feat(core): initialize SQLite off-main-thread (denoland#18401)

This gets SQLite off the flamegraph and reduces initialization time by
somewhere between 0.2ms and 0.5ms. In addition, I took the opportunity
to move all the cache management code to a single place and reduce
duplication. While the PR has a net gain of lines, much of that is just
being a bit more deliberate with how we're recovering from errors.

The existing caches had various policies for dealing with cache
corruption, so I've unified them and tried to isolate the decisions we
make for recovery in a single place (see `open_connection` in
`CacheDB`). The policy I chose was:

 1. Retry twice to open on-disk caches
 2. If that fails, try to delete the file and recreate it on-disk
3. If we fail to delete the file or re-create a new cache, use a
fallback strategy that can be chosen per-cache: InMemory (temporary
cache for the process run), BlackHole (ignore writes, return empty
reads), or Error (fail on every operation).

The caches all use the same general code now, and share the cache
failure recovery policy.

In addition, it cleans up a TODO in the `NodeAnalysisCache`.

cli/cache/caches.rs

@@ -0,0 +1,75 @@
+// Copyright 2018-2023 the Deno authors. All rights reserved. MIT license.
+
+use std::path::PathBuf;
+use std::sync::Arc;
+
+use once_cell::sync::OnceCell;
+
+use super::cache_db::CacheDB;
+use super::cache_db::CacheDBConfiguration;
+use super::check::TYPE_CHECK_CACHE_DB;
+use super::incremental::INCREMENTAL_CACHE_DB;
+use super::node::NODE_ANALYSIS_CACHE_DB;
+use super::parsed_source::PARSED_SOURCE_CACHE_DB;
+use super::DenoDir;
+
+#[derive(Clone, Default)]
+pub struct Caches {
+ fmt_incremental_cache_db: Arc<OnceCell<CacheDB>>,
+ lint_incremental_cache_db: Arc<OnceCell<CacheDB>>,
+ dep_analysis_db: Arc<OnceCell<CacheDB>>,
+ node_analysis_db: Arc<OnceCell<CacheDB>>,
+ type_checking_cache_db: Arc<OnceCell<CacheDB>>,
+}
+
+impl Caches {
+ fn make_db(
+ cell: &Arc<OnceCell<CacheDB>>,
+ config: &'static CacheDBConfiguration,
+ path: PathBuf,
+ ) -> CacheDB {
+ cell
+ .get_or_init(|| CacheDB::from_path(config, path, crate::version::deno()))
+ .clone()
+ }
+
+ pub fn fmt_incremental_cache_db(&self, dir: &DenoDir) -> CacheDB {
+ Self::make_db(
+ &self.fmt_incremental_cache_db,
+ &INCREMENTAL_CACHE_DB,
+ dir.fmt_incremental_cache_db_file_path(),
+ )
+ }
+
+ pub fn lint_incremental_cache_db(&self, dir: &DenoDir) -> CacheDB {
+ Self::make_db(
+ &self.lint_incremental_cache_db,
+ &INCREMENTAL_CACHE_DB,
+ dir.lint_incremental_cache_db_file_path(),
+ )
+ }
+
+ pub fn dep_analysis_db(&self, dir: &DenoDir) -> CacheDB {
+ Self::make_db(
+ &self.dep_analysis_db,
+ &PARSED_SOURCE_CACHE_DB,
+ dir.dep_analysis_db_file_path(),
+ )
+ }
+
+ pub fn node_analysis_db(&self, dir: &DenoDir) -> CacheDB {
+ Self::make_db(
+ &self.node_analysis_db,
+ &NODE_ANALYSIS_CACHE_DB,
+ dir.node_analysis_db_file_path(),
+ )
+ }
+
+ pub fn type_checking_cache_db(&self, dir: &DenoDir) -> CacheDB {
+ Self::make_db(
+ &self.type_checking_cache_db,
+ &TYPE_CHECK_CACHE_DB,
+ dir.type_checking_cache_db_file_path(),
+ )
+ }
+}

cli/cache/check.rs

@@ -1,68 +1,40 @@
 // Copyright 2018-2023 the Deno authors. All rights reserved. MIT license.
 
