theater/cabletest

T

flamingcow 36c2ff34fe Paint timeline slots spent in a cable measure blue

2026-08-30 16:50:06 -07:00

Drop goals and open-questions docs, cut PTP from the appliance kernel

2026-08-28 21:59:41 -07:00

Render the live status to the framebuffer in Atkinson Hyperlegible Mono

2026-07-25 19:41:53 -07:00

Stream the panel as masked png deltas and seqlock the scanout reads

2026-08-30 16:45:19 -07:00

Swap in the Riverdi RVT70 panel, flip the rotation, and stop ^C orphaning harness runs

2026-08-28 21:04:55 -07:00

Turn off the x86 instruction decoder selftest so mkimage skips posttest

2026-08-28 22:14:19 -07:00

.gitignore

Nothing lives outside the repo: the bench ixgbe driver source moves in tracked at kernel/ixgbe-sff/ (build artifacts gitignored, load-ixgbe repointed, module build verified from the new home — the reg_ops/ts_bench bench extras are finally versioned), the ice sff_i2c patch is vendored at docs/nics/e810/ice-sff-i2c.patch with its out-of-tree build recipe (TRACE_INCLUDE_PATH fixup, version-matched Makefile), the e810 NVM flash logistics land in the e810 README (exact package name, rollback -m B49691D578EC into the inactive bank, post-flash inventory failing until power cycle is the expected tell), and every ~/work/phydiag-work pointer becomes an inline statement; phydiag-work, ice-sff, e810nvm and the wiitekphy CLI (a duplicate of documented transport+protocol) are deleted

2026-08-19 16:27:48 -07:00

bcm.go

Bugs-evident cleanup: cable-measure failure is fatal with its reason (failed/fail pretend-path and dead relinked plumbing deleted), bcm.command refuses partial SETs (0 or 5 params, stale-DATA trap made structural), BOOT gains a patched-ixgbe check naming stock-driver boots, panel fault rows matched by label and console rows strict on column count, phy.go split into sff/bcm/rollball and main.go's direction machinery into direction.go (pure moves), input events decoded via encoding/binary, gofmt; verified on hardware (20.00G, zero errors, ECD 41m)

2026-08-17 18:56:55 -07:00

counters.go

Hardware lock-in cleanup: ice RX counter set and by-driver map collapsed to the single ixgbe nicRxFields list (newNICPoller drops its redundant driver gate), checkCoalesce loses the try-tx-then-EINVAL fallback and sets rx-usecs shared with tx as the one path, ice-attributed comments neutralized, dead errs.total removed with its redundant test assertion; ice counter reference set recorded in docs/nics/e810; verified on hardware (20.00G, zero errors, ECD 41m)

2026-08-17 12:45:41 -07:00

direction.go

Bugs-evident cleanup: cable-measure failure is fatal with its reason (failed/fail pretend-path and dead relinked plumbing deleted), bcm.command refuses partial SETs (0 or 5 params, stale-DATA trap made structural), BOOT gains a patched-ixgbe check naming stock-driver boots, panel fault rows matched by label and console rows strict on column count, phy.go split into sff/bcm/rollball and main.go's direction machinery into direction.go (pure moves), input events decoded via encoding/binary, gofmt; verified on hardware (20.00G, zero errors, ECD 41m)

2026-08-17 18:56:55 -07:00

fatal.go

Panel rebuilt as the heat matrix: verdict header (ring + newest fault named with its age, cable cell top right), one row per check with NOW / last-90s-at-5s-a-cell / since-reset columns, noise as a blue context lane; history.go keeps the 18-slot wall-time ring (per-class fault deltas, SNR minima, rate verdicts, measuring gray) plus since-reset aggregates and fault ages that clear on reset; noise cycle grid-locked to the shared slot clock so cells never straddle a phase and flicker, presence granted at boot until the first up-phase verdict; check/cross stroked as capsule marks (font has neither), line rate is a pure verdict everywhere, footer is a three-line stat stack beside hold-to-reset; old panel/chip/stat-grid machinery deleted; mockups/ gitignored

2026-08-17 12:22:55 -07:00

fb.go

Swap in the Riverdi RVT70 panel, flip the rotation, and stop ^C orphaning harness runs

