Files
firezone/rust/Cargo.toml
Thomas Eizinger 90cf191a7c feat(linux): multi-threaded TUN device operations (#7449)
## Context

At present, we only have a single thread that reads and writes to the
TUN device on all platforms. On Linux, it is possible to open the file
descriptor of a TUN device multiple times by setting the
`IFF_MULTI_QUEUE` option using `ioctl`. Using multi-queue, we can then
spawn multiple threads that concurrently read and write to the TUN
device. This is critical for achieving a better throughput.

## Solution

`IFF_MULTI_QUEUE` is a Linux-only thing and therefore only applies to
headless-client, GUI-client on Linux and the Gateway (it may also be
possible on Android, I haven't tried). As such, we need to first change
our internal abstractions a bit to move the creation of the TUN thread
to the `Tun` abstraction itself. For this, we change the interface of
`Tun` to the following:

- `poll_recv_many`: An API, inspired by tokio's `mpsc::Receiver` where
multiple items in a channel can be batch-received.
- `poll_send_ready`: Mimics the API of `Sink` to check whether more
items can be written.
- `send`: Mimics the API of `Sink` to actually send an item.

With these APIs in place, we can implement various (performance)
improvements for the different platforms.

- On Linux, this allows us to spawn multiple threads to read and write
from the TUN device and send all packets into the same channel. The `Io`
component of `connlib` then uses `poll_recv_many` to read batches of up
to 100 packets at once. This ties in well with #7210 because we can then
use GSO to send the encrypted packets in single syscalls to the OS.
- On Windows, we already have a dedicated recv thread because `WinTun`'s
most-convenient API uses blocking IO. As such, we can now also tie into
that by batch-receiving from this channel.
- In addition to using multiple threads, this API now also uses correct
readiness checks on Linux, Darwin and Android to uphold backpressure in
case we cannot write to the TUN device.

## Configuration

Local testing has shown that 2 threads give the best performance for a
local `iperf3` run. I suspect this is because there is only so much
traffic that a single application (i.e. `iperf3`) can generate. With
more than 2 threads, the throughput actually drops drastically because
`connlib`'s main thread is too busy with lock-contention and triggering
`Waker`s for the TUN threads (which mostly idle around if there are 4+
of them). I've made it configurable on the Gateway though so we can
experiment with this during concurrent speedtests etc.

In addition, switching `connlib` to a single-threaded tokio runtime
further increased the throughput. I suspect due to less task / context
switching.

## Results

Local testing with `iperf3` shows some very promising results. We now
achieve a throughput of 2+ Gbit/s.

```
Connecting to host 172.20.0.110, port 5201
Reverse mode, remote host 172.20.0.110 is sending
[  5] local 100.80.159.34 port 57040 connected to 172.20.0.110 port 5201
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-1.00   sec   274 MBytes  2.30 Gbits/sec
[  5]   1.00-2.00   sec   279 MBytes  2.34 Gbits/sec
[  5]   2.00-3.00   sec   216 MBytes  1.82 Gbits/sec
[  5]   3.00-4.00   sec   224 MBytes  1.88 Gbits/sec
[  5]   4.00-5.00   sec   234 MBytes  1.96 Gbits/sec
[  5]   5.00-6.00   sec   238 MBytes  2.00 Gbits/sec
[  5]   6.00-7.00   sec   229 MBytes  1.92 Gbits/sec
[  5]   7.00-8.00   sec   222 MBytes  1.86 Gbits/sec
[  5]   8.00-9.00   sec   223 MBytes  1.87 Gbits/sec
[  5]   9.00-10.00  sec   217 MBytes  1.82 Gbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec  2.30 GBytes  1.98 Gbits/sec  22247             sender
[  5]   0.00-10.00  sec  2.30 GBytes  1.98 Gbits/sec                  receiver

iperf Done.
```

