cabletest/phy.go

package main

import (
	"fmt"
	"os"
	"path/filepath"
	"strings"
	"sync"
	"sync/atomic"
	"time"

	"golang.org/x/sys/unix"
)

const (
	bcmI2CWrite = 0xAC
	bcmI2CRead  = 0xAD

	bcmMMDVendor uint16 = 0x1E
	bcmRegCmd    uint16 = 0x4005
	bcmRegStatus uint16 = 0x4037
	bcmRegData1  uint16 = 0x4038

	bcmStInProgress uint16 = 0x0002
	bcmStPass       uint16 = 0x0004
	bcmStError      uint16 = 0x0008
	bcmStBusy       uint16 = 0xBBBB

	bcmCmdGetPairSwap uint16 = 0x8000
	bcmCmdSetEEEMode  uint16 = 0x8009
	bcmCmdSetJumbo    uint16 = 0x801C

	bcmRegECDCtrl   uint16 = 0x4006
	bcmRegECDResult uint16 = 0xA896
	bcmRegECDLen    uint16 = 0xA897

	bcmPHYIDHi = 0x3590
	bcmPHYIDLo = 0x5081

	bcmCmdResidentTemp uint16 = 0x0031

	bcmReadDelayUs  = 3000
	bcmRetryDelayUs = 10000

	bcmWindow    = 3400 * time.Millisecond
	bcmWindowFit = 100 * time.Millisecond
	bcmFlipPoll  = 10 * time.Millisecond
	bcmFlipWait  = 5 * time.Second

	bcmStatusPoll = 100 * time.Millisecond
	// Covers the handler's documented 2 s freeze during 10GBASE-T training.
	bcmStatusTimeout = 3 * time.Second

	ecdPoll     = 200 * time.Millisecond
	ecdDeadline = 50 * time.Second

	pairIdentityMap = 0xE4

	fsVendorPN     = "SFP-10G-T-100"
	wiitekVendorPN = "UF-RJ45-10G-100"
)

const (
	pairOK    = 1
	pairOpen  = 2
	pairShort = 3
	pairXtalk = 4
)

var pairVerdicts = map[int]string{
	pairOK: "ok", pairOpen: "OPEN", pairShort: "SHORT", pairXtalk: "XTALK",
}

type sff struct {
	ifname string
	path   string

	// Every transport touch happens on the loop goroutine: requests execute
	// one at a time, each admitted by the module type's own gate first.
	reqs  chan func()
	admit func()
}

func openSFF(ifname string) (*sff, error) {
	devLink, err := os.Readlink("/sys/class/net/" + ifname + "/device")
	if err != nil {
		return nil, fmt.Errorf("%s: %w", ifname, err)
	}
	drv, err := ifDriver(ifname)
	if err != nil {
		return nil, fmt.Errorf("%s: %w", ifname, err)
	}
	if drv != "ixgbe" {
		return nil, fmt.Errorf("%s: no module I2C transport for driver %s", ifname, drv)
	}
	s := &sff{
		ifname: ifname,
		path:   "/sys/kernel/debug/ixgbe/" + filepath.Base(devLink) + "/sff_i2c",
		reqs:   make(chan func()),
		admit:  func() {},
	}
	if _, err := os.Stat(s.path); err != nil {
		return nil, fmt.Errorf("%s: %w (patched ixgbe?)", ifname, err)
	}
	go s.loop()
	return s, nil
}

func (s *sff) loop() {
	defer holdPanic()
	for fn := range s.reqs {
		s.admit()
		fn()
	}
}

func (s *sff) exec(fn func()) {
	done := make(chan struct{})
	s.reqs <- func() { fn(); close(done) }
	<-done
}

func (s *sff) name() string { return s.ifname }

func (s *sff) op(cmd string) (string, error) {
	fd, err := unix.Open(s.path, unix.O_RDWR, 0)
	if err != nil {
		return "", fmt.Errorf("%s: %w", s.path, err)
	}
	defer unix.Close(fd)
	if _, err := unix.Write(fd, []byte(cmd)); err != nil {
		return "", fmt.Errorf("%s %q: %w", s.ifname, cmd, err)
	}
	buf := make([]byte, 256)
	n, err := unix.Read(fd, buf)
	if err != nil {
		return "", fmt.Errorf("%s %q: %w", s.ifname, cmd, err)
	}
	resp := strings.TrimSpace(string(buf[:n]))
	if !strings.HasPrefix(resp, "ok") {
		return "", fmt.Errorf("%s %q: %s", s.ifname, cmd, resp)
	}
	return strings.TrimSpace(resp[2:]), nil
}

func parseHexBytes(s string, n int) ([]byte, error) {
	fields := strings.Fields(s)
	if len(fields) != n {
		return nil, fmt.Errorf("want %d bytes, got %q", n, s)
	}
	out := make([]byte, n)
	for i, f := range fields {
		var v byte
		if _, err := fmt.Sscanf(f, "%x", &v); err != nil {
			return nil, fmt.Errorf("byte %q in %q", f, s)
		}
		out[i] = v
	}
	return out, nil
}

