gaspersic/solana
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

AcctIdx: introduce BucketApi for access to a specific bucket (#20359)

Browse Source
This commit is contained in:
Jeff Washington (jwash)
2021-10-04 12:48:09 -04:00
committed by GitHub
parent fb8a7cfa92
commit 8da2eb980a
4 changed files with 194 additions and 113 deletions
Download Patch File Download Diff File Expand all files Collapse all files

134
bucket_map/src/bucket_api.rs Normal file
View File

@@ -0,0 +1,134 @@
use crate::bucket::Bucket;
use crate::bucket_item::BucketItem;
use crate::bucket_map::BucketMapError;
use crate::bucket_stats::BucketMapStats;
use crate::{MaxSearch, RefCount};
use solana_sdk::pubkey::Pubkey;
use std::ops::RangeBounds;
use std::path::PathBuf;
use std::sync::Arc;
use std::sync::{RwLock, RwLockWriteGuard};
type LockedBucket<T> = Arc<RwLock<Option<Bucket<T>>>>;
pub struct BucketApi<T: Clone + Copy> {
drives: Arc<Vec<PathBuf>>,
max_search: MaxSearch,
pub stats: Arc<BucketMapStats>,
bucket: LockedBucket<T>,
}
impl<T: Clone + Copy> BucketApi<T> {
pub fn new(
drives: Arc<Vec<PathBuf>>,
max_search: MaxSearch,
stats: Arc<BucketMapStats>,
) -> Self {
Self {
drives,
max_search,
stats,
bucket: Arc::default(),
}
}
/// Get the items for bucket
pub fn items_in_range<R>(&self, range: &Option<&R>) -> Vec<BucketItem<T>>
where
R: RangeBounds<Pubkey>,
{
self.bucket
.read()
.unwrap()
.as_ref()
.map(|bucket| bucket.items_in_range(range))
.unwrap_or_default()
}
/// Get the Pubkeys
pub fn keys(&self) -> Vec<Pubkey> {
self.bucket
.read()
.unwrap()
.as_ref()
.map_or_else(Vec::default, |bucket| bucket.keys())
}
/// Get the values for Pubkey `key`
pub fn read_value(&self, key: &Pubkey) -> Option<(Vec<T>, RefCount)> {
self.bucket.read().unwrap().as_ref().and_then(|bucket| {
bucket
.read_value(key)
.map(|(value, ref_count)| (value.to_vec(), ref_count))
})
}
pub fn bucket_len(&self) -> u64 {
self.bucket
.read()
.unwrap()
.as_ref()
.map(|bucket| bucket.bucket_len())
.unwrap_or_default()
}
pub fn delete_key(&self, key: &Pubkey) {
let mut bucket = self.get_write_bucket();
if let Some(bucket) = bucket.as_mut() {
bucket.delete_key(key)
}
}
fn get_write_bucket(&self) -> RwLockWriteGuard<Option<Bucket<T>>> {
let mut bucket = self.bucket.write().unwrap();
if bucket.is_none() {
*bucket = Some(Bucket::new(
Arc::clone(&self.drives),
self.max_search,
Arc::clone(&self.stats),
));
}
bucket
}
pub fn addref(&self, key: &Pubkey) -> Option<RefCount> {
self.get_write_bucket()
.as_mut()
.and_then(|bucket| bucket.addref(key))
}
pub fn unref(&self, key: &Pubkey) -> Option<RefCount> {
self.get_write_bucket()
.as_mut()
.and_then(|bucket| bucket.unref(key))
}
pub fn insert(&self, pubkey: &Pubkey, value: (&[T], RefCount)) {
let mut bucket = self.get_write_bucket();
bucket.as_mut().unwrap().insert(pubkey, value)
}
pub fn grow(&self, err: BucketMapError) {
let mut bucket = self.get_write_bucket();
bucket.as_mut().unwrap().grow(err)
}
pub fn update<F>(&self, key: &Pubkey, updatefn: F)
where
F: Fn(Option<(&[T], RefCount)>) -> Option<(Vec<T>, RefCount)>,
{
let mut bucket = self.get_write_bucket();
bucket.as_mut().unwrap().update(key, updatefn)
}
pub fn try_write(
&self,
pubkey: &Pubkey,
value: (&[T], RefCount),
) -> Result<(), BucketMapError> {
let mut bucket = self.get_write_bucket();
bucket.as_mut().unwrap().try_write(pubkey, value.0, value.1)
}
}