-use std::path::Path;
-
+use super::cache_db::CacheDB;
+use super::cache_db::CacheDBConfiguration;
+use super::cache_db::CacheFailure;
 use deno_ast::ModuleSpecifier;
 use deno_core::error::AnyError;
 use deno_runtime::deno_webstorage::rusqlite::params;
-use deno_runtime::deno_webstorage::rusqlite::Connection;
 
-use super::common::INITIAL_PRAGMAS;
+pub static TYPE_CHECK_CACHE_DB: CacheDBConfiguration = CacheDBConfiguration {
+ table_initializer: concat!(
+ "CREATE TABLE IF NOT EXISTS checkcache (
+ check_hash TEXT PRIMARY KEY
+ );",
+ "CREATE TABLE IF NOT EXISTS tsbuildinfo (
+ specifier TEXT PRIMARY KEY,
+ text TEXT NOT NULL
+ );",
+ ),
+ on_version_change: concat!(
+ "DELETE FROM checkcache;",
+ "DELETE FROM tsbuildinfo;"
+ ),
+ preheat_queries: &[],
+ // If the cache fails, just ignore all caching attempts
+ on_failure: CacheFailure::Blackhole,
+};
 
 /// The cache used to tell whether type checking should occur again.
 ///
 /// This simply stores a hash of the inputs of each successful type check
 /// and only clears them out when changing CLI versions.
-pub struct TypeCheckCache(Option<Connection>);
+pub struct TypeCheckCache(CacheDB);
 
 impl TypeCheckCache {
- pub fn new(db_file_path: &Path) -> Self {
- log::debug!("Loading type check cache.");
- match Self::try_new(db_file_path) {
- Ok(cache) => cache,
- Err(err) => {
- log::debug!(
- concat!(
- "Failed loading internal type checking cache. ",
- "Recreating...\n\nError details:\n{:#}",
- ),
- err
- );
- // Maybe the cache file is corrupt. Attempt to remove the cache file
- // then attempt to recreate again. Otherwise, use null object pattern.
- match std::fs::remove_file(db_file_path) {
- Ok(_) => match Self::try_new(db_file_path) {
- Ok(cache) => cache,
- Err(err) => {
- log::debug!(
- concat!(
- "Unable to load internal cache for type checking. ",
- "This will reduce the performance of type checking.\n\n",
- "Error details:\n{:#}",
- ),
- err
- );
- Self(None)
- }
- },
- Err(_) => Self(None),
- }
- }
- }
- }
-
- fn try_new(db_file_path: &Path) -> Result<Self, AnyError> {
- let conn = Connection::open(db_file_path)?;
- Self::from_connection(conn, crate::version::deno())
- }
-
- fn from_connection(
- conn: Connection,
- cli_version: &'static str,
- ) -> Result<Self, AnyError> {
- initialize(&conn, cli_version)?;
-
- Ok(Self(Some(conn)))
+ pub fn new(db: CacheDB) -> Self {
+ Self(db)
  }
 
  pub fn has_check_hash(&self, hash: u64) -> bool {
@@ -81,13 +53,10 @@ impl TypeCheckCache {
  }
 
  fn hash_check_hash_result(&self, hash: u64) -> Result<bool, AnyError> {
- let conn = match &self.0 {
- Some(conn) => conn,
- None => return Ok(false),
- };
- let query = "SELECT * FROM checkcache WHERE check_hash=?1 LIMIT 1";
- let mut stmt = conn.prepare_cached(query)?;
- Ok(stmt.exists(params![hash.to_string()])?)
+ self.0.exists(
+ "SELECT * FROM checkcache WHERE check_hash=?1 LIMIT 1",
+ params![hash.to_string()],
+ )
  }
 
  pub fn add_check_hash(&self, check_hash: u64) {
@@ -101,32 +70,24 @@ impl TypeCheckCache {
  }
 
  fn add_check_hash_result(&self, check_hash: u64) -> Result<(), AnyError> {
- let conn = match &self.0 {
- Some(conn) => conn,
- None => return Ok(()),
- };
  let sql = "
  INSERT OR REPLACE INTO
  checkcache (check_hash)
  VALUES
  (?1)";
- let mut stmt = conn.prepare_cached(sql)?;
- stmt.execute(params![&check_hash.to_string(),])?;
+ self.0.execute(sql, params![&check_hash.to_string(),])?;
  Ok(())
  }
 
  pub fn get_tsbuildinfo(&self, specifier: &ModuleSpecifier) -> Option<String> {
- let conn = match &self.0 {
- Some(conn) => conn,
- None => return None,
- };
- let mut stmt = conn
- .prepare_cached("SELECT text FROM tsbuildinfo WHERE specifier=?1 LIMIT 1")
- .ok()?;
- let mut rows = stmt.query(params![specifier.to_string()]).ok()?;
- let row = rows.next().ok().flatten()?;
-
- row.get(0).ok()
+ self
+ .0
+ .query_row(
+ "SELECT text FROM tsbuildinfo WHERE specifier=?1 LIMIT 1",
+ params![specifier.to_string()],
+ |row| Ok(row.get::<_, String>(0)?),
+ )
+ .ok()?
  }
 
