Update README.md

.gitignore

@@ -1,4 +1,6 @@
-data/config.json
-data/cache_db/**
-aleesa-webapp-go
-flickr_init
+/data/config.json
+/data/cache_db/**
+/aleesa-webapp-go
+/flickr_init
+/flickr_populate
+/flickr_test

Makefile

@@ -1,23 +1,147 @@
 #!/usr/bin/env gmake -f
 
-BUILDOPTS=-ldflags='-s -w' -a -gcflags=all=-l -trimpath
+GOOPTS=CGO_ENABLED=0
+BUILDOPTS=-ldflags="-s -w" -a -gcflags=all=-l -trimpath -buildvcs=false
+
+MYNAME=aleesa-webapp-go
+BINARY=${MYNAME}
+UNIX_BINARY=${MYNAME}
+WINDOWS_BINARY=${MYNAME}.exe
+
+MYNAME2=flickr_init
+BINARY2=${MYNAME2}
+UNIX_BINARY2=${MYNAME2}
+WINDOWS_BINARY2=${MYNAME2}.exe
+
+MYNAME3=flickr_populate
+BINARY3=${MYNAME3}
+UNIX_BINARY3=${MYNAME3}
+WINDOWS_BINARY3=${MYNAME3}.exe
+
+MYNAME4=flickr_test
+BINARY4=${MYNAME4}
+UNIX_BINARY4=${MYNAME4}
+WINDOWS_BINARY4=${MYNAME4}.exe
+
+RMCMD=rm -rf
+
+# На windows имя бинарника может зависеть не только от платформы, но и от выбранной цели, для linux-а суффикс .exe
+# не нужен
+ifeq ($(OS),Windows_NT)
+ifdef GOOS
+ifeq ($(GOOS),windows)
+BINARY=${WINDOWS_BINARY}
+BINARY2=${WINDOWS_BINARY2}
+BINARY3=${WINDOWS_BINARY3}
+BINARY4=${WINDOWS_BINARY4}
+else  # not ifeq ($(GOOS),windows)
+BINARY=${MYNAME}
+BINARY2=${MYNAME2}
+BINARY3=${MYNAME3}
+BINARY4=${MYNAME4}
+endif # ifeq ($(GOOS),windows)
+else  # not ifdef GOOS
+BINARY=${WINDOWS_BINARY}
+BINARY2=${WINDOWS_BINARY}
+BINARY3=${WINDOWS_BINARY}
+BINARY4=${WINDOWS_BINARY}
+endif # ifdef GOOS
+ifeq ($(SHELL), sh.exe)
+RMCMD=DEL /Q /F
+endif
+endif
+
+# Явно определяем символ новой строки, чтобы избежать неоднозначности на windows
+define IFS
+
+endef
+
 
 all: clean build
 
+
 build:
-	CGO_ENABLED=0 go build ${BUILDOPTS} -o aleesa-webapp-go ./cmd/aleesa-webapp-go
+ifeq ($(OS),Windows_NT)
+# Looks like on windows gnu make explicitly set SHELL to sh.exe, if it was not set.
+ifeq ($(SHELL), sh.exe)
+#       # Vanilla cmd.exe / powershell.
+	SET "CGO_ENABLED=0"
+	go build ${BUILDOPTS} -o ${BINARY} ./cmd/${MYNAME}
+else ifeq (,$(findstring(Git/usr/bin/sh.exe, $(SHELL))))
+#       # git-bash
+	CGO_ENABLED=0 go build ${BUILDOPTS} -o ${BINARY} ./cmd/${MYNAME}
+else  # not ifeq (,$(findstring(Git/usr/bin/sh.exe, $(SHELL))))
+#       # Some other shell.
+#       # TODO: handle it.
+	$(info "-- Dunno how to handle this shell: ${SHELL}")
+endif # ifeq (,$(findstring(Git/usr/bin/sh.exe, $(SHELL))))
+else  # not  ($(OS),Windows_NT)
+	CGO_ENABLED=0 go build ${BUILDOPTS} -o ${BINARY} ./cmd/${MYNAME}
+endif # ifeq ($(OS),Windows_NT)
 
 buildutils:
-	CGO_ENABLED=0 go build ${BUILDOPTS} -o flickr_init ./cmd/flickr_init
-	CGO_ENABLED=0 go build ${BUILDOPTS} -o flickr_populate ./cmd/flickr_populate
-	CGO_ENABLED=0 go build ${BUILDOPTS} -o flickr_test ./cmd/flickr_test
+ifeq ($(OS),Windows_NT)
+# Looks like on windows gnu make explicitly set SHELL to sh.exe, if it was not set.
+ifeq ($(SHELL), sh.exe)
+#       # Vanilla cmd.exe / powershell.
+	SET "CGO_ENABLED=0"
+	go build ${BUILDOPTS} -o ${BINARY2} ./cmd/${MYNAME2}
+	go build ${BUILDOPTS} -o ${BINARY3} ./cmd/${MYNAME3}
+	go build ${BUILDOPTS} -o ${BINARY4} ./cmd/${MYNAME4}
+else ifeq (,$(findstring(Git/usr/bin/sh.exe, $(SHELL))))
+#       # git-bash
+	CGO_ENABLED=0 go build ${BUILDOPTS} -o ${BINARY2} ./cmd/${MYNAME2}
+	CGO_ENABLED=0 go build ${BUILDOPTS} -o ${BINARY3} ./cmd/${MYNAME3}
+	CGO_ENABLED=0 go build ${BUILDOPTS} -o ${BINARY4} ./cmd/${MYNAME4}
+else  # not ifeq (,$(findstring(Git/usr/bin/sh.exe, $(SHELL))))
+#       # Some other shell.
+#       # TODO: handle it.
+	$(info "-- Dunno how to handle this shell: ${SHELL}")
+endif # ifeq (,$(findstring(Git/usr/bin/sh.exe, $(SHELL))))
+else  # not  ($(OS),Windows_NT)
+	CGO_ENABLED=0 go build ${BUILDOPTS} -o ${BINARY2} ./cmd/${MYNAME2}
+	CGO_ENABLED=0 go build ${BUILDOPTS} -o ${BINARY3} ./cmd/${MYNAME3}
+	CGO_ENABLED=0 go build ${BUILDOPTS} -o ${BINARY4} ./cmd/${MYNAME4}
+endif # ifeq ($(OS),Windows_NT)
+
 
