[IR] Change pragma convention, enable pass unroll option via pragma (#1112)

* [IR] Change pragma scope convention, enable pass unroll option via pragma

* add coverage test

* add explicit unroll as option

include/tvm/ir.h

@@ -177,8 +177,8 @@ constexpr const char* device_context_type = "device_context_type";
 constexpr const char* loop_scope = "loop_scope";
 /*! \brief Mark of reduce scope */
 constexpr const char* reduce_scope = "reduce_scope";
-/*! \brief Mark region is guarded by the pragma */
-constexpr const char* pragma_scope = "pragma_scope";
+/*! \brief Mark region is guarded by the pragma extension */
+constexpr const char* pragma_scope_prefix = "pragma_";
 /*!
  * \brief Mark of prefetch scope, value=offset,
  *  run prefetch of Tensor on the current loop scope
@@ -233,6 +233,16 @@ constexpr const char* pipeline_exec_scope = "pipeline_exec_scope";
  * Store statement.
  */
 constexpr const char* opengl_stage_scope = "opengl_stage_scope";
+
+/*!
+ * \brief Check if attr_key is a pragma key extension
+ * \param attr_key The attr key to be compared
+ * \return true if it is a pragma key
+ */
+inline bool IsPragmaKey(const std::string& attr_key) {
+  return attr_key.compare(0, 7, "pragma_") == 0;
+}
+
 }  // namespace attr
 
 /*! \brief namespace of TVM Intrinsic functions */

include/tvm/schedule.h

@@ -185,10 +185,13 @@ class Stage : public NodeRef {
    *
    * \param var The axis to be parallelized.
    * \param pragma_type The pragma type.
+   * \param pragma_value The pragma value
    *
    * \return reference to self.
    */
-  EXPORT Stage& pragma(IterVar var, const std::string& pragma_type);   // NOLINT(*)
+  EXPORT Stage& pragma(IterVar var,
+                       const std::string& pragma_type,
+                       const Expr& pragma_value = Expr());   // NOLINT(*)
   /*!
    * \brief Fetch data in advance.
    * \param domain the tensor to be prefetched
@@ -539,9 +542,13 @@ class IterVarAttrNode : public Node {
   /*! \brief Alignment offset of buffer dimension */
   int dim_align_offset{0};
   /*!
-   * \brief Additional pragmas, array of StringImm
+   * \brief Additional pragma keys, array of StringImm
    */
-  Array<Expr> pragmas;
+  Array<Expr> pragma_keys;
+  /*!
+   * \brief Additional values of pragma, if any
+   */
+  Array<Expr> pragma_values;
 
   void VisitAttrs(AttrVisitor* v) final {
     v->Visit("iter_type", &iter_type);
@@ -551,7 +558,8 @@ class IterVarAttrNode : public Node {
     v->Visit("tensor_intrin", &tensor_intrin);
     v->Visit("dim_align_factor", &dim_align_factor);
     v->Visit("dim_align_offset", &dim_align_offset);
-    v->Visit("pragmas", &pragmas);
+    v->Visit("pragma_keys", &pragma_keys);
+    v->Visit("pragma_values", &pragma_values);
   }
 
   static constexpr const char* _type_key = "IterVarAttr";

python/tvm/schedule.py

@@ -552,7 +552,7 @@ class Stage(NodeBase):
         """
         _api_internal._StageParallel(self, var)
 
-    def pragma(self, var, pragma_type):
+    def pragma(self, var, pragma_type, pragma_value=None):
         """Annotate the iteration with pragma
 
         This will translate to a pragma_scope surrounding
@@ -567,6 +567,9 @@ class Stage(NodeBase):
         pragma_type : str
              The pragma string to be annotated
 
+        pragma_value : Expr, optional
+             The pragma value to pass along the pragma
+
         Note
         ----
         Most pragmas are advanced/experimental features
@@ -597,7 +600,7 @@ class Stage(NodeBase):
           :code:`for (int i = task_id; i < end; i += num_task)`
 
