[RISCV][VP] Introduce vp saturating addition/subtraction and RISC-V support.

llvm/docs/LangRef.rst

@@ -16718,6 +16718,7 @@ an operation is greater than the maximum value, the result is set (or
 "clamped") to this maximum. If it is below the minimum, it is clamped to this
 minimum.
 
+.. _int_sadd_sat:
 
 '``llvm.sadd.sat.*``' Intrinsics
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
@@ -16767,6 +16768,8 @@ Examples
       %res = call i4 @llvm.sadd.sat.i4(i4 -4, i4 -5)  ; %res = -8
 
 
+.. _int_uadd_sat:
+
 '``llvm.uadd.sat.*``' Intrinsics
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
@@ -16814,6 +16817,8 @@ Examples
       %res = call i4 @llvm.uadd.sat.i4(i4 8, i4 8)  ; %res = 15
 
 
+.. _int_ssub_sat:
+
 '``llvm.ssub.sat.*``' Intrinsics
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
@@ -16862,6 +16867,8 @@ Examples
       %res = call i4 @llvm.ssub.sat.i4(i4 4, i4 -5)  ; %res = 7
 
 
+.. _int_usub_sat:
+
 '``llvm.usub.sat.*``' Intrinsics
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
@@ -23579,6 +23586,202 @@ Examples:
       %also.r = select <4 x i1> %mask, <4 x i32> %t, <4 x i32> poison
 