2026-08-28 21:04:55 -07:00

frame_test.go

Checksum the header and refuse sequence numbers the sender never sent

2026-08-04 20:41:16 -07:00

frame.go

Check the header version on arrival and say what the error window actually spans

2026-08-04 21:29:03 -07:00

go.mod

Bump the go directive to 1.27.0

2026-08-30 16:46:01 -07:00

go.sum

Sweep the superseded x/image and x/text hashes from go.sum

2026-08-28 22:17:37 -07:00

history.go

Noise ripped out after the bench verdict (tightly-wrapped unshielded cables move victim SNR only 0.2-0.3 dB, recorded in measurement.md dead ends): noise.go deleted with the noiser, phase clock and panel lane; configureNoise/checkPrivFlag gone so the X710 pair is untouched; console noise column, NOISE links rows and lane colors removed; driverPair rehomed to system.go, carrierUp to phy.go; history anchors to run start (frame-rate sampling and draw-clock roll kept — they fix display latency for every lane); goals/hardware/state/nics docs updated; verified on hardware (20.00G, zero errors, ECD 41m, SNR +7.6)

2026-08-17 21:37:18 -07:00

init.go

Bugs-evident cleanup: cable-measure failure is fatal with its reason (failed/fail pretend-path and dead relinked plumbing deleted), bcm.command refuses partial SETs (0 or 5 params, stale-DATA trap made structural), BOOT gains a patched-ixgbe check naming stock-driver boots, panel fault rows matched by label and console rows strict on column count, phy.go split into sff/bcm/rollball and main.go's direction machinery into direction.go (pure moves), input events decoded via encoding/binary, gofmt; verified on hardware (20.00G, zero errors, ECD 41m)

2026-08-17 18:56:55 -07:00

input.go

Bugs-evident cleanup: cable-measure failure is fatal with its reason (failed/fail pretend-path and dead relinked plumbing deleted), bcm.command refuses partial SETs (0 or 5 params, stale-DATA trap made structural), BOOT gains a patched-ixgbe check naming stock-driver boots, panel fault rows matched by label and console rows strict on column count, phy.go split into sff/bcm/rollball and main.go's direction machinery into direction.go (pure moves), input events decoded via encoding/binary, gofmt; verified on hardware (20.00G, zero errors, ECD 41m)

2026-08-17 18:56:55 -07:00

load-ixgbe

Nothing lives outside the repo: the bench ixgbe driver source moves in tracked at kernel/ixgbe-sff/ (build artifacts gitignored, load-ixgbe repointed, module build verified from the new home — the reg_ops/ts_bench bench extras are finally versioned), the ice sff_i2c patch is vendored at docs/nics/e810/ice-sff-i2c.patch with its out-of-tree build recipe (TRACE_INCLUDE_PATH fixup, version-matched Makefile), the e810 NVM flash logistics land in the e810 README (exact package name, rollback -m B49691D578EC into the inactive bank, post-flash inventory failing until power cycle is the expected tell), and every ~/work/phydiag-work pointer becomes an inline statement; phydiag-work, ice-sff, e810nvm and the wiitekphy CLI (a duplicate of documented transport+protocol) are deleted

2026-08-19 16:27:48 -07:00

loss_test.go

Cable verdict folded into one console cell (length green, fault name red, xover amber, fail; diag rule lines and counters-reset line gone), reset fires at the press not diag completion, and the measure's blip is suppressed at every counter source instead of counted-hidden-reverted: worker error adds, loss-window write-offs (quiet slides state without charging; leaving the quiet era presumes in-window holes delivered so the gap's tail is never charged late), NIC poller tracks raw without accumulating, socket drops discarded, modules un-primed by the measure itself; display hold deleted so pre-existing errors no longer blink out during a measure

2026-08-17 11:25:31 -07:00

loss.go

Cable verdict folded into one console cell (length green, fault name red, xover amber, fail; diag rule lines and counters-reset line gone), reset fires at the press not diag completion, and the measure's blip is suppressed at every counter source instead of counted-hidden-reverted: worker error adds, loss-window write-offs (quiet slides state without charging; leaving the quiet era presumes in-window holes delivered so the gap's tail is never charged late), NIC poller tracks raw without accumulating, socket drops discarded, modules un-primed by the measure itself; display hold deleted so pre-existing errors no longer blink out during a measure

2026-08-17 11:25:31 -07:00

main_test.go

Hardware lock-in cleanup: ice RX counter set and by-driver map collapsed to the single ixgbe nicRxFields list (newNICPoller drops its redundant driver gate), checkCoalesce loses the try-tx-then-EINVAL fallback and sets rx-usecs shared with tx as the one path, ice-attributed comments neutralized, dead errs.total removed with its redundant test assertion; ice counter reference set recorded in docs/nics/e810; verified on hardware (20.00G, zero errors, ECD 41m)