         """
-        _api_internal._StagePragma(self, var, pragma_type)
+        _api_internal._StagePragma(self, var, pragma_type, pragma_value)
 
     def prefetch(self, tensor, var, offset):
         """Prefetch the specified variable

src/api/api_lang.cc

@@ -380,7 +380,7 @@ TVM_REGISTER_API("_StageParallel")
 TVM_REGISTER_API("_StagePragma")
   .set_body([](TVMArgs args, TVMRetValue* ret) {
     args[0].operator Stage()
-        .pragma(args[1], args[2]);
+        .pragma(args[1], args[2], args[3]);
   });
 
 TVM_REGISTER_API("_StagePrefetch")

src/codegen/llvm/codegen_cpu.cc

@@ -683,16 +683,15 @@ void CodeGenCPU::VisitStmt_(const AttrStmt* op) {
     this->CreateStaticInit(op->value.as<StringImm>()->value, op->body);
   } else  if (op->attr_key == ir::attr::compute_scope) {
     this->CreateComputeScope(op);
-  } else if (op->attr_key == ir::attr::pragma_scope) {
-    const std::string& pname = op->value.as<StringImm>()->value;
-    if (pname == "parallel_stride_pattern") {
+  } else if (attr::IsPragmaKey(op->attr_key)) {
+    if (op->attr_key == "pragma_parallel_stride_pattern") {
       CHECK(parallel_env_.penv != nullptr)
           << "Pragma parallel_stride_pattern only valid in parallel launch";
       parallel_env_.stride_pattern = true;
       this->VisitStmt(op->body);
-    } else if (pname == "parallel_launch_point") {
+    } else if (op->attr_key == "pragma_parallel_launch_point") {
       CreateParallelLaunch(op->body, 0);
-    } else if (pname == "parallel_barrier_when_finish") {
+    } else if (op->attr_key == "pragma_parallel_barrier_when_finish") {
       CHECK(parallel_env_.penv != nullptr)
           << "Cannot run barrier without parallel environment";
       CHECK(!parallel_env_.in_parallel_loop)
@@ -703,7 +702,7 @@ void CodeGenCPU::VisitStmt_(const AttrStmt* op) {
           RuntimeTVMParallelBarrier(),
           {MakeValue(parallel_env_.task_id),  parallel_env_.penv});
     } else {
-      LOG(WARNING) << "Unknown pragma " << pname;
+      LOG(WARNING) << "Unknown pragma " << op->attr_key;
       this->VisitStmt(op->body);
     }
   } else {

src/codegen/spirv/codegen_spirv.cc

@@ -77,7 +77,7 @@ spirv::Value CodeGenSPIRV::GetThreadIndex(
   spirv::Value v;
   if (ts.rank == 1) {
     v = builder_->GetLocalID(ts.dim_index);
-    int size;
+    int size = 0;
     CHECK(arith::GetConstInt(extent, &size))
         << "SPIRV only allows constant thread group size " << " get " << extent;
     CHECK_LT(ts.dim_index, 3);

src/op/op_util.cc

@@ -71,9 +71,15 @@ MakeLoopNest(const Stage& stage,
                               << it_attr->iter_type
                               << " in the iter_var_attrs";
         }
-        for (Expr p : it_attr->pragmas) {
+        CHECK_EQ(it_attr->pragma_keys.size(), it_attr->pragma_values.size());
+        for (size_t k = 0; k < it_attr->pragma_keys.size(); ++k) {
+          const std::string& pkey = it_attr->pragma_keys[k].as<StringImm>()->value;
+          Expr pvalue = it_attr->pragma_values[k];
+          if (!pvalue.defined()) {
+            pvalue = make_const(Int(32), 1);
+          }
           nest[i + 1].emplace_back(
-              AttrStmt::make(iv, ir::attr::pragma_scope, p, no_op));
+              AttrStmt::make(iv, ir::attr::pragma_scope_prefix + pkey, pvalue, no_op));
         }
       }
       if (!debug_keep_trivial_loop && is_one(dom->extent)) {

src/pass/inject_copy_intrin.cc

@@ -17,7 +17,7 @@ class CopyIntrinInjector : public IRMutator {
  public:
   CopyIntrinInjector(const std::string& pragma_key,
                      const PackedFunc& flower_copy_fromto)
-      : pragma_key_(pragma_key),
+      : pragma_key_(attr::pragma_scope_prefix+  pragma_key),
         flower_copy_fromto_(flower_copy_fromto) {
   }
 