 
+.. _int_vp_sadd_sat:
+
+'``llvm.vp.sadd.sat.*``' Intrinsics
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+      declare <16 x i32>  @llvm.vp.sadd.sat.v16i32 (<16 x i32> <left_op> <16 x i32> <right_op>, <16 x i1> <mask>, i32 <vector_length>)
+      declare <vscale x 4 x i32>  @llvm.vp.sadd.sat.nxv4i32 (<vscale x 4 x i32> <left_op>, <vscale x 4 x i32> <right_op>, <vscale x 4 x i1> <mask>, i32 <vector_length>)
+      declare <256 x i64>  @llvm.vp.sadd.sat.v256i64 (<256 x i64> <left_op>, <256 x i64> <right_op>, <256 x i1> <mask>, i32 <vector_length>)
+
+Overview:
+"""""""""
+
+Predicated signed saturating addition of two vectors of integers.
+
+
+Arguments:
+""""""""""
+
+The first two operands and the result have the same vector of integer type. The
+third operand is the vector mask and has the same number of elements as the
+result vector type. The fourth operand is the explicit vector length of the
+operation.
+
+Semantics:
+""""""""""
+
+The '``llvm.vp.sadd.sat``' intrinsic performs sadd.sat (:ref:`sadd.sat <int_sadd_sat>`)
+of the first and second vector operands on each enabled lane. The result on
+disabled lanes is a :ref:`poison value <poisonvalues>`.
+
+
+Examples:
+"""""""""
+
+.. code-block:: llvm
+
+      %r = call <4 x i32> @llvm.vp.sadd.sat.v4i32(<4 x i32> %a, <4 x i32> %b, <4 x i1> %mask, i32 %evl)
+      ;; For all lanes below %evl, %r is lane-wise equivalent to %also.r
+
+      %t = call <4 x i32> @llvm.sadd.sat.v4i32(<4 x i32> %a, <4 x i32> %b)
+      %also.r = select <4 x i1> %mask, <4 x i32> %t, <4 x i32> poison
+
+
+.. _int_vp_uadd_sat:
+
+'``llvm.vp.uadd.sat.*``' Intrinsics
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+      declare <16 x i32>  @llvm.vp.uadd.sat.v16i32 (<16 x i32> <left_op> <16 x i32> <right_op>, <16 x i1> <mask>, i32 <vector_length>)
+      declare <vscale x 4 x i32>  @llvm.vp.uadd.sat.nxv4i32 (<vscale x 4 x i32> <left_op>, <vscale x 4 x i32> <right_op>, <vscale x 4 x i1> <mask>, i32 <vector_length>)
+      declare <256 x i64>  @llvm.vp.uadd.sat.v256i64 (<256 x i64> <left_op>, <256 x i64> <right_op>, <256 x i1> <mask>, i32 <vector_length>)
+
+Overview:
+"""""""""
+
+Predicated unsigned saturating addition of two vectors of integers.
+
+
+Arguments:
+""""""""""
+
+The first two operands and the result have the same vector of integer type. The
+third operand is the vector mask and has the same number of elements as the
+result vector type. The fourth operand is the explicit vector length of the
+operation.
+
+Semantics:
+""""""""""
+
+The '``llvm.vp.uadd.sat``' intrinsic performs uadd.sat (:ref:`uadd.sat <int_uadd_sat>`)
+of the first and second vector operands on each enabled lane. The result on
+disabled lanes is a :ref:`poison value <poisonvalues>`.
+
+
+Examples:
+"""""""""
+
+.. code-block:: llvm
+
+      %r = call <4 x i32> @llvm.vp.uadd.sat.v4i32(<4 x i32> %a, <4 x i32> %b, <4 x i1> %mask, i32 %evl)
+      ;; For all lanes below %evl, %r is lane-wise equivalent to %also.r
+
+      %t = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> %a, <4 x i32> %b)
+      %also.r = select <4 x i1> %mask, <4 x i32> %t, <4 x i32> poison
+
+
+.. _int_vp_ssub_sat:
+
+'``llvm.vp.ssub.sat.*``' Intrinsics
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+      declare <16 x i32>  @llvm.vp.ssub.sat.v16i32 (<16 x i32> <left_op> <16 x i32> <right_op>, <16 x i1> <mask>, i32 <vector_length>)
+      declare <vscale x 4 x i32>  @llvm.vp.ssub.sat.nxv4i32 (<vscale x 4 x i32> <left_op>, <vscale x 4 x i32> <right_op>, <vscale x 4 x i1> <mask>, i32 <vector_length>)
+      declare <256 x i64>  @llvm.vp.ssub.sat.v256i64 (<256 x i64> <left_op>, <256 x i64> <right_op>, <256 x i1> <mask>, i32 <vector_length>)
+
+Overview:
+"""""""""
+
+Predicated signed saturating subtraction of two vectors of integers.
+
+
+Arguments:
+""""""""""
+
+The first two operands and the result have the same vector of integer type. The
+third operand is the vector mask and has the same number of elements as the
+result vector type. The fourth operand is the explicit vector length of the
+operation.
+
+Semantics:
+""""""""""
+
+The '``llvm.vp.ssub.sat``' intrinsic performs ssub.sat (:ref:`ssub.sat <int_ssub_sat>`)
+of the first and second vector operands on each enabled lane. The result on
+disabled lanes is a :ref:`poison value <poisonvalues>`.
+
+
+Examples:
+"""""""""
+
+.. code-block:: llvm
+
+      %r = call <4 x i32> @llvm.vp.ssub.sat.v4i32(<4 x i32> %a, <4 x i32> %b, <4 x i1> %mask, i32 %evl)
+      ;; For all lanes below %evl, %r is lane-wise equivalent to %also.r
+
+      %t = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> %a, <4 x i32> %b)
+      %also.r = select <4 x i1> %mask, <4 x i32> %t, <4 x i32> poison
+
+
+.. _int_vp_usub_sat:
+
+'``llvm.vp.usub.sat.*``' Intrinsics
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+      declare <16 x i32>  @llvm.vp.usub.sat.v16i32 (<16 x i32> <left_op> <16 x i32> <right_op>, <16 x i1> <mask>, i32 <vector_length>)
+      declare <vscale x 4 x i32>  @llvm.vp.usub.sat.nxv4i32 (<vscale x 4 x i32> <left_op>, <vscale x 4 x i32> <right_op>, <vscale x 4 x i1> <mask>, i32 <vector_length>)
+      declare <256 x i64>  @llvm.vp.usub.sat.v256i64 (<256 x i64> <left_op>, <256 x i64> <right_op>, <256 x i1> <mask>, i32 <vector_length>)
+
+Overview:
+"""""""""
+
+Predicated unsigned saturating subtraction of two vectors of integers.
+
+
+Arguments:
+""""""""""
+
+The first two operands and the result have the same vector of integer type. The
+third operand is the vector mask and has the same number of elements as the
+result vector type. The fourth operand is the explicit vector length of the
+operation.
+
+Semantics:
+""""""""""
+
+The '``llvm.vp.usub.sat``' intrinsic performs usub.sat (:ref:`usub.sat <int_usub_sat>`)
+of the first and second vector operands on each enabled lane. The result on
+disabled lanes is a :ref:`poison value <poisonvalues>`.
+
+
+Examples:
+"""""""""
+
+.. code-block:: llvm
+
+      %r = call <4 x i32> @llvm.vp.usub.sat.v4i32(<4 x i32> %a, <4 x i32> %b, <4 x i1> %mask, i32 %evl)
+      ;; For all lanes below %evl, %r is lane-wise equivalent to %also.r
+
+      %t = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> %a, <4 x i32> %b)
+      %also.r = select <4 x i1> %mask, <4 x i32> %t, <4 x i32> poison
+
+
 .. _int_vp_fshl:
 
