Commit Graph
38 Commits
Author SHA1 Message Date
MrBruh c5fcfbd46a Kick speed hacks and add bans 2026-10-01 13:12:54 +13:00
MrBruh 71ae1f47bc Fix seventh netcode review 2026-10-01 12:03:52 +13:00
MrBruh e233e1c1bb Fix fifth netcode review 2026-10-01 09:12:07 +13:00
MrBruh 671cc9ed91 Fix fourth netcode review 2026-10-01 06:23:16 +13:00
MrBruh 4f3ce00a8e Fix third netcode review 2026-10-01 04:57:51 +13:00
MrBruh 0d548798e5 Fix second netcode review 2026-10-01 03:54:02 +13:00
MrBruh e7d1ed3a67 Harden and fix the netcode 2026-10-01 01:22:46 +13:00
MrBruh a3ed50c56d Remove the lockstep netcode 2026-09-30 22:12:05 +13:00
MrBruh 4adc3a87a5 Remove matching tooling and HALO_LINUX ifdefs 2026-09-30 21:49:46 +13:00
MrBruh a17d89c6f1 Merge bnunu/halo-1 main 2026-09-30 18:13:05 +13:00
MrBruh 30709dbf0f Fix late joiners missing players and crashing 2026-09-30 13:53:49 +13:00
MrBruh 2ff6516f02 Fix joining games in progress 2026-09-30 03:29:38 +13:00
Jonas Volman f3aadeab98 Reconcile five Halo functions and approved declaration batches
Integrate coherent B3, P2/K17 prerequisites, completed K1-K16 + S, and the independent profile/effects rulings. Five gains and two explicitly approved B3 debits: net +3 strict functions, +2770 meaningful/+2800 padded bytes (+2774 progress credit). No additional losses; Halo objects 394 to 396 and data unchanged.

Clean rebuild, identical 8252-row stable sweep, 68 valid parks, admission/provider checks, 467-unit warning comparison, unchanged fake scan, and 1318 tests passed. Preserve canonical helper recoveries and unrelated work. Full evidence and disclosures in docs/object_matching_logs/astra_five_function_reconciliation_20260928.md.
2026-09-28 23:17:10 -07:00
MrBruh 5c867dc46e Fix teams and player slots for mid-match joins 2026-09-29 15:49:22 +13:00
MrBruh c0c86bda26 Allow joining games in progress 2026-09-29 13:53:13 +13:00
MrBruh 087c1f5685 Improve netcode, add network versions 2026-09-29 11:32:57 +13:00
MrBruh 1bcd57afe4 Merge branch 'bnunu:main' into main 2026-09-28 14:46:17 +13:00
Jonas Volman 709bfc3a3c Reconcile approved frontier ownership and admission packets
Import only independently reviewed production changes. Preserve canonical physics and virtual-keyboard repairs; keep B3 and all unresolved holds out.

Record exactly four approved canary debits (6987 meaningful bytes), six Halo admissions and two revocations. Keep vendor ioinit/storage work separate. Full build, 8252-row sweep, 1318 tests, park/admission/fake audits and 621-object binding pass within approved debit bounds.

Do not publish donor research dumps or private assets. README and unrelated user research remain untouched. See frontier_publication_reconciliation_20260927.md for complete evidence limits and accounting.
2026-09-27 17:12:48 -07:00
Jonas Volman 59757f4263 Reconcile approved type ownership, static locals and source disclosures
Independently gate storage + K, OQ-4 P-a/P-b1, four static-local packets and comment-only disclosures. Preserve all 7641 exact owner rows, 71 valid parks and 1161 passing tests; code/data credit unchanged. G and game_engine admission remain held because current canonical loses populate_statistic_buffer. Existing collision static-local repair is not duplicated. No donor tree, held candidate, scorer or private asset imported.
2026-09-27 05:17:27 -07:00
MrBruh 755f610025 Fix 128-player multiplayer issues 2026-09-27 18:17:24 +13:00
Jonas VolmanandClaude Opus 5.5 5d93d24e87 Networking: keep a host of 128 machines responsive
Found by running a 128-machine session with tools/system_link_bots.py:

- The listen backlog was 32 connections, so machines joining at once were
  refused; it is now the machine limit.
- Socket buffers are at least 1 MB. A host's connection to a machine must
  hold every update the machine has not acknowledged, up to 128 ticks of
  3.9 KB with 128 players; a full buffer blocked the host for the write
  timeout. The host's connection to its own client must also hold all it
  sends in one frame.
- Each machine that finished loading logged a line for every machine
  still loading, 8,000 lines as 128 machines load; it now logs a count.
- Removing a machine dumped the whole server state to the log; when many
  machines left at once the host wrote dumps for minutes.
- Machines that had loaded heard nothing while the others loaded, and a
  client drops a connection that is silent for 15 seconds, less than the
  60 seconds the host waits for slow machines. The host now sends its
  pregame keep-alive every 5 seconds while machines load; a client in game
  ignores it.