// A single bus hold; split write/read ops would let the driver's own SFP
// traffic consume the bridge's pending read.
func (s *sff) compound(waddr, raddr byte, delayUs, n int, wdata []byte) ([]byte, error) {
	var sb strings.Builder
	fmt.Fprintf(&sb, "x %02x %02x %d %x", waddr, raddr, delayUs, n)
	for _, v := range wdata {
		fmt.Fprintf(&sb, " %02x", v)
	}
	resp, err := s.op(sb.String())
	if err != nil {
		return nil, err
	}
	return parseHexBytes(resp, n)
}

func (s *sff) eeprom(off byte, n int) ([]byte, error) {
	return s.compound(0xA0, 0xA1, 500, n, []byte{off})
}

func (s *sff) vendorPN() (string, error) {
	pn, err := s.eeprom(40, 16)
	if err != nil {
		return "", err
	}
	return strings.TrimSpace(string(pn)), nil
}

type mdioDev interface {
	name() string
	exec(func())
	mdioRead(devad, reg uint16) (uint16, error)
	mdioWrite(devad, reg, val uint16) error
}

type phyDev interface {
	mdioDev
	identify() (string, error)
}

type snrSource interface {
	snrMargins() ([4]float64, error)
}

// PMA 1.1 latches low, so the first read reports any drop since it was last
// read and the second reports the wire as it is now.
func devLinkUp(d mdioDev) (up bool, raw uint16, err error) {
	d.exec(func() {
		if _, err = d.mdioRead(1, 1); err != nil {
			return
		}
		if raw, err = d.mdioRead(1, 1); err != nil {
			return
		}
		up = raw&0x0004 != 0
	})
	return
}

func devPCSLatch(d mdioDev) (blocks, ber uint64, raw uint16, err error) {
	d.exec(func() {
		if raw, err = d.mdioRead(3, 33); err != nil {
			return
		}
		blocks, ber = uint64(raw&0xFF), uint64((raw>>8)&0x3F)
	})
	return
}

func devFastRetrain(d mdioDev) (count, raw uint16, err error) {
	d.exec(func() {
		if raw, err = d.mdioRead(1, 147); err != nil {
			return
		}
		count = raw >> 11
	})
	return
}

// AN enable is forced alongside the restart: the ECD can leave the BCM with
// 7.0.12 cleared (proven live — no AN pulses, both ends deaf, link down until
// power cycle), and a bare restart bit preserves the cleared enable.
func devRestartAN(d mdioDev) (err error) {
	d.exec(func() {
		var v uint16
		if v, err = d.mdioRead(7, 0); err != nil {
			return
		}
		err = d.mdioWrite(7, 0, v|0x1200)
	})
	return
}

func devEEEAdvert(d mdioDev) (v uint16, err error) {
	d.exec(func() { v, err = d.mdioRead(7, 60) })
	return
}

type bcm struct {
	*sff
	windowEnd time.Time
}

func newBCM(t *sff) *bcm {
	b := &bcm{sff: t}
	t.exec(func() { t.admit = b.window })
	return b
}

// The firmware's internal temp poll serves stale bridge reads for ~50 ms
// around it; work stays inside 3.4 s of an observed poll. A resident 0x0031 at
// expiry means the phase is unknown, so re-lock: arm, then take the true edge.
// Each taken edge immediately re-arms — the one CMD write per window lands at
// the start of the quiet period, maximally far from the next poll (writes near
// the poll are the µC-wedge risk), and every later expiry reads the phase
// without writing.
func (b *bcm) window() {
	if time.Now().Add(bcmWindowFit).Before(b.windowEnd) {
		return
	}
	armed := false
	deadline := time.Now().Add(bcmFlipWait)
	for {
		v, err := b.mdioRead(bcmMMDVendor, bcmRegCmd)
		if err != nil {
			panic(fmt.Sprintf("%s: heartbeat poll: %v", b.ifname, err))
		}
		if v == bcmCmdResidentTemp {
			if armed {
				b.windowEnd = time.Now().Add(bcmWindow)
				b.rearm()
				return
			}
			b.rearm()
			armed = true
			deadline = time.Now().Add(bcmFlipWait)
			continue
		}
		armed = true
		if time.Now().After(deadline) {
			b.windowEnd = time.Now().Add(bcmWindow)
			return
		}
		time.Sleep(bcmFlipPoll)
	}
}