 clean:
-	$(RM) -rf aleesa-webapp-go flickr_init flickr_populate flickr_test
+ifeq ($(OS),Windows_NT)
+ifeq ($(SHELL),sh.exe)
+#	# Vanilla cmd.exe / powershell.
+	if exist ${WINDOWS_BINARY} ${RMCMD} ${WINDOWS_BINARY}
+	if exist ${UNIX_BINARY} ${RMCMD} ${UNIX_BINARY}
+
+	if exist ${WINDOWS_BINARY2} ${RMCMD} ${WINDOWS_BINARY2}
+	if exist ${UNIX_BINARY2} ${RMCMD} ${UNIX_BINARY2}
+
+	if exist ${WINDOWS_BINARY3} ${RMCMD} ${WINDOWS_BINARY3}
+	if exist ${UNIX_BINARY3} ${RMCMD} ${UNIX_BINARY3}
+
+	if exist ${WINDOWS_BINARY4} ${RMCMD} ${WINDOWS_BINARY4}
+	if exist ${UNIX_BINARY4} ${RMCMD} ${UNIX_BINARY4}
+else  # not ifeq ($(SHELL),sh.exe)
+	${RMCMD} ./${WINDOWS_BINARY}
+	${RMCMD} ./${UNIX_BINARY}
+
+	${RMCMD} ./${WINDOWS_BINARY2}
+	${RMCMD} ./${UNIX_BINARY2}
+
+	${RMCMD} ./${WINDOWS_BINARY3}
+	${RMCMD} ./${UNIX_BINARY3}
+
+	${RMCMD} ./${WINDOWS_BINARY4}
+	${RMCMD} ./${UNIX_BINARY4}
+endif # ifeq ($(SHELL),sh.exe)
+else  # not ifeq ($(OS),Windows_NT)
+	${RMCMD} ./${BINARY}
+	${RMCMD} ./${BINARY2}
+	${RMCMD} ./${BINARY3}
+	${RMCMD} ./${BINARY4}
+endif
 
 upgrade:
-	$(RM) -rf vendor
-	go get -d -u -t ./...
+	go get -u ./...
 	go mod tidy
 	go mod vendor
 

README.md

@@ -14,6 +14,18 @@
 различных UA, по количеству дней в месяце. В общем случае делается произвольная выборка из этого списка, но для
 некоторых сервисов сделана привязка ко дню недели.
 
+## Как это собрать?
+
+Понадобится компилятор go версии 1.25.1 или более новый.
+
+Сам процесс сборки простой, достаточно запустить make или gmake, в зависимости от платформы:
+
+```bash
+make
+```
+
+и через некоторое время мы получаем бинарник **aleesa-irc-go**. Это и есть наше приложение.
+
 ### Как это запустить?
 
 Приложение тестировалось на alpine linux x86_64 3.16, но должно работать на любом (более свежем) linux-е.

cmd/aleesa-webapp-go/main.go

@@ -0,0 +1,130 @@
+package main
+
+import (
+	"context"
+	"fmt"
+	"os"
+	"os/signal"
+	"syscall"
+	"time"
+
+	"aleesa-webapp-go/internal/config"
+	"aleesa-webapp-go/internal/log"
+	"aleesa-webapp-go/internal/webapp"
+
+	"github.com/go-redis/redis/v8"
+)
+
+var (
+	// ctx контекст редиски.
+	Ctx = context.Background()
+
+	// Subscriber объектик сабскрайбера редиски.
+	Subscriber *redis.PubSub
+
+	// Shutdown ставится в true, если мы получили сигнал на выключение.
+	Shutdown = false
+
+	// SigChan канал, в который приходят уведомления для хэндлера сигналов от траппера сигналов.
+	SigChan = make(chan os.Signal, 1)
+)
+
+// Основная функция программы, не добавить и не убавить.
+func main() {
+	var (
+		logfile *os.File
+	)
+
+	err := webapp.ReadConfig()
+
+	if err != nil {
+		log.Errorf("Unable to parse config: %s", err)
+		os.Exit(1)
+	}
+
+	// Откроем лог и скормим его логгеру.
+	if webapp.Config.Log != "" {
+		logfile, err = os.OpenFile(webapp.Config.Log, os.O_WRONLY|os.O_APPEND|os.O_CREATE, 0644)
+
+		if err != nil {
+			log.Errorf("Unable to open log file %s: %s", webapp.Config.Log, err)
+			os.Exit(1)
+		}
+	}
+
+	log.Init(webapp.Config.Loglevel, logfile)
+
+	// Иницализируем клиента Редиски.
+	webapp.Config.RedisClient = redis.NewClient(&redis.Options{
+		Addr: fmt.Sprintf("%s:%d", webapp.Config.Server, webapp.Config.Port),
+	}).WithContext(Ctx).WithTimeout(time.Duration(webapp.Config.Timeout) * time.Second)
+
+	// Обозначим, что хотим после соединения подписаться на события из канала config.Channel.
+	Subscriber = webapp.Config.RedisClient.Subscribe(Ctx, webapp.Config.Channel)
+
+	// Самое время поставить трапы на сигналы.
+	signal.Notify(SigChan,
+		syscall.SIGINT,
+		syscall.SIGTERM,
+		syscall.SIGQUIT)
+
+	// Запустим обработчик сигналов.
+	go SigHandler(webapp.Config)
+
+	// Начнём выгребать события из редиски (длина конвеера/буфера канала по-умолчанию - 100 сообщений).
+	ch := Subscriber.Channel()
+
+	log.Info("Service started.")
+
+	for msg := range ch {
+		if !Shutdown {
+			webapp.MsgParser(webapp.Config, Ctx, msg.Payload)
+		}
+	}
+}
+
+// SigHandler хэндлер сигналов закрывает все бд и сваливает из приложения.
+func SigHandler(cfg *config.MyConfig) {
+	// TODO: утащить отсюда в модули.
+
+	var err error
+
+	for {
+		var s = <-SigChan
+		switch s {
+		case syscall.SIGINT:
+			log.Info("Got SIGINT, quitting")
+		case syscall.SIGTERM:
+			log.Info("Got SIGTERM, quitting")
+		case syscall.SIGQUIT:
+			log.Info("Got SIGQUIT, quitting")
+
+		// Заходим на новую итерацию, если у нас "неинтересный" сигнал
+		default:
+			continue
+		}
+
+		Shutdown = true
+
+		// Отпишемся от всех каналов и закроем коннект к редиске
+		if err = Subscriber.Unsubscribe(Ctx); err != nil {
+			log.Errorf("Unable to unsubscribe from redis channels cleanly: %s", err)
+		}
+
+		if err = Subscriber.Close(); err != nil {
+			log.Errorf("Unable to close redis connection cleanly: %s", err)
+		}
+
+		if len(cfg.PcacheDB) > 0 {
+			log.Debug("Closing persistent cache db")
+
+			for _, db := range cfg.PcacheDB {
+				_ = db.Close()
+			}
+		}
+
+		os.Exit(0)
+	}
+}
+
+/* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

cmd/flickr_init/main.go

@@ -0,0 +1,22 @@
+package main
+
+import (
+	"aleesa-webapp-go/internal/flickr"
+	"aleesa-webapp-go/internal/webapp"
+	"fmt"
+	"os"
+)
+
+func main() {
+	if err := webapp.ReadConfig(); err != nil {
+		fmt.Fprintln(os.Stderr, err)
+		os.Exit(1)
+	} else {
+		if err := flickr.APIClientInit(webapp.Config); err != nil {
+			fmt.Fprintln(os.Stderr, err)
+			os.Exit(1)
+		}
+	}
+}
+
+/* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

