                block_indices[b, t, h, :len(i_i)] = i_i
+    block_indices = block_indices.sort(-1)[0]
+    block_counts = torch.randint(1, selected_blocks + 1, (batch, seq_len, head_kv), device='cuda')
+
+    out = kernel(Q, K, V, block_indices.to(torch.int32))
+
+    ref = naive_nsa_ref(
+        q=Q,
+        k=K,
+        v=V,
+        g_slc=g_slc,
+        g_swa=g_swa,
+        block_indices=block_indices,
+        block_counts=block_counts,
+        block_size=block_size,
+        scale=scale,
+    )
+    torch.testing.assert_close(ref, out, atol=1e-2, rtol=1e-2)
+
+
+def test_tilelang_kernel_deepseek_nsa():
+    # disable pipeline
+    run_native_sparse_attention(
+        batch=2,
+        heads=64,
+        seq_len=1,
+        dim=16,
+        is_causal=True,
+        scale=None,
+        block_size=32,
+        groups=16,
+        selected_blocks=16,
+        num_stages=0,
+        threads=32)
+    # enable pipeline
+    run_native_sparse_attention(
+        batch=2,
+        heads=64,
+        seq_len=1,
+        dim=16,
+        is_causal=True,
+        scale=None,
+        block_size=32,
+        groups=16,
+        selected_blocks=16,
+        num_stages=2,
+        threads=32)
+
+
+if __name__ == "__main__":
+    tilelang.testing.main()