func (b *bcm) rearm() {
	if _, err := b.waitStatus(func(st uint16) bool {
		return st != bcmStInProgress && st != bcmStBusy
	}); err != nil {
		panic(fmt.Sprintf("%s: rearm: %v", b.ifname, err))
	}
	if err := b.mdioWrite(bcmMMDVendor, bcmRegCmd, bcmCmdGetPairSwap); err != nil {
		panic(fmt.Sprintf("%s: rearm: %v", b.ifname, err))
	}
	if _, err := b.waitStatus(func(st uint16) bool {
		return st == bcmStPass || st == bcmStError
	}); err != nil {
		panic(fmt.Sprintf("%s: rearm: %v", b.ifname, err))
	}
}

func (b *bcm) mdioReadDelay(devad, reg uint16, delayUs int) (uint16, error) {
	d, err := b.compound(bcmI2CWrite, bcmI2CRead, delayUs, 2,
		[]byte{0x20 | byte(devad), byte(reg >> 8), byte(reg)})
	if err != nil {
		return 0, err
	}
	return uint16(d[0])<<8 | uint16(d[1]), nil
}

// 0x0000 is also the bridge's not-ready signature, so a zero is read again at
// a longer delay before being believed.
func (b *bcm) mdioRead(devad, reg uint16) (uint16, error) {
	v, err := b.mdioReadDelay(devad, reg, bcmReadDelayUs)
	if err != nil || v != 0 {
		return v, err
	}
	return b.mdioReadDelay(devad, reg, bcmRetryDelayUs)
}

func (b *bcm) mdioWrite(devad, reg, val uint16) error {
	_, err := b.op(fmt.Sprintf("w %02x %02x %02x %02x %02x %02x",
		bcmI2CWrite, byte(devad), byte(reg>>8), byte(reg), byte(val>>8), byte(val)))
	return err
}

// The datasheet's completion handshake: poll STATUS on its 100 ms cadence
// until the wanted state, bounded by a deadline.
func (b *bcm) waitStatus(want func(uint16) bool) (uint16, error) {
	deadline := time.Now().Add(bcmStatusTimeout)
	for {
		st, err := b.mdioRead(bcmMMDVendor, bcmRegStatus)
		if err != nil {
			return 0, err
		}
		if want(st) {
			return st, nil
		}
		if time.Now().After(deadline) {
			return 0, fmt.Errorf("%s: command handler stuck, status %#04x", b.ifname, st)
		}
		time.Sleep(bcmStatusPoll)
	}
}

// GETs must be invoked bare (pre-writing any DATA register leaves the handler
// executing as a no-op); SETs must pass their full parameter set (the handler
// executes stale DATA).
func (b *bcm) command(code uint16, params ...uint16) (data [5]uint16, err error) {
	b.exec(func() {
		if _, err = b.waitStatus(func(st uint16) bool {
			return st != bcmStInProgress && st != bcmStBusy
		}); err != nil {
			return
		}
		for i, p := range params {
			if err = b.mdioWrite(bcmMMDVendor, bcmRegData1+uint16(i), p); err != nil {
				return
			}
		}
		if err = b.mdioWrite(bcmMMDVendor, bcmRegCmd, code); err != nil {
			return
		}
		var st uint16
		if st, err = b.waitStatus(func(st uint16) bool {
			return st == bcmStPass || st == bcmStError
		}); err != nil {
			return
		}
		if st == bcmStError {
			err = fmt.Errorf("%s: command %#04x returned ERROR", b.ifname, code)
			return
		}
		if len(params) > 0 {
			return
		}
		for i := range data {
			if data[i], err = b.mdioRead(bcmMMDVendor, bcmRegData1+uint16(i)); err != nil {
				return
			}
		}
	})
	return
}

func (b *bcm) identify() (ident string, err error) {
	b.exec(func() {
		var hi, lo uint16
		if hi, err = b.mdioRead(1, 2); err != nil {
			return
		}
		if lo, err = b.mdioRead(1, 3); err != nil {
			return
		}
		if hi != bcmPHYIDHi || lo != bcmPHYIDLo {
			err = fmt.Errorf("%s: PHY ID %#04x:%#04x, want %#04x:%#04x",
				b.ifname, hi, lo, bcmPHYIDHi, bcmPHYIDLo)
			return
		}
		var sn []byte
		if sn, err = b.eeprom(68, 16); err != nil {
			return
		}
		ident = "BCM84891L sn " + strings.TrimSpace(string(sn))
	})
	return
}

func (b *bcm) forceEEEOff() error {
	_, err := b.command(bcmCmdSetEEEMode, 0x0000, 0x0000, 0x7A12, 0x0480, 0x0000)
	return err
}

func (b *bcm) forceJumbo() error {
	_, err := b.command(bcmCmdSetJumbo, 1, 0, 0, 0, 0)
	return err
}