cmd/flickr_populate/main.go

@@ -1,18 +1,18 @@
 package main
 
 import (
-	"aleesa-webapp-go/internal/config"
 	"aleesa-webapp-go/internal/flickr"
+	"aleesa-webapp-go/internal/webapp"
 	"fmt"
 	"os"
 )
 
 func main() {
-	if c, err := config.ReadConfig(); err != nil {
+	if err := webapp.ReadConfig(); err != nil {
 		fmt.Fprintln(os.Stderr, err)
 		os.Exit(1)
 	} else {
-		if err := flickr.Populate(c); err != nil {
+		if err := flickr.Populate(webapp.Config); err != nil {
 			fmt.Fprintln(os.Stderr, err)
 			os.Exit(1)
 		}

cmd/flickr_test/main.go

@@ -1,18 +1,18 @@
 package main
 
 import (
-	"aleesa-webapp-go/internal/config"
 	"aleesa-webapp-go/internal/flickr"
+	"aleesa-webapp-go/internal/webapp"
 	"fmt"
 	"os"
 )
 
 func main() {
-	if c, err := config.ReadConfig(); err != nil {
+	if err := webapp.ReadConfig(); err != nil {
 		fmt.Fprintln(os.Stderr, err)
 		os.Exit(1)
 	} else {
-		result, err := flickr.SearchByTags(c, []string{"snail", "slug"})
+		result, err := flickr.SearchByTags(webapp.Config, []string{"snail", "slug"})
 
 		if err != nil {
 			fmt.Fprintln(os.Stderr, err)

go.mod

@@ -1,43 +1,43 @@
 module aleesa-webapp-go
 
-go 1.22.5
+go 1.25.1
 
 require (
 	github.com/briandowns/openweathermap v0.21.1
-	github.com/carlmjohnson/requests v0.24.3
-	github.com/cockroachdb/errors v1.11.3
-	github.com/cockroachdb/pebble v1.1.4
+	github.com/carlmjohnson/requests v0.25.1
+	github.com/cockroachdb/errors v1.12.0
+	github.com/cockroachdb/pebble v1.1.5
 	github.com/go-redis/redis/v8 v8.11.5
 	github.com/hjson/hjson-go v3.3.0+incompatible
-	github.com/sirupsen/logrus v1.9.3
-	golang.org/x/net v0.34.0
-	gopkg.in/masci/flickr.v3 v3.0.0-20230428071620-b971d524ac6f
+	golang.org/x/net v0.46.0
+	gopkg.in/masci/flickr.v3 v3.0.0-20250416134523-515bc5586967
 )
 
 require (
-	github.com/DataDog/zstd v1.5.6 // indirect
+	github.com/DataDog/zstd v1.4.5 // indirect
 	github.com/beorn7/perks v1.0.1 // indirect
-	github.com/cespare/xxhash/v2 v2.3.0 // indirect
-	github.com/cockroachdb/fifo v0.0.0-20240816210425-c5d0cb0b6fc0 // indirect
-	github.com/cockroachdb/logtags v0.0.0-20241215232642-bb51bb14a506 // indirect
-	github.com/cockroachdb/redact v1.1.6 // indirect
+	github.com/cespare/xxhash/v2 v2.2.0 // indirect
+	github.com/cockroachdb/fifo v0.0.0-20240606204812-0bbfbd93a7ce // indirect
+	github.com/cockroachdb/logtags v0.0.0-20230118201751-21c54148d20b // indirect
+	github.com/cockroachdb/redact v1.1.5 // indirect
 	github.com/cockroachdb/tokenbucket v0.0.0-20230807174530-cc333fc44b06 // indirect
 	github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f // indirect
-	github.com/getsentry/sentry-go v0.31.1 // indirect
+	github.com/getsentry/sentry-go v0.27.0 // indirect
 	github.com/gogo/protobuf v1.3.2 // indirect
+	github.com/golang/protobuf v1.5.3 // indirect
 	github.com/golang/snappy v0.0.4 // indirect
-	github.com/klauspost/compress v1.17.11 // indirect
+	github.com/klauspost/compress v1.16.0 // indirect
 	github.com/kr/pretty v0.3.1 // indirect
 	github.com/kr/text v0.2.0 // indirect
-	github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect
+	github.com/matttproud/golang_protobuf_extensions v1.0.4 // indirect
 	github.com/pkg/errors v0.9.1 // indirect
-	github.com/prometheus/client_golang v1.20.5 // indirect
-	github.com/prometheus/client_model v0.6.1 // indirect
-	github.com/prometheus/common v0.62.0 // indirect
-	github.com/prometheus/procfs v0.15.1 // indirect
-	github.com/rogpeppe/go-internal v1.13.1 // indirect
-	golang.org/x/exp v0.0.0-20250207012021-f9890c6ad9f3 // indirect
-	golang.org/x/sys v0.30.0 // indirect
-	golang.org/x/text v0.22.0 // indirect
-	google.golang.org/protobuf v1.36.5 // indirect
+	github.com/prometheus/client_golang v1.15.0 // indirect
+	github.com/prometheus/client_model v0.3.0 // indirect
+	github.com/prometheus/common v0.42.0 // indirect
+	github.com/prometheus/procfs v0.9.0 // indirect
+	github.com/rogpeppe/go-internal v1.9.0 // indirect
+	golang.org/x/exp v0.0.0-20230626212559-97b1e661b5df // indirect
+	golang.org/x/sys v0.37.0 // indirect
+	golang.org/x/text v0.30.0 // indirect
+	google.golang.org/protobuf v1.33.0 // indirect
 )

