src/README.md: map the source folder, its threads and its locks

src/README.md

@@ -0,0 +1,112 @@
+<!--
+### Copyright (c) 2026
+
+Author(s):
+* mcfnord
+* The Jamulus Development Team
+
+This program is free software: you can redistribute it and/or modify
+it under the terms of the GNU Affero General Public License as published by
+the Free Software Foundation, either version 3 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU Affero General Public License for more details.
+
+You should have received a copy of the GNU Affero General Public License
+along with this program.  If not, see <https://www.gnu.org/licenses/>.
+-->
+
+# Main Jamulus codebase
+
+Licensed under the AGPL 3.0 or any later version; full text in [../COPYING](../COPYING).
+
+This directory contains the main code of Jamulus.
+
+Code used by both client and server:
+
+- [main.cpp](main.cpp) parses the command line and constructs a `CClient` or a `CServer`.
+- [protocol.cpp](protocol.cpp) — `CProtocol`: message framing, acknowledgement and
+  retransmission. Wire format: [../docs/JAMULUS_PROTOCOL.md](../docs/JAMULUS_PROTOCOL.md).
+- [channel.cpp](channel.cpp) — `CChannel`: the connection and its receive jitter buffer, used by
+  both client and server.
+- [socket.cpp](socket.cpp) — `CSocket` and `CHighPrioSocket`: the UDP socket shared by all
+  sending and receiving, with its receive thread.
+- [buffer.h](buffer.h) — `CNetBuf` and `CNetBufWithStats`: the jitter buffer, including the
+  automatic size decision.
+- [util.h](util.h) / [util.cpp](util.cpp) — `CHighPrecisionTimer`, the server's frame clock, and
+  assorted helpers.
+
+Client only:
+
+- [client.cpp](client.cpp) — `CClient`: the client's audio path and its one channel.
+- [sound/](sound/) — the sound layer, one backend per platform. See [sound/README.md](sound/README.md).
+- [clientdlg.cpp](clientdlg.cpp), [clientsettingsdlg.cpp](clientsettingsdlg.cpp),
+  [audiomixerboard.cpp](audiomixerboard.cpp), [connectdlg.cpp](connectdlg.cpp),
+  [chatdlg.cpp](chatdlg.cpp) — the GUI.
+- [clientrpc.cpp](clientrpc.cpp) — the client half of the JSON-RPC API.
+
+Server only:
+
+- [server.cpp](server.cpp) — `CServer`: the channels and the mix.
+- [serverlist.cpp](serverlist.cpp) — directory registration and the server list.
+- [recorder/](recorder/) — `CJamController` and `CJamRecorder`.
+- [serverlogging.cpp](serverlogging.cpp) — the connection log.
+- [serverrpc.cpp](serverrpc.cpp) — the server half of the JSON-RPC API.
+- [serverdlg.cpp](serverdlg.cpp) — the server GUI.
+
+The JSON-RPC API ([rpcserver.cpp](rpcserver.cpp), [clientrpc.cpp](clientrpc.cpp),
+[serverrpc.cpp](serverrpc.cpp)) is documented in [../docs/JSON-RPC.md](../docs/JSON-RPC.md).
+
+## Jamulus Architecture
+
+Jamulus is a client/server system: each client encodes the audio from its sound device and sends
+it to the server over UDP, and the server decodes every client's stream, builds a separate mix for
+each connected client from that client's own fader gains, re-encodes it and sends it back to be
+decoded and played out. Each receiving end holds arriving packets in a jitter buffer first.
+
+### Threading
+
+| thread | exists | started from | what runs on it |
+|---|---|---|---|
+| Qt main thread | always | — | the GUI; every protocol message body, parsed and created, on client and server; directory registration; JSON-RPC; and the server's complete frame cycle (see below) |
+| `CSocketThread` | always | `CHighPrioSocket::Start()`, at `QThread::TimeCriticalPriority` | a blocking UDP receive loop. Audio packets are decoded into the jitter buffer synchronously, in `CChannel::PutAudioData` (client) or `CServer::PutAudioData` (server). Protocol messages are split across the two threads: `CProtocol::ParseMessageFrame` validates the frame here, then the body is re-emitted as a queued signal and `ParseMessageBody` runs it on the main thread. |
+| audio driver threads | client | the sound driver | the backend callback, which runs `CClient::AudioCallback`: Opus decode of the received stream, Opus encode of the sound card input, and the UDP send of the encoded packet |
+| `CHighPrecisionTimer` | server, except on Windows | `CHighPrecisionTimer::Start()`, at `QThread::TimeCriticalPriority` | only `emit timeout()` once per frame, plus the absolute-time sleep that paces it |
+| `CThreadPool` workers | server with `--multithreading`, on more than one core | `CServer`'s constructor | Opus decode and mix/encode/send work, in per-block chunks handed out by `CServer::OnTimer` |
+| recorder thread | server with recording | `CJamController` | `CJamRecorder`, fed by queued `AudioFrame` signals from the frame cycle |
+| `QThreadPool` global pool | client GUI | the connect dialog | one task per listed server for the ping/info fan-out (`QtConcurrent::run`) |
+
+**The server's frame cycle runs on the main thread.** The `CHighPrecisionTimer` thread only
+emits `timeout()`; the slot behind that queued connection, `CServer::OnTimer`, does the jitter
+buffer drain, decode, mix, encode and transmit — on the main thread. The TODO in
+[util.cpp](util.cpp) notes the same escape from the timer thread. On Windows the pacer is a
+plain `QTimer`, also main thread. With `--multithreading` the heavy blocks go to the pool, but
+`OnTimer` waits for them — and on a machine reporting one core, `CServer`'s constructor turns the
+option back off, so no pool thread is created at all.
+
+### Locks
+
+The locks taken from more than one thread:
+
+| lock | protects | taken from |
+|---|---|---|
+| `CChannel::MutexSocketBuf` | the jitter buffer | put on `CSocketThread`; get from the client's driver callback or the server's frame cycle; re-init from the main thread |
+| `CSocket::Mutex` | the send path of the shared UDP socket | every `SendPacket()` call: the driver callback (client), the frame cycle and pool workers (server), and protocol code on the main thread |
+| `CServer::Mutex` | connect and disconnect of channels against the frame cycle | `CServer::OnTimer` holds it while it collects the connected channels and drains and decodes their jitter buffers, and releases it before mix and send; `CServer::PutAudioData` (`CSocketThread`) and the protocol slots (main thread) take it too |
+| `CChannel::Mutex` | per-channel state: the enable flag, gain and pan tables, name | setters in protocol slots on the main thread; getters in the server's frame cycle |
+| `CChannel::MutexConvBuf` | the send-side conversion buffer | `PrepAndSendPacket()` on the sending thread; re-init from the main thread |
+
+**Smaller locks:**
+
+- `CProtocol::Mutex` — queue of sent but not yet acknowledged messages
+- `CServer::MutexChanOrder` — channel allocation in `FindChannel` and `FreeChannel`
+- `CServer::MutexWelcomeMessage` — the welcome message string. Taken in `OnNewConnection` and
+  `SetWelcomeMessage` only; the other readers (`OnCLReqServerFeatures`, `OnCLReqWelcomeMessage`,
+  `GetWelcomeMessage`) do not take it, and every one of them is reached on the main thread.
+- `CClient::MutexChannels` — client-side channel number map
+- `CClient::MutexGainOrPan` — gain/pan message rate limiter
+- `CClient::MutexDriverReinit` — serializes sound device re-initialization
+- Sound layer locks (`MutexAudioProcessCallback`, `MutexDevProperties`, per-backend) — see [sound/README.md](sound/README.md)