// Left to AN, master/slave is a per-training lottery and each training's DSP
// convergence moves per-pair SNR by up to ~3.6 dB; pinned roles at least keep
// every session measured under identical conditions.
func (b *bcm) forceRole(master bool) (err error) {
	b.exec(func() {
		var v uint16
		if v, err = b.mdioRead(7, 32); err != nil {
			return
		}
		v |= 0x8000
		if master {
			v |= 0x4000
		} else {
			v &^= 0x4000
		}
		err = b.mdioWrite(7, 32, v)
	})
	return
}

func (b *bcm) pairMap() (byte, error) {
	d, err := b.command(bcmCmdGetPairSwap)
	if err != nil {
		return 0, err
	}
	return byte(d[1]), nil
}

const (
	rbI2CWrite = 0xA2
	rbI2CRead  = 0xA3

	rbOffPassword byte = 0x7B
	rbOffPage     byte = 0x7F
	rbOffCmd      byte = 0x80
	rbOffDevad    byte = 0x81
	rbOffValHi    byte = 0x84
	rbOffPartNum  byte = 0xFA

	rbPageMailbox byte = 3

	rbCmdWrite byte = 0x01
	rbCmdRead  byte = 0x02
	rbCmdDone  byte = 0x04

	rbReadDelayUs = 500
	rbCmdPoll     = 20 * time.Millisecond
	// Matches the BCM allowance for a handler frozen by 10GBASE-T training.
	rbCmdTimeout = 3 * time.Second

	rbPHYIDHi uint16 = 0x002B
	rbPHYIDLo uint16 = 0x0BF4

	// IEEE margins land near 7-9 dB on a healthy short cable; far outside is
	// another register's data.
	rbGhostLow  = -10.0
	rbGhostHigh = 25.0
)

// Only the registers proven safe on this PHY (docs/modules/wiitek/): single
// reads in the vendor windows brick the µC permanently, so everything else
// refuses before touching hardware.
var rbReadSafe = map[uint16]map[uint16]bool{
	1: {1: true, 2: true, 3: true, 133: true, 134: true, 135: true, 136: true, 147: true},
	3: {32: true, 33: true},
	7: {0: true, 33: true, 60: true},
}

var rbWriteSafe = map[uint16]map[uint16]bool{
	7: {0: true},
}

type rollball struct {
	*sff
}

func (r *rollball) i2cWrite(off byte, data ...byte) error {
	var sb strings.Builder
	fmt.Fprintf(&sb, "w %02x %02x", rbI2CWrite, off)
	for _, v := range data {
		fmt.Fprintf(&sb, " %02x", v)
	}
	_, err := r.op(sb.String())
	return err
}

func (r *rollball) i2cRead(off byte, n int) ([]byte, error) {
	return r.compound(rbI2CWrite, rbI2CRead, rbReadDelayUs, n, []byte{off})
}

func (r *rollball) unlock() error {
	if err := r.i2cWrite(rbOffPage, rbPageMailbox); err != nil {
		return err
	}
	return r.i2cWrite(rbOffPassword, 0xFF, 0xFF, 0xFF, 0xFF)
}

func (r *rollball) mbox(cmd byte, devad, reg, val uint16) error {
	if err := r.unlock(); err != nil {
		return err
	}
	if err := r.i2cWrite(rbOffDevad, byte(devad), byte(reg>>8), byte(reg)); err != nil {
		return err
	}
	if cmd == rbCmdWrite {
		if err := r.i2cWrite(rbOffValHi, byte(val>>8), byte(val)); err != nil {
			return err
		}
	}
	if err := r.i2cWrite(rbOffCmd, cmd); err != nil {
		return err
	}
	deadline := time.Now().Add(rbCmdTimeout)
	for {
		d, err := r.i2cRead(rbOffCmd, 1)
		if err != nil {
			return err
		}
		if d[0] == rbCmdDone {
			return nil
		}
		if time.Now().After(deadline) {
			return fmt.Errorf("%s: mailbox %d.%#04x stuck at %#02x", r.ifname, devad, reg, d[0])
		}
		time.Sleep(rbCmdPoll)
	}
}

