(MOVHZreg y:(MOV(H|B)Zreg _)) => y // repeat
(MOVHZreg y:(MOVHBRload _ _)) => y

(MOVHreg y:(MOV(H|B)reg _)) => y // repeat

(MOV(H|HZ)reg y:(MOV(HZ|H)reg x)) => (MOV(H|HZ)reg x)

// W - there are more combinations than these

(MOV(WZ|WZ|WZ|W|W|W)reg y:(MOV(WZ|HZ|BZ|W|H|B)reg _)) => y // repeat
(MOVWZreg y:(MOV(H|W)BRload _ _)) => y

(MOV(W|WZ)reg y:(MOV(WZ|W)reg x)) => (MOV(W|WZ)reg x)

(S(LD|RD|RAD|LW|RW|RAW) x (ANDWconst [c] y)) && c&63 == 63
  => (S(LD|RD|RAD|LW|RW|RAW) x y)
(SLD  x (MOV(W|H|B|WZ|HZ|BZ)reg y)) => (SLD  x y)
(SRD  x (MOV(W|H|B|WZ|HZ|BZ)reg y)) => (SRD  x y)
(SRAD x (MOV(W|H|B|WZ|HZ|BZ)reg y)) => (SRAD x y)
(SLW  x (MOV(W|H|B|WZ|HZ|BZ)reg y)) => (SLW  x y)
(SRW  x (MOV(W|H|B|WZ|HZ|BZ)reg y)) => (SRW  x y)
(SRAW x (MOV(W|H|B|WZ|HZ|BZ)reg y)) => (SRAW x y)

// Match rotate by constant.

	"regexp"
	"runtime"
	"runtime/debug"
	"strings"
	"sync"
	"sync/atomic"
	"testing"
	"time"
	_ "unsafe"
)

func cpuHogger(f func(x int) int, y *int, dur time.Duration) {
	// We only need to get one 100 Hz clock tick, so we've got
	// a large safety buffer.
	// But do at least 500 iterations (which should take about 100ms),
	// otherwise TestCPUProfileMultithreaded can fail if only one