Index: include/polly/CodeGen/IslAst.h
===================================================================
--- include/polly/CodeGen/IslAst.h
+++ include/polly/CodeGen/IslAst.h
@@ -126,6 +126,9 @@
   /// @brief Is this loop a reduction parallel loop?
   static bool isReductionParallel(__isl_keep isl_ast_node *Node);
 
+  /// @brief Will the loop be run as thread parallel?
+  static bool runAsThreadParallel(__isl_keep isl_ast_node *Node);
+
   /// @brief Get the nodes schedule or a nullptr if not available.
   static __isl_give isl_union_map *getSchedule(__isl_keep isl_ast_node *Node);
 
Index: lib/CodeGen/IslAst.cpp
===================================================================
--- lib/CodeGen/IslAst.cpp
+++ lib/CodeGen/IslAst.cpp
@@ -42,6 +42,11 @@
 
 using IslAstUserPayload = IslAstInfo::IslAstUserPayload;
 
+static cl::opt<bool>
+    UseThreads("polly-parallel",
+               cl::desc("Generate thread parallel code (isl codegen only)"),
+               cl::init(false), cl::ZeroOrMore, cl::cat(PollyCategory));
+
 static cl::opt<bool> UseContext("polly-ast-use-context",
                                 cl::desc("Use context"), cl::Hidden,
                                 cl::init(false), cl::ZeroOrMore,
@@ -148,6 +153,7 @@
 
   isl_pw_aff *DD = IslAstInfo::getMinimalDependenceDistance(Node);
   const std::string BrokenReductionsStr = getBrokenReductionsStr(Node);
+  const std::string KnownParallelStr = "#pragma known-parallel";
   const std::string DepDisPragmaStr = "#pragma minimal dependence distance: ";
   const std::string SimdPragmaStr = "#pragma simd";
   const std::string OmpPragmaStr = "#pragma omp parallel for";
@@ -158,8 +164,10 @@
   if (IslAstInfo::isInnermostParallel(Node))
     Printer = printLine(Printer, SimdPragmaStr + BrokenReductionsStr);
 
-  if (IslAstInfo::isOutermostParallel(Node))
-    Printer = printLine(Printer, OmpPragmaStr + BrokenReductionsStr);
+  if (IslAstInfo::runAsThreadParallel(Node))
+    Printer = printLine(Printer, OmpPragmaStr);
+  else if (IslAstInfo::isOutermostParallel(Node))
+    Printer = printLine(Printer, KnownParallelStr + BrokenReductionsStr);
 
   isl_pw_aff_free(DD);
   return isl_ast_node_for_print(Node, Printer, Options);
@@ -357,7 +365,8 @@
   isl_union_map *Schedule =
       isl_union_map_intersect_domain(S->getSchedule(), S->getDomains());
 
-  if (DetectParallel || PollyVectorizerChoice != VECTORIZER_NONE) {
+  if (UseThreads || DetectParallel ||
+      PollyVectorizerChoice != VECTORIZER_NONE) {
     BuildInfo.Deps = &D;
     BuildInfo.InParallelFor = 0;
 
@@ -444,6 +453,10 @@
   return Payload && Payload->IsReductionParallel;
 }
 
+bool IslAstInfo::runAsThreadParallel(__isl_keep isl_ast_node *Node) {
+  return isOutermostParallel(Node) && !isReductionParallel(Node) && UseThreads;
+}
+
 isl_union_map *IslAstInfo::getSchedule(__isl_keep isl_ast_node *Node) {
   IslAstUserPayload *Payload = getNodePayload(Node);
   return Payload ? isl_ast_build_get_schedule(Payload->Build) : nullptr;
Index: test/Isl/Ast/OpenMP/multiple_loops_outer_parallel.ll
===================================================================
--- test/Isl/Ast/OpenMP/multiple_loops_outer_parallel.ll
+++ test/Isl/Ast/OpenMP/multiple_loops_outer_parallel.ll
@@ -1,4 +1,4 @@
-; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
+; RUN: opt %loadPolly -polly-ast -polly-parallel -analyze < %s | FileCheck %s
 ;
 ;       void jd(int *A) {
 ; CHECK:  #pragma omp parallel for
Index: test/Isl/Ast/OpenMP/nested_loop_both_parallel.ll
===================================================================
--- test/Isl/Ast/OpenMP/nested_loop_both_parallel.ll
+++ test/Isl/Ast/OpenMP/nested_loop_both_parallel.ll
@@ -1,4 +1,4 @@
-; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
+; RUN: opt %loadPolly -polly-ast -polly-parallel -analyze < %s | FileCheck %s
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
 target triple = "x86_64-pc-linux-gnu"
 