145
bucket_map/src/bucket_map.rs
View File

@@ -1,17 +1,14 @@
//! BucketMap is a mostly contention free concurrent map backed by MmapMut
use crate::bucket::Bucket;
use crate::bucket_item::BucketItem;
use crate::bucket_api::BucketApi;
use crate::bucket_stats::BucketMapStats;
use crate::{MaxSearch, RefCount};
use solana_sdk::pubkey::Pubkey;
use std::convert::TryInto;
use std::fmt::Debug;
use std::fs;
use std::ops::RangeBounds;
use std::path::PathBuf;
use std::sync::Arc;
use std::sync::{RwLock, RwLockWriteGuard};
use tempfile::TempDir;
#[derive(Debug, Default, Clone)]
@@ -33,10 +30,9 @@ impl BucketMapConfig {
}
pub struct BucketMap<T: Clone + Copy + Debug> {
buckets: Vec<RwLock<Option<Bucket<T>>>>,
buckets: Vec<Arc<BucketApi<T>>>,
drives: Arc<Vec<PathBuf>>,
max_buckets_pow2: u8,
max_search: MaxSearch,
pub stats: Arc<BucketMapStats>,
pub temp_dir: Option<TempDir>,
}
@@ -71,8 +67,6 @@ impl<T: Clone + Copy + Debug> BucketMap<T> {
config.max_buckets.is_power_of_two(),
"Max number of buckets must be a power of two"
);
let mut buckets = Vec::with_capacity(config.max_buckets);
buckets.resize_with(config.max_buckets, || RwLock::new(None));
let stats = Arc::new(BucketMapStats::default());
// this should be <= 1 << DEFAULT_CAPACITY or we end up searching the same items over and over - probably not a big deal since it is so small anyway
const MAX_SEARCH: MaxSearch = 32;
@@ -88,6 +82,15 @@ impl<T: Clone + Copy + Debug> BucketMap<T> {
});
let drives = Arc::new(drives);
let mut buckets = Vec::with_capacity(config.max_buckets);
buckets.resize_with(config.max_buckets, || {
Arc::new(BucketApi::new(
Arc::clone(&drives),
max_search,
Arc::clone(&stats),
))
});
// A simple log2 function that is correct if x is a power of two
let log2 = |x: usize| usize::BITS - x.leading_zeros() - 1;
@@ -96,7 +99,6 @@ impl<T: Clone + Copy + Debug> BucketMap<T> {
drives,
max_buckets_pow2: log2(config.max_buckets) as u8,
stats,
max_search,
temp_dir,
}
}
@@ -112,92 +114,24 @@ impl<T: Clone + Copy + Debug> BucketMap<T> {
self.buckets.len()
}
pub fn bucket_len(&self, ix: usize) -> u64 {
self.buckets[ix]
.read()
.unwrap()
.as_ref()
.map(|bucket| bucket.bucket_len())
.unwrap_or_default()
}
/// Get the items for bucket `ix` in `range`
pub fn items_in_range<R>(&self, ix: usize, range: &Option<&R>) -> Vec<BucketItem<T>>
where
R: RangeBounds<Pubkey>,
{
self.buckets[ix]
.read()
.unwrap()
.as_ref()
.map(|bucket| bucket.items_in_range(range))
.unwrap_or_default()
}
/// Get the Pubkeys for bucket `ix`
pub fn keys(&self, ix: usize) -> Vec<Pubkey> {
self.buckets[ix]
.read()
.unwrap()
.as_ref()
.map_or_else(Vec::default, |bucket| bucket.keys())
}
/// Get the values for Pubkey `key`
pub fn read_value(&self, key: &Pubkey) -> Option<(Vec<T>, RefCount)> {
let ix = self.bucket_ix(key);
self.buckets[ix]
.read()
.unwrap()
.as_ref()
.and_then(|bucket| {
bucket
.read_value(key)
.map(|(value, ref_count)| (value.to_vec(), ref_count))
})
self.get_bucket(key).read_value(key)
}
/// Delete the Pubkey `key`
pub fn delete_key(&self, key: &Pubkey) {
let ix = self.bucket_ix(key);
if let Some(bucket) = self.buckets[ix].write().unwrap().as_mut() {
bucket.delete_key(key);
}
self.get_bucket(key).delete_key(key);
}
/// Update Pubkey `key`'s value with 'value'
pub fn insert(&self, ix: usize, key: &Pubkey, value: (&[T], RefCount)) {
let mut bucket = self.get_bucket(ix);
bucket.as_mut().unwrap().insert(key, value)
}
fn get_bucket(&self, ix: usize) -> RwLockWriteGuard<Option<Bucket<T>>> {
let mut bucket = self.buckets[ix].write().unwrap();
if bucket.is_none() {
*bucket = Some(Bucket::new(
Arc::clone(&self.drives),
self.max_search,
Arc::clone(&self.stats),
));
}
bucket
pub fn insert(&self, key: &Pubkey, value: (&[T], RefCount)) {
self.get_bucket(key).insert(key, value)
}
/// Update Pubkey `key`'s value with 'value'
pub fn try_insert(
&self,
ix: usize,
key: &Pubkey,
value: (&[T], RefCount),
) -> Result<(), BucketMapError> {
let mut bucket = self.get_bucket(ix);
