[performance] overhaul struct (+ result) caching library for simplicity, performance and multiple-result lookups (#2535)

* rewrite cache library as codeberg.org/gruf/go-structr, implement in gotosocial

* use actual go-structr release version (not just commit hash)

* revert go toolchain changes (damn you go for auto changing this)

* fix go mod woes

* ensure %w is used in calls to errs.Appendf()

* fix error checking

* fix possible panic

* remove unnecessary start/stop functions, move to main Cache{} struct, add note regarding which caches require start/stop

* fix copy-paste artifact... 😇

* fix all comment copy-paste artifacts

* remove dropID() function, now we can just use slices.DeleteFunc()

* use util.Deduplicate() instead of collate(), move collate to util

* move orderByIDs() to util package and "generify"

* add a util.DeleteIf() function, use this to delete entries on failed population

* use slices.DeleteFunc() instead of util.DeleteIf() (i had the logic mixed up in my head somehow lol)

* add note about how collate differs from deduplicate
This commit is contained in:
kim 2024-01-19 12:57:29 +00:00 committed by GitHub
commit 7ec1e1332e
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
66 changed files with 4038 additions and 2711 deletions

30
internal/cache/ap.go vendored
View file

@ -1,30 +0,0 @@
// GoToSocial
// Copyright (C) GoToSocial Authors admin@gotosocial.org
// SPDX-License-Identifier: AGPL-3.0-or-later
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package cache
type APCaches struct{}
// Init will initialize all the ActivityPub caches in this collection.
// NOTE: the cache MUST NOT be in use anywhere, this is not thread-safe.
func (c *APCaches) Init() {}
// Start will attempt to start all of the ActivityPub caches, or panic.
func (c *APCaches) Start() {}
// Stop will attempt to stop all of the ActivityPub caches, or panic.
func (c *APCaches) Stop() {}