Index: test/Isl/Ast/OpenMP/nested_loop_both_parallel_parametric.ll
===================================================================
--- test/Isl/Ast/OpenMP/nested_loop_both_parallel_parametric.ll
+++ test/Isl/Ast/OpenMP/nested_loop_both_parallel_parametric.ll
@@ -1,4 +1,4 @@
-; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze -polly-delinearize < %s | FileCheck %s
+; RUN: opt %loadPolly -polly-ast -polly-parallel -analyze -polly-delinearize < %s | FileCheck %s
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
 target triple = "x86_64-pc-linux-gnu"
 ; int A[1024][1024];
Index: test/Isl/Ast/OpenMP/nested_loop_inner_parallel.ll
===================================================================
--- test/Isl/Ast/OpenMP/nested_loop_inner_parallel.ll
+++ test/Isl/Ast/OpenMP/nested_loop_inner_parallel.ll
@@ -1,4 +1,4 @@
-; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
+; RUN: opt %loadPolly -polly-ast -polly-parallel -analyze < %s | FileCheck %s
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
 target triple = "x86_64-pc-linux-gnu"
 
Index: test/Isl/Ast/OpenMP/nested_loop_outer_parallel.ll
===================================================================
--- test/Isl/Ast/OpenMP/nested_loop_outer_parallel.ll
+++ test/Isl/Ast/OpenMP/nested_loop_outer_parallel.ll
@@ -1,4 +1,4 @@
-; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
+; RUN: opt %loadPolly -polly-ast -polly-parallel -analyze < %s | FileCheck %s
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
 target triple = "x86_64-pc-linux-gnu"
 
Index: test/Isl/Ast/OpenMP/single_loop_param_non_parallel.ll
===================================================================
--- test/Isl/Ast/OpenMP/single_loop_param_non_parallel.ll
+++ test/Isl/Ast/OpenMP/single_loop_param_non_parallel.ll
@@ -1,4 +1,4 @@
-; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
+; RUN: opt %loadPolly -polly-ast -polly-parallel -analyze < %s | FileCheck %s
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
 target triple = "x86_64-pc-linux-gnu"
 
Index: test/Isl/Ast/OpenMP/single_loop_param_parallel.ll
===================================================================
--- test/Isl/Ast/OpenMP/single_loop_param_parallel.ll
+++ test/Isl/Ast/OpenMP/single_loop_param_parallel.ll
@@ -1,4 +1,4 @@
-; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
+; RUN: opt %loadPolly -polly-ast -polly-parallel -analyze < %s | FileCheck %s
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
 target triple = "x86_64-pc-linux-gnu"
 
Index: test/Isl/Ast/OpenMP/single_loop_param_parallel_computeout.ll
===================================================================
--- test/Isl/Ast/OpenMP/single_loop_param_parallel_computeout.ll
+++ test/Isl/Ast/OpenMP/single_loop_param_parallel_computeout.ll
@@ -1,4 +1,4 @@
-; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -polly-dependences-computeout=1 -analyze < %s | FileCheck %s
+; RUN: opt %loadPolly -polly-ast -polly-parallel -polly-dependences-computeout=1 -analyze < %s | FileCheck %s
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
 target triple = "x86_64-pc-linux-gnu"
 