bucket.as_mut().unwrap().try_write(key, value.0, value.1)
}
/// if err is a grow error, then grow the appropriate piece
pub fn grow(&self, ix: usize, err: BucketMapError) {
let mut bucket = self.get_bucket(ix);
bucket.as_mut().unwrap().grow(err);
pub fn try_insert(&self, key: &Pubkey, value: (&[T], RefCount)) -> Result<(), BucketMapError> {
self.get_bucket(key).try_write(key, value)
}
/// Update Pubkey `key`'s value with function `updatefn`
@@ -205,9 +139,15 @@ impl<T: Clone + Copy + Debug> BucketMap<T> {
where
F: Fn(Option<(&[T], RefCount)>) -> Option<(Vec<T>, RefCount)>,
{
let ix = self.bucket_ix(key);
let mut bucket = self.get_bucket(ix);
bucket.as_mut().unwrap().update(key, updatefn)
self.get_bucket(key).update(key, updatefn)
}
pub fn get_bucket(&self, key: &Pubkey) -> &Arc<BucketApi<T>> {
self.get_bucket_from_index(self.bucket_ix(key))
}
pub fn get_bucket_from_index(&self, ix: usize) -> &Arc<BucketApi<T>> {
&self.buckets[ix]
}
/// Get the bucket index for Pubkey `key`
@@ -223,15 +163,15 @@ impl<T: Clone + Copy + Debug> BucketMap<T> {
/// Increment the refcount for Pubkey `key`
pub fn addref(&self, key: &Pubkey) -> Option<RefCount> {
let ix = self.bucket_ix(key);
let mut bucket = self.buckets[ix].write().unwrap();
bucket.as_mut()?.addref(key)
let bucket = &self.buckets[ix];
bucket.addref(key)
}
/// Decrement the refcount for Pubkey `key`
pub fn unref(&self, key: &Pubkey) -> Option<RefCount> {
let ix = self.bucket_ix(key);
let mut bucket = self.buckets[ix].write().unwrap();
bucket.as_mut()?.unref(key)
let bucket = &self.buckets[ix];
bucket.unref(key)
}
}
@@ -247,6 +187,7 @@ mod tests {
use rand::thread_rng;
use rand::Rng;
use std::collections::HashMap;
use std::sync::RwLock;
#[test]
fn bucket_map_test_insert() {
@@ -263,21 +204,21 @@ mod tests {
let key = Pubkey::new_unique();
let config = BucketMapConfig::new(1 << 1);
let index = BucketMap::new(config);
let ix = index.bucket_ix(&key);
let bucket = index.get_bucket(&key);
if pass == 0 {
index.insert(ix, &key, (&[0], 0));
index.insert(&key, (&[0], 0));
} else {
let result = index.try_insert(ix, &key, (&[0], 0));
let result = index.try_insert(&key, (&[0], 0));
assert!(result.is_err());
assert_eq!(index.read_value(&key), None);
if pass == 2 {
// another call to try insert again - should still return an error
let result = index.try_insert(ix, &key, (&[0], 0));
let result = index.try_insert(&key, (&[0], 0));
assert!(result.is_err());
assert_eq!(index.read_value(&key), None);
}
index.grow(ix, result.unwrap_err());
let result = index.try_insert(ix, &key, (&[0], 0));
bucket.grow(result.unwrap_err());
let result = index.try_insert(&key, (&[0], 0));
assert!(result.is_ok());
}
assert_eq!(index.read_value(&key), Some((vec![0], 0)));
@@ -289,9 +230,9 @@ mod tests {
let key = Pubkey::new_unique();
let config = BucketMapConfig::new(1 << 1);
let index = BucketMap::new(config);
index.insert(index.bucket_ix(&key), &key, (&[0], 0));
index.insert(&key, (&[0], 0));
assert_eq!(index.read_value(&key), Some((vec![0], 0)));
index.insert(index.bucket_ix(&key), &key, (&[1], 0));
index.insert(&key, (&[1], 0));
assert_eq!(index.read_value(&key), Some((vec![1], 0)));
}
@@ -465,8 +406,8 @@ mod tests {
let mut r = vec![];
for bin in 0..map.num_buckets() {
r.append(
&mut map
.items_in_range(bin, &None::<&std::ops::RangeInclusive<Pubkey>>),
&mut map.buckets[bin]
.items_in_range(&None::<&std::ops::RangeInclusive<Pubkey>>),
);
}
r
@@ -505,7 +446,7 @@ mod tests {
let insert = thread_rng().gen_range(0, 2) == 0;
maps.iter().for_each(|map| {
if insert {
map.insert(map.bucket_ix(&k), &k, (&v.0, v.1))
map.insert(&k, (&v.0, v.1))
} else {
map.update(&k, |current| {
assert!(current.is_none());
@@ -524,7 +465,7 @@ mod tests {
let insert = thread_rng().gen_range(0, 2) == 0;
maps.iter().for_each(|map| {
if insert {
map.insert(map.bucket_ix(&k), &k, (&v, rc))
map.insert(&k, (&v, rc))
} else {
map.update(&k, |current| {
assert_eq!(current, v_old.map(|(v, rc)| (&v[..], *rc)), "{}", k);