go.sum

@@ -1,59 +1,64 @@
-github.com/DataDog/zstd v1.5.6 h1:LbEglqepa/ipmmQJUDnSsfvA8e8IStVcGaFWDuxvGOY=
-github.com/DataDog/zstd v1.5.6/go.mod h1:g4AWEaM3yOg3HYfnJ3YIawPnVdXJh9QME85blwSAmyw=
+github.com/DataDog/zstd v1.4.5 h1:EndNeuB0l9syBZhut0wns3gV1hL8zX8LIu6ZiVHWLIQ=
+github.com/DataDog/zstd v1.4.5/go.mod h1:1jcaCB/ufaK+sKp1NBhlGmpz41jOoPQ35bpF36t7BBo=
 github.com/beorn7/perks v1.0.1 h1:VlbKKnNfV8bJzeqoa4cOKqO6bYr3WgKZxO8Z16+hsOM=
 github.com/beorn7/perks v1.0.1/go.mod h1:G2ZrVWU2WbWT9wwq4/hrbKbnv/1ERSJQ0ibhJ6rlkpw=
 github.com/briandowns/openweathermap v0.21.1 h1:TPbuixuF+aGJP1mpgTNny6eUkdbvj7gqODGXkwhss48=
 github.com/briandowns/openweathermap v0.21.1/go.mod h1:0GLnknqicWxXnGi1IqoOaZIw+kIe5hkt+YM5WY3j8+0=
-github.com/carlmjohnson/requests v0.24.3 h1:LYcM/jVIVPkioigMjEAnBACXl2vb42TVqiC8EYNoaXQ=
-github.com/carlmjohnson/requests v0.24.3/go.mod h1:duYA/jDnyZ6f3xbcF5PpZ9N8clgopubP2nK5i6MVMhU=
-github.com/cespare/xxhash/v2 v2.3.0 h1:UL815xU9SqsFlibzuggzjXhog7bL6oX9BbNZnL2UFvs=
-github.com/cespare/xxhash/v2 v2.3.0/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs=
+github.com/carlmjohnson/requests v0.25.1 h1:17zNRLecxtAjhtdEIV+F+wrYfe+AGZUjWJtpndcOUYA=
+github.com/carlmjohnson/requests v0.25.1/go.mod h1:z3UEf8IE4sZxZ78spW6/tLdqBkfCu1Fn4RaYMnZ8SRM=
+github.com/cespare/xxhash/v2 v2.2.0 h1:DC2CZ1Ep5Y4k3ZQ899DldepgrayRUGE6BBZ/cd9Cj44=
+github.com/cespare/xxhash/v2 v2.2.0/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs=
 github.com/cockroachdb/datadriven v1.0.3-0.20230413201302-be42291fc80f h1:otljaYPt5hWxV3MUfO5dFPFiOXg9CyG5/kCfayTqsJ4=
 github.com/cockroachdb/datadriven v1.0.3-0.20230413201302-be42291fc80f/go.mod h1:a9RdTaap04u637JoCzcUoIcDmvwSUtcUFtT/C3kJlTU=
-github.com/cockroachdb/errors v1.11.3 h1:5bA+k2Y6r+oz/6Z/RFlNeVCesGARKuC6YymtcDrbC/I=
-github.com/cockroachdb/errors v1.11.3/go.mod h1:m4UIW4CDjx+R5cybPsNrRbreomiFqt8o1h1wUVazSd8=
-github.com/cockroachdb/fifo v0.0.0-20240816210425-c5d0cb0b6fc0 h1:pU88SPhIFid6/k0egdR5V6eALQYq2qbSmukrkgIh/0A=
-github.com/cockroachdb/fifo v0.0.0-20240816210425-c5d0cb0b6fc0/go.mod h1:9/y3cnZ5GKakj/H4y9r9GTjCvAFta7KLgSHPJJYc52M=
-github.com/cockroachdb/logtags v0.0.0-20241215232642-bb51bb14a506 h1:ASDL+UJcILMqgNeV5jiqR4j+sTuvQNHdf2chuKj1M5k=
-github.com/cockroachdb/logtags v0.0.0-20241215232642-bb51bb14a506/go.mod h1:Mw7HqKr2kdtu6aYGn3tPmAftiP3QPX63LdK/zcariIo=
-github.com/cockroachdb/pebble v1.1.4 h1:5II1uEP4MyHLDnsrbv/EZ36arcb9Mxg3n+owhZ3GrG8=
-github.com/cockroachdb/pebble v1.1.4/go.mod h1:4exszw1r40423ZsmkG/09AFEG83I0uDgfujJdbL6kYU=
-github.com/cockroachdb/redact v1.1.6 h1:zXJBwDZ84xJNlHl1rMyCojqyIxv+7YUpQiJLQ7n4314=
-github.com/cockroachdb/redact v1.1.6/go.mod h1:BVNblN9mBWFyMyqK1k3AAiSxhvhfK2oOZZ2lK+dpvRg=
+github.com/cockroachdb/errors v1.12.0 h1:d7oCs6vuIMUQRVbi6jWWWEJZahLCfJpnJSVobd1/sUo=
+github.com/cockroachdb/errors v1.12.0/go.mod h1:SvzfYNNBshAVbZ8wzNc/UPK3w1vf0dKDUP41ucAIf7g=
+github.com/cockroachdb/fifo v0.0.0-20240606204812-0bbfbd93a7ce h1:giXvy4KSc/6g/esnpM7Geqxka4WSqI1SZc7sMJFd3y4=
+github.com/cockroachdb/fifo v0.0.0-20240606204812-0bbfbd93a7ce/go.mod h1:9/y3cnZ5GKakj/H4y9r9GTjCvAFta7KLgSHPJJYc52M=
+github.com/cockroachdb/logtags v0.0.0-20230118201751-21c54148d20b h1:r6VH0faHjZeQy818SGhaone5OnYfxFR/+AzdY3sf5aE=
+github.com/cockroachdb/logtags v0.0.0-20230118201751-21c54148d20b/go.mod h1:Vz9DsVWQQhf3vs21MhPMZpMGSht7O/2vFW2xusFUVOs=
+github.com/cockroachdb/pebble v1.1.5 h1:5AAWCBWbat0uE0blr8qzufZP5tBjkRyy/jWe1QWLnvw=
+github.com/cockroachdb/pebble v1.1.5/go.mod h1:17wO9el1YEigxkP/YtV8NtCivQDgoCyBg5c4VR/eOWo=
+github.com/cockroachdb/redact v1.1.5 h1:u1PMllDkdFfPWaNGMyLD1+so+aq3uUItthCFqzwPJ30=
+github.com/cockroachdb/redact v1.1.5/go.mod h1:BVNblN9mBWFyMyqK1k3AAiSxhvhfK2oOZZ2lK+dpvRg=
 github.com/cockroachdb/tokenbucket v0.0.0-20230807174530-cc333fc44b06 h1:zuQyyAKVxetITBuuhv3BI9cMrmStnpT18zmgmTxunpo=
 github.com/cockroachdb/tokenbucket v0.0.0-20230807174530-cc333fc44b06/go.mod h1:7nc4anLGjupUW/PeY5qiNYsdNXj7zopG+eqsS7To5IQ=
 github.com/creack/pty v1.1.9/go.mod h1:oKZEueFk5CKHvIhNR5MUki03XCEU+Q6VDXinZuGJ33E=
-github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f h1:lO4WD4F/rVNCu3HqELle0jiPLLBs70cWOduZpkS1E78=
 github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f/go.mod h1:cuUVRXasLTGF7a8hSLbxyZXjz+1KgoB3wDUb6vlszIc=
 github.com/fsnotify/fsnotify v1.4.9 h1:hsms1Qyu0jgnwNXIxa+/V/PDsU6CfLf6CNO8H7IWoS4=
 github.com/fsnotify/fsnotify v1.4.9/go.mod h1:znqG4EE+3YCdAaPaxE2ZRY/06pZUdp0tY4IgpuI1SZQ=
-github.com/getsentry/sentry-go v0.31.1 h1:ELVc0h7gwyhnXHDouXkhqTFSO5oslsRDk0++eyE0KJ4=
-github.com/getsentry/sentry-go v0.31.1/go.mod h1:CYNcMMz73YigoHljQRG+qPF+eMq8gG72XcGN/p71BAY=
+github.com/getsentry/sentry-go v0.27.0 h1:Pv98CIbtB3LkMWmXi4Joa5OOcwbmnX88sF5qbK3r3Ps=
+github.com/getsentry/sentry-go v0.27.0/go.mod h1:lc76E2QywIyW8WuBnwl8Lc4bkmQH4+w1gwTf25trprY=
 github.com/go-errors/errors v1.4.2 h1:J6MZopCL4uSllY1OfXM374weqZFFItUbrImctkmUxIA=
 github.com/go-errors/errors v1.4.2/go.mod h1:sIVyrIiJhuEF+Pj9Ebtd6P/rEYROXFi3BopGUQ5a5Og=
 github.com/go-redis/redis/v8 v8.11.5 h1:AcZZR7igkdvfVmQTPnu9WE37LRrO/YrBH5zWyjDC0oI=
 github.com/go-redis/redis/v8 v8.11.5/go.mod h1:gREzHqY1hg6oD9ngVRbLStwAWKhA0FEgq8Jd4h5lpwo=
 github.com/gogo/protobuf v1.3.2 h1:Ov1cvc58UF3b5XjBnZv7+opcTcQFZebYjWzi34vdm4Q=
 github.com/gogo/protobuf v1.3.2/go.mod h1:P1XiOD3dCwIKUDQYPy72D8LYyHL2YPYrpS2s69NZV8Q=
+github.com/golang/protobuf v1.2.0/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