View file

@ -18,8 +18,9 @@
package cache
import (
"time"
"github.com/superseriousbusiness/gotosocial/internal/cache/headerfilter"
"github.com/superseriousbusiness/gotosocial/internal/gtsmodel"
"github.com/superseriousbusiness/gotosocial/internal/log"
)
@ -49,198 +50,59 @@ type Caches struct {
func (c *Caches) Init() {
log.Infof(nil, "init: %p", c)
c.GTS.Init()
c.Visibility.Init()
// Setup cache invalidate hooks.
// !! READ THE METHOD COMMENT
c.setuphooks()
c.initAccount()
c.initAccountNote()
c.initApplication()
c.initBlock()
c.initBlockIDs()
c.initBoostOfIDs()
c.initDomainAllow()
c.initDomainBlock()
c.initEmoji()
c.initEmojiCategory()
c.initFollow()
c.initFollowIDs()
c.initFollowRequest()
c.initFollowRequestIDs()
c.initInReplyToIDs()
c.initInstance()
c.initList()
c.initListEntry()
c.initMarker()
c.initMedia()
c.initMention()
c.initNotification()
c.initPoll()
c.initPollVote()
c.initPollVoteIDs()
c.initReport()
c.initStatus()
c.initStatusFave()
c.initTag()
c.initThreadMute()
c.initStatusFaveIDs()
c.initTombstone()
c.initUser()
c.initWebfinger()
c.initVisibility()
}
// Start will start both the GTS and AP cache collections.
// Start will start any caches that require a background
// routine, which usually means any kind of TTL caches.
func (c *Caches) Start() {
log.Infof(nil, "start: %p", c)
c.GTS.Start()
c.Visibility.Start()
tryUntil("starting *gtsmodel.Webfinger cache", 5, func() bool {
return c.GTS.Webfinger.Start(5 * time.Minute)
})
}
// Stop will stop both the GTS and AP cache collections.
// Stop will stop any caches that require a background
// routine, which usually means any kind of TTL caches.
func (c *Caches) Stop() {
log.Infof(nil, "stop: %p", c)
c.GTS.Stop()
c.Visibility.Stop()
}
// setuphooks sets necessary cache invalidation hooks between caches,
// as an invalidation indicates a database INSERT / UPDATE / DELETE.
// NOTE THEY ARE ONLY CALLED WHEN THE ITEM IS IN THE CACHE, SO FOR
// HOOKS TO BE CALLED ON DELETE YOU MUST FIRST POPULATE IT IN THE CACHE.
func (c *Caches) setuphooks() {
c.GTS.Account().SetInvalidateCallback(func(account *gtsmodel.Account) {
// Invalidate account ID cached visibility.
c.Visibility.Invalidate("ItemID", account.ID)
c.Visibility.Invalidate("RequesterID", account.ID)
// Invalidate this account's
// following / follower lists.
// (see FollowIDs() comment for details).
c.GTS.FollowIDs().InvalidateAll(
">"+account.ID,
"l>"+account.ID,
"<"+account.ID,
"l<"+account.ID,
)
// Invalidate this account's
// follow requesting / request lists.
// (see FollowRequestIDs() comment for details).
c.GTS.FollowRequestIDs().InvalidateAll(
">"+account.ID,
"<"+account.ID,
)
// Invalidate this account's block lists.
c.GTS.BlockIDs().Invalidate(account.ID)
})
c.GTS.Block().SetInvalidateCallback(func(block *gtsmodel.Block) {
// Invalidate block origin account ID cached visibility.
c.Visibility.Invalidate("ItemID", block.AccountID)
c.Visibility.Invalidate("RequesterID", block.AccountID)
// Invalidate block target account ID cached visibility.
c.Visibility.Invalidate("ItemID", block.TargetAccountID)
c.Visibility.Invalidate("RequesterID", block.TargetAccountID)
// Invalidate source account's block lists.
c.GTS.BlockIDs().Invalidate(block.AccountID)
})
c.GTS.EmojiCategory().SetInvalidateCallback(func(category *gtsmodel.EmojiCategory) {
// Invalidate any emoji in this category.
c.GTS.Emoji().Invalidate("CategoryID", category.ID)
})
c.GTS.Follow().SetInvalidateCallback(func(follow *gtsmodel.Follow) {
// Invalidate follow request with this same ID.
c.GTS.FollowRequest().Invalidate("ID", follow.ID)
// Invalidate any related list entries.
c.GTS.ListEntry().Invalidate("FollowID", follow.ID)
// Invalidate follow origin account ID cached visibility.
c.Visibility.Invalidate("ItemID", follow.AccountID)
c.Visibility.Invalidate("RequesterID", follow.AccountID)
// Invalidate follow target account ID cached visibility.
c.Visibility.Invalidate("ItemID", follow.TargetAccountID)
c.Visibility.Invalidate("RequesterID", follow.TargetAccountID)
// Invalidate source account's following
// lists, and destination's follwer lists.
// (see FollowIDs() comment for details).
c.GTS.FollowIDs().InvalidateAll(
">"+follow.AccountID,
"l>"+follow.AccountID,
"<"+follow.AccountID,
"l<"+follow.AccountID,
"<"+follow.TargetAccountID,
"l<"+follow.TargetAccountID,
">"+follow.TargetAccountID,
"l>"+follow.TargetAccountID,
)
})
c.GTS.FollowRequest().SetInvalidateCallback(func(followReq *gtsmodel.FollowRequest) {
// Invalidate follow with this same ID.
