Update dependencies

[ghuls]

No description provided.

[ghuls]

cranelift can also be updated, but requires different code to print IR.
bumpalo can be updated without problems, but there is some new stuff that could be useful: https://github.com/fitzgen/bumpalo/blob/main/CHANGELOG.md

mimalloc could also be added as alternative allocator.

❯  git diff Cargo.toml src/main.rs src/codegen/clif.rs
diff --git a/Cargo.toml b/Cargo.toml
index 19bbb3b..e149875 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -24,6 +24,7 @@ itoa = "1.0"
 ryu = "1.0"
 libc = "0.2"
 tikv-jemallocator = { version = "0.4", optional = true }
+mimalloc = { version = "0.1", default-features = false, optional = true }
 rand = "0.8.4"
 lalrpop-util = "0.19.6"
 unicode-xid = "0.2.2"
@@ -39,13 +40,13 @@ termcolor = "1.1"
 itertools = "0.10"
 assert_cmd = "2.0.3"
 paste = "1.0"
-cranelift = "0.75.0"
-cranelift-codegen = "0.75.0"
-cranelift-frontend = "0.75.0"
-cranelift-module = "0.75.0"
-cranelift-jit = "0.75.0"
+cranelift = "0.80.0"
+cranelift-codegen = "0.80.0"
+cranelift-frontend = "0.80.0"
+cranelift-module = "0.80.0"
+cranelift-jit = "0.80.0"
 fast-float = "0.2"
-bumpalo = { version = "3.6", features = ["collections"] }
+bumpalo = { version = "3.9", features = ["collections"] }
 target-lexicon = "0.12.2"
 
 [dev-dependencies]
@@ -56,6 +57,7 @@ tempfile = "3.3"
 [features]
 default = ["use_jemalloc", "allow_avx2", "llvm_backend", "unstable"]
 use_jemalloc = ["tikv-jemallocator"]
+use_mimalloc = ["mimalloc"]
 # Certain features leverage the AVX2 instruction set, but AVX2 can often make
 # the entire application slightly slower, even on chips that support it. For
 # those cases, consider disabling allow_avx2.
diff --git a/src/codegen/clif.rs b/src/codegen/clif.rs
index 94f534f..4a4a88c 100644
--- a/src/codegen/clif.rs
+++ b/src/codegen/clif.rs
@@ -573,7 +573,7 @@ impl<'a> View<'a> {
         self.builder.seal_all_blocks();
         self.builder.finalize();
         if DUMP_IR {
-            eprintln!("{}", self.builder.display(None));
+            //eprintln!("{}", self.builder.display(None));
         }
         Ok(())
     }
diff --git a/src/main.rs b/src/main.rs
index 75991aa..dc871b6 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -54,6 +54,10 @@ use std::mem;
 #[global_allocator]
 static ALLOC: tikv_jemallocator::Jemalloc = tikv_jemallocator::Jemalloc;
 
+#[cfg(feature = "use_mimalloc")]
+#[global_allocator]
+static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
+
 macro_rules! fail {
     ($($t:tt)*) => {{
         eprintln_ignore!($($t)*);

[ezrosent]

Thanks for the PR! I should have time to review this soon but I expect to merge it in the next couple of days.

[ezrosent]

I think there are a couple more things to clean up here on the LLVM side, but I can push changes post-merge. Thanks again!