Auto merge of #83682 - bjorn3:mmap_wrapper, r=cjgillot

Add an Mmap wrapper to rustc_data_structures

This wrapper implements StableAddress and falls back to directly reading the file on wasm32.

Taken from #83640, which I will close due to the perf regression.

Cargo.lock

@@ -3812,7 +3812,6 @@ dependencies = [
  "itertools 0.9.0",
  "jobserver",
  "libc",
- "memmap2",
  "pathdiff",
  "rustc_apfloat",
  "rustc_ast",
@@ -3847,6 +3846,7 @@ dependencies = [
  "jobserver",
  "libc",
  "measureme",
+ "memmap2",
  "parking_lot",
  "rustc-hash",
  "rustc-rayon",
@@ -4146,7 +4146,6 @@ name = "rustc_metadata"
 version = "0.0.0"
 dependencies = [
  "libc",
- "memmap2",
  "rustc_ast",
  "rustc_attr",
  "rustc_data_structures",

compiler/rustc_codegen_cranelift/Cargo.lock

@@ -240,15 +240,6 @@ dependencies = [
  "libc",
 ]
 
-[[package]]
-name = "memmap2"
-version = "0.2.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "04e3e85b970d650e2ae6d70592474087051c11c54da7f7b4949725c5735fbcc6"
-dependencies = [
- "libc",
-]
-
 [[package]]
 name = "object"
 version = "0.23.0"
@@ -319,7 +310,6 @@ dependencies = [
  "gimli",
  "indexmap",
  "libloading",
- "memmap2",
  "object",
  "smallvec",
  "target-lexicon",

compiler/rustc_codegen_cranelift/Cargo.toml

@@ -22,7 +22,6 @@ ar = { git = "https://github.com/bjorn3/rust-ar.git", branch = "do_not_remove_cg
 indexmap = "1.0.2"
 libloading = { version = "0.6.0", optional = true }
 smallvec = "1.6.1"
-memmap2 = "0.2.1"
 
 # Uncomment to use local checkout of cranelift
 #[patch."https://github.com/bytecodealliance/wasmtime/"]

compiler/rustc_codegen_cranelift/src/metadata.rs

@@ -1,11 +1,11 @@
 //! Reading and writing of the rustc metadata for rlibs and dylibs
 
 use std::fs::File;
-use std::ops::Deref;
 use std::path::Path;
 
 use rustc_codegen_ssa::METADATA_FILENAME;
-use rustc_data_structures::owning_ref::{OwningRef, StableAddress};
+use rustc_data_structures::memmap::Mmap;
+use rustc_data_structures::owning_ref::OwningRef;
 use rustc_data_structures::rustc_erase_owner;
 use rustc_data_structures::sync::MetadataRef;
 use rustc_middle::middle::cstore::{EncodedMetadata, MetadataLoader};
@@ -17,26 +17,13 @@
 
 pub(crate) struct CraneliftMetadataLoader;
 
-struct StableMmap(memmap2::Mmap);
-
-impl Deref for StableMmap {
-    type Target = [u8];
-
-    fn deref(&self) -> &[u8] {
-        &*self.0
-    }
-}
-
-unsafe impl StableAddress for StableMmap {}
-
 fn load_metadata_with(
     path: &Path,
     f: impl for<'a> FnOnce(&'a [u8]) -> Result<&'a [u8], String>,
 ) -> Result<MetadataRef, String> {
     let file = File::open(path).map_err(|e| format!("{:?}", e))?;
-    let data = unsafe { memmap2::MmapOptions::new().map_copy_read_only(&file) }
-        .map_err(|e| format!("{:?}", e))?;
-    let metadata = OwningRef::new(StableMmap(data)).try_map(f)?;
+    let data = unsafe { Mmap::map(file) }.map_err(|e| format!("{:?}", e))?;
+    let metadata = OwningRef::new(data).try_map(f)?;
     return Ok(rustc_erase_owner!(metadata.map_owner_box()));
 }
 

compiler/rustc_codegen_ssa/Cargo.toml

@@ -11,7 +11,6 @@ test = false
 bitflags = "1.2.1"
 cc = "1.0.1"
 itertools = "0.9"
-memmap2 = "0.2.1"
 tracing = "0.1"
 libc = "0.2.50"
 jobserver = "0.1.11"

compiler/rustc_codegen_ssa/src/back/lto.rs

@@ -2,6 +2,7 @@
 use crate::traits::*;
 use crate::ModuleCodegen;
 
+use rustc_data_structures::memmap::Mmap;
 use rustc_errors::FatalError;
 
 use std::ffi::CString;
@@ -93,7 +94,7 @@ pub fn cost(&self) -> u64 {
 pub enum SerializedModule<M: ModuleBufferMethods> {
     Local(M),
     FromRlib(Vec<u8>),
-    FromUncompressedFile(memmap2::Mmap),
+    FromUncompressedFile(Mmap),
 }
 
 impl<M: ModuleBufferMethods> SerializedModule<M> {

compiler/rustc_codegen_ssa/src/back/write.rs

@@ -10,6 +10,7 @@
 use crate::traits::*;
 use jobserver::{Acquired, Client};
 use rustc_data_structures::fx::FxHashMap;
+use rustc_data_structures::memmap::Mmap;
 use rustc_data_structures::profiling::SelfProfilerRef;
 use rustc_data_structures::profiling::TimingGuard;
 use rustc_data_structures::profiling::VerboseTimingGuard;
@@ -1958,7 +1959,7 @@ pub fn submit_pre_lto_module_to_llvm<B: ExtraBackendMethods>(
         .unwrap_or_else(|e| panic!("failed to open bitcode file `{}`: {}", bc_path.display(), e));
 
     let mmap = unsafe {
-        memmap2::Mmap::map(&file).unwrap_or_else(|e| {
+        Mmap::map(file).unwrap_or_else(|e| {
             panic!("failed to mmap bitcode file `{}`: {}", bc_path.display(), e)
         })
     };

compiler/rustc_data_structures/Cargo.toml

@@ -36,3 +36,6 @@ features = ["nightly"]
 