This is a pretty solid improvement over what is in `main`:

```
Connecting to host 172.20.0.110, port 5201
[  5] local 100.65.159.3 port 56970 connected to 172.20.0.110 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-1.00   sec  90.4 MBytes   758 Mbits/sec  1800    106 KBytes
[  5]   1.00-2.00   sec  93.4 MBytes   783 Mbits/sec  1550   51.6 KBytes
[  5]   2.00-3.00   sec  92.6 MBytes   777 Mbits/sec  1350   76.8 KBytes
[  5]   3.00-4.00   sec  92.9 MBytes   779 Mbits/sec  1800   56.4 KBytes
[  5]   4.00-5.00   sec  93.4 MBytes   783 Mbits/sec  1650   69.6 KBytes
[  5]   5.00-6.00   sec  90.6 MBytes   760 Mbits/sec  1500   73.2 KBytes
[  5]   6.00-7.00   sec  87.6 MBytes   735 Mbits/sec  1400   76.8 KBytes
[  5]   7.00-8.00   sec  92.6 MBytes   777 Mbits/sec  1600   82.7 KBytes
[  5]   8.00-9.00   sec  91.1 MBytes   764 Mbits/sec  1500   70.8 KBytes
[  5]   9.00-10.00  sec  92.0 MBytes   771 Mbits/sec  1550   85.1 KBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec   917 MBytes   769 Mbits/sec  15700             sender
[  5]   0.00-10.00  sec   916 MBytes   768 Mbits/sec                  receiver

iperf Done.
```
2024-12-05 00:18:20 +00:00

