MSP: report the u-blox satellites in MSP_GPSSVINFO, in Betaflight's layout

[MrScothh]

MSP_GPSSVINFO has been a stub in INAV: one channel holding the HDOP. The u-blox driver already keeps the satellites the receiver reports, from NAV-SIG or from NAV-SAT converted, for the gpssats CLI command and the OSD; this sends them, in the layout Betaflight uses for the same message, so one decoder reads both.

The Configurator's GPS tab shows them in iNavFlight/inav-configurator#2819.

What changes

MSP: report the u-blox satellites in MSP_GPSSVINFO, as Betaflight does

• The reply is a count, then 4 bytes per entry: GNSS id, satellite id, quality (bits 0-2 the u-blox quality indicator, bit 3 "used in the navigation solution"), C/N0 in dB-Hz.
• Betaflight tells this layout from its older per-channel one by a count above 16, so a shorter list is padded to 17 entries with GNSS id 255 and satellite id 0, as Betaflight pads its own (it always sends 32).
• Entries the receiver has not filled are left out, and the list stops when the reply buffer is full. Without a u-blox receiver the count is 0.
• The entries are the driver's: one per signal with NAV-SIG, so a dual-band receiver lists a satellite once per band.
• NAV-SAT, which u-blox 8 receivers send instead of NAV-SIG, now sets the "pseudorange used" bit from its own "used" flag, so bit 3 is right with both messages. gpssats shows that bit for those receivers too.
• docs/development/msp/msp_messages.json describes the new reply and README.md is regenerated with src/utils/gen_msp_md.py. The doc version goes to 2.1.2, which I read as a content change inside the current schema; if the changed reply counts as a compatibility change, it is a minor bump instead. GPS: report what the receiver actually has, and allow NavIC where it exists #11979 takes the version to 2.2.0, so whichever of the two merges second needs the number adjusted.

gps_ublox: clear the satellite table when NAV-SIG reports no signal

• A NAV-SIG with no signals left the table as it was, so a receiver that lost every satellite kept reporting the last ones, used flags included: to gpssats, to the OSD's GPS extra stats element, and now to the reply above. NAV-SAT already cleared it.

Tested

• Unit test in src/test/unit/gps_ublox_unittest.cc: the NAV-SAT "used" flag becomes the pseudorange-used bit, and nothing else does.
• SITL with an emulated receiver on UART2, reading MSP_GPSSVINFO and MSP_RAW_GPS every 2 s:
  • M8 sending NAV-SAT: 19 satellites, 10 of them used, count 19.
  • M10 sending NAV-SIG with 6 satellites: count 17, 6 satellites and 11 padding entries.
  • The same M10 sending no signals while it keeps its fix: without the second commit the 6 satellites stay in the reply for as long as I watched (40 s); with it the list is empty from the next NAV-SIG on, while the fix and its 16 satellites stay.
  • Receiver disconnected: count 17, all padding.
• Real receivers (comment of 2026-10-06): a NEO-F10N on a TBS Lucid H7 Wing with a 3D fix, 30 entries matching gpssats signal by signal; a SAM-M10Q on a Holybro Kakute H7 Mini indoors with nothing tracked, 17 padding entries.
• Size against maintenance-10.x: flash +112 B on MATEKF722SE (98.46 %), +160 B on MATEKF405, +176 B on KAKUTEF7 and KAKUTEF7HDV, +192 B on MATEKH743. RAM and ITCM unchanged (KAKUTEF7HDV stays at 16264 B of ITCM); the linker shows 8 B more RAM on MATEKF722SE, which is alignment, since the RAM symbols are the same.

Test merges against the open PRs: the only conflict is the doc version with #11979.

[qodo-code-review]

ⓘ Qodo reviews are paused because the subscription is no longer active. Ask your workspace admin to reactivate the subscription to resume reviews. Manage billing

[qodo-free-for-open-source-projects]

PR Summary by Qodo

Report u-blox satellites through Betaflight-compatible MSP_GPSSVINFO

✨ Enhancement 🐞 Bug fix 🧪 Tests 📝 Documentation 🕐 20-40 Minutes

AI Description

• Replace the GPS satellite-info stub with u-blox signal data in Betaflight’s layout for shared
 decoding.
• Clear stale NAV-SIG entries and preserve NAV-SAT used flags so reported satellites remain
 accurate.
• Document the new reply and test NAV-SAT flag conversion.

Diagram

graph TD
  receiver["u-blox receiver"] --> sig["NAV-SIG parser"] --> table["Satellite table"] --> msp["MSP_GPSSVINFO"] --> client["GPS tab"]
  receiver --> sat["NAV-SAT converter"] --> table --> display["CLI and OSD"]

Loading

High-Level Assessment

The following are alternative approaches to this PR:

1. Add a separate MSP satellite command

• ➕ Preserves the existing MSP_GPSSVINFO reply for legacy consumers.
• ➖ Requires clients to implement another command instead of sharing Betaflight’s decoder.
• ➖ Adds protocol surface for a reply that was previously only a stub.

Recommendation: Reuse MSP_GPSSVINFO as proposed: its previous reply was a stub, and matching Betaflight lets clients share a decoder. Confirm whether changing that reply warrants a minor rather than patch documentation-version bump.