 [target.'cfg(windows)'.dependencies]
 winapi = { version = "0.3", features = ["fileapi", "psapi"] }
+
+[target.'cfg(not(target_arch = "wasm32"))'.dependencies]
+memmap2 = "0.2.1"

compiler/rustc_data_structures/src/lib.rs

@@ -84,6 +84,7 @@ macro_rules! unlikely {
 pub mod stable_map;
 pub mod svh;
 pub use ena::snapshot_vec;
+pub mod memmap;
 pub mod sorted_map;
 pub mod stable_set;
 #[macro_use]

compiler/rustc_data_structures/src/memmap.rs

@@ -0,0 +1,47 @@
+use std::fs::File;
+use std::io;
+use std::ops::Deref;
+
+use crate::owning_ref::StableAddress;
+
+/// A trivial wrapper for [`memmap2::Mmap`] that implements [`StableAddress`].
+#[cfg(not(target_arch = "wasm32"))]
+pub struct Mmap(memmap2::Mmap);
+
+#[cfg(target_arch = "wasm32")]
+pub struct Mmap(Vec<u8>);
+
+#[cfg(not(target_arch = "wasm32"))]
+impl Mmap {
+    #[inline]
+    pub unsafe fn map(file: File) -> io::Result<Self> {
+        memmap2::Mmap::map(&file).map(Mmap)
+    }
+}
+
+#[cfg(target_arch = "wasm32")]
+impl Mmap {
+    #[inline]
+    pub unsafe fn map(mut file: File) -> io::Result<Self> {
+        use std::io::Read;
+
+        let mut data = Vec::new();
+        file.read_to_end(&mut data)?;
+        Ok(Mmap(data))
+    }
+}
+
+impl Deref for Mmap {
+    type Target = [u8];
+
+    #[inline]
+    fn deref(&self) -> &[u8] {
+        &*self.0
+    }
+}
+
+// SAFETY: On architectures other than WASM, mmap is used as backing storage. The address of this
+// memory map is stable. On WASM, `Vec<u8>` is used as backing storage. The `Mmap` type doesn't
+// export any function that can cause the `Vec` to be re-allocated. As such the address of the
+// bytes inside this `Vec` is stable.
+unsafe impl StableAddress for Mmap {}

compiler/rustc_metadata/Cargo.toml

@@ -11,7 +11,6 @@ doctest = false
 libc = "0.2"
 snap = "1"
 tracing = "0.1"
-memmap2 = "0.2.1"
 smallvec = { version = "1.6.1", features = ["union", "may_dangle"] }
 rustc_middle = { path = "../rustc_middle" }
 rustc_attr = { path = "../rustc_attr" }

compiler/rustc_metadata/src/locator.rs

@@ -216,6 +216,7 @@
 use crate::rmeta::{rustc_version, MetadataBlob, METADATA_HEADER};
 
 use rustc_data_structures::fx::{FxHashMap, FxHashSet};
+use rustc_data_structures::memmap::Mmap;
 use rustc_data_structures::owning_ref::OwningRef;
 use rustc_data_structures::svh::Svh;
 use rustc_data_structures::sync::MetadataRef;
@@ -232,7 +233,6 @@
 
 use snap::read::FrameDecoder;
 use std::io::{Read, Result as IoResult, Write};
-use std::ops::Deref;
 use std::path::{Path, PathBuf};
 use std::{cmp, fmt, fs};
 use tracing::{debug, info, warn};
@@ -727,19 +727,6 @@ fn find_commandline_library(&mut self) -> Result<Option<Library>, CrateError> {
     }
 }
 
-/// A trivial wrapper for `Mmap` that implements `StableDeref`.
-struct StableDerefMmap(memmap2::Mmap);
-
-impl Deref for StableDerefMmap {
-    type Target = [u8];
-
-    fn deref(&self) -> &[u8] {
-        self.0.deref()
-    }
-}
-
-unsafe impl stable_deref_trait::StableDeref for StableDerefMmap {}
-
 fn get_metadata_section(
     target: &Target,
     flavor: CrateFlavor,
@@ -779,11 +766,11 @@ fn get_metadata_section(
             // mmap the file, because only a small fraction of it is read.
             let file = std::fs::File::open(filename)
                 .map_err(|_| format!("failed to open rmeta metadata: '{}'", filename.display()))?;
-            let mmap = unsafe { memmap2::Mmap::map(&file) };
+            let mmap = unsafe { Mmap::map(file) };
             let mmap = mmap
                 .map_err(|_| format!("failed to mmap rmeta metadata: '{}'", filename.display()))?;
 
-            rustc_erase_owner!(OwningRef::new(StableDerefMmap(mmap)).map_owner_box())
+            rustc_erase_owner!(OwningRef::new(mmap).map_owner_box())
         }
     };
     let blob = MetadataBlob::new(raw_bytes);