func rbGuard(safe map[uint16]map[uint16]bool, ifname, what string, devad, reg uint16) {
	if !safe[devad][reg] {
		panic(fmt.Sprintf("%s: refusing MDIO %s %d.%#04x: outside the proven-safe set",
			ifname, what, devad, reg))
	}
}

func (r *rollball) mdioRead(devad, reg uint16) (uint16, error) {
	rbGuard(rbReadSafe, r.ifname, "read", devad, reg)
	if err := r.mbox(rbCmdRead, devad, reg, 0); err != nil {
		return 0, err
	}
	d, err := r.i2cRead(rbOffValHi, 2)
	if err != nil {
		return 0, err
	}
	return uint16(d[0])<<8 | uint16(d[1]), nil
}

func (r *rollball) mdioWrite(devad, reg, val uint16) error {
	rbGuard(rbWriteSafe, r.ifname, "write", devad, reg)
	return r.mbox(rbCmdWrite, devad, reg, val)
}

func (r *rollball) identify() (ident string, err error) {
	r.exec(func() {
		var hi, lo uint16
		if hi, err = r.mdioRead(1, 2); err != nil {
			return
		}
		if lo, err = r.mdioRead(1, 3); err != nil {
			return
		}
		if hi != rbPHYIDHi || lo != rbPHYIDLo {
			err = fmt.Errorf("%s: PHY ID %#04x:%#04x, want %#04x:%#04x",
				r.ifname, hi, lo, rbPHYIDHi, rbPHYIDLo)
			return
		}
		if err = r.unlock(); err != nil {
			return
		}
		var part []byte
		if part, err = r.i2cRead(rbOffPartNum, 1); err != nil {
			return
		}
		var sn []byte
		if sn, err = r.eeprom(68, 16); err != nil {
			return
		}
		ident = fmt.Sprintf("CUX3610 sn %s (A2.250=%d)", strings.TrimSpace(string(sn)), part[0])
	})
	return
}

func (r *rollball) snrMargins() (out [4]float64, err error) {
	r.exec(func() {
		for i := range out {
			var v uint16
			if v, err = r.mdioRead(1, uint16(133+i)); err != nil {
				return
			}
			m := (float64(v) - 0x8000) / 10
			if m < rbGhostLow || m > rbGhostHigh {
				panic(fmt.Sprintf("%s: ghost SNR margin %.1f dB (1.%d=%#04x)", r.ifname, m, 133+i, v))
			}
			out[i] = m
		}
	})
	return
}

const (
	phyInterval  = time.Second
	phyStale     = 5 * time.Second
	phyMaxDark   = 30
	linkWaitSpan = 25 * time.Second
	linkWaitPoll = time.Second

	snrGoodMargin = 3.0
	snrWarnMargin = 1.0
)

type phyModule struct {
	dev  phyDev
	busy atomic.Bool

	mu           sync.Mutex
	sampled      bool
	lastOK       time.Time
	link         bool
	haveSNR      bool
	margins      [4]float64
	blocks       uint64
	ber          uint64
	retrains     uint64
	recentDelta  uint64
	primed       bool
	retrainCount uint16
	notes        []string
}

// Silent while a measure owns the module.
func (m *phyModule) poll() error {
	if m.busy.Load() {
		return nil
	}
	link, linkRaw, err := devLinkUp(m.dev)
	if err != nil {
		return err
	}
	var margins [4]float64
	haveSNR := false
	if src, ok := m.dev.(snrSource); ok && link {
		if margins, err = src.snrMargins(); err != nil {
			return err
		}
		haveSNR = true
	}
	blocks, ber, pcsRaw, err := devPCSLatch(m.dev)
	if err != nil {
		return err
	}
	count, frRaw, err := devFastRetrain(m.dev)
	if err != nil {
		return err
	}

	m.mu.Lock()
	if m.link && !link {
		m.notes = append(m.notes,
			fmt.Sprintf("%s link read down: 1.1=0x%04x", m.dev.name(), linkRaw))
	}
	m.sampled = true
	m.lastOK = time.Now()
	m.link = link
	m.haveSNR = haveSNR
	m.margins = margins
	// The first poll after a baseline drains latches from the bringup/diag
	// retrain era, so it only sets the origin; the retrain counter is 5 bits.
	if m.primed {
		rt := uint64((count - m.retrainCount) & 0x1F)
		delta := blocks + ber + rt
		if delta > 0 {
			m.notes = append(m.notes, fmt.Sprintf(
				"%s corrected +%d raw: 3.33=0x%04x (blocks %d ber %d) 1.147=0x%04x (retrain +%d) 1.1=0x%04x",
				m.dev.name(), delta, pcsRaw, blocks, ber, frRaw, rt, linkRaw))
		}
		m.blocks += blocks
		m.ber += ber
		m.retrains += rt
		m.recentDelta = delta
	} else {
		m.recentDelta = 0
		m.primed = true
	}
	m.retrainCount = count
	m.mu.Unlock()
	return nil
}

