[dev.simd] cmd/compile, simd: update DotProd to DotProduct

API naming changes.

This CL also remove AddDotProductPairsSaturated.

Change-Id: I02e6d45268704f3ed4eaf62f0ecb7dc936b42124
Reviewed-on: https://go-review.googlesource.com/c/go/+/710935
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: David Chase <drchase@google.com>

3 files changed, 41 insertions, 70 deletions

diff --git a/src/simd/_gen/simdgen/ops/MLOps/categories.yaml b/src/simd/_gen/simdgen/ops/MLOps/categories.yaml
index 772a7b3cf6..0317b42c6a 100644
--- a/src/simd/_gen/simdgen/ops/MLOps/categories.yaml
+++ b/src/simd/_gen/simdgen/ops/MLOps/categories.yaml
@@ -1,38 +1,34 @@
 !sum
-- go: DotProdPairs
+- go: DotProductPairs
   commutative: false
   documentation: !string |-
     // NAME multiplies the elements and add the pairs together,
     // yielding a vector of half as many elements with twice the input element size.
 # TODO: maybe simplify this name within the receiver-type + method-naming scheme we use.
-- go: DotProdPairsSaturated
+- go: DotProductPairsSaturated
   commutative: false
   documentation: !string |-
     // NAME multiplies the elements and add the pairs together with saturation,
     // yielding a vector of half as many elements with twice the input element size.
-# QuadDotProd, i.e. VPDPBUSD(S) are operations with src/dst on the same register, we are not supporting this as of now.
-# - go: DotProdBroadcast
+# QuadDotProduct, i.e. VPDPBUSD(S) are operations with src/dst on the same register, we are not supporting this as of now.
+# - go: DotProductBroadcast
 #   commutative: true
 # #   documentation: !string |-
 #     // NAME multiplies all elements and broadcasts the sum.
-- go: AddDotProdQuadruple
+- go: AddDotProductQuadruple
   commutative: false
   documentation: !string |-
     // NAME performs dot products on groups of 4 elements of x and y and then adds z.
-- go: AddDotProdQuadrupleSaturated
+- go: AddDotProductQuadrupleSaturated
   commutative: false
   documentation: !string |-
     // NAME multiplies performs dot products on groups of 4 elements of x and y and then adds z.
-- go: AddDotProdPairs
+- go: AddDotProductPairs
   commutative: false
   noTypes: "true"
   noGenericOps: "true"
   documentation: !string |-
     // NAME performs dot products on pairs of elements of y and z and then adds x.
-- go: AddDotProdPairsSaturated
-  commutative: false
-  documentation: !string |-
-    // NAME performs dot products on pairs of elements of y and z and then adds x.
 - go: MulAdd
   commutative: false
   documentation: !string |-
diff --git a/src/simd/_gen/simdgen/ops/MLOps/go.yaml b/src/simd/_gen/simdgen/ops/MLOps/go.yaml
index 5c2009dcf8..162c47ea0e 100644
--- a/src/simd/_gen/simdgen/ops/MLOps/go.yaml
+++ b/src/simd/_gen/simdgen/ops/MLOps/go.yaml
@@ -1,5 +1,5 @@
 !sum
-- go: DotProdPairs
+- go: DotProductPairs
   asm: VPMADDWD
   in:
   - &int
@@ -10,7 +10,7 @@
   - &int2 # The elemBits are different
     go: $t2
     base: int
-- go: DotProdPairsSaturated
+- go: DotProductPairsSaturated
   asm: VPMADDUBSW
   in:
   - &uint
@@ -23,7 +23,7 @@
     overwriteElementBits: 8
   out:
   - *int2
-# - go: DotProdBroadcast
+# - go: DotProductBroadcast
 #   asm: VDPP[SD]
 #   in:
 #   - &dpb_src
@@ -33,7 +33,7 @@
 #     const: 127
 #   out:
 #   - *dpb_src
-- go: AddDotProdQuadruple
+- go: AddDotProductQuadruple
   asm: "VPDPBUSD"
   operandOrder: "31" # switch operand 3 and 1
   in:
@@ -51,7 +51,7 @@
     overwriteElementBits: 8
   out:
   - *qdpa_acc
-- go: AddDotProdQuadrupleSaturated
+- go: AddDotProductQuadrupleSaturated
   asm: "VPDPBUSDS"
   operandOrder: "31" # switch operand 3 and 1
   in:
@@ -60,7 +60,7 @@
   - *qdpa_src2
   out:
   - *qdpa_acc
-- go: AddDotProdPairs
+- go: AddDotProductPairs
   asm: "VPDPWSSD"
   in:
   - &pdpa_acc
@@ -77,14 +77,6 @@
     overwriteElementBits: 16
   out:
   - *pdpa_acc
-- go: AddDotProdPairsSaturated
-  asm: "VPDPWSSDS"
-  in:
-  - *pdpa_acc
-  - *pdpa_src1
-  - *pdpa_src2
-  out:
-  - *pdpa_acc
 - go: MulAdd
   asm: "VFMADD213PS|VFMADD213PD"
   in:
diff --git a/src/simd/ops_amd64.go b/src/simd/ops_amd64.go
index 8956c2e077..2331622361 100644
--- a/src/simd/ops_amd64.go
+++ b/src/simd/ops_amd64.go
@@ -314,56 +314,39 @@ func (x Uint64x4) Add(y Uint64x4) Uint64x4
 // Asm: VPADDQ, CPU Feature: AVX512
 func (x Uint64x8) Add(y Uint64x8) Uint64x8
 
