Commit Graph
100 Commits
Author SHA1 Message Date
flamingcow 4e5ba0bccf Sharpen MCIA open question from mlx5_ifc.h register format; drop offline-rig and instrumentation-flow framing 2026-08-12 21:37:02 -07:00
flamingcow 65b1e18004 Dissolve resolved analyses into standalone device docs; open-questions.md keeps only the four live problems; discard NIC-timestamp length path (BCM ECD is proven) and Aquantia length trivia (30m part, oracle-only) 2026-08-12 21:30:16 -07:00
flamingcow 8b021d186b Add full transcription of Marvell Alaska 88E151x public datasheet (168pp, 89 register tables) as CUX3610 family reference 2026-08-12 21:15:05 -07:00
flamingcow 1afe36a3c8 Remove WX1820: never had the hardware, evaluation was paper-only 2026-08-12 21:10:38 -07:00
flamingcow b6e0fe8978 Add NIC datasheets: 82599 rev3.5, 710-series v4.1, E810 full datasheet, ConnectX-5 user manual + mlx5 PRM 2026-08-12 21:09:08 -07:00
flamingcow 9cb82ca3fa Re-transcribe BCM84891L MDIO command and SFP register excerpts in full (all 45 commands, complete serial-ID table); scrub doc-provenance attribution and label all our interjections as field notes 2026-08-12 21:02:30 -07:00
flamingcow 5588ebec6f Resolve X520 open questions on hardware: ETQF raw-L2 steering and single-hold compound I2C proven; 82599 has no all-packet exact-timestamp path so it is disqualified as product NIC (diagnostics-only), counter route rejected for a cable tester 2026-08-12 20:36:31 -07:00
flamingcow 910379e614 Correct Wiitek danger analysis: whole 3.0x8000+/31.0xF000+ window is lethal, not just *.??64 -- 88Q2220 recipe is IN the danger zone; record live baseline and no-safe-VCT-path finding 2026-08-12 19:50:15 -07:00
flamingcow ad9bc9fae4 BCM ECD recovered from OpenBCM SDK and proven: 1E.4006 start, 1.A896 verdicts, 1.A897-9A per-pair meters -- meter-accurate on a 45m cable, healthy pairs report length 2026-08-12 19:37:54 -07:00
flamingcow c57816345f Restructure docs: per-device directories under modules/ and nics/ with co-located datasheets, state centralized in state.md, prose tightened into lists and tables 2026-08-12 19:23:02 -07:00
flamingcow dc1fa6538d Record FS probe sweep: AutogrEEEn found and forced off (defensive at bringup), identity pair map, FR count as metric, firmware DATA1 reliability caveat 2026-08-12 18:41:32 -07:00
flamingcow ff64b59fe0 Transcribe BCM PDFs to markdown visually (tables intact); catalog probe candidates: EEE advertised in reference bringup, EMI cmds are fast-retrain enables, 1588/loopback/ECD chapters missing 2026-08-12 18:33:10 -07:00
flamingcow 4023d7ccf0 Add FS/BCM capability inventory; drop SNR/power-backoff length proxies -- length is PHY features or NIC timestamps only 2026-08-12 18:23:28 -07:00
flamingcow 01e00dfffe BCM path proven on X520: fix multi-byte i2c read (SDA release), GET_SNR is 0x8030 invoked bare, IEEE SNR regs dead on BCM, module pause forced off 2026-08-12 17:50:14 -07:00
flamingcow 6f787c0873 Record X520 install: FS+Wiitek seated and linked, FS's honest EEPROM is what trips ixgbe qualification, uplink moved to enp88s0 2026-08-12 17:35:25 -07:00
flamingcow 5f4a8ceb57 Vendor AQR datasheet, correct timestamp model (hw rx stamps are a hard requirement), order ConnectX-5, map 82599 bucket-timestamp routes 2026-08-10 12:50:57 -07:00
flamingcow 73074040e0 Answer open questions from desk analysis: ETQF steering, E820 is post-FEC, compound I2C op, forced co-location 2026-08-10 11:51:39 -07:00