func (m *phyModule) takeNotes() []string {
	m.mu.Lock()
	defer m.mu.Unlock()
	n := m.notes
	m.notes = nil
	return n
}

func (m *phyModule) run(done *atomic.Bool) {
	tick := time.NewTicker(phyInterval)
	defer tick.Stop()

	dark := 0
	var lastErr error
	for !done.Load() {
		<-tick.C
		if err := m.poll(); err != nil {
			dark++
			lastErr = err
			if dark >= phyMaxDark {
				panic(fmt.Sprintf("module diagnostics dark for %d polls: %v", dark, lastErr))
			}
			continue
		}
		dark = 0
	}
}

func (m *phyModule) reset() {
	m.mu.Lock()
	m.blocks, m.ber, m.retrains, m.recentDelta = 0, 0, 0, 0
	m.primed = false
	m.mu.Unlock()
}

type phyModView struct {
	fresh   bool
	link    bool
	haveSNR bool
	margins [4]float64
	blocks  uint64
	ber     uint64
	retrain uint64
	recent  uint64
}

func (m *phyModule) view() phyModView {
	m.mu.Lock()
	defer m.mu.Unlock()
	v := phyModView{
		fresh:   m.sampled && time.Since(m.lastOK) < phyStale,
		link:    m.link,
		haveSNR: m.haveSNR,
		margins: m.margins,
		blocks:  m.blocks,
		ber:     m.ber,
		retrain: m.retrains,
	}
	if v.fresh {
		v.recent = m.recentDelta
	}
	return v
}

type cableInfo struct {
	ecd      ecdResult
	maps     [2]byte
	haveMaps [2]bool
}

func (c cableInfo) metresString() string {
	sum, n := 0, 0
	for i, v := range c.ecd.verdicts {
		if v == pairOK {
			sum += c.ecd.metres[i]
			n++
		}
	}
	if n == 0 {
		return "-"
	}
	return fmt.Sprintf("%d", (sum+n/2)/n)
}

const (
	clsNone = iota
	clsGood
	clsWarn
	clsBad
)

func snrClass(margin float64) int {
	switch {
	case margin >= snrGoodMargin:
		return clsGood
	case margin >= snrWarnMargin:
		return clsWarn
	default:
		return clsBad
	}
}

type phyDisplay struct {
	haveSNR     bool
	worstMargin float64
	corrected   uint64
	recent      uint64
	metres      string
	metresClass int
}

func pairLetter(i int) string { return string(rune('A' + i)) }

// Each end resolves MDI on its own, so a swap at either known end counts; an
// end with no readable map abstains.
func pairSwapped(i int, c cableInfo) bool {
	for e := range c.maps {
		if c.haveMaps[e] && int(c.maps[e]>>(2*i))&3 != i {
			return true
		}
	}
	return false
}

func cableSummary(cable cableInfo, measuring bool) (string, int) {
	if measuring {
		return "...", clsNone
	}
	anyData, anyFault, anySwap := false, false, false
	for i, v := range cable.ecd.verdicts {
		if v != 0 {
			anyData = true
		}
		if v != 0 && v != pairOK {
			anyFault = true
		}
		if pairSwapped(i, cable) {
			anySwap = true
		}
	}
	s := cable.metresString()
	switch {
	case !anyData:
		return "-", clsNone
	case anyFault:
		return s, clsBad
	case anySwap:
		return s, clsWarn
	}
	return s, clsGood
}

// The margin is the worst pair across the ends that measure SNR (the Wiitek's
// IEEE 1.133–136), gated on the whole pair being fresh and linked.
func phyDisplayFrom(cable cableInfo, measuring bool, a, b phyModView) phyDisplay {
	d := phyDisplay{
		haveSNR:   a.fresh && b.fresh && a.link && b.link && (a.haveSNR || b.haveSNR),
		corrected: a.blocks + a.ber + a.retrain + b.blocks + b.ber + b.retrain,
		recent:    a.recent + b.recent,
	}
	if d.haveSNR {
		first := true
		for _, v := range []phyModView{a, b} {
			if !v.haveSNR {
				continue
			}
			for _, m := range v.margins {
				if first || m < d.worstMargin {
					d.worstMargin = m
					first = false
				}
			}
		}
	}
	d.metres, d.metresClass = cableSummary(cable, measuring)
	return d
}