Files changed (6) +99 / -57

Enhancement (1) +23 / -8

fc_msp.cSerialize u-blox signals in MSP_GPSSVINFO +23/-8

Serialize u-blox signals in MSP_GPSSVINFO

• Writes each available signal’s GNSS ID, satellite ID, quality and used bit, and C/N0, stopping at buffer capacity. Pads u-blox replies shorter than 17 entries for Betaflight layout detection; non-u-blox receivers return a zero count.

src/main/fc/fc_msp.c

Bug fix (2) +12 / -11

gps_ublox.cClear the signal table on empty NAV-SIG reports +8/-10

Clear the signal table on empty NAV-SIG reports

• Runs the unused-entry clearing pass even when NAV-SIG reports zero signals. This prevents stale satellites from remaining visible to MSP, the CLI, and the OSD.

src/main/io/gps_ublox.c

gps_ublox_utils.cCarry NAV-SAT’s used flag into converted signal records +4/-1

Carry NAV-SAT’s used flag into converted signal records

• Maps the NAV-SAT satellite-used flag to the NAV-SIG pseudorange-used bit. M8 receivers can therefore report the used state through existing signal consumers and MSP.

src/main/io/gps_ublox_utils.c

Tests (1) +21 / -0

gps_ublox_unittest.ccTest NAV-SAT used-flag conversion +21/-0

Test NAV-SAT used-flag conversion

• Adds a test confirming that conversion preserves satellite fields and sets the pseudorange-used bit only when NAV-SAT marks the satellite as used.

src/test/unit/gps_ublox_unittest.cc

Documentation (2) +43 / -38

README.mdDocument the satellite-info reply +9/-9

Document the satellite-info reply

• Replaces the stub description with the count and four-byte per-signal fields. Explains Betaflight-compatible padding, u-blox message sources, and the previous reply.

docs/development/msp/README.md

msp_messages.jsonSpecify the variable-length MSP_GPSSVINFO payload +34/-29

Specify the variable-length MSP_GPSSVINFO payload

• Changes the reply schema from HDOP stub fields to repeating satellite records and marks its length as variable. Bumps the specification version from 2.1.1 to 2.1.2.

docs/development/msp/msp_messages.json

[qodo-free-for-open-source-projects]

Code Review by Qodo

🐞 Bugs (0) 📘 Rule violations (0) 📎 Requirement gaps (0)

Great, no issues found!

Qodo reviewed your code and found no material issues that require review

Tip of the day

💡 Did you know, you can keep summaries lean with Findings visible per group, which tucks the rest behind a View link

More tips ↗ | Customize Qodo ↗ | Qodo docs ↗

[github-actions]

RAM / Flash usage vs. base commit e87050f — commit 15ce6d0

Using the nearest available size baseline — the PR's exact base commit has no stored baseline yet.

Target	Flash Δ	RAM Δ
MATEKF405	+112 B (+0.02%)	CCM: ±0 B (±0.00%) RAM: +8 B (+0.01%)
MATEKF722	+184 B (+0.04%)	ITCM_RAM: ±0 B (±0.00%) RAM: ±0 B (±0.00%) TCM: ±0 B (±0.00%)
MATEKF765	+280 B (+0.04%)	DTCM_RAM: ±0 B (±0.00%) SRAM1: ±0 B (±0.00%)
MATEKH743	+208 B (+0.03%)	D2_RAM: ±0 B (±0.00%) DTCM_RAM: ±0 B (±0.00%) ITCM_RAM: ±0 B (±0.00%) RAM: ±0 B (±0.00%)

See RAM/flash optimization guide for techniques to reduce usage.

[github-actions]

Test firmware build ready — commit 15ce6d0

Download firmware for PR #12091

251 targets built. Find your board's .hex file by name on that page (e.g. MATEKF405SE.hex). Files are individually downloadable — no GitHub login required.

Development build for testing only. Use Full Chip Erase when flashing.

[MrScothh]

Tested today on a real NEO-F10N (SPGL1L5 6.00) on a TBS Lucid H7 Wing, this branch merged on the current maintenance-10.x, with a 3D fix on 10 satellites, reading MSP_GPSSVINFO from the host and gpssats from the CLI seconds apart.

The reply held 30 entries, every one with a satellite id, 14 to 16 of them flagged used against the 10 satellites of MSP_RAW_GPS: the list is per signal, so a satellite tracked on L1 C/A and L5 comes twice, which is what the Configurator side (#2819) folds into one row. The quality byte matched NAV-SIG as gpssats prints it, with "used" in bit 3: satellite 10 on L5, 19 dBHz, quality 4, used (0x0C); satellite 13 on L1, 9 dBHz, quality 3, not used (0x03); satellite 14 on L5 still searching (0x01, C/N0 0) and on L1 at 15 dBHz, quality 4, not used (0x04). Strengths agreed within 1 to 2 dBHz between the two reads.

An M10 (SAM-M10Q) on a Holybro Kakute H7 Mini, indoors without a fix and tracking nothing: the reply holds 17 padding entries (GNSS id 0xFF, satellite id 0), the count above 16 that tells Betaflight's tools the new layout apart, and gpssats lists nothing either. M8 and M9 not tested here.