211 lines
6.8 KiB
TOML

[workspace]
members = [
"bin-shared",
"connlib/clients/android",
"connlib/clients/apple",
"connlib/clients/shared",
"connlib/model",
"connlib/snownet",
"connlib/tunnel",
"dns-over-tcp",
"gateway",
"gui-client/src-common",
"gui-client/src-tauri",
"headless-client",
"ip-packet",
"logging",
"phoenix-channel",
"relay",
"socket-factory",
"telemetry",
"tests/gui-smoke-test",
"tests/http-test-server",
"tun",
]
resolver = "2"
[workspace.package]
license = "Apache-2.0"
edition = "2021"
[workspace.dependencies]
anyhow = "1.0.93"
arboard = { version = "3.4.0", default-features = false }
async-trait = { version = "0.1", default-features = false }
atomicwrites = "0.4.4"
axum = { version = "0.7.7", default-features = false }
backoff = { version = "0.4", features = ["tokio"] }
base64 = { version = "0.22.1", default-features = false }
bimap = "0.6"
boringtun = { version = "0.6", default-features = false }
bytecodec = "0.4.15"
bytes = { version = "1.9.0", default-features = false }
chrono = { version = "0.4", default-features = false, features = ["std", "clock", "oldtime", "serde"] }
clap = "4.5.21"
derive_more = "1.0.0"
difference = "2.0.0"
dirs = "5.0.1"
divan = "0.1.14"
dns-lookup = "2.0"
domain = { version = "0.10", features = ["serde"] }
either = "1"
env_logger = "0.11.3"
etherparse = "0.16"
futures = { version = "0.3.31", default-features = false }
futures-bounded = "0.2.1"
futures-util = { version = "0.3", default-features = false }
glob = "0.3.1"
hex = "0.4.3"
hex-display = "0.3.0"
hex-literal = "0.4.1"
humantime = "2.1"
ip_network = { version = "0.4", default-features = false }
ip_network_table = { version = "0.2", default-features = false }
itertools = "0.13"
jni = "0.21.1"
keyring = "3.2.1"
known-folders = "1.2.0"
libc = "0.2.150"
log = "0.4"
lru = "0.12.5"
mio = "1.0.1"
native-dialog = "0.7.0"
nix = "0.29.0"
nu-ansi-term = "0.50"
once_cell = "1.17.1"
opentelemetry = "0.26.0"
opentelemetry-otlp = "0.26.0"
opentelemetry_sdk = "0.26.0"
os_info = { version = "3", default-features = false }
output_vt100 = "0.1"
png = "0.17.13"
proptest = "1"
proptest-state-machine = "0.3"
quinn-udp = { version = "0.5.7", features = ["fast-apple-datapath"] }
rand = "0.8.5"
rand_core = "0.6.4"
rangemap = "1.5.1"
reqwest = { version = "0.12.5", default-features = false }
rtnetlink = { version = "0.14.1", default-features = false, features = ["tokio_socket"] }
rustls = { version = "0.23.10", default-features = false, features = ["ring"] }
sadness-generator = "0.6.0"
secrecy = "0.8"
semver = "1.0.22"
sentry = { version = "0.35.0", default-features = false }
sentry-tracing = "0.35.0"
serde = "1.0.215"
serde_json = "1.0.133"
serde_variant = "0.1.3"
sha2 = "0.10.8"
smallvec = "1.13.2"
smbios-lib = "0.9.2"
smoltcp = { version = "0.11", default-features = false }
static_assertions = "1.1.0"
str0m = { version = "0.6.3", default-features = false, features = ["sha1"] }
stun_codec = "0.3.4"
subprocess = "0.2.9"
subtle = "2.5.0"
swift-bridge = "0.1.57"
swift-bridge-build = "0.1.57"
tauri = "2.0.3"
tauri-build = "2.0.1"
tauri-plugin-dialog = "2.0.1"
tauri-plugin-notification = "2.0.1"
tauri-plugin-shell = "2.0.2"
tauri-runtime = "2.1.0"
tauri-utils = "2.0.1"
tempfile = "3.13.0"
test-case = "3.3.1"
test-strategy = "0.4.0"
thiserror = "1.0.68"
time = "0.3.36"
tokio = "1.41"
tokio-stream = "0.1.16"
flume = { version = "0.11.1", features = ["async"] }
tokio-tungstenite = "0.23.1"
tokio-util = "0.7.11"
tracing = { version = "0.1.40" }
tracing-appender = "0.2.3"
tracing-core = "0.1.31"
tracing-log = "0.2.0"
tracing-macros = { git = "https://github.com/tokio-rs/tracing", branch = "v0.1.x" } # Contains `dbg!` but for `tracing`.
tracing-opentelemetry = "0.27.0"
tracing-stackdriver = "0.11.0"
tracing-subscriber = { version = "0.3.17", features = ["parking_lot"] }
trackable = "1.3.0"
url = "2.5.2"
uuid = "1.10.0"
windows = "0.58.0"
winreg = "0.52.0"
zip = { version = "2", default-features = false }
connlib-client-android = { path = "connlib/clients/android" }
connlib-client-apple = { path = "connlib/clients/apple" }
connlib-client-shared = { path = "connlib/clients/shared" }
firezone-bin-shared = { path = "bin-shared" }
firezone-logging = { path = "logging" }
firezone-telemetry = { path = "telemetry" }
firezone-headless-client = { path = "headless-client" }
firezone-gui-client-common = { path = "gui-client/src-common" }
snownet = { path = "connlib/snownet" }
dns-over-tcp = { path = "dns-over-tcp" }
firezone-relay = { path = "relay" }
connlib-model = { path = "connlib/model" }
firezone-tunnel = { path = "connlib/tunnel" }
phoenix-channel = { path = "phoenix-channel" }
ip-packet = { path = "ip-packet" }
socket-factory = { path = "socket-factory" }
tun = { path = "tun" }
socket2 = { version = "0.5" }
[workspace.lints.clippy]
dbg_macro = "warn"
print_stdout = "warn"
print_stderr = "warn"
unnecessary_wraps = "warn"
unused_async = "warn"
wildcard_enum_match_arm = "warn" # Ensures we match on all combinations of `Poll`, preventing erroneous suspensions.
redundant_else = "warn"
redundant_clone = "warn"
unwrap_in_result = "warn"
unwrap_used = "warn"
[workspace.lints.rustdoc]
private-intra-doc-links = "allow" # We don't publish any of our docs but want to catch dead links.
[patch.crates-io]
smoltcp = { git = "https://github.com/smoltcp-rs/smoltcp", branch = "main" }
quinn-udp = { git = "https://github.com/quinn-rs/quinn", branch = "main" }
boringtun = { git = "https://github.com/firezone/boringtun", branch = "master" }
str0m = { git = "https://github.com/algesten/str0m", branch = "main" }
ip_network = { git = "https://github.com/JakubOnderka/ip_network", branch = "master" } # Waiting for release.
ip_network_table = { git = "https://github.com/edmonds/ip_network_table", branch = "some-useful-traits" } # For `Debug` and `Clone`
proptest = { git = "https://github.com/proptest-rs/proptest", branch = "main" }
proptest-state-machine = { git = "https://github.com/proptest-rs/proptest", branch = "main" }
tracing-stackdriver = { git = "https://github.com/thomaseizinger/tracing-stackdriver", branch = "bump-otel-0.26" } # Waiting for release.
sentry = { git = "https://github.com/getsentry/sentry-rust", branch = "master" }
sentry-tracing = { git = "https://github.com/getsentry/sentry-rust", branch = "master" }
# Enforce `tracing-macros` to have released `tracing` version.
[patch.'https://github.com/tokio-rs/tracing']
tracing = "0.1.41"
[profile.release]
# Full link-time optimization. Reduces binaries by up to 3x on some platforms.
lto = "fat"
# Increases the compiler's ability to produce smaller, optimized code
# at the expense of compilation time
codegen-units = 1
# Override build settings just for the GUI client, so we get a pdb/dwp
# Cargo ignores profile settings if they're not in the workspace's Cargo.toml
[profile.dev.package.firezone-gui-client]
debug = "full"
split-debuginfo = "packed"
[profile.release.package.firezone-gui-client]
debug = "full"
split-debuginfo = "packed"