type ecdResult struct {
	verdicts [4]int
	metres   [4]int
}

func (b *bcm) cableDiag() (res ecdResult, err error) {
	b.exec(func() {
		var ctrl uint16
		if ctrl, err = b.mdioRead(bcmMMDVendor, bcmRegECDCtrl); err != nil {
			return
		}
		if err = b.mdioWrite(bcmMMDVendor, bcmRegECDCtrl, ctrl&^0xF400|0x8400); err != nil {
			return
		}
		deadline := time.Now().Add(ecdDeadline)
		for {
			if ctrl, err = b.mdioRead(bcmMMDVendor, bcmRegECDCtrl); err != nil {
				return
			}
			if ctrl&0x0800 == 0 {
				break
			}
			if time.Now().After(deadline) {
				err = fmt.Errorf("%s: cable diag still busy after %s", b.ifname, ecdDeadline)
				return
			}
			time.Sleep(ecdPoll)
		}
		b.window()
		var v uint16
		if v, err = b.mdioRead(1, bcmRegECDResult); err != nil {
			return
		}
		for i := range res.verdicts {
			res.verdicts[i] = int(v>>(4*i)) & 0xF
			if res.verdicts[i] > pairXtalk {
				panic(fmt.Sprintf("%s: ghost ECD verdict %#04x", b.ifname, v))
			}
			var m uint16
			if m, err = b.mdioRead(1, bcmRegECDLen+uint16(i)); err != nil {
				return
			}
			res.metres[i] = int(m)
		}
	})
	return
}

func bcmEnd(mods []*phyModule) *bcm {
	for _, m := range mods {
		if b, ok := m.dev.(*bcm); ok {
			return b
		}
	}
	panic("no BCM module in the pair: the ECD is the only length path")
}

// The pollers are held silent throughout; pair maps are read after the
// relink, so the MDI resolution is the fresh one.
func measureCable(mods []*phyModule, waitRelink bool, done *atomic.Bool) (cableInfo, bool, error) {
	for _, m := range mods {
		m.busy.Store(true)
	}
	defer func() {
		for _, m := range mods {
			m.busy.Store(false)
		}
	}()

	var c cableInfo
	var err error
	end := bcmEnd(mods)
	c.ecd, err = end.cableDiag()
	if err != nil {
		return c, false, err
	}
	if err = devRestartAN(end); err != nil {
		return c, false, err
	}
	relinked := false
	if waitRelink {
		names := [2]string{mods[0].dev.name(), mods[1].dev.name()}
		_, relinked = waitCarrier(names, done)
	}
	for i, m := range mods {
		b, ok := m.dev.(*bcm)
		if !ok {
			continue
		}
		if c.maps[i], err = b.pairMap(); err != nil {
			return c, false, err
		}
		c.haveMaps[i] = true
	}
	return c, relinked, nil
}

type cableDiag struct {
	mods      []*phyModule
	completed chan error

	mu      sync.Mutex
	info    cableInfo
	running bool
}

func newCableDiag(mods []*phyModule, info cableInfo) *cableDiag {
	return &cableDiag{mods: mods, completed: make(chan error, 1), info: info}
}

func (c *cableDiag) snapshot() (cableInfo, bool) {
	c.mu.Lock()
	defer c.mu.Unlock()
	return c.info, c.running
}

func (c *cableDiag) kick(done *atomic.Bool) bool {
	c.mu.Lock()
	if c.running {
		c.mu.Unlock()
		return false
	}
	c.running = true
	c.mu.Unlock()

	go func() {
		defer holdPanic()
		info, _, err := measureCable(c.mods, true, done)
		if err != nil {
			info = cableInfo{}
		}
		c.mu.Lock()
		c.info = info
		c.running = false
		c.mu.Unlock()
		select {
		case c.completed <- err:
		default:
		}
	}()
	return true
}

func mapString(m byte, have bool) string {
	if !have {
		return "unread"
	}
	if m == pairIdentityMap {
		return "straight"
	}
	out := make([]string, 4)
	for i := range out {
		out[i] = pairLetter(int(m>>(2*i)) & 3)
	}
	return "swapped to " + strings.Join(out, "")
}

func verdictString(r ecdResult) string {
	bad := []string{}
	for i, v := range r.verdicts {
		if v != pairOK {
			s, ok := pairVerdicts[v]
			if !ok {
				s = fmt.Sprintf("%d", v)
			}
			bad = append(bad, fmt.Sprintf("%s %s at %dm", pairLetter(i), s, r.metres[i]))
		}
	}
	if len(bad) > 0 {
		return strings.Join(bad, ", ")
	}
	return fmt.Sprintf("all pairs ok, %d/%d/%d/%d m",
		r.metres[0], r.metres[1], r.metres[2], r.metres[3])
}