c.GTS.Follow().Invalidate("ID", followReq.ID)
// Invalidate source account's followreq
// lists, and destinations follow req lists.
// (see FollowRequestIDs() comment for details).
c.GTS.FollowRequestIDs().InvalidateAll(
">"+followReq.AccountID,
"<"+followReq.AccountID,
">"+followReq.TargetAccountID,
"<"+followReq.TargetAccountID,
)
})
c.GTS.List().SetInvalidateCallback(func(list *gtsmodel.List) {
// Invalidate all cached entries of this list.
c.GTS.ListEntry().Invalidate("ListID", list.ID)
})
c.GTS.Media().SetInvalidateCallback(func(media *gtsmodel.MediaAttachment) {
if *media.Avatar || *media.Header {
// Invalidate cache of attaching account.
c.GTS.Account().Invalidate("ID", media.AccountID)
}
if media.StatusID != "" {
// Invalidate cache of attaching status.
c.GTS.Status().Invalidate("ID", media.StatusID)
}
})
c.GTS.Poll().SetInvalidateCallback(func(poll *gtsmodel.Poll) {
// Invalidate all cached votes of this poll.
c.GTS.PollVote().Invalidate("PollID", poll.ID)
// Invalidate cache of poll vote IDs.
c.GTS.PollVoteIDs().Invalidate(poll.ID)
})
c.GTS.PollVote().SetInvalidateCallback(func(vote *gtsmodel.PollVote) {
// Invalidate cached poll (contains no. votes).
c.GTS.Poll().Invalidate("ID", vote.PollID)
// Invalidate cache of poll vote IDs.
c.GTS.PollVoteIDs().Invalidate(vote.PollID)
})
c.GTS.Status().SetInvalidateCallback(func(status *gtsmodel.Status) {
// Invalidate status ID cached visibility.
c.Visibility.Invalidate("ItemID", status.ID)
for _, id := range status.AttachmentIDs {
// Invalidate each media by the IDs we're aware of.
// This must be done as the status table is aware of
// the media IDs in use before the media table is
// aware of the status ID they are linked to.
//
// c.GTS.Media().Invalidate("StatusID") will not work.
c.GTS.Media().Invalidate("ID", id)
}
if status.BoostOfID != "" {
// Invalidate boost ID list of the original status.
c.GTS.BoostOfIDs().Invalidate(status.BoostOfID)
}
if status.InReplyToID != "" {
// Invalidate in reply to ID list of original status.
c.GTS.InReplyToIDs().Invalidate(status.InReplyToID)
}
if status.PollID != "" {
// Invalidate cache of attached poll ID.
c.GTS.Poll().Invalidate("ID", status.PollID)
}
})
c.GTS.StatusFave().SetInvalidateCallback(func(fave *gtsmodel.StatusFave) {
// Invalidate status fave ID list for this status.
c.GTS.StatusFaveIDs().Invalidate(fave.StatusID)
})
c.GTS.User().SetInvalidateCallback(func(user *gtsmodel.User) {
// Invalidate local account ID cached visibility.
c.Visibility.Invalidate("ItemID", user.AccountID)
c.Visibility.Invalidate("RequesterID", user.AccountID)
})
tryUntil("stopping *gtsmodel.Webfinger cache", 5, c.GTS.Webfinger.Stop)
}
// Sweep will sweep all the available caches to ensure none
@ -250,30 +112,30 @@ func (c *Caches) setuphooks() {
// require an eviction on every single write, which adds
// significant overhead to all cache writes.
func (c *Caches) Sweep(threshold float64) {
c.GTS.Account().Trim(threshold)
c.GTS.AccountNote().Trim(threshold)
c.GTS.Block().Trim(threshold)
c.GTS.BlockIDs().Trim(threshold)
c.GTS.Emoji().Trim(threshold)
c.GTS.EmojiCategory().Trim(threshold)
c.GTS.Follow().Trim(threshold)
c.GTS.FollowIDs().Trim(threshold)
c.GTS.FollowRequest().Trim(threshold)
c.GTS.FollowRequestIDs().Trim(threshold)
c.GTS.Instance().Trim(threshold)
c.GTS.List().Trim(threshold)
c.GTS.ListEntry().Trim(threshold)
c.GTS.Marker().Trim(threshold)
c.GTS.Media().Trim(threshold)
c.GTS.Mention().Trim(threshold)
c.GTS.Notification().Trim(threshold)
c.GTS.Poll().Trim(threshold)
c.GTS.Report().Trim(threshold)
c.GTS.Status().Trim(threshold)
c.GTS.StatusFave().Trim(threshold)
c.GTS.Tag().Trim(threshold)
c.GTS.ThreadMute().Trim(threshold)
c.GTS.Tombstone().Trim(threshold)
c.GTS.User().Trim(threshold)
c.GTS.Account.Trim(threshold)
c.GTS.AccountNote.Trim(threshold)
c.GTS.Block.Trim(threshold)
c.GTS.BlockIDs.Trim(threshold)
c.GTS.Emoji.Trim(threshold)
c.GTS.EmojiCategory.Trim(threshold)
c.GTS.Follow.Trim(threshold)
c.GTS.FollowIDs.Trim(threshold)
c.GTS.FollowRequest.Trim(threshold)
c.GTS.FollowRequestIDs.Trim(threshold)
c.GTS.Instance.Trim(threshold)
c.GTS.List.Trim(threshold)
c.GTS.ListEntry.Trim(threshold)
c.GTS.Marker.Trim(threshold)
c.GTS.Media.Trim(threshold)
c.GTS.Mention.Trim(threshold)
c.GTS.Notification.Trim(threshold)
c.GTS.Poll.Trim(threshold)
c.GTS.Report.Trim(threshold)
c.GTS.Status.Trim(threshold)
c.GTS.StatusFave.Trim(threshold)
c.GTS.Tag.Trim(threshold)
c.GTS.ThreadMute.Trim(threshold)
c.GTS.Tombstone.Trim(threshold)
c.GTS.User.Trim(threshold)
c.Visibility.Trim(threshold)
}