-/* AddDotProdPairsSaturated */
+/* AddDotProductQuadruple */
 
-// AddDotProdPairsSaturated performs dot products on pairs of elements of y and z and then adds x.
-//
-// Asm: VPDPWSSDS, CPU Feature: AVXVNNI
-func (x Int32x4) AddDotProdPairsSaturated(y Int16x8, z Int16x8) Int32x4
-
-// AddDotProdPairsSaturated performs dot products on pairs of elements of y and z and then adds x.
-//
-// Asm: VPDPWSSDS, CPU Feature: AVXVNNI
-func (x Int32x8) AddDotProdPairsSaturated(y Int16x16, z Int16x16) Int32x8
-
-// AddDotProdPairsSaturated performs dot products on pairs of elements of y and z and then adds x.
-//
-// Asm: VPDPWSSDS, CPU Feature: AVX512VNNI
-func (x Int32x16) AddDotProdPairsSaturated(y Int16x32, z Int16x32) Int32x16
-
-/* AddDotProdQuadruple */
-
-// AddDotProdQuadruple performs dot products on groups of 4 elements of x and y and then adds z.
+// AddDotProductQuadruple performs dot products on groups of 4 elements of x and y and then adds z.
 //
 // Asm: VPDPBUSD, CPU Feature: AVXVNNI
-func (x Int8x16) AddDotProdQuadruple(y Uint8x16, z Int32x4) Int32x4
+func (x Int8x16) AddDotProductQuadruple(y Uint8x16, z Int32x4) Int32x4
 
-// AddDotProdQuadruple performs dot products on groups of 4 elements of x and y and then adds z.
+// AddDotProductQuadruple performs dot products on groups of 4 elements of x and y and then adds z.
 //
 // Asm: VPDPBUSD, CPU Feature: AVXVNNI
-func (x Int8x32) AddDotProdQuadruple(y Uint8x32, z Int32x8) Int32x8
+func (x Int8x32) AddDotProductQuadruple(y Uint8x32, z Int32x8) Int32x8
 
-// AddDotProdQuadruple performs dot products on groups of 4 elements of x and y and then adds z.
+// AddDotProductQuadruple performs dot products on groups of 4 elements of x and y and then adds z.
 //
 // Asm: VPDPBUSD, CPU Feature: AVX512VNNI
-func (x Int8x64) AddDotProdQuadruple(y Uint8x64, z Int32x16) Int32x16
+func (x Int8x64) AddDotProductQuadruple(y Uint8x64, z Int32x16) Int32x16
 
-/* AddDotProdQuadrupleSaturated */
+/* AddDotProductQuadrupleSaturated */
 
-// AddDotProdQuadrupleSaturated multiplies performs dot products on groups of 4 elements of x and y and then adds z.
+// AddDotProductQuadrupleSaturated multiplies performs dot products on groups of 4 elements of x and y and then adds z.
 //
 // Asm: VPDPBUSDS, CPU Feature: AVXVNNI
-func (x Int8x16) AddDotProdQuadrupleSaturated(y Uint8x16, z Int32x4) Int32x4
+func (x Int8x16) AddDotProductQuadrupleSaturated(y Uint8x16, z Int32x4) Int32x4
 