flamingcow 8c8e0c7d32 Add open-questions doc for pre-hardware deep-dive 2026-08-10 11:31:19 -07:00
flamingcow 44e010caa9 Note Wiitek ixgbe RollBall client and flag ixgbe RX-steering as open investigation 2026-08-10 11:29:05 -07:00
flamingcow 3f1bb03e52 Record X520 ixgbe raw-i2c patch and BCM client prep in state doc 2026-08-10 11:21:49 -07:00
flamingcow 6597a11c33 Add in-repo project docs: goals, state, hardware, module-PHY transports, module diagnostics, and measurement lessons 2026-08-10 10:34:26 -07:00
flamingcow 6d73221123 Hold a fatal error on the framebuffer and reboot in five minutes rather than exiting, since init's death drops the kernel's panic screen over the cause 2026-08-05 21:46:15 -07:00
flamingcow cd259556d4 Build i40e into the appliance kernel, whose forbidden list predated the noise pair that needs it 2026-08-05 21:46:15 -07:00
flamingcow e7c73ca0a9 Reveal the build's commit hash in place of the reset label while a finger holds the top-right corner, and dev where go run left no stamp 2026-08-05 16:13:20 -07:00
flamingcow 2fcab83756 Drive a noise cable on the i40e pair, cycling its link down and up since the module PHYs ignore everything softer, and flag its absence beside the error columns 2026-08-05 10:49:23 -07:00
flamingcow 88f4f7da80 Poll the nic on the sampler's cadence, and reboot rather than exit when a clean return reaches PID 1 2026-08-05 08:18:44 -07:00
flamingcow ab60cc1cfc Cap cpu idle exit latency for the life of the run, deep enough to keep C1 and shallow enough to block the rest 2026-08-04 22:50:22 -07:00
flamingcow 49eaa8a8a8 Take the rate epoch from the slowest stream's frontier, since reading a bucket claims every stream has delivered through it 2026-08-04 22:38:10 -07:00
flamingcow 7b2601a02e Fail loudly on a touch read, stop the workers before their sockets close, and file down the small edges 2026-08-04 22:26:07 -07:00
flamingcow 12cef6a89b Fold the shutdown flags into one, take the smallest frame to the wire minimum, drop the speed column 2026-08-04 22:12:07 -07:00
flamingcow 19b0f1d2ea Refuse a filter location something else holds, and the driver's word on how many it wrote 2026-08-04 22:12:03 -07:00
flamingcow f5d05d1fc5 Fail loudly on a drm read instead of leaving the panel frozen on its last frame 2026-08-04 22:11:58 -07:00
flamingcow dd94f4cab5 Set the socket buffers through the force options alone instead of falling back to the capped ones 2026-08-04 22:01:10 -07:00
flamingcow 4c7062d600 Take the line rate as a constant so nothing has to be plugged in, and drop the config box 2026-08-04 21:58:52 -07:00
flamingcow 01d673dc83 Check the header version on arrival and say what the error window actually spans 2026-08-04 21:29:03 -07:00
flamingcow 49de171ad0 Commit the receive counters once per batch instead of once per frame 2026-08-04 21:25:04 -07:00
flamingcow 21788fdd4f Judge a frame against the stream its socket receives rather than the one its header claims 2026-08-04 21:19:08 -07:00
flamingcow 51467231ee Fill what a rounded rectangle contains outright as spans instead of pixel by pixel 2026-08-04 21:09:12 -07:00
flamingcow ce44fe9866 Walk the control buffer by hand instead of allocating a message list per frame 2026-08-04 20:48:17 -07:00
flamingcow 1d54a1a1f6 Checksum the header and refuse sequence numbers the sender never sent 2026-08-04 20:41:16 -07:00
