mention enable-dfa-jump-thread in readme

Cargo.toml

@@ -1,3 +1,11 @@
 [workspace]
 members = ["simdnbt", "simdnbt-derive", "fuzz"]
 resolver = "2"
+
+[profile.release]
+lto = true
+debug = false
+
+[profile.bench]
+lto = true
+debug = true

simdnbt/Cargo.toml

@@ -17,6 +17,7 @@ thiserror = "2.0.12"
 
 [dev-dependencies]
 criterion = { version = "0.5.1", features = ["html_reports"] }
+mimalloc = "0.1.43"
 
 graphite_binary = "0.1.0"
 valence_nbt = { version = "0.8.0", features = ["binary"] }
@@ -24,8 +25,6 @@ fastnbt = "2.5.0"
 azalea-nbt = { git = "https://github.com/azalea-rs/azalea", rev = "84e036ce3752ecf57904b0f5aff1f33d43e95a32" }
 hematite-nbt = { version = "0.5.2", default-features = false }
 shen-nbt5 = "0.4.6"
-
-mimalloc = "0.1.43"
 crab_nbt = "0.2.9"
 ussr-nbt = "0.2.1"
 

simdnbt/README.md

@@ -54,7 +54,11 @@ nbt.write(&mut buffer);
 
 Use the borrow variant of `Nbt` if possible, and avoid allocating unnecessarily (for example, keep strings as `Cow<str>` if you can).
 
-If you're using the owned variant of Simdnbt, switching to a faster allocator like [mimalloc](https://docs.rs/mimalloc/latest/mimalloc/) may help a decent amount (it's ~20% faster on my machine). Setting `RUSTFLAGS='-C target-cpu=native'` when running your code may also help or hurt performance.
+If you're using the owned variant of Simdnbt, switching to a faster allocator like [mimalloc](https://docs.rs/mimalloc/latest/mimalloc/) may help a decent amount (it's ~20% faster on my machine).
+
+Using `RUSTFLAGS="-Cllvm-args=-enable-dfa-jump-thread"` makes Simdnbt about 4% faster in exchange for potentially slightly longer compile times and compiler instability.
+
+Setting `RUSTFLAGS='-C target-cpu=native'` when running your code may help or hurt performance, depending on your computer and program.
 
 ## Implementation details