 '``llvm.vp.fshl.*``' Intrinsics

llvm/include/llvm/IR/Intrinsics.td

@@ -1933,6 +1933,26 @@ let IntrProperties = [IntrNoMem, IntrNoSync, IntrWillReturn] in {
                                LLVMMatchType<0>,
                                LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                llvm_i32_ty]>;
+  def int_vp_sadd_sat : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
+                             [ LLVMMatchType<0>,
+                               LLVMMatchType<0>,
+                               LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                               llvm_i32_ty]>;
+  def int_vp_uadd_sat : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
+                             [ LLVMMatchType<0>,
+                               LLVMMatchType<0>,
+                               LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                               llvm_i32_ty]>;
+  def int_vp_ssub_sat : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
+                             [ LLVMMatchType<0>,
+                               LLVMMatchType<0>,
+                               LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                               llvm_i32_ty]>;
+  def int_vp_usub_sat : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
+                             [ LLVMMatchType<0>,
+                               LLVMMatchType<0>,
+                               LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                               llvm_i32_ty]>;
 
   // Floating-point arithmetic
   def int_vp_fadd : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],

llvm/include/llvm/IR/VPIntrinsics.def

@@ -293,6 +293,30 @@ BEGIN_REGISTER_VP(vp_fshr, 3, 4, VP_FSHR, -1)
 VP_PROPERTY_FUNCTIONAL_INTRINSIC(fshr)
 VP_PROPERTY_FUNCTIONAL_SDOPC(FSHR)
 END_REGISTER_VP(vp_fshr, VP_FSHR)
+
+// llvm.vp.sadd.sat(x,y,mask,vlen)
+BEGIN_REGISTER_VP(vp_sadd_sat, 2, 3, VP_SADDSAT, -1)
+VP_PROPERTY_FUNCTIONAL_INTRINSIC(sadd_sat)
+VP_PROPERTY_FUNCTIONAL_SDOPC(SADDSAT)
+END_REGISTER_VP(vp_sadd_sat, VP_SADDSAT)
+
+// llvm.vp.uadd.sat(x,y,mask,vlen)
+BEGIN_REGISTER_VP(vp_uadd_sat, 2, 3, VP_UADDSAT, -1)
+VP_PROPERTY_FUNCTIONAL_INTRINSIC(uadd_sat)
+VP_PROPERTY_FUNCTIONAL_SDOPC(UADDSAT)
+END_REGISTER_VP(vp_uadd_sat, VP_UADDSAT)
+
+// llvm.vp.ssub.sat(x,y,mask,vlen)
+BEGIN_REGISTER_VP(vp_ssub_sat, 2, 3, VP_SSUBSAT, -1)
+VP_PROPERTY_FUNCTIONAL_INTRINSIC(ssub_sat)
+VP_PROPERTY_FUNCTIONAL_SDOPC(SSUBSAT)
+END_REGISTER_VP(vp_ssub_sat, VP_SSUBSAT)
+
+// llvm.vp.usub.sat(x,y,mask,vlen)
+BEGIN_REGISTER_VP(vp_usub_sat, 2, 3, VP_USUBSAT, -1)
+VP_PROPERTY_FUNCTIONAL_INTRINSIC(usub_sat)
+VP_PROPERTY_FUNCTIONAL_SDOPC(USUBSAT)
+END_REGISTER_VP(vp_usub_sat, VP_USUBSAT)
 ///// } Integer Arithmetic
 
 ///// Floating-Point Arithmetic {