+github.com/golang/protobuf v1.3.5/go.mod h1:6O5/vntMXwX2lRkT1hjjk0nAC1IDOTvTlVgjlRvqsdk=
+github.com/golang/protobuf v1.5.0/go.mod h1:FsONVRAS9T7sI+LIUmWTfcYkHO4aIWwzhcaSAoJOfIk=
+github.com/golang/protobuf v1.5.3 h1:KhyjKVUg7Usr/dYsdSqoFveMYd5ko72D+zANwlG1mmg=
+github.com/golang/protobuf v1.5.3/go.mod h1:XVQd3VNwM+JqD3oG2Ue2ip4fOMUkwXdXDdiuN0vRsmY=
 github.com/golang/snappy v0.0.4 h1:yAGX7huGHXlcLOEtBnF4w7FQwA26wojNCwOYAEhLjQM=
 github.com/golang/snappy v0.0.4/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
-github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
-github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
+github.com/google/go-cmp v0.5.5/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
+github.com/google/go-cmp v0.5.9 h1:O2Tfq5qg4qc4AmwVlvv0oLiVAGB7enBSJ2x2DqQFi38=
+github.com/google/go-cmp v0.5.9/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
 github.com/hjson/hjson-go v3.3.0+incompatible h1:Rqr+Ya+0aCJMjaE4s8E9YKvuJLuLVpEvz4ONum52vnI=
 github.com/hjson/hjson-go v3.3.0+incompatible/go.mod h1:qsetwF8NlsTsOTwZTApNlTCerV+b2GjYRRcIk4JMFio=
 github.com/kisielk/errcheck v1.5.0/go.mod h1:pFxgyoBC7bSaBwPgfKdkLd5X25qrDl4LWUI2bnpBCr8=
 github.com/kisielk/gotool v1.0.0/go.mod h1:XhKaO+MFFWcvkIS/tQcRk01m1F5IRFswLeQ+oQHNcck=
-github.com/klauspost/compress v1.17.11 h1:In6xLpyWOi1+C7tXUUWv2ot1QvBjxevKAaI6IXrJmUc=
-github.com/klauspost/compress v1.17.11/go.mod h1:pMDklpSncoRMuLFrf1W9Ss9KT+0rH90U12bZKk7uwG0=
+github.com/klauspost/compress v1.16.0 h1:iULayQNOReoYUe+1qtKOqw9CwJv3aNQu8ivo7lw1HU4=
+github.com/klauspost/compress v1.16.0/go.mod h1:ntbaceVETuRiXiv4DpjP66DpAtAGkEQskQzEyD//IeE=
 github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
 github.com/kr/pretty v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
 github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
 github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
-github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 h1:C3w9PqII01/Oq1c1nUAm88MOHcQC9l5mIlSMApZMrHA=
-github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822/go.mod h1:+n7T8mK8HuQTcFwEeznm/DIxMOiR9yIdICNftLE1DvQ=
+github.com/matttproud/golang_protobuf_extensions v1.0.4 h1:mmDVorXM7PCGKw94cs5zkfA9PSy5pEvNWRP0ET0TIVo=
+github.com/matttproud/golang_protobuf_extensions v1.0.4/go.mod h1:BSXmuO+STAnVfrANrmjBb36TMTDstsz7MSK+HVaYKv4=
 github.com/nxadm/tail v1.4.8 h1:nPr65rt6Y5JFSKQO7qToXr7pePgD6Gwiw05lkbyAQTE=
 github.com/nxadm/tail v1.4.8/go.mod h1:+ncqLTQzXmGhMZNUePPaPqPvBxHAIsmXswZKocGu+AU=
 github.com/onsi/ginkgo v1.16.5 h1:8xi0RTUf59SOSfEtZMvwTvXYMzG4gV23XVHOZiXNtnE=
@@ -67,55 +72,48 @@ github.com/pkg/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4=
 github.com/pkg/errors v0.9.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
-github.com/prometheus/client_golang v1.20.5 h1:cxppBPuYhUnsO6yo/aoRol4L7q7UFfdm+bR9r+8l63Y=
-github.com/prometheus/client_golang v1.20.5/go.mod h1:PIEt8X02hGcP8JWbeHyeZ53Y/jReSnHgO035n//V5WE=
-github.com/prometheus/client_model v0.6.1 h1:ZKSh/rekM+n3CeS952MLRAdFwIKqeY8b62p8ais2e9E=
-github.com/prometheus/client_model v0.6.1/go.mod h1:OrxVMOVHjw3lKMa8+x6HeMGkHMQyHDk9E3jmP2AmGiY=
-github.com/prometheus/common v0.62.0 h1:xasJaQlnWAeyHdUBeGjXmutelfJHWMRr+Fg4QszZ2Io=
-github.com/prometheus/common v0.62.0/go.mod h1:vyBcEuLSvWos9B1+CyL7JZ2up+uFzXhkqml0W5zIY1I=
-github.com/prometheus/procfs v0.15.1 h1:YagwOFzUgYfKKHX6Dr+sHT7km/hxC76UB0learggepc=
-github.com/prometheus/procfs v0.15.1/go.mod h1:fB45yRUv8NstnjriLhBQLuOUt+WW4BsoGhij/e3PBqk=
+github.com/prometheus/client_golang v1.15.0 h1:5fCgGYogn0hFdhyhLbw7hEsWxufKtY9klyvdNfFlFhM=
+github.com/prometheus/client_golang v1.15.0/go.mod h1:e9yaBhRPU2pPNsZwE+JdQl0KEt1N9XgF6zxWmaC0xOk=
+github.com/prometheus/client_model v0.3.0 h1:UBgGFHqYdG/TPFD1B1ogZywDqEkwp3fBMvqdiQ7Xew4=
+github.com/prometheus/client_model v0.3.0/go.mod h1:LDGWKZIo7rky3hgvBe+caln+Dr3dPggB5dvjtD7w9+w=
+github.com/prometheus/common v0.42.0 h1:EKsfXEYo4JpWMHH5cg+KOUWeuJSov1Id8zGR8eeI1YM=
+github.com/prometheus/common v0.42.0/go.mod h1:xBwqVerjNdUDjgODMpudtOMwlOwf2SaTr1yjz4b7Zbc=
+github.com/prometheus/procfs v0.9.0 h1:wzCHvIvM5SxWqYvwgVL7yJY8Lz3PKn49KQtpgMYJfhI=
+github.com/prometheus/procfs v0.9.0/go.mod h1:+pB4zwohETzFnmlpe6yd2lSc+0/46IYZRB/chUwxUZY=
+github.com/rogpeppe/go-internal v1.9.0 h1:73kH8U+JUqXU8lRuOHeVHaa/SZPifC7BkcraZVejAe8=
 github.com/rogpeppe/go-internal v1.9.0/go.mod h1:WtVeX8xhTBvf0smdhujwtBcq4Qrzq/fJaraNFVN+nFs=
-github.com/rogpeppe/go-internal v1.13.1 h1:KvO1DLK/DRN07sQ1LQKScxyZJuNnedQ5/wKSR38lUII=
-github.com/rogpeppe/go-internal v1.13.1/go.mod h1:uMEvuHeurkdAXX61udpOXGD/AzZDWNMNyH2VO9fmH0o=
-github.com/sirupsen/logrus v1.9.3 h1:dueUQJ1C2q9oE3F7wvmSGAaVtTmUizReu6fjN8uqzbQ=
-github.com/sirupsen/logrus v1.9.3/go.mod h1:naHLuLoDiP4jHNo9R0sCBMtWGeIprob74mVsIT4qYEQ=
-github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
-github.com/stretchr/testify v1.7.0/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
-github.com/stretchr/testify v1.10.0 h1:Xv5erBjTwe/5IxqUQTdXv5kgmIvbHo3QQyRwhJsOfJA=
-github.com/stretchr/testify v1.10.0/go.mod h1:r2ic/lqez/lEtzL7wO/rwa5dbSLXVDPFyf8C91i36aY=
+github.com/stretchr/testify v1.9.0 h1:HtqpIVDClZ4nwg75+f6Lvsy/wHu+3BoSGCbBAcpTsTg=
+github.com/stretchr/testify v1.9.0/go.mod h1:r2ic/lqez/lEtzL7wO/rwa5dbSLXVDPFyf8C91i36aY=
 github.com/yuin/goldmark v1.1.27/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
