cmd/compile: add SSA config options useAvg and useHmul

This commit allows architectures to disable optimizations that need the
Avg* and Hmul* operations.

WebAssembly has no such operations, so using them as an optimization
but then having to emulate them with multiple instructions makes no
sense, especially since the WebAssembly compiler may do the same
optimizations internally.

Updates #18892

Change-Id: If57b59e3235482a9e0ec334a7312b3e3b5fc2b61
Reviewed-on: https://go-review.googlesource.com/103256
Run-TryBot: Brad Fitzpatrick <[email protected]>
TryBot-Result: Gobot Gobot <[email protected]>
Reviewed-by: Josh Bleecher Snyder <[email protected]>

Author: neelance

src/cmd/compile/internal/ssa/config.go

@@ -34,6 +34,8 @@ type Config struct {
 	optimize        bool          // Do optimization
 	noDuffDevice    bool          // Don't use Duff's device
 	useSSE          bool          // Use SSE for non-float operations
+	useAvg          bool          // Use optimizations that need Avg* operations
+	useHmul         bool          // Use optimizations that need Hmul* operations
 	nacl            bool          // GOOS=nacl
 	use387          bool          // GO386=387
 	SoftFloat       bool          //
@@ -190,6 +192,8 @@ const (
 // NewConfig returns a new configuration object for the given architecture.
 func NewConfig(arch string, types Types, ctxt *obj.Link, optimize bool) *Config {
 	c := &Config{arch: arch, Types: types}
+	c.useAvg = true
+	c.useHmul = true
 	switch arch {
 	case "amd64":
 		c.PtrSize = 8

src/cmd/compile/internal/ssa/gen/generic.rules

@@ -845,7 +845,7 @@
         (Const32 <typ.UInt32> [int64(1<<15+(umagic(16,c).m+1)/2)])
         (Rsh32Ux64 <typ.UInt32> (ZeroExt16to32 x) (Const64 <typ.UInt64> [1])))
       (Const64 <typ.UInt64> [16+umagic(16,c).s-2])))
-(Div16u x (Const16 [c])) && umagicOK(16, c) && config.RegSize == 4 ->
+(Div16u x (Const16 [c])) && umagicOK(16, c) && config.RegSize == 4 && config.useAvg ->
   (Trunc32to16
     (Rsh32Ux64 <typ.UInt32>
       (Avg32u
@@ -856,19 +856,19 @@
       (Const64 <typ.UInt64> [16+umagic(16,c).s-1])))
 
 // For 32-bit divides on 32-bit machines
-(Div32u x (Const32 [c])) && umagicOK(32, c) && config.RegSize == 4 && umagic(32,c).m&1 == 0 ->
+(Div32u x (Const32 [c])) && umagicOK(32, c) && config.RegSize == 4 && umagic(32,c).m&1 == 0 && config.useHmul ->
   (Rsh32Ux64 <typ.UInt32>
     (Hmul32u <typ.UInt32>
       (Const32 <typ.UInt32> [int64(int32(1<<31+umagic(32,c).m/2))])
       x)
     (Const64 <typ.UInt64> [umagic(32,c).s-1]))
-(Div32u x (Const32 [c])) && umagicOK(32, c) && config.RegSize == 4 && c&1 == 0 ->
+(Div32u x (Const32 [c])) && umagicOK(32, c) && config.RegSize == 4 && c&1 == 0 && config.useHmul ->
   (Rsh32Ux64 <typ.UInt32>
     (Hmul32u <typ.UInt32>
       (Const32 <typ.UInt32> [int64(int32(1<<31+(umagic(32,c).m+1)/2))])
       (Rsh32Ux64 <typ.UInt32> x (Const64 <typ.UInt64> [1])))
     (Const64 <typ.UInt64> [umagic(32,c).s-2]))
-(Div32u x (Const32 [c])) && umagicOK(32, c) && config.RegSize == 4 ->
+(Div32u x (Const32 [c])) && umagicOK(32, c) && config.RegSize == 4 && config.useAvg && config.useHmul ->
   (Rsh32Ux64 <typ.UInt32>
     (Avg32u
       x
@@ -893,7 +893,7 @@
         (Const64 <typ.UInt64> [int64(1<<31+(umagic(32,c).m+1)/2)])
         (Rsh64Ux64 <typ.UInt64> (ZeroExt32to64 x) (Const64 <typ.UInt64> [1])))
       (Const64 <typ.UInt64> [32+umagic(32,c).s-2])))
-(Div32u x (Const32 [c])) && umagicOK(32, c) && config.RegSize == 8 ->
+(Div32u x (Const32 [c])) && umagicOK(32, c) && config.RegSize == 8 && config.useAvg ->
   (Trunc64to32
     (Rsh64Ux64 <typ.UInt64>
       (Avg64u
@@ -905,19 +905,19 @@
 