@@ -25,15 +25,12 @@ class CopyIntrinInjector : public IRMutator {
     if (op->attr_key == attr::storage_scope) {
       const Variable* buf = op->node.as<Variable>();
       storage_scope_[buf] = op->value.as<StringImm>()->value;
-    } else if (op->attr_key == ir::attr::pragma_scope) {
-      const std::string& pname = op->value.as<StringImm>()->value;
-      if (pname == pragma_key_) {
+    } else if (op->attr_key == pragma_key_) {
       Stmt ret;
       CHECK(MatchCopyPattern(op->body, &ret))
           << "Cannot match copy pattern of " << op->body;
       return ret;
     }
-    }
     return IRMutator::Mutate_(op, s);
   }
 

src/pass/remove_no_op.cc

@@ -20,12 +20,9 @@ class NoOpRemover : public IRMutator {
     return is_no_op(op->body) ? MakeEvaluate(op->value) : stmt;
   }
   Stmt Mutate_(const AttrStmt* op, const Stmt& s) final {
-    if (op->attr_key == ir::attr::pragma_scope) {
-      const std::string& pname = op->value.as<StringImm>()->value;
-      if (pname == "debug_skip_region") {
+    if (op->attr_key == "pragma_debug_skip_region") {
       return MakeEvaluate(0);
     }
-    }
     Stmt stmt = IRMutator::Mutate_(op, s);
     op = stmt.as<AttrStmt>();
     return is_no_op(op->body) ? MakeEvaluate(op->value) : stmt;

src/pass/storage_rewrite.cc

@@ -401,7 +401,7 @@ class StoragePlanRewriter : public IRMutator {
       return this->Mutate(op->body);
     } else if (op->attr_key == attr::thread_extent ||
                op->attr_key == attr::virtual_thread ||
-               op->attr_key == attr::pragma_scope) {
+               attr::IsPragmaKey(op->attr_key)) {
       // remake all the allocation at the attach scope.
       if (attach_map_.count(op)) {
         auto& svec = attach_map_[op];
@@ -737,8 +737,8 @@ class StoragePlanRewriter : public IRMutator {
       if (s.stmt->is_type<AttrStmt>()) {
         const auto* op = static_cast<const AttrStmt*>(s.stmt);
         if (op->attr_key == attr::thread_extent ||
-            op->attr_key == attr::pragma_scope ||
-            op->attr_key == attr::virtual_thread) {
+            op->attr_key == attr::virtual_thread ||
+            attr::IsPragmaKey(op->attr_key)) {
           PlanNewScope(op);
         } else {
           CHECK(op->attr_key == attr::extern_scope);

src/pass/unroll_loop.cc

@@ -27,6 +27,27 @@ class LoopUnroller : public IRMutator {
         explicit_unroll_(explicit_unroll) {
   }
 
+  Stmt Mutate_(const AttrStmt* op, const Stmt& stmt) final {
+    if (op->attr_key == "pragma_auto_unroll_max_step") {
+      int value;
+      CHECK(arith::GetConstInt(op->value, &value));
+      std::swap(value, auto_max_step_);
+      Stmt ret = this->Mutate(op->body);
+      std::swap(value, auto_max_step_);
+      return ret;
+    } else if (op->attr_key == "pragma_unroll_explicit") {
+      int value;
+      CHECK(arith::GetConstInt(op->value, &value));
+      bool explicit_unroll = value;
+      std::swap(explicit_unroll, explicit_unroll_);
+      Stmt ret = this->Mutate(op->body);
+      std::swap(explicit_unroll, explicit_unroll_);
+      return ret;
+    } else {
+      return IRMutator::Mutate_(op, stmt);
+    }
+  }
+
   Stmt Mutate_(const For* op, const Stmt& s) {
     Stmt stmt = IRMutator::Mutate_(op, s);
     op = stmt.as<For>();

src/schedule/schedule_lang.cc

@@ -350,14 +350,18 @@ Stage& Stage::parallel(IterVar var) {   // NOLINT(*)
   return *this;
 }
 