 github.com/yuin/goldmark v1.2.1/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
-go.uber.org/goleak v1.3.0 h1:2K3zAYmnTNqV73imy9J1T3WC+gmCePx2hEGkimedGto=
-go.uber.org/goleak v1.3.0/go.mod h1:CoHD4mav9JJNrW/WLlf7HGZPjdw8EucARQHekz1X6bE=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
 golang.org/x/crypto v0.0.0-20191011191535-87dc89f01550/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
 golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPhUkYOs5KpL4U8rLKemX1yGLhDgUto=
-golang.org/x/exp v0.0.0-20250207012021-f9890c6ad9f3 h1:qNgPs5exUA+G0C96DrPwNrvLSj7GT/9D+3WMWUcUg34=
-golang.org/x/exp v0.0.0-20250207012021-f9890c6ad9f3/go.mod h1:tujkw807nyEEAamNbDrEGzRav+ilXA7PCRAd6xsmwiU=
+golang.org/x/exp v0.0.0-20230626212559-97b1e661b5df h1:UA2aFVmmsIlefxMk29Dp2juaUSth8Pyn3Tq5Y5mJGME=
+golang.org/x/exp v0.0.0-20230626212559-97b1e661b5df/go.mod h1:FXUEEKJgO7OQYeo8N01OfiKP8RXMtf6e8aTskBGqWdc=
 golang.org/x/mod v0.2.0/go.mod h1:s0Qsj1ACt9ePp/hMypM3fl4fZqREWJwdYDEqhRiZZUA=
 golang.org/x/mod v0.3.0/go.mod h1:s0Qsj1ACt9ePp/hMypM3fl4fZqREWJwdYDEqhRiZZUA=
 golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
 golang.org/x/net v0.0.0-20190620200207-3b0461eec859/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
 golang.org/x/net v0.0.0-20200226121028-0de0cce0169b/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
 golang.org/x/net v0.0.0-20201021035429-f5854403a974/go.mod h1:sp8m0HH+o8qH0wwXwYZr8TS3Oi6o0r6Gce1SSxlDquU=
-golang.org/x/net v0.34.0 h1:Mb7Mrk043xzHgnRM88suvJFwzVrRfHEHJEl5/71CKw0=
-golang.org/x/net v0.34.0/go.mod h1:di0qlW3YNM5oh6GqDGQr92MyTozJPmybPK4Ev/Gm31k=
+golang.org/x/net v0.46.0 h1:giFlY12I07fugqwPuWJi68oOnpfqFnJIJzaIIm2JVV4=
+golang.org/x/net v0.46.0/go.mod h1:Q9BGdFy1y4nkUwiLvT5qtyhAnEHgnQ/zd8PfU6nc210=
+golang.org/x/sync v0.0.0-20181221193216-37e7f081c4d4/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20190911185100-cd5d95a43a6e/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.0.0-20201020160332-67f06af15bc9/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
-golang.org/x/sync v0.11.0 h1:GGz8+XQP4FvTTrjZPzNKTMFtSXH80RAzG+5ghFPgK9w=
-golang.org/x/sync v0.11.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
+golang.org/x/sync v0.17.0 h1:l60nONMj9l5drqw6jlhIELNv9I0A4OFgRsG9k2oT9Ug=
+golang.org/x/sync v0.17.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
 golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20190412213103-97732733099d/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20200930185726-fdedc70b468f/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
-golang.org/x/sys v0.0.0-20220715151400-c0bba94af5f8/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.30.0 h1:QjkSwP/36a20jFYWkSue1YwXzLmsV5Gfq7Eiy72C1uc=
-golang.org/x/sys v0.30.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
+golang.org/x/sys v0.37.0 h1:fdNQudmxPjkdUTPnLn5mdQv7Zwvbvpaxqs831goi9kQ=
+golang.org/x/sys v0.37.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
 golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
-golang.org/x/text v0.22.0 h1:bofq7m3/HAFvbF51jz3Q9wLg3jkvSPuiZu/pD1XwgtM=
-golang.org/x/text v0.22.0/go.mod h1:YRoo4H8PVmsu+E3Ou7cqLVH8oXWIHVoX0jqUWALQhfY=
+golang.org/x/text v0.30.0 h1:yznKA/E9zq54KzlzBEAWn1NXSQ8DIp/NYMy88xJjl4k=
+golang.org/x/text v0.30.0/go.mod h1:yDdHFIX9t+tORqspjENWgzaCVXgk0yYnYuSZ8UzzBVM=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
 golang.org/x/tools v0.0.0-20191119224855-298f0cb1881e/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
 golang.org/x/tools v0.0.0-20200619180055-7c47624df98f/go.mod h1:EkVYQZoAsY45+roYkvgYkIh4xh/qjgUK9TdY2XT94GE=
@@ -124,15 +122,15 @@ golang.org/x/xerrors v0.0.0-20190717185122-a985d3407aa7/go.mod h1:I/5z698sn9Ka8T
 golang.org/x/xerrors v0.0.0-20191011141410-1b5146add898/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
-google.golang.org/protobuf v1.36.5 h1:tPhr+woSbjfYvY6/GPufUoYizxw1cF/yFoxJ2fmpwlM=
-google.golang.org/protobuf v1.36.5/go.mod h1:9fA7Ob0pmnwhb644+1+CVWFRbNajQ6iRojtC/QF5bRE=
-gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
-gopkg.in/masci/flickr.v3 v3.0.0-20230428071620-b971d524ac6f h1:RUQkjvXxx0U73EFbcOE6lz0mjwHyOCpkIkY8Toitzlk=
-gopkg.in/masci/flickr.v3 v3.0.0-20230428071620-b971d524ac6f/go.mod h1:2fXoNIK2lULBoY5w6sID6+4rwMujRHgsKIHTMYSu+TU=
+google.golang.org/protobuf v1.26.0-rc.1/go.mod h1:jlhhOSvTdKEhbULTjvd4ARK9grFBp09yW+WbY/TyQbw=
+google.golang.org/protobuf v1.26.0/go.mod h1:9q0QmTI4eRPtz6boOQmLYwt+qCgq0jsYwAQnmE0givc=
+google.golang.org/protobuf v1.33.0 h1:uNO2rsAINq/JlFpSdYEKIZ0uKD/R9cpdv0T+yoGwGmI=
+google.golang.org/protobuf v1.33.0/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos=
+gopkg.in/masci/flickr.v3 v3.0.0-20250416134523-515bc5586967 h1:oFthoMSKQ8bHGNGF/KmUwkECdBjx/Odxqnli3zhDD+A=
+gopkg.in/masci/flickr.v3 v3.0.0-20250416134523-515bc5586967/go.mod h1:2fXoNIK2lULBoY5w6sID6+4rwMujRHgsKIHTMYSu+TU=
 gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7 h1:uRGJdciOHaEIrze2W8Q3AKkepLTh2hOroT7a+7czfdQ=
 gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7/go.mod h1:dt/ZhP58zS4L8KSrWDmTeBkI65Dw0HsyUHuEVlX15mw=
 gopkg.in/yaml.v2 v2.4.0 h1:D8xgwECY7CYvx+Y2n4sBz93Jn9JRvxdiyyo8CTfuKaY=
 gopkg.in/yaml.v2 v2.4.0/go.mod h1:RDklbk79AGWmwhnvt/jBztapEOGDOx6ZbXqjP6csGnQ=
-gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
 gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
 gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=