 // For 64-bit divides on 64-bit machines
 // (64-bit divides on 32-bit machines are lowered to a runtime call by the walk pass.)
-(Div64u x (Const64 [c])) && umagicOK(64, c) && config.RegSize == 8 && umagic(64,c).m&1 == 0 ->
+(Div64u x (Const64 [c])) && umagicOK(64, c) && config.RegSize == 8 && umagic(64,c).m&1 == 0 && config.useHmul ->
   (Rsh64Ux64 <typ.UInt64>
     (Hmul64u <typ.UInt64>
       (Const64 <typ.UInt64> [int64(1<<63+umagic(64,c).m/2)])
       x)
     (Const64 <typ.UInt64> [umagic(64,c).s-1]))
-(Div64u x (Const64 [c])) && umagicOK(64, c) && config.RegSize == 8 && c&1 == 0 ->
+(Div64u x (Const64 [c])) && umagicOK(64, c) && config.RegSize == 8 && c&1 == 0 && config.useHmul ->
   (Rsh64Ux64 <typ.UInt64>
     (Hmul64u <typ.UInt64>
       (Const64 <typ.UInt64> [int64(1<<63+(umagic(64,c).m+1)/2)])
       (Rsh64Ux64 <typ.UInt64> x (Const64 <typ.UInt64> [1])))
     (Const64 <typ.UInt64> [umagic(64,c).s-2]))
-(Div64u x (Const64 [c])) && umagicOK(64, c) && config.RegSize == 8 ->
+(Div64u x (Const64 [c])) && umagicOK(64, c) && config.RegSize == 8 && config.useAvg && config.useHmul ->
   (Rsh64Ux64 <typ.UInt64>
     (Avg64u
       x
@@ -992,7 +992,7 @@
     (Rsh64x64 <t>
       (SignExt32to64 x)
       (Const64 <typ.UInt64> [63])))
-(Div32 <t> x (Const32 [c])) && smagicOK(32,c) && config.RegSize == 4 && smagic(32,c).m&1 == 0 ->
+(Div32 <t> x (Const32 [c])) && smagicOK(32,c) && config.RegSize == 4 && smagic(32,c).m&1 == 0 && config.useHmul ->
   (Sub32 <t>
     (Rsh32x64 <t>
       (Hmul32 <t>
@@ -1002,7 +1002,7 @@
     (Rsh32x64 <t>
       x
       (Const64 <typ.UInt64> [31])))
-(Div32 <t> x (Const32 [c])) && smagicOK(32,c) && config.RegSize == 4 && smagic(32,c).m&1 != 0 ->
+(Div32 <t> x (Const32 [c])) && smagicOK(32,c) && config.RegSize == 4 && smagic(32,c).m&1 != 0 && config.useHmul ->
   (Sub32 <t>
     (Rsh32x64 <t>
       (Add32 <t>
@@ -1014,7 +1014,7 @@
     (Rsh32x64 <t>
       x
       (Const64 <typ.UInt64> [31])))
-(Div64 <t> x (Const64 [c])) && smagicOK(64,c) && smagic(64,c).m&1 == 0 ->
+(Div64 <t> x (Const64 [c])) && smagicOK(64,c) && smagic(64,c).m&1 == 0 && config.useHmul ->
   (Sub64 <t>
     (Rsh64x64 <t>
       (Hmul64 <t>
@@ -1024,7 +1024,7 @@
     (Rsh64x64 <t>
       x
       (Const64 <typ.UInt64> [63])))
-(Div64 <t> x (Const64 [c])) && smagicOK(64,c) && smagic(64,c).m&1 != 0 ->
+(Div64 <t> x (Const64 [c])) && smagicOK(64,c) && smagic(64,c).m&1 != 0 && config.useHmul ->
   (Sub64 <t>
     (Rsh64x64 <t>
       (Add64 <t>