-// AddDotProdQuadrupleSaturated multiplies performs dot products on groups of 4 elements of x and y and then adds z.
+// AddDotProductQuadrupleSaturated multiplies performs dot products on groups of 4 elements of x and y and then adds z.
 //
 // Asm: VPDPBUSDS, CPU Feature: AVXVNNI
-func (x Int8x32) AddDotProdQuadrupleSaturated(y Uint8x32, z Int32x8) Int32x8
+func (x Int8x32) AddDotProductQuadrupleSaturated(y Uint8x32, z Int32x8) Int32x8
 
-// AddDotProdQuadrupleSaturated multiplies performs dot products on groups of 4 elements of x and y and then adds z.
+// AddDotProductQuadrupleSaturated multiplies performs dot products on groups of 4 elements of x and y and then adds z.
 //
 // Asm: VPDPBUSDS, CPU Feature: AVX512VNNI
-func (x Int8x64) AddDotProdQuadrupleSaturated(y Uint8x64, z Int32x16) Int32x16
+func (x Int8x64) AddDotProductQuadrupleSaturated(y Uint8x64, z Int32x16) Int32x16
 
 /* AddPairs */
 
@@ -2143,45 +2126,45 @@ func (x Float64x4) Div(y Float64x4) Float64x4
 // Asm: VDIVPD, CPU Feature: AVX512
 func (x Float64x8) Div(y Float64x8) Float64x8
 
-/* DotProdPairs */
+/* DotProductPairs */
 
-// DotProdPairs multiplies the elements and add the pairs together,
+// DotProductPairs multiplies the elements and add the pairs together,
 // yielding a vector of half as many elements with twice the input element size.
 //
 // Asm: VPMADDWD, CPU Feature: AVX
-func (x Int16x8) DotProdPairs(y Int16x8) Int32x4
+func (x Int16x8) DotProductPairs(y Int16x8) Int32x4
 
-// DotProdPairs multiplies the elements and add the pairs together,
+// DotProductPairs multiplies the elements and add the pairs together,
 // yielding a vector of half as many elements with twice the input element size.
 //
 // Asm: VPMADDWD, CPU Feature: AVX2
-func (x Int16x16) DotProdPairs(y Int16x16) Int32x8
+func (x Int16x16) DotProductPairs(y Int16x16) Int32x8
 
-// DotProdPairs multiplies the elements and add the pairs together,
+// DotProductPairs multiplies the elements and add the pairs together,
 // yielding a vector of half as many elements with twice the input element size.
 //
 // Asm: VPMADDWD, CPU Feature: AVX512
-func (x Int16x32) DotProdPairs(y Int16x32) Int32x16
+func (x Int16x32) DotProductPairs(y Int16x32) Int32x16
 
-/* DotProdPairsSaturated */
+/* DotProductPairsSaturated */
 
-// DotProdPairsSaturated multiplies the elements and add the pairs together with saturation,
+// DotProductPairsSaturated multiplies the elements and add the pairs together with saturation,
 // yielding a vector of half as many elements with twice the input element size.
 //
 // Asm: VPMADDUBSW, CPU Feature: AVX
-func (x Uint8x16) DotProdPairsSaturated(y Int8x16) Int16x8
+func (x Uint8x16) DotProductPairsSaturated(y Int8x16) Int16x8
 
-// DotProdPairsSaturated multiplies the elements and add the pairs together with saturation,
+// DotProductPairsSaturated multiplies the elements and add the pairs together with saturation,
 // yielding a vector of half as many elements with twice the input element size.
 //
 // Asm: VPMADDUBSW, CPU Feature: AVX2
-func (x Uint8x32) DotProdPairsSaturated(y Int8x32) Int16x16
+func (x Uint8x32) DotProductPairsSaturated(y Int8x32) Int16x16
 
-// DotProdPairsSaturated multiplies the elements and add the pairs together with saturation,
+// DotProductPairsSaturated multiplies the elements and add the pairs together with saturation,
 // yielding a vector of half as many elements with twice the input element size.
 //
 // Asm: VPMADDUBSW, CPU Feature: AVX512
-func (x Uint8x64) DotProdPairsSaturated(y Int8x64) Int16x32
+func (x Uint8x64) DotProductPairsSaturated(y Int8x64) Int16x32
 
 /* Equal */