  pub fn set_tsbuildinfo(&self, specifier: &ModuleSpecifier, text: &str) {
@@ -145,67 +106,22 @@ impl TypeCheckCache {
  specifier: &ModuleSpecifier,
  text: &str,
  ) -> Result<(), AnyError> {
- let conn = match &self.0 {
- Some(conn) => conn,
- None => return Ok(()),
- };
- let mut stmt = conn.prepare_cached(
+ self.0.execute(
  "INSERT OR REPLACE INTO tsbuildinfo (specifier, text) VALUES (?1, ?2)",
+ params![specifier.to_string(), text],
  )?;
- stmt.execute(params![specifier.to_string(), text])?;
  Ok(())
  }
 }
 
-fn initialize(
- conn: &Connection,
- cli_version: &'static str,
-) -> Result<(), AnyError> {
- // INT doesn't store up to u64, so use TEXT for check_hash
- let query = format!(
- "{INITIAL_PRAGMAS}
- CREATE TABLE IF NOT EXISTS checkcache (
- check_hash TEXT PRIMARY KEY
- );
- CREATE TABLE IF NOT EXISTS tsbuildinfo (
- specifier TEXT PRIMARY KEY,
- text TEXT NOT NULL
- );
- CREATE TABLE IF NOT EXISTS info (
- key TEXT PRIMARY KEY,
- value TEXT NOT NULL
- );
- ",
- );
- conn.execute_batch(&query)?;
-
- // delete the cache when the CLI version changes
- let data_cli_version: Option<String> = conn
- .query_row(
- "SELECT value FROM info WHERE key='CLI_VERSION' LIMIT 1",
- [],
- |row| row.get(0),
- )
- .ok();
- if data_cli_version.as_deref() != Some(cli_version) {
- conn.execute("DELETE FROM checkcache", params![])?;
- conn.execute("DELETE FROM tsbuildinfo", params![])?;
- let mut stmt = conn
- .prepare("INSERT OR REPLACE INTO info (key, value) VALUES (?1, ?2)")?;
- stmt.execute(params!["CLI_VERSION", cli_version])?;
- }
-
- Ok(())
-}
-
 #[cfg(test)]
 mod test {
  use super::*;
 
  #[test]
  pub fn check_cache_general_use() {
- let conn = Connection::open_in_memory().unwrap();
- let cache = TypeCheckCache::from_connection(conn, "1.0.0").unwrap();
+ let conn = CacheDB::in_memory(&TYPE_CHECK_CACHE_DB, "1.0.0");
+ let cache = TypeCheckCache::new(conn);
 
  assert!(!cache.has_check_hash(1));
  cache.add_check_hash(1);
@@ -218,8 +134,9 @@ mod test {
  assert_eq!(cache.get_tsbuildinfo(&specifier1), Some("test".to_string()));
 
  // try changing the cli version (should clear)
- let conn = cache.0.unwrap();
- let cache = TypeCheckCache::from_connection(conn, "2.0.0").unwrap();
+ let conn = cache.0.recreate_with_version("2.0.0");
+ let cache = TypeCheckCache::new(conn);
+
  assert!(!cache.has_check_hash(1));
  cache.add_check_hash(1);
  assert!(cache.has_check_hash(1));
@@ -228,8 +145,9 @@ mod test {
  assert_eq!(cache.get_tsbuildinfo(&specifier1), Some("test".to_string()));
 
  // recreating the cache should not remove the data because the CLI version is the same
- let conn = cache.0.unwrap();
- let cache = TypeCheckCache::from_connection(conn, "2.0.0").unwrap();
+ let conn = cache.0.recreate_with_version("2.0.0");
+ let cache = TypeCheckCache::new(conn);
+
  assert!(cache.has_check_hash(1));
  assert!(!cache.has_check_hash(2));
  assert_eq!(cache.get_tsbuildinfo(&specifier1), Some("test".to_string()));

cli/cache/common.rs

@@ -43,15 +43,3 @@ impl FastInsecureHasher {
  self.0.finish()
  }
 }
-
-/// Disable write-ahead-logging and tweak some other stuff.
-/// We want to favor startup time over cache performance and
-/// creating a WAL is expensive on startup.
-pub static INITIAL_PRAGMAS: &str = "
- PRAGMA journal_mode=OFF;
- PRAGMA synchronous=NORMAL;
- PRAGMA temp_store=memory;
- PRAGMA page_size=4096;
- PRAGMA mmap_size=6000000;
- PRAGMA optimize;
-";