Index: test/Isl/Ast/reduction_clauses_multidimensional_access.ll
===================================================================
--- test/Isl/Ast/reduction_clauses_multidimensional_access.ll
+++ test/Isl/Ast/reduction_clauses_multidimensional_access.ll
@@ -1,6 +1,6 @@
 ; RUN: opt %loadPolly -polly-delinearize -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
-; CHECK: #pragma omp parallel for reduction (^ : sum)
+; CHECK: #pragma known-parallel reduction (^ : sum)
 ;        void f(int N, int M, int P, int sum[P][M]) {
 ;          for (int i = 0; i < N; i++)
 ;            for (int j = 0; j < P; j++)
Index: test/Isl/Ast/reduction_clauses_onedimensional_access.ll
===================================================================
--- test/Isl/Ast/reduction_clauses_onedimensional_access.ll
+++ test/Isl/Ast/reduction_clauses_onedimensional_access.ll
@@ -1,6 +1,6 @@
 ; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
-; CHECK: #pragma omp parallel for reduction (^ : sum)
+; CHECK: #pragma known-parallel reduction (^ : sum)
 ;        void f(int N, int M, int *sum) {
 ;          for (int i = 0; i < N; i++)
 ; CHECK:    #pragma simd
Index: test/Isl/Ast/reduction_different_reduction_clauses.ll
===================================================================
--- test/Isl/Ast/reduction_different_reduction_clauses.ll
+++ test/Isl/Ast/reduction_different_reduction_clauses.ll
@@ -1,7 +1,7 @@
 ; RUN: opt %loadPolly -basicaa -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
 ; CHECK: #pragma simd reduction (+ : sum{{[1,2]}}, sum{{[1,2]}}) reduction (* : prod) reduction (| : or) reduction (& : and)
-; CHECK: #pragma omp parallel for reduction (+ : sum{{[1,2]}}, sum{{[1,2]}}) reduction (* : prod) reduction (| : or) reduction (& : and)
+; CHECK: #pragma known-parallel reduction (+ : sum{{[1,2]}}, sum{{[1,2]}}) reduction (* : prod) reduction (| : or) reduction (& : and)
 ; CHECK: for (int c1 = 0; c1 < N; c1 += 1)
 ; CHECK:   Stmt_for_body(c1);
 ;
Index: test/Isl/Ast/reduction_in_one_dimension.ll
===================================================================
--- test/Isl/Ast/reduction_in_one_dimension.ll
+++ test/Isl/Ast/reduction_in_one_dimension.ll
@@ -2,7 +2,7 @@
 ;
 ; Verify that we won't privatize anything in the outer dimension
 ;
-; CHECK:    #pragma omp parallel for
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c1 = 0; c1 < 2 * n; c1 += 1)
 ; CHECK:      #pragma simd reduction
 ; CHECK:      for (int c3 = 0; c3 <= 1023; c3 += 1)
Index: test/Isl/Ast/reduction_modulo_and_loop_reversal_schedule_2.ll
===================================================================
--- test/Isl/Ast/reduction_modulo_and_loop_reversal_schedule_2.ll
+++ test/Isl/Ast/reduction_modulo_and_loop_reversal_schedule_2.ll
@@ -1,6 +1,6 @@
 ; RUN: opt %loadPolly -polly-import-jscop-dir=%S -polly-import-jscop -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
-; CHECK:    #pragma omp parallel for reduction
+; CHECK:    #pragma known-parallel reduction
 ; CHECK:    for (int c0 = 0; c0 <= 2; c0 += 1) {
 ; CHECK:      if (c0 == 2) {
 ; CHECK:        #pragma simd reduction
Index: test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions.ll
===================================================================
--- test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions.ll
+++ test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions.ll
@@ -1,6 +1,6 @@
 ; RUN: opt %loadPolly -polly-import-jscop-dir=%S -polly-import-jscop -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
-; CHECK:    #pragma omp parallel for
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c0 = 0; c0 <= 1; c0 += 1) {
 ; CHECK:      if (c0 == 1) {
 ; CHECK:        for (int c1 = 1; c1 < 2 * n; c1 += 2)
Index: test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_2.ll
===================================================================
--- test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_2.ll
+++ test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_2.ll
@@ -2,8 +2,8 @@
 ;
 ; Verify that the outer dimension doesnt't carry reduction dependences
 ;
-; CHECK-NOT:#pragma omp parallel for reduction
-; CHECK:    #pragma omp parallel for
+; CHECK-NOT:#pragma known-parallel reduction
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c1 = 0; c1 < 2 * n; c1 += 1) {
 ; CHECK:      if (c1 % 2 == 0) {
 ; CHECK:        #pragma simd reduction
Index: test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_3.ll
===================================================================
--- test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_3.ll
+++ test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_3.ll
@@ -2,8 +2,8 @@
 ;
 ; Verify that the outer dimension doesnt't carry reduction dependences
 ;
-; CHECK-NOT:#pragma omp parallel for reduction
-; CHECK:    #pragma omp parallel for
+; CHECK-NOT:#pragma known-parallel reduction
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c1 = 0; c1 < 2 * n; c1 += 1)
 ; CHECK:      #pragma simd reduction
 ; CHECK:      for (int c3 = 0; c3 <= 1023; c3 += 1) {
Index: test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_4.ll
===================================================================
--- test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_4.ll
+++ test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_4.ll
@@ -2,8 +2,8 @@
 ;
 ; Verify that the outer dimension doesnt't carry reduction dependences
 ;
-; CHECK-NOT:#pragma omp parallel for reduction
-; CHECK:    #pragma omp parallel for
+; CHECK-NOT:#pragma known-parallel reduction
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c1 = 0; c1 < 2 * n; c1 += 1)
 ; CHECK:      #pragma simd reduction
 ; CHECK:      for (int c3 = -1023; c3 <= 1023; c3 += 1) {
Index: test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_5.ll
===================================================================
--- test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_5.ll
+++ test/Isl/Ast/reduction_modulo_schedule_multiple_dimensions_5.ll
@@ -2,7 +2,7 @@
 ;
 ; Verify that only the outer dimension needs privatization
 ;
-; CHECK:    #pragma omp parallel for reduction
+; CHECK:    #pragma known-parallel reduction
 ; CHECK:    for (int c1 = 0; c1 <= 1023; c1 += 1) {
 ; CHECK:      if (c1 % 2 == 0) {
 ; CHECK-NOT:    #pragma simd reduction
Index: test/Isl/Ast/reduction_multiple_dimensions.ll
===================================================================
--- test/Isl/Ast/reduction_multiple_dimensions.ll
+++ test/Isl/Ast/reduction_multiple_dimensions.ll
@@ -1,7 +1,7 @@
 ; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
-; CHECK-NOT:#pragma omp parallel for reduction
-; CHECK:    #pragma omp parallel for
+; CHECK-NOT:#pragma known-parallel reduction
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c1 = 0; c1 <= 2047; c1 += 1)
 ; CHECK:      for (int c3 = 0; c3 <= 1023; c3 += 1)
 ; CHECK:        #pragma simd reduction
Index: test/Isl/Ast/reduction_multiple_dimensions_2.ll
===================================================================
--- test/Isl/Ast/reduction_multiple_dimensions_2.ll
+++ test/Isl/Ast/reduction_multiple_dimensions_2.ll
@@ -1,7 +1,7 @@
 ; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
-; CHECK-NOT:#pragma omp parallel for reduction
-; CHECK:    #pragma omp parallel for
+; CHECK-NOT:#pragma known-parallel reduction
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c1 = 0; c1 <= 2047; c1 += 1)
 ; CHECK:      for (int c3 = 0; c3 <= 1023; c3 += 1)
 ; CHECK:        #pragma simd reduction
Index: test/Isl/Ast/reduction_multiple_dimensions_3.ll
===================================================================
--- test/Isl/Ast/reduction_multiple_dimensions_3.ll
+++ test/Isl/Ast/reduction_multiple_dimensions_3.ll
@@ -1,7 +1,7 @@
 ; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
-; CHECK-NOT:#pragma omp parallel for reduction
-; CHECK:    #pragma omp parallel for
+; CHECK-NOT:#pragma known-parallel reduction
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c1 = 0; c1 <= 2047; c1 += 1)
 ; CHECK:      for (int c3 = 0; c3 <= 1023; c3 += 1)
 ; CHECK:        #pragma simd reduction
Index: test/Isl/Ast/reduction_multiple_dimensions_4.ll
===================================================================
--- test/Isl/Ast/reduction_multiple_dimensions_4.ll
+++ test/Isl/Ast/reduction_multiple_dimensions_4.ll
@@ -1,7 +1,7 @@
 ; RUN: opt %loadPolly -polly-ast -polly-ast-detect-parallel -analyze < %s | FileCheck %s
 ;
-; CHECK-NOT:#pragma omp parallel for reduction
-; CHECK:    #pragma omp parallel for
+; CHECK-NOT:#pragma known-parallel reduction
+; CHECK:    #pragma known-parallel
 ; CHECK:    for (int c1 = 0; c1 <= 2047; c1 += 1)
 ; CHECK:      for (int c3 = 0; c3 <= 1023; c3 += 1)
 ; CHECK:        #pragma simd reduction
Index: test/Isl/Ast/single_loop_strip_mine.ll
===================================================================
--- test/Isl/Ast/single_loop_strip_mine.ll
+++ test/Isl/Ast/single_loop_strip_mine.ll
@@ -32,7 +32,7 @@
 ; CHECK: for (int c1 = 0; c1 <= 1023; c1 += 1)
 ; CHECK:     Stmt_for_body(c1);
 
