Building MafiaNet¶

Linux / macOS¶

# Configure
mkdir build && cd build
cmake ..

# Build
cmake --build .

# Or for release build
cmake --build . --config Release

Windows (Visual Studio)¶

# Generate Visual Studio 2022 solution
cmake -G "Visual Studio 17 2022" -A x64 -B build

# Build from command line
cmake --build build --config Release

# Or open build/MafiaNet.sln in Visual Studio

Build Options¶

Configure build options with CMake:

Option	Default	Description
`MAFIANET_BUILD_SHARED`	ON	Build shared library (.dll/.so/.dylib)
`MAFIANET_BUILD_STATIC`	ON	Build static library (.lib/.a)
`MAFIANET_BUILD_SAMPLES`	OFF	Build sample applications
`MAFIANET_BUILD_TESTS`	OFF	Build GoogleTest suite (requires `MAFIANET_BUILD_STATIC=ON`)

Example with options:

cmake -DMAFIANET_BUILD_SAMPLES=ON -DMAFIANET_BUILD_TESTS=ON ..

Batched Datagram I/O¶

On Linux, MafiaNet coalesces multiple datagrams into a single recvmmsg(2) / sendmmsg(2) system call instead of one recvfrom/sendto per packet. On a server pushing high packet rates this removes most of the per-datagram syscall overhead; at low rates it changes nothing measurable.

There is nothing to configure. Batching is a platform capability, not a build option: it is always on where the syscalls exist, and every other platform compiles the portable per-datagram paths instead. Delivery semantics are identical either way – the same datagrams arrive, in the same order, with the same reliability, and nothing differs that application code can observe.

Two internals do differ: the number of system calls, and how a transient send failure is reported inside RakNetSocket2::SendBatch. The batched override classifies errno and can report “nothing sent, retry later”; the portable loop cannot read errno portably through Send() and reports every failure as permanent. Either way the datagrams are dropped and the reliability layer resends them.

Measured on the 2560-message reliable-ordered burst in Tests/Integration/MmsgBatchLiveTests.cpp (Linux, Release build, strace -c, median of 3 runs):

Syscall	Per-datagram	Batched
`sendto`	2907	35
`sendmmsg`	0	58
`recvfrom`	2618	0
`recvmmsg`	0	85
Total	5525	178

~31x fewer system calls. Up to MMSG_BATCH_MAX (64) datagrams are coalesced per call.

Counts vary by a percent or two between runs – congestion control decides how many datagrams are ready on each tick – so the ratio is the result, not the exact figures. Reproduce with:

strace -f -c -e trace=sendmmsg,sendto,recvmmsg,recvfrom \
  ./build/Tests/IntegrationTests \
  --gtest_filter='MmsgBatchLive.LargeReliableOrderedBurstArrivesIntactAndInOrder'

The per-datagram column is the code every non-Linux platform runs. To reproduce it on Linux, flip the #if defined(__linux__) batching guards in RakNetSocket2.cpp, RakNetSocket2_Berkley.cpp, ReliabilityLayer.cpp and socket2.h to #if 0.

Running Tests¶

Build with tests enabled, then drive the GoogleTest suites through CTest:

cmake -B build -DMAFIANET_BUILD_TESTS=ON
cmake --build build

ctest --test-dir build --output-on-failure   # everything
ctest --test-dir build -L unit               # hermetic unit suite only
ctest --test-dir build -L integration        # loopback integration suite only
ctest --test-dir build -R "DispatcherLive"   # by name pattern

# Or run a binary directly with a filter (useful for debugging):
./build/Tests/IntegrationTests --gtest_filter='DispatcherLive.*'

See Contributing for the testing guidelines (unit vs. integration, port allocation, condition-based waiting).

Linking Your Project¶

CMake (recommended):

find_package(MafiaNet REQUIRED)
target_link_libraries(your_target MafiaNet::MafiaNet)

Manual linking:

Include path: <install_prefix>/include
Library path: <install_prefix>/lib
Link against: mafianet (shared) or MafiaNetStatic (static)