internal/config/types.go

@@ -0,0 +1,37 @@
+// Этот package нужен, чтобы разбить кольцевую зависимость.
+package config
+
+import (
+	"github.com/cockroachdb/pebble"
+	"github.com/go-redis/redis/v8"
+)
+
+// MyConfig структурка, описывающая конфиг.
+type MyConfig struct {
+	Server         string `json:"server,omitempty"`
+	Port           int    `json:"port,omitempty"`
+	Timeout        int    `json:"timeout,omitempty"`
+	Loglevel       string `json:"loglevel,omitempty"`
+	Log            string `json:"log,omitempty"`
+	Channel        string `json:"channel,omitempty"`
+	DataDir        string `json:"datadir,omitempty"`
+	Csign          string `json:"csign,omitempty"`
+	ForwardsMax    int64  `json:"forwards_max,omitempty"`
+	OpenWeatherMap struct {
+		Enabled bool   `json:"enabled,omitempty"`
+		Country bool   `json:"country,omitempty"`
+		Appid   string `json:"appid,omitempty"`
+	} `json:"openweathermap,omitempty"`
+	Flickr struct {
+		Enabled          bool   `json:"enabled,omitempty"`
+		Key              string `json:"key,omitempty"`
+		Secret           string `json:"secret,omitempty"`
+		OAuthToken       string `json:"oauth_token,omitempty"`
+		OAuthTokenSecret string `json:"oauth_token_secret,omitempty"`
+	} `json:"flickr,omitempty"`
+	UserAgents  []string `json:"user_agents,omitempty"`
+	PcacheDB    map[string]*pebble.DB
+	RedisClient *redis.Client
+}
+
+/* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

internal/log/log.go

@@ -0,0 +1,106 @@
+package log
+
+import (
+	"fmt"
+	"log/slog"
+	"os"
+	"time"
+)
+
+// Init setup logger stuff.
+// level can be error, warn, info, debug if something other supplied info level selected.
+// fileDescriptor should be opened before supplying it to Init().
+func Init(level string, fileDescriptor *os.File) {
+	var loglevel slog.Level
+
+	// no panic, no trace.
+	switch level {
+	case "error":
+		loglevel = slog.LevelError
+
+	case "warn":
+		loglevel = slog.LevelWarn
+
+	case "info":
+		loglevel = slog.LevelInfo
+
+	case "debug":
+		loglevel = slog.LevelDebug
+
+	default:
+		loglevel = slog.LevelInfo
+
+	}
+
+	opts := &slog.HandlerOptions{
+		// Use the ReplaceAttr function on the handler options
+		// to be able to replace any single attribute in the log output
+		ReplaceAttr: func(groups []string, a slog.Attr) slog.Attr {
+			// check that we are handling the time key
+			if a.Key != slog.TimeKey {
+				return a
+			}
+
+			t := a.Value.Time()
+
+			// change the value from a time.Time to a String
+			// where the string has the correct time format.
+			a.Value = slog.StringValue(t.Format(time.DateTime))
+
+			return a
+		},
+
+		Level: loglevel,
+	}
+
+	slog.SetDefault(
+		slog.New(
+			slog.NewTextHandler(
+				fileDescriptor,
+				opts,
+			),
+		),
+	)
+}
+
+// Error passes message directly to slog.Error().
+func Error(message string) {
+	slog.Error(message)
+}
+
+// Errorf logs message on error log level and allows to supply arguments in printf() style.
+func Errorf(format string, a ...any) {
+	slog.Error(fmt.Sprintf(format, a...))
+}
+
+// Warn passes message directly to slog.Warn().
+func Warn(message string) {
+	slog.Warn(message)
+}
+
+// Warnf logs message on warn log level and allows to supply arguments in printf() style.
+func Warnf(format string, a ...any) {
+	slog.Warn(fmt.Sprintf(format, a...))
+}
+
+// Info passes message directly to slog.Info().
+func Info(message string) {
+	slog.Info(message)
+}
+
+// Infof logs message on info log level and allows to supply arguments in printf() style.
+func Infof(format string, a ...any) {
+	slog.Info(fmt.Sprintf(format, a...))
+}
+
+// Debug passes message directly to slog.Debug().
+func Debug(message string) {
+	slog.Debug(message)
+}
+
+// Debugf logs message on debug log level and allows to supply arguments in printf() style.
+func Debugf(format string, a ...any) {
+	slog.Debug(fmt.Sprintf(format, a...))
+}
+
+/* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