All under HALO_LINUX; the MSVC build is unchanged.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-26 10:18:07 -07:00
Jonas VolmanandClaude Opus 5.5 db207ea1a4 Networking: 128 players on up to 128 machines in the native builds
The Xbox game allows 16 players on 4 machines (4 split screen players
each). The native builds raise both limits to 128, keeping 4 players per
machine, and leave the byte-matching build unchanged: every change is
under HALO_LINUX (the MSVC objects compare identical).

- Limits: every local copy of the player and machine limits, and the
  bare 4s and 16s that stood for them, follow halo_port_limits.h. The
  player data array, the machine-to-player table (and its reset) and the
  per-tick update arrays grow with them.
- Records: the copies of struct network_game agree on the larger layout
  and check it against halo_port_limits.h; maximum_player_count becomes
  a byte (128 does not fit a signed char); network_player_is_valid
  checks the machine against the machine limit, not the split screen
  one; a machine's input carries MAXIMUM_LOCAL_PLAYERS actions (16/4
  would have become 128/128).
- Protocol: a message holds at most 4095 bytes (its header has a 12-bit
  length), so the 13 KB game settings record is sent in 3.5 KB pieces
  that clients reassemble; the per-tick update of 128 players (3,857
  bytes) fits. Buffers and the packet codec's limits grow to match, and
  the game search uses protocol version 2 so these builds and the Xbox
  game never see each other's games.
- Connections: the host accepts a client connection per machine (the
  split screen count stood in for it), FD_SETSIZE is 256, stream writes
  send all of a message or drop a peer that stops reading instead of
  corrupting the stream, and the input and update queues are larger.
- Host: lobby changes are sent at most every 250 ms instead of on every
  change (a full lobby would otherwise send ~650 MB), in-game joins wait
  in a queue instead of being dropped, the random name and colour
  pickers stop looping forever once names or colours run out, machines
  get 60 s to load a map, and the debug dump lists only the slots in
  use (128 empty slots took seconds to log).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-26 10:18:07 -07:00
Jonas VolmanandClaude Opus 5.5 f7cd2e722e Batch 4a: admit player_ui and sound_manager with their shared-header packets
player_ui (header prerequisite + admission; reviewer-amended):
- game_engine.h: the opaque game_variant view becomes the recovered January
  layout (universal/ctf/slayer/oddball/king/race variant members, HCEX names
  where January's offsets attest them); member renames across the game
  engine variants, network managers and playlist_profile.
- player_ui.h: player_ui_get_edit_playlist_profile returns struct game_variant *;
  ui_widget_game_data_input_functions adopts it (reviewer patch 07 removes the
  two C4133 warnings the worker's packet introduced).
- player_ui.c drops its partial playlist_profile view; symbols.json names
  player1_profile_path; rejection retired; status Matching.

sound_manager (shared header sound_definitions.h, 18 consumers):
- sound_definition_find_pitch_range_by_pitch takes HCEX `short old_range_index`
  with the /Od 0x877810 single-exit body (prototype and definition together).
- update_channel_for_looping_sound (1360) and sound_set_definition_end (208) are
  exact in the /Od-attested shape; 45 static flags per cachebeta publics; parks
  retired; status Matching.

Gate (full ninja, objdiff 3.3.1): Halo objects 375 -> 377; stable diff +2
strict / 1,568 padded, 0 regressions across both header consumer sets; parks
87/0/0; admission 11/0/4/0; fake scan 26 (unchanged); pytest 1157 passed.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-24 02:21:50 -07:00
Jonas Volman 105d9acaac Reconcile 10K exact-byte Claude batch 2026-09-09 00:07:34 -07:00
Jonas Volman 303c1d55a7 Reconcile Bink UI network HS and sound graphs 2026-09-08 12:49:14 -07:00
Jonas Volman 883e7c0621 Restore defined countdown increment with runtime regression coverage 2026-09-06 17:24:34 -07:00
Jonas Volman 2b2589b5cb Expose verified server idle protocol prerequisites 2026-09-06 17:01:01 -07:00
Jonas Volman 52612df62d Reconstruct server creation and rejection with safe disposal 2026-09-06 16:25:54 -07:00
Jonas Volman 8549cb79af Reconstruct server reset and playlist setup with verified owner contracts 2026-09-06 15:39:13 -07:00
Jonas Volman d574f4ff45 Restore five source-attested network server functions 2026-09-06 11:04:14 -07:00
Jonas Volman 081fec4f88 Reconstruct small network server manager leaves 2026-09-04 05:49:03 -07:00
Jonas Volman fc057d673a Reconcile Opus network server manager reconstruction 2026-09-04 04:17:10 -07:00
Jonas Volman 10e330359a Reconcile Opus network server message handlers 2026-09-04 02:46:43 -07:00
Jonas Volman 4097c72e1f Match network pregame readiness pair 2026-09-04 00:15:19 -07:00
Jonas Volman ec28ac4e29 Advance network server manager reconstruction 2026-08-20 16:31:43 -07:00
Jonas Volman 60c1f7e6f4 Advance network server manager reconstruction 2026-08-20 12:21:33 -07:00
Jonas Volman 018aa03f36 Checkpoint network server manager reconstruction 2026-08-20 11:23:39 -07:00
punpckhdq e830efbd7c Initial commit 2025-11-27 02:27:27 +00:00