1
bucket_map/src/lib.rs
View File

@@ -1,5 +1,6 @@
#![allow(clippy::integer_arithmetic)]
mod bucket;
pub mod bucket_api;
mod bucket_item;
pub mod bucket_map;
mod bucket_stats;

27
runtime/src/in_mem_accounts_index.rs
View File

@@ -6,6 +6,7 @@ use crate::bucket_map_holder::{Age, BucketMapHolder};
use crate::bucket_map_holder_stats::BucketMapHolderStats;
use rand::thread_rng;
use rand::Rng;
use solana_bucket_map::bucket_api::BucketApi;
use solana_measure::measure::Measure;
use solana_sdk::{clock::Slot, pubkey::Pubkey};
use std::collections::{hash_map::Entry, HashMap};
@@ -28,6 +29,8 @@ pub struct InMemAccountsIndex<T: IndexValue> {
storage: Arc<BucketMapHolder<T>>,
bin: usize,
bucket: Option<Arc<BucketApi<SlotT<T>>>>,
// pubkey ranges that this bin must hold in the cache while the range is present in this vec
pub(crate) cache_ranges_held: CacheRangesHeld,
// true while ranges are being manipulated. Used to keep an async flush from removing things while a range is being held.
@@ -51,6 +54,11 @@ impl<T: IndexValue> InMemAccountsIndex<T> {
map_internal: RwLock::default(),
storage: Arc::clone(storage),
bin,
bucket: storage
.disk
.as_ref()
.map(|disk| disk.get_bucket_from_index(bin))
.map(Arc::clone),
cache_ranges_held: CacheRangesHeld::default(),
stop_flush: AtomicU64::default(),
bin_dirty: AtomicBool::default(),
@@ -111,7 +119,7 @@ impl<T: IndexValue> InMemAccountsIndex<T> {
}
fn load_from_disk(&self, pubkey: &Pubkey) -> Option<(SlotList<T>, RefCount)> {
self.storage.disk.as_ref().and_then(|disk| {
self.bucket.as_ref().and_then(|disk| {
let m = Measure::start("load_disk_found_count");
let entry_disk = disk.read_value(pubkey);
match &entry_disk {
@@ -209,7 +217,7 @@ impl<T: IndexValue> InMemAccountsIndex<T> {
}
fn delete_disk_key(&self, pubkey: &Pubkey) {
if let Some(disk) = self.storage.disk.as_ref() {
if let Some(disk) = self.bucket.as_ref() {
disk.delete_key(pubkey)
}
}
@@ -579,8 +587,8 @@ impl<T: IndexValue> InMemAccountsIndex<T> {
let m = Measure::start("range");
// load from disk
if let Some(disk) = self.storage.disk.as_ref() {
let items = disk.items_in_range(self.bin, range);
if let Some(disk) = self.bucket.as_ref() {
let items = disk.items_in_range(range);
let mut map = self.map().write().unwrap();
let future_age = self.storage.future_age_to_flush();
for item in items {
@@ -668,7 +676,7 @@ impl<T: IndexValue> InMemAccountsIndex<T> {
loop {
let mut removes = Vec::default();
let mut removes_random = Vec::default();
let disk = self.storage.disk.as_ref().unwrap();
let disk = self.bucket.as_ref().unwrap();
let mut updates = Vec::default();
let m = Measure::start("flush_scan");
@@ -706,11 +714,8 @@ impl<T: IndexValue> InMemAccountsIndex<T> {
continue; // marked dirty after we grabbed it above, so handle this the next time this bucket is flushed
}
flush_entries_updated_on_disk += 1;
disk_resize = disk.try_insert(
self.bin,
&k,
(&v.slot_list.read().unwrap(), v.ref_count()),
);
disk_resize =
disk.try_write(&k, (&v.slot_list.read().unwrap(), v.ref_count()));
}
if disk_resize.is_err() {
// disk needs to resize, so mark all unprocessed items as dirty again so we pick them up after the resize
@@ -745,7 +750,7 @@ impl<T: IndexValue> InMemAccountsIndex<T> {
Err(err) => {
// grow the bucket, outside of all in-mem locks.
// then, loop to try again
disk.grow(self.bin, err);
disk.grow(err);
}
}
}