internal/openweathermap/openweathermap.go

@@ -2,6 +2,7 @@ package openweathermap
 
 import (
 	"aleesa-webapp-go/internal/config"
+	"aleesa-webapp-go/internal/log"
 	"aleesa-webapp-go/internal/pcachedb"
 	"crypto/sha256"
 	"fmt"
@@ -11,7 +12,6 @@ import (
 	"time"
 
 	owm "github.com/briandowns/openweathermap"
-	log "github.com/sirupsen/logrus"
 )
 
 // Структурка для передачи в обрабтчик запросов в openweathermap api.
@@ -319,7 +319,7 @@ func QueryOwmCache(cfg *config.MyConfig, item OwmItem) (string, error) {
 
 		// Если всё хорошо, надо обновить кэш.
 		if err := UpdateOwmCache(cfg, tsNow, key, answer); err != nil {
-			log.Error(err)
+			log.Errorf("%s", err)
 		}
 
 		return answer, err

internal/pcachedb/pcachedb.go

@@ -2,6 +2,7 @@ package pcachedb
 
 import (
 	"aleesa-webapp-go/internal/config"
+	"aleesa-webapp-go/internal/log"
 	"crypto/sha256"
 	"errors"
 	"fmt"
@@ -9,7 +10,6 @@ import (
 
 	"github.com/cockroachdb/errors/oserror"
 	"github.com/cockroachdb/pebble"
-	log "github.com/sirupsen/logrus"
 )
 
 // StoreKV сохраняет в указанной бд ключ и значение.

internal/prazdnikisegodnyaru/prazdnikisegodnyaru.go

@@ -2,6 +2,7 @@ package prazdnikisegodnyaru
 
 import (
 	"aleesa-webapp-go/internal/config"
+	"aleesa-webapp-go/internal/log"
 	"aleesa-webapp-go/internal/pcachedb"
 	"context"
 	"errors"
@@ -12,7 +13,6 @@ import (
 	"time"
 
 	"github.com/carlmjohnson/requests"
-	log "github.com/sirupsen/logrus"
 	"golang.org/x/net/html"
 )
 
@@ -36,7 +36,7 @@ func PsrClient(cfg *config.MyConfig) (string, error) {
 		}
 
 		if err := UpdatePsrCache(cfg, eTime, answer); err != nil {
-			log.Error(err)
+			log.Errorf("%s", err)
 		}
 
 		return answer, nil
@@ -56,7 +56,7 @@ func PsrClient(cfg *config.MyConfig) (string, error) {
 
 		// Если всё хорошо, надо обновить кэш
 		if err := UpdatePsrCache(cfg, eTime, answer); err != nil {
-			log.Error(err)
+			log.Errorf("%s", err)
 		}
 
 		return answer, nil
@@ -78,7 +78,7 @@ func PsrClient(cfg *config.MyConfig) (string, error) {
 
 		// Если всё хорошо, надо обновить кэш
 		if err := UpdatePsrCache(cfg, eTime, answer); err != nil {
-			log.Error(err)
+			log.Errorf("%s", err)
 		}
 
 		return answer, nil

internal/webapp/MsgParser.go

@@ -1,10 +1,11 @@
-package lib
+package webapp
 
 import (
 	"aleesa-webapp-go/internal/anekdotru"
 	"aleesa-webapp-go/internal/bunicomic"
 	"aleesa-webapp-go/internal/config"
 	"aleesa-webapp-go/internal/flickr"
+	"aleesa-webapp-go/internal/log"
 	"aleesa-webapp-go/internal/monkeyuser"
 	"aleesa-webapp-go/internal/oboobs"
 	"aleesa-webapp-go/internal/obutts"
@@ -12,7 +13,6 @@ import (
 	"aleesa-webapp-go/internal/prazdnikisegodnyaru"
 	"aleesa-webapp-go/internal/randomfox"
 	"aleesa-webapp-go/internal/thecatapi"
-	"aleesa-webapp-go/internal/types"
 	"aleesa-webapp-go/internal/xkcdru"
 	"context"
 	"encoding/json"
@@ -20,8 +20,6 @@ import (
 	"math/rand/v2"
 	"regexp"
 	"time"
-
-	log "github.com/sirupsen/logrus"
 )
 
 // MsgParser горутинка, которая парсит json-чики прилетевшие из REDIS-ки.
@@ -29,7 +27,7 @@ func MsgParser(cfg *config.MyConfig, ctx context.Context, msg string) {
 	var (
 		sendTo string
 		answer string
-		j      types.RMsg
+		j      RMsg
 		err    error
 	)
 
@@ -46,7 +44,7 @@ func MsgParser(cfg *config.MyConfig, ctx context.Context, msg string) {
 	j, err = ValidateRmsg(cfg, j, msg)
 
 	if err != nil {
-		log.Warn(err)
+		log.Warnf("%s", err)
 
 		return
 	}
@@ -417,7 +415,7 @@ func MsgParser(cfg *config.MyConfig, ctx context.Context, msg string) {
 			city := openweathermap.QueryOwmUserCache(cfg, j.Chatid, j.Userid)
 
 			if city == "" {
-				j.Message = fmt.Sprint("Не припоминаю, какой город вас интересовал в прошлый раз.")
+				j.Message = "Не припоминаю, какой город вас интересовал в прошлый раз."
 			} else {
 				if answer, err := openweathermap.OwmClient(cfg, city, 0); err != nil {
 					log.Errorf("Unable to handle city %s in openweartermap api: %s", city, err)
@@ -470,7 +468,7 @@ func MsgParser(cfg *config.MyConfig, ctx context.Context, msg string) {
 	}
 
 	// Настало время формировать json и засылать его в дальше
-	var message types.SMsg
+	var message SMsg
 	message.From = j.