flamingcow 703039587b Bucket received frames by their mac receive stamp and read one bucket back 2026-08-04 20:04:30 -07:00
flamingcow 8db5c29882 Cover the loss window, rate window, error taxonomy and formatters with tests 2026-08-04 15:46:12 -07:00
flamingcow 63196eb691 Mirror the panel in the console table as one summed row in two sections 2026-08-04 14:02:53 -07:00
flamingcow fbfcabda7e Draw every panel figure from the latest sample instead of holding the totals 2026-08-04 13:49:21 -07:00
flamingcow 4e60757937 Rate from the newest bucket, stamped when the counters are actually read 2026-08-04 13:45:16 -07:00
flamingcow 2035d42734 Fold the leftover formatters into the SI ones and stop reallocating the draw scratch 2026-08-04 13:37:19 -07:00
flamingcow f5a6420cfa Drop the fatal flag now that every failed check is fatal 2026-08-04 13:33:08 -07:00
flamingcow 39c27166dc Hardcode the settled sizes, streams and batch instead of carrying flags for them 2026-08-04 13:31:14 -07:00
flamingcow 05e1958710 Cut the drm package comments back to what is not evident from the code 2026-08-04 13:23:41 -07:00
flamingcow 4fd3abaf25 Move the shared drm ioctl abi into internal/drm 2026-08-04 13:21:50 -07:00
flamingcow 66bfb88dd7 Make every failed check and unreadable counter fatal instead of degrading 2026-08-04 13:03:07 -07:00
flamingcow fb6a0d2c1b Push the new origin into the sample ring so a reset cannot underflow the totals 2026-08-04 12:59:50 -07:00
flamingcow a285b8999c Drop the counter padding that was never buying a cache line 2026-08-04 12:53:53 -07:00
flamingcow 06592cbc81 Stop counting a partly filled send vector as a failure 2026-08-04 12:45:12 -07:00
flamingcow 1f9b5ea62e Look the expected checksum up instead of carrying it in the frame 2026-08-04 12:45:04 -07:00
flamingcow 03b7ff4954 Drop the write-only direction fields and the unused pattern names 2026-08-04 12:26:10 -07:00
flamingcow 21b20d0071 Scale integer counts on the panel like every other figure 2026-08-04 11:27:35 -07:00
flamingcow 49a1a2f829 Read rx errors from the driver stats array and add the pcs counters 2026-08-04 11:27:30 -07:00
flamingcow 5eea6856fb Cut the comments back to the genuinely subtle ones 2026-08-01 16:14:12 -07:00
flamingcow 726c4c2445 Rework the panel onto SI figures, labelled chips and one spacing scale 2026-08-01 16:07:44 -07:00
flamingcow de2ff13eb6 Collapse errors into lost, corrupt, link and internal 2026-07-31 17:10:25 -07:00
flamingcow 3c8637f931 Show recent errors as labelled chips and rebuild the reset button to match 2026-07-31 17:05:31 -07:00
flamingcow f512391ee4 Capture the panel by reading the displayed drm buffer instead of fbdev 2026-07-31 16:46:11 -07:00
flamingcow 1c5bda37bf Drive the panel through drm and page flip between four scanout buffers 2026-07-31 16:46:11 -07:00
flamingcow 57d7c91df1 Sample the rate buckets on their own clock instead of the display refresh 2026-07-31 16:15:16 -07:00
flamingcow 5d2dbffb4a Add a harness that runs cabletest on a timer and captures the panel to png 2026-07-31 16:15:16 -07:00
flamingcow c94b7ae33d Show median bucket rates, slide the error window, and poll sysfs off the draw loop 2026-07-31 15:45:03 -07:00
flamingcow cf204f5a23 Add mkimage to build and install the image from the latest signed kernel tag 2026-07-26 17:07:09 -07:00
flamingcow ab0a8c8759 Take ns/m to 4.85, measured against a 6m cable 2026-07-26 15:51:52 -07:00