func openModules(names [2]string) ([]*phyModule, [2]string, error) {
	mods := make([]*phyModule, 0, 2)
	var idents [2]string
	for i, name := range names {
		t, err := openSFF(name)
		if err != nil {
			return nil, idents, err
		}
		pn, err := t.vendorPN()
		if err != nil {
			return nil, idents, err
		}
		var dev phyDev
		switch pn {
		case fsVendorPN:
			dev = newBCM(t)
		case wiitekVendorPN:
			dev = &rollball{sff: t}
		default:
			return nil, idents, fmt.Errorf("%s: unknown module PN %q", name, pn)
		}
		idents[i], err = dev.identify()
		if err != nil {
			return nil, idents, err
		}
		m := &phyModule{dev: dev}
		// Born busy: the pollers stay silent through bringup's SETs and
		// retrains until the first measure completes and lifts the gate.
		m.busy.Store(true)
		mods = append(mods, m)
	}
	return mods, idents, nil
}

func waitCarrier(names [2]string, done *atomic.Bool) (time.Duration, bool) {
	start := time.Now()
	deadline := start.Add(linkWaitSpan)
	for {
		if carrierUp(names[0]) && carrierUp(names[1]) {
			return time.Since(start), true
		}
		if time.Now().After(deadline) || (done != nil && done.Load()) {
			return time.Since(start), false
		}
		time.Sleep(linkWaitPoll)
	}
}

// No trustworthy config readback exists (DATA1 is firmware scratch) and no
// cable is guaranteed at bringup, so both settings are forced every boot: the
// one deterministic assurance. The handler freezes during training, so the
// carrier settles — the host checks just reset the links — before any command.
// Never waits for a link: there may be no cable, and forcing config needs
// none — the AN restart applies it whenever training next happens.
func moduleChecks(mods []*phyModule, names [2]string) []checkResult {
	var out []checkResult
	fail := func(item string, err error) []checkResult {
		return append(out, checkResult{item: item, err: err})
	}

	// With a non-BCM partner the BCM is forced slave: the partner's manual
	// config is unreachable, and auto-resolves-master against manual-slave is
	// the combination proven to link.
	mixed := false
	for _, m := range mods {
		if _, ok := m.dev.(*bcm); !ok {
			mixed = true
		}
	}
	master := !mixed
	for i, m := range mods {
		b, ok := m.dev.(*bcm)
		if !ok {
			out = append(out, checkResult{item: names[i] + " role", state: "auto"})
			continue
		}

		res := checkResult{item: names[i] + " eee", state: "forced off"}
		if err := b.forceEEEOff(); err != nil {
			return fail(res.item, err)
		}
		out = append(out, res)

		res = checkResult{item: names[i] + " jumbo", state: "forced on"}
		if err := b.forceJumbo(); err != nil {
			return fail(res.item, err)
		}
		out = append(out, res)

		res = checkResult{item: names[i] + " role", state: "forced master"}
		if !master {
			res.state = "forced slave"
		}
		if err := b.forceRole(master); err != nil {
			return fail(res.item, err)
		}
		master = false
		out = append(out, res)
	}

	// Both modules configured and verified before either AN restart: the
	// modules link to each other, so one restart puts both µCs into training,
	// and no read should race that. The restarts fire last, nothing after.
	res := checkResult{item: "eee advert"}
	var adv [2]uint16
	var advs [2]string
	for i, m := range mods {
		v, err := devEEEAdvert(m.dev)
		if err != nil {
			return fail(res.item, err)
		}
		adv[i] = v
		advs[i] = fmt.Sprintf("%#04x", v)
		if _, ok := m.dev.(*bcm); ok && v != 0 {
			res.err = fmt.Errorf("%s still advertises EEE %#04x", names[i], v)
		}
	}
	if res.err == nil && adv[0]&adv[1] != 0 {
		res.err = fmt.Errorf("EEE would negotiate: common ability %#04x", adv[0]&adv[1])
	}
	res.state = advs[0] + "/" + advs[1]
	out = append(out, res)

	for i, m := range mods {
		if err := devRestartAN(m.dev); err != nil {
			return fail(names[i]+" retrain", err)
		}
	}
	return out
}

func cableLine(c cableInfo) string {
	return fmt.Sprintf("%s; map %s / %s",
		verdictString(c.ecd), mapString(c.maps[0], c.haveMaps[0]), mapString(c.maps[1], c.haveMaps[1]))
}