1071
internal/cache/db.go vendored Normal file

File diff suppressed because it is too large Load diff

1119
internal/cache/gts.go vendored

File diff suppressed because it is too large Load diff

192
internal/cache/invalidate.go vendored Normal file
View file

@ -0,0 +1,192 @@
// GoToSocial
// Copyright (C) GoToSocial Authors admin@gotosocial.org
// SPDX-License-Identifier: AGPL-3.0-or-later
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package cache
import (
"github.com/superseriousbusiness/gotosocial/internal/gtsmodel"
)
// Below are cache invalidation hooks between other caches,
// as an invalidation indicates a database INSERT / UPDATE / DELETE.
// NOTE THEY ARE ONLY CALLED WHEN THE ITEM IS IN THE CACHE, SO FOR
// HOOKS TO BE CALLED ON DELETE YOU MUST FIRST POPULATE IT IN THE CACHE.
func (c *Caches) OnInvalidateAccount(account *gtsmodel.Account) {
// Invalidate account ID cached visibility.
c.Visibility.Invalidate("ItemID", account.ID)
c.Visibility.Invalidate("RequesterID", account.ID)
// Invalidate this account's
// following / follower lists.
// (see FollowIDs() comment for details).
c.GTS.FollowIDs.InvalidateAll(
">"+account.ID,
"l>"+account.ID,
"<"+account.ID,
"l<"+account.ID,
)
// Invalidate this account's
// follow requesting / request lists.
// (see FollowRequestIDs() comment for details).
c.GTS.FollowRequestIDs.InvalidateAll(
">"+account.ID,
"<"+account.ID,
)
// Invalidate this account's block lists.
c.GTS.BlockIDs.Invalidate(account.ID)
}
func (c *Caches) OnInvalidateBlock(block *gtsmodel.Block) {
// Invalidate block origin account ID cached visibility.
c.Visibility.Invalidate("ItemID", block.AccountID)
c.Visibility.Invalidate("RequesterID", block.AccountID)
// Invalidate block target account ID cached visibility.
c.Visibility.Invalidate("ItemID", block.TargetAccountID)
c.Visibility.Invalidate("RequesterID", block.TargetAccountID)
// Invalidate source account's block lists.
c.GTS.BlockIDs.Invalidate(block.AccountID)
}
func (c *Caches) OnInvalidateEmojiCategory(category *gtsmodel.EmojiCategory) {
// Invalidate any emoji in this category.
c.GTS.Emoji.Invalidate("CategoryID", category.ID)
}
func (c *Caches) OnInvalidateFollow(follow *gtsmodel.Follow) {
// Invalidate follow request with this same ID.
c.GTS.FollowRequest.Invalidate("ID", follow.ID)
// Invalidate any related list entries.
c.GTS.ListEntry.Invalidate("FollowID", follow.ID)
// Invalidate follow origin account ID cached visibility.
c.Visibility.Invalidate("ItemID", follow.AccountID)
c.Visibility.Invalidate("RequesterID", follow.AccountID)
// Invalidate follow target account ID cached visibility.
c.Visibility.Invalidate("ItemID", follow.TargetAccountID)
c.Visibility.Invalidate("RequesterID", follow.TargetAccountID)
// Invalidate source account's following
// lists, and destination's follwer lists.
// (see FollowIDs() comment for details).
c.GTS.FollowIDs.InvalidateAll(
">"+follow.AccountID,
"l>"+follow.AccountID,
"<"+follow.AccountID,
"l<"+follow.AccountID,
"<"+follow.TargetAccountID,
"l<"+follow.TargetAccountID,
">"+follow.TargetAccountID,
"l>"+follow.TargetAccountID,
)
}
func (c *Caches) OnInvalidateFollowRequest(followReq *gtsmodel.FollowRequest) {
// Invalidate follow with this same ID.
c.GTS.Follow.Invalidate("ID", followReq.ID)
// Invalidate source account's followreq
// lists, and destinations follow req lists.
// (see FollowRequestIDs() comment for details).
c.GTS.FollowRequestIDs.InvalidateAll(
">"+followReq.AccountID,
"<"+followReq.AccountID,
">"+followReq.TargetAccountID,
"<"+followReq.TargetAccountID,
)
}
func (c *Caches) OnInvalidateList(list *gtsmodel.List) {
// Invalidate all cached entries of this list.
c.GTS.ListEntry.Invalidate("ListID", list.ID)
}
func (c *Caches) OnInvalidateMedia(media *gtsmodel.MediaAttachment) {
if (media.Avatar != nil && *media.Avatar) ||
(media.Header != nil && *media.Header) {
// Invalidate cache of attaching account.
c.GTS.Account.Invalidate("ID", media.AccountID)
}
if media.StatusID != "" {
// Invalidate cache of attaching status.
c.GTS.Status.Invalidate("ID", media.StatusID)
}
}
func (c *Caches) OnInvalidatePoll(poll *gtsmodel.Poll) {
// Invalidate all cached votes of this poll.
c.GTS.PollVote.Invalidate("PollID", poll.ID)
// Invalidate cache of poll vote IDs.
c.GTS.PollVoteIDs.Invalidate(poll.ID)
}
func (c *Caches) OnInvalidatePollVote(vote *gtsmodel.PollVote) {
// Invalidate cached poll (contains no. votes).
c.GTS.Poll.Invalidate("ID", vote.PollID)
// Invalidate cache of poll vote IDs.
c.GTS.PollVoteIDs.Invalidate(vote.PollID)
}
func (c *Caches) OnInvalidateStatus(status *gtsmodel.Status) {
// Invalidate status ID cached visibility.
c.Visibility.Invalidate("ItemID", status.ID)
for _, id := range status.AttachmentIDs {
// Invalidate each media by the IDs we're aware of.
// This must be done as the status table is aware of
// the media IDs in use before the media table is
// aware of the status ID they are linked to.
//
// c.GTS.Media().Invalidate("StatusID") will not work.
c.GTS.Media.Invalidate("ID", id)
}
if status.BoostOfID != "" {
// Invalidate boost ID list of the original status.
c.GTS.BoostOfIDs.Invalidate(status.BoostOfID)
}
if status.InReplyToID != "" {
// Invalidate in reply to ID list of original status.
c.GTS.InReplyToIDs.Invalidate(status.InReplyToID)
}
if status.PollID != "" {
// Invalidate cache of attached poll ID.
c.GTS.Poll.Invalidate("ID", status.PollID)
}
}
func (c *Caches) OnInvalidateStatusFave(fave *gtsmodel.StatusFave) {
// Invalidate status fave ID list for this status.
c.GTS.StatusFaveIDs.Invalidate(fave.StatusID)
}
func (c *Caches) OnInvalidateUser(user *gtsmodel.User) {
// Invalidate local account ID cached visibility.
c.Visibility.Invalidate("ItemID", user.AccountID)
c.Visibility.Invalidate("RequesterID", user.AccountID)
}

