Add configuration for cargo flamegprah and docs for profiling

.gitignore

@@ -10,4 +10,6 @@
 /target
 /tracker.*
 /tracker.toml
-callgrind.out
+callgrind.out
+flamegraph.svg
+perf.data*

Cargo.toml

@@ -103,3 +103,7 @@ opt-level = 1
 debug = 1
 lto = "fat"
 opt-level = 3
+
+[target.x86_64-unknown-linux-gnu]
+linker = "/usr/bin/clang"
+rustflags = ["-Clink-arg=-fuse-ld=lld", "-Clink-arg=-Wl,--no-rosegment"]

cSpell.json

@@ -18,6 +18,7 @@
         "binstall",
         "Bitflu",
         "bools",
+        "Bragilevsky",
         "bufs",
         "Buildx",
         "byteorder",
@@ -45,10 +46,12 @@
         "dtolnay",
         "elif",
         "filesd",
+        "flamegraph",
         "Freebox",
         "gecos",
         "Grcov",
         "hasher",
+        "heaptrack",
         "hexlify",
         "hlocalhost",
         "Hydranode",
@@ -105,6 +108,7 @@
         "rerequests",
         "ringbuf",
         "rngs",
+        "rosegment",
         "routable",
         "rusqlite",
         "RUSTDOCFLAGS",
@@ -138,11 +142,13 @@
         "uroot",
         "Vagaa",
         "valgrind",
+        "Vitaly",
         "Vuze",
         "Weidendorfer",
         "Werror",
         "whitespaces",
         "XBTT",
+        "Xdebug",
         "Xeon",
         "Xtorrent",
         "Xunlei",

docs/profiling.md

@@ -0,0 +1,59 @@
+# Profiling
+
+## Using flamegraph
+
+```console
+TORRUST_TRACKER_PATH_CONFIG="./share/default/config/tracker.udp.benchmarking.toml" cargo flamegraph --bin=profiling -- 60
+```
+
+![flamegraph](./media/flamegraph.svg)
+
+## Using valgrind and kcachegrind
+
+You need to:
+
+1. Build an run the tracker for profiling.
+2. Make requests to the tracker while it's running.
+
+Build and the binary for profiling:
+
+```console
+RUSTFLAGS='-g' cargo build --release --bin profiling \
+   && export TORRUST_TRACKER_PATH_CONFIG="./share/default/config/tracker.udp.benchmarking.toml" \
+   && valgrind \
+     --tool=callgrind \
+     --callgrind-out-file=callgrind.out \
+     --collect-jumps=yes \
+     --simulate-cache=yes \
+     ./target/release/profiling 60
+```
+
+> NOTICE: You should make requests to the services you want to profile. For example, using the [UDP load test](./benchmarking.md#run-udp-load-test).
+
+After running the tracker with `<valgrind` it generates a file `callgrind.out`
+that you can open with `kcachegrind`.
+
+```console
+kcachegrind callgrind.out
+```
+
+![kcachegrind screenshot](./media/kcachegrind-screenshot.png)
+
+## Links
+
+Profiling tools:
+
+- [valgrind](https://valgrind.org/).
+- [kcachegrind](https://kcachegrind.github.io/).
+- [flamegraph](https://github.com/flamegraph-rs/flamegraph).
+
+Talks about profiling:
+
+- [Profiling Rust Programs with valgrind, heaptrack, and hyperfine](https://www.youtube.com/watch?v=X6Xz4CRd6kw&t=191s).
+- [RustConf 2023 - Profiling async applications in Rust by Vitaly Bragilevsky](https://www.youtube.com/watch?v=8FAdY_0DpkM).
+- [Profiling Code in Rust - by Vitaly Bragilevsky - Rust Linz, December 2022](https://www.youtube.com/watch?v=JRMOIE_wAFk&t=8s).
+- [Xdebug 3 Profiling: 2. KCachegrind tour](https://www.youtube.com/watch?v=h-0HpCblt3A).
+
+## Acknowledgments
+
+Many thanks to [Vitaly Bragilevsky](https://github.com/bravit) and others for sharing the talks about profiling.