-; CHECK-VECTOR: #pragma omp parallel for
+; CHECK-VECTOR: #pragma known-parallel
 ; CHECK-VECTOR: for (int c0 = 0; c0 <= 1023; c0 += 4)
 ; CHECK-VECTOR:     #pragma simd
 ; CHECK-VECTOR:     for (int c1 = c0; c1 <= c0 + 3; c1 += 1)
Index: test/ScheduleOptimizer/prevectorization.ll
===================================================================
--- test/ScheduleOptimizer/prevectorization.ll
+++ test/ScheduleOptimizer/prevectorization.ll
@@ -54,7 +54,7 @@
 
 attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
 
-; CHECK: #pragma omp parallel for
+; CHECK: #pragma known-parallel
 ; CHECK: for (int c1 = 0; c1 <= 1535; c1 += 32)
 ; CHECK:   for (int c2 = 0; c2 <= 1535; c2 += 32)
 ; CHECK:     for (int c3 = c1; c3 <= c1 + 31; c3 += 1)
@@ -62,7 +62,7 @@
 ; CHECK:         #pragma simd
 ; CHECK:         for (int c5 = c4; c5 <= c4 + 3; c5 += 1)
 ; CHECK:           Stmt_for_body3(c3, c5);
-; CHECK: #pragma omp parallel for
+; CHECK: #pragma known-parallel
 ; CHECK: for (int c1 = 0; c1 <= 1535; c1 += 32)
 ; CHECK:   for (int c2 = 0; c2 <= 1535; c2 += 32)
 ; CHECK:     for (int c3 = 0; c3 <= 1535; c3 += 32)