-Stage& Stage::pragma(IterVar var, const std::string& pragma_type) {   // NOLINT(*)
+Stage& Stage::pragma(IterVar var,
+                     const std::string& pragma_type,
+                     const Expr& pragma_value) {   // NOLINT(*)
   if (pragma_type == "unroll") {
     this->unroll(var);
   } else if (pragma_type == "vectorize") {
     this->vectorize(var);
   } else {
-    UpdateIterVarAttr(operator->(), var, [pragma_type](IterVarAttrNode* n) {
-        n->pragmas.push_back(ir::StringImm::make(pragma_type));
+    UpdateIterVarAttr(
+        operator->(), var, [pragma_type, pragma_value](IterVarAttrNode* n) {
+          n->pragma_keys.push_back(ir::StringImm::make(pragma_type));
+          n->pragma_values.push_back(pragma_value);
         });
   }
   return *this;

tests/python/unittest/test_lang_schedule.py

@@ -110,7 +110,7 @@ def test_pragma():
     s[T].pragma(xo, "pragma1")
     s[T].pragma(xi, "vectorize")
     VECTORIZE = tvm.schedule.IterVar.Vectorized
-    assert s[T].iter_var_attrs[xo].pragmas[0].value == "pragma1"
+    assert s[T].iter_var_attrs[xo].pragma_keys[0].value == "pragma1"
     assert s[T].iter_var_attrs[xi].iter_type == VECTORIZE
 
 

tests/python/unittest/test_pass_unroll.py

 import tvm
 import os
 
+
 def test_unroll_loop():
+    ib = tvm.ir_builder.create()
     dtype = 'int64'
     n = tvm.var('n')
     Ab = tvm.decl_buffer((n, ), dtype)
-    i = tvm.var('i')
-    j = tvm.var('j')
+    Aptr = ib.buffer_ptr(Ab)
     # for i in 0 to n-1:
-    stmt = tvm.make.For(
-        i, n, 2, 0, 0,
-        tvm.make.For(j, 0, 8, 3, 0,
-                     tvm.make.Store(Ab.data,
-                                    tvm.make.Load(dtype, Ab.data, i) + 1,
-                                    j + 1)))
+    with ib.for_range(n, n + 2, name="i") as i:
+        with ib.for_range(0, 8, name="i", for_type="unroll") as j:
+            Aptr[j + 1] = Aptr[i] + 1
+
+    stmt = ib.get()
     assert isinstance(stmt, tvm.stmt.For)
     ret = tvm.ir_pass.UnrollLoop(stmt, 16, 8, 0, True)
     assert not isinstance(ret, tvm.stmt.For)
@@ -23,23 +23,18 @@ def test_unroll_loop():
     assert isinstance(ret, tvm.stmt.For)
     assert ret.for_type == tvm.stmt.For.Unrolled
 
+    ib = tvm.ir_builder.create()
+    ib.scope_attr(tvm.const(0), "pragma_auto_unroll_max_step", 16)
+    ib.emit(stmt)
+    wrapped = ib.get()
+    wrapped = tvm.make.Block(wrapped, stmt)
+    assert isinstance(ret, tvm.stmt.For)
+    ret = tvm.ir_pass.UnrollLoop(wrapped, 0, 8, 0, False)
+    assert isinstance(ret.first, tvm.stmt.For)
+    assert ret.first.for_type == tvm.stmt.For.Unrolled
+    assert isinstance(ret.rest, tvm.stmt.For)
+    assert ret.rest.for_type != tvm.stmt.For.Unrolled
+
 
 if __name__ == "__main__":
-    with tvm.build_config(dump_pass_ir=True):
     test_unroll_loop()
-
-    def end_with(*suffix):
-        ends = suffix
-        def run(s):
-            f = map(s.endswith, ends)
-            if True in f: return s
-        return run
-
-    file_list = os.listdir('./')
-    cc_file = end_with('.cc')
-    cc_file = filter(cc_file, file_list)
-    cc_file = [f for f in cc_file]
-    assert len(cc_file) == 3
-    for i in cc_file:
-        os.remove(i)
-    

tests/python/unittest/test_schedule_schedule_ops.py

@@ -20,6 +20,7 @@ def test_schedule1():
 
     s = tvm.create_schedule(A1.op)
     xo, xi = s[A1].split(A1.op.axis[0], 8)
+    s[A1].pragma(xo, "auto_unroll_max_step", 10)
     bounds = tvm.schedule.InferBound(s)
     assert isinstance(bounds, tvm.container.Map)
     stmt = tvm.schedule.ScheduleOps(s, bounds)