uhid: emit per-button Usage items instead of Usage Min/Max range

[VaisVaisov]

Problem

buildFromOutput emitted Usage Minimum (1) / Usage Maximum (N) for the button collection. This causes the kernel to assign evdev codes in a fixed pattern — BTN_GAMEPAD for usages 1-16, then BTN_MISC for 17-32 — regardless of what is declared in [output.buttons]. Extra buttons like M1-M4 ended up on wrong BTN_* codes and were not visible to SDL/Steam.

Fix

Replace the range with one Usage (N) item per button, where N is derived from the button's configured BTN_* code via the kernel's reverse mapping:

Usage range	evdev range
1–16	BTN_GAMEPAD (0x130–0x13F)
17–32	BTN_MISC (0x100–0x10F)
33–48	BTN_MOUSE (0x110–0x11F)
49–64	BTN_JOYSTICK(0x120–0x12F)

Codes outside these ranges (e.g. BTN_TRIGGER_HAPPY) receive the lowest free Usage number so the descriptor remains valid.

Adds btnCodeToHidUsage() to implement the reverse BTN_* → HID Button Page Usage mapping.

Tested

Verified with evtest and Steam on Linux (CachyOS, kernel 7.0.5) using a Flydigi Vader 5 Pro. All buttons including extra paddle/misc buttons now map to the correct BTN_* codes.

Summary by CodeRabbit

• Bug Fixes
  • Improved gamepad descriptor generation to assign per-button HID usages (with safe fallbacks) so controllers report consistent, valid button layouts.
• Chores
  • Standardized Flydigi Vader 5 Pro button mappings for M1–M4, O, C, Z, LM, RM to stable hardware codes for more reliable input behavior.

[qodo-code-review]

ⓘ You've reached your Qodo monthly free-tier limit. Reviews pause until next month — upgrade your plan to continue now, or link your paid account if you already have one.

[coderabbitai]

Warning

Review limit reached

@BANANASJIM, we couldn't start this review because you've used your available PR reviews for now.

Your plan includes 1 review of capacity. Refill in 10 minutes and 54 seconds.

Your organization has run out of usage credits. Purchase more in the billing tab.

⌛ How to resolve this issue?

After more review capacity refills, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans have higher rate limits than trial, open-source, and free plans. In all cases, review capacity refills continuously over time.

Please see our FAQ for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: c3449dd9-56ef-4c15-896f-096bc6751f4e

📥 Commits

Reviewing files that changed from the base of the PR and between 7a0464b and af8c195.

⛔ Files ignored due to path filters (1)

• src/test/fixtures/golden/steam_deck_hid_descriptor.bin is excluded by !**/*.bin

📒 Files selected for processing (3)

• devices/flydigi/vader5.toml
• src/io/uhid_descriptor.zig
• src/test/fixtures/golden/steam_deck_hid_descriptor.annotated.txt

📝 Walkthrough

Walkthrough

Emit HID Button Page usages per-button by resolving configured evdev BTN_* codes via a new mapper, assign fallback usages for unresolved buttons, and update Flydigi Vader5 device output button mappings to explicit BTN_* codes.

Changes

Button Descriptor Per-Usage Resolution

Layer / File(s)	Summary
Import input_codes & per-button descriptor emission src/io/uhid_descriptor.zig	Adds input_codes import and reworks face-button HID emission to resolve each ButtonId to a HID Usage, pack usages in declaration order, track allocated usages with a bitset, fill unresolved slots with lowest free usages (1..64), and emit one Usage (u8) item per active button slot instead of a contiguous Usage Minimum/Maximum range.
HID button code-to-usage mapper src/io/uhid_descriptor.zig	New btnCodeToHidUsage(code: u16) ?u8 helper maps specific Linux BTN_* ranges to HID Button Page usage numbers 1–64 and returns null for unsupported codes.

Flydigi Vader5 device mapping

Layer / File(s)	Summary
Flydigi Vader5 TOML button mappings devices/flydigi/vader5.toml	Replaces BTN_TRIGGER_HAPPY*-based mappings with explicit standard BTN_0–BTN_8 and BTN_4–BTN_7 assignments for multiple output buttons, and removes the commented remapping guidance.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~40 minutes

Possibly related PRs

• BANANASJIM/padctl#132: Modifies src/io/uhid_descriptor.zig and earlier descriptor emission logic related to button usages.
• BANANASJIM/padctl#36: Extends input_codes resolution for BTN_* codes that this PR consumes.

Poem

🐰 I hopped through mappings, one by one,
Turning BTN to HID 'til the job was done.
Usages placed in tidy rows,
Fallbacks found where mystery grows,
A happy hop—descriptor fun!

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: replacing Usage Min/Max range with per-button Usage items in HID button descriptors.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@src/io/uhid_descriptor.zig`:
- Around line 247-269: The button_usages array is declared uninitialized and
later fully iterated, which can cause undefined reads; zero-initialize
button_usages (or explicitly fill it with 0) before the enum walk so any slots
not assigned remain 0 and the subsequent `for (button_usages[0..button_count])`
loop and `next_free` allocation see defined values; update the declaration/early
code around `var button_usages: [64]u8`, `occupied`, and the enum `inline for`
block to ensure `button_usages` is initialized before partial population.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 3e63ff8e-3994-4bd5-afa5-28c814c48593

📥 Commits

Reviewing files that changed from the base of the PR and between fe6166f and e0d45fe.

📒 Files selected for processing (1)

• src/io/uhid_descriptor.zig

[BANANASJIM]

Thanks @VaisVaisov. The per-button Usage approach is the right direction. One thing to fix before it lands.

The paddle Usage numbers are off. padctl emits a Game Pad collection, so the kernel HID_GD_GAMEPAD path is code = usage - 1, then +BTN_GAMEPAD if code <= 0xf else +0x2b0. M1 to M4 (Usage 17 to 20) therefore land on BTN_TRIGGER_HAPPY1..4, but SDL only detects Xbox Elite paddles at BTN_TRIGGER_HAPPY5..8 (SDL_sysjoystick.c). So paddles will not be recognized as written. (The btnCodeToHidUsage comment about a BTN_MISC/MOUSE/JOYSTICK split is also inaccurate; the gamepad page has no such split.)

Minimal fix: map M1 to M4 to Usage 21 to 24 (these resolve to BTN_TRIGGER_HAPPY5..8). Add a BTN_TRIGGER_HAPPY branch to btnCodeToHidUsage, and set the [output.buttons] paddles in vader5.toml to BTN_TRIGGER_HAPPY5..8 (already in input_codes.zig). Also please refresh the stale [report.button_group] comment that still says C/Z/LM/RM are disabled.