2026-08-17 12:45:41 -07:00

main.go

Power: 2 streams and a 2.0 GHz clock pinned by a new checkMaxFreq SETTINGS check (performance governor + scaling_max_freq = fixed clock) take line rate from 80 W to 29.7 W package, qualified 96 s zero-loss; measured on the bench: streams buy scheduling headroom not watts (C1-parked threads, lonelier spinners boost harder), relaxed cpu_dma_latency never reaches deep idle (scheduler spreads wakeups, ~2% residency), offlining E-cores saves 1.7 W and is wire-invisible mid-run, the cliff is TX starvation below 1.6 GHz (1.2 GHz sags to ~18 G, zero errors)

2026-08-19 14:11:19 -07:00

mkimage

Drop goals and open-questions docs, cut PTP from the appliance kernel

2026-08-28 21:59:41 -07:00

phy_test.go

Bugs-evident cleanup: cable-measure failure is fatal with its reason (failed/fail pretend-path and dead relinked plumbing deleted), bcm.command refuses partial SETs (0 or 5 params, stale-DATA trap made structural), BOOT gains a patched-ixgbe check naming stock-driver boots, panel fault rows matched by label and console rows strict on column count, phy.go split into sff/bcm/rollball and main.go's direction machinery into direction.go (pure moves), input events decoded via encoding/binary, gofmt; verified on hardware (20.00G, zero errors, ECD 41m)

2026-08-17 18:56:55 -07:00

phy.go

Appliance image builds and boots for the locked-in hardware: kernel/ixgbe-sff.patch carries the product driver subset (raw I2C + sff_i2c, ETQF steering with rule round-trip, FCoE slot guard — bench-only reg_ops/ts_bench excluded), ported to v7.2 and compile-verified; mkimage applies it or dies, builds IXGBE=y (ICE/I40E out with the ice.pkg blob) and puts ixgbe.allow_unsupported_sfp=1 on the cmdline; load-ixgbe joins the repo anchored to its own path; module bringup ordered after the host checks because a down port's TX_DISABLE powers the PHY core down while the mailbox keeps answering (proven by flipping admin state: down reads 0xffff everywhere, core wakes 0.07-0.24s after the raise) with identify settling 3s bounded; cold-boot condition reproduced on the bench (ports downed first) and passing, confirmed on the appliance

2026-08-18 09:41:38 -07:00

render_test.go

Panel cleanup: chip padding up a step and a tighter chip line gap, corrected/SNR numbers neutral until amber or red, dB margin casing, compound elapsed units (1m39s) ending the minutes/metres m collision, noise chip shows presence only

2026-08-14 23:47:45 -07:00

render.go

Noise ripped out after the bench verdict (tightly-wrapped unshielded cables move victim SNR only 0.2-0.3 dB, recorded in measurement.md dead ends): noise.go deleted with the noiser, phase clock and panel lane; configureNoise/checkPrivFlag gone so the X710 pair is untouched; console noise column, NOISE links rows and lane colors removed; driverPair rehomed to system.go, carrierUp to phy.go; history anchors to run start (frame-rate sampling and draw-clock roll kept — they fix display latency for every lane); goals/hardware/state/nics docs updated; verified on hardware (20.00G, zero errors, ECD 41m, SNR +7.6)

2026-08-17 21:37:18 -07:00

rollball.go

Ghost SNR check removed as asked: no bounds, no panic, no latch re-read — a raw 0x0000 in 1.133-136 decodes as no-margin-this-poll (the untrained resting value, which appears before 1.1 reports a drop, so no read ordering could prove a ghost); panel value columns bounded by scalePanelCount (4 chars forever) with RESET narrowed to 5 chars so counts can never reach the timeline; rows and heat cells stretch proportionally to fill the band between header and footer instead of leaving the bottom half dead

2026-08-17 21:48:29 -07:00

rx.go

Cable verdict folded into one console cell (length green, fault name red, xover amber, fail; diag rule lines and counters-reset line gone), reset fires at the press not diag completion, and the measure's blip is suppressed at every counter source instead of counted-hidden-reverted: worker error adds, loss-window write-offs (quiet slides state without charging; leaving the quiet era presumes in-window holes delivered so the gap's tail is never charged late), NIC poller tracks raw without accumulating, socket drops discarded, modules un-primed by the measure itself; display hold deleted so pre-existing errors no longer blink out during a measure

2026-08-17 11:25:31 -07:00

sff.go

Bugs-evident cleanup: cable-measure failure is fatal with its reason (failed/fail pretend-path and dead relinked plumbing deleted), bcm.command refuses partial SETs (0 or 5 params, stale-DATA trap made structural), BOOT gains a patched-ixgbe check naming stock-driver boots, panel fault rows matched by label and console rows strict on column count, phy.go split into sff/bcm/rollball and main.go's direction machinery into direction.go (pure moves), input events decoded via encoding/binary, gofmt; verified on hardware (20.00G, zero errors, ECD 41m)

2026-08-17 18:56:55 -07:00

sock.go

X520 live as the product NIC: rate buckets re-keyed to read time (one shared clock, one commit per drained batch) with the backward smear as a settled window — each completed bucket enters once, donates excess to the nearest earlier deficits once, locks for display when nothing later can refill it (per-sample recompute double-counted excess and read >20G on a 20G wire); timestamp machinery deleted (ts.go, SO_TIMESTAMPING, rx_filter=ALL check) after the audit confirmed the buckets were its sole consumer; testDriver ixgbe; RollBall mailbox hardened against orphaned-command completions (never sample status before a poll gap, value from the same block read, devad/reg echo required — a killed session's 7.60 read answered a PHY ID as 0x0006); full hardware pass: 20.00G flat both directions, zero errors, ECD 42m, SNR +7.4dB

2026-08-17 09:35:58 -07:00

system.go

Power: 2 streams and a 2.0 GHz clock pinned by a new checkMaxFreq SETTINGS check (performance governor + scaling_max_freq = fixed clock) take line rate from 80 W to 29.7 W package, qualified 96 s zero-loss; measured on the bench: streams buy scheduling headroom not watts (C1-parked threads, lonelier spinners boost harder), relaxed cpu_dma_latency never reaches deep idle (scheduler spreads wakeups, ~2% residency), offlining E-cores saves 1.7 W and is wire-invisible mid-run, the cliff is TX starvation below 1.6 GHz (1.2 GHz sags to ~18 G, zero errors)

2026-08-19 14:11:19 -07:00

text.go

Cut the comments back to the genuinely subtle ones

2026-08-01 16:14:12 -07:00

tx.go

Cable verdict folded into one console cell (length green, fault name red, xover amber, fail; diag rule lines and counters-reset line gone), reset fires at the press not diag completion, and the measure's blip is suppressed at every counter source instead of counted-hidden-reverted: worker error adds, loss-window write-offs (quiet slides state without charging; leaving the quiet era presumes in-window holes delivered so the gap's tail is never charged late), NIC poller tracks raw without accumulating, socket drops discarded, modules un-primed by the measure itself; display hold deleted so pre-existing errors no longer blink out during a measure

2026-08-17 11:25:31 -07:00

ui_test.go

Footer SI prefix joins the unit segment (30.04 bright, M pkts muted; 50.66 + GB) so GB is never split across colours; the 90s ago/now axis line removed as a duplicate of the RESET | LAST 90 S | NOW column headers, rows stretching into the freed space

2026-08-18 09:56:40 -07:00

ui.go

Paint timeline slots spent in a cable measure blue

2026-08-30 16:50:06 -07:00

docs/README.md

cabletest docs

Project knowledge that moves and versions with the repo.

cabletest finds Ethernet cables that misbehave under sustained full-duplex 10 Gbit load. It exchanges raw Ethernet frames (no IP, no ARP) across a variety of sizes and payloads between two directly-cabled ports on one host, and attributes every problem it can to the cable rather than the host. It runs as PID 1 on a dedicated appliance with a framebuffer UI; development happens on the same machine under Arch. The target of measurement is the cable, not throughput — line rate is a means to stress the physical layer.

General:

state.md — where things are now: the box and what's in it, the bench driver, every-boot steps. All current-placement state lives here and only here.
hardware.md — host rules: interface conventions, tuning, how to run.
measurement.md — how to measure correctly; performance results and dead ends.
xdp.md — the AF_XDP datapath experiment: what it proved, what rebuilding it needs, why it can't run on the X520.

Per-device (each directory: device doc + its datasheets):

nics/ — the transport rule, the comparison table, and one directory per NIC: x520, connectx-5, x710, e810.
modules/ — module inventory, media-lies table, the shared IEEE register set, RollBall protocol, cable-length physics, and one directory per module: fs, wiitek, fibergaga.