View file

@ -18,18 +18,16 @@
package cache
import (
"codeberg.org/gruf/go-cache/v3/result"
"codeberg.org/gruf/go-structr"
"github.com/superseriousbusiness/gotosocial/internal/config"
"github.com/superseriousbusiness/gotosocial/internal/log"
)
type VisibilityCache struct {
*result.Cache[*CachedVisibility]
structr.Cache[*CachedVisibility]
}
// Init will initialize the visibility cache in this collection.
// NOTE: the cache MUST NOT be in use anywhere, this is not thread-safe.
func (c *VisibilityCache) Init() {
func (c *Caches) initVisibility() {
// Calculate maximum cache size.
cap := calculateResultCacheMax(
sizeofVisibility(), // model in-mem size.
@ -38,25 +36,22 @@ func (c *VisibilityCache) Init() {
log.Infof(nil, "Visibility cache size = %d", cap)
c.Cache = result.New([]result.Lookup{
{Name: "ItemID", Multi: true},
{Name: "RequesterID", Multi: true},
{Name: "Type.RequesterID.ItemID"},
}, func(v1 *CachedVisibility) *CachedVisibility {
copyF := func(v1 *CachedVisibility) *CachedVisibility {
v2 := new(CachedVisibility)
*v2 = *v1
return v2
}, cap)
}
c.Cache.IgnoreErrors(ignoreErrors)
}
// Start will attempt to start the visibility cache, or panic.
func (c *VisibilityCache) Start() {
}
// Stop will attempt to stop the visibility cache, or panic.
func (c *VisibilityCache) Stop() {
c.Visibility.Init(structr.Config[*CachedVisibility]{
Indices: []structr.IndexConfig{
{Fields: "ItemID", Multiple: true},
{Fields: "RequesterID", Multiple: true},
{Fields: "Type,RequesterID,ItemID"},
},
MaxSize: cap,
IgnoreErr: ignoreErrors,
CopyValue: copyF,
})
}
// VisibilityType represents a visibility lookup type.