flamingcow 5676b2b581 Detect mounts by mount root, wait for the touchscreen, and default to eth0/eth1 2026-07-26 15:51:34 -07:00
flamingcow 832a2b42d6 Mount the filesystems we need so this can run as PID 1 2026-07-26 10:54:19 -07:00
flamingcow 5e2b36bdaa Clear staticcheck, gopls and go.mod warnings 2026-07-26 10:33:38 -07:00
flamingcow 102d53fb9b Turn the display portrait and split it into live and since-reset panels 2026-07-26 10:27:41 -07:00
flamingcow 4d102f377e Take the cable zero from the shortest delay seen rather than a flag 2026-07-26 09:43:03 -07:00
flamingcow e77574712d Cycle every payload pattern per packet and always run both directions 2026-07-25 23:03:31 -07:00
flamingcow 0f16edfe61 Count link and syscall failures instead of printing them 2026-07-25 22:54:04 -07:00
flamingcow 1b322bb32b Measure cable length from hardware transmit and receive timestamps 2026-07-25 22:38:52 -07:00
flamingcow 09411aebdf Reset everything from a held on-screen button or space 2026-07-25 21:41:21 -07:00
flamingcow b2441ba64a Redraw at 60Hz with adaptive rate smoothing, console output at 1Hz 2026-07-25 21:30:37 -07:00
flamingcow 12958ced56 Take the console into graphics mode while drawing 2026-07-25 21:30:37 -07:00
flamingcow 892e855207 Group the display into instant and cumulative sections on a shared grid 2026-07-25 19:48:15 -07:00
flamingcow 7a9567da55 Clear the framebuffer on exit 2026-07-25 19:42:45 -07:00
flamingcow 419bdc4bb7 Render the live status to the framebuffer in Atkinson Hyperlegible Mono 2026-07-25 19:41:53 -07:00
flamingcow 16416bc1b0 Draw the counter reset as a labelled rule across the table 2026-07-25 19:31:03 -07:00
flamingcow 269199366d Reset cumulative error counts on space 2026-07-25 19:29:00 -07:00
flamingcow abb8b81919 Remove CPU pinning flags that measured slower than the scheduler 2026-07-25 19:24:10 -07:00
flamingcow 32c49ccc4b Widen rate thresholds so interval jitter is not flagged 2026-07-25 19:17:25 -07:00
flamingcow ed0f1d11e6 Default to seven steered streams across the full frame size range 2026-07-25 19:15:20 -07:00
flamingcow 7e77d22f36 Give each stream its own ethertype steered to its own rx queue 2026-07-25 19:08:11 -07:00
flamingcow 0af94219a7 Show cumulative error counts in the streaming rows 2026-07-25 18:45:47 -07:00
flamingcow e0781b3334 Default to worker counts and frame sizes that measure the cable, not the host 2026-07-25 18:40:37 -07:00
flamingcow 7c09c34564 Stream forever with eviction-based per-stream loss detection 2026-07-25 18:33:36 -07:00
flamingcow 2516d908b9 Remove unmanaged check; nothing else will own these interfaces 2026-07-25 18:23:01 -07:00
flamingcow 36711522a2 Drop binary gitignore now that go run is the only entry point 2026-07-25 18:21:53 -07:00
flamingcow 3ba68952bb Always color output and fold settled values into constants 2026-07-25 18:20:34 -07:00
flamingcow 29c6ce0b56 Drop drift detection; host tuning is one-shot system startup 2026-07-25 18:18:00 -07:00
flamingcow 773724368e Always correct host settings instead of gating fixes behind flags 2026-07-25 18:16:34 -07:00
flamingcow 27f89d9916 Render output as colored tables with pass/fail verdict 2026-07-25 18:13:58 -07:00
flamingcow 7f671a262e Check and correct host tuning at startup and on drift 2026-07-25 18:07:25 -07:00
flamingcow befd732406 Add full-duplex raw-Ethernet 10G cable stress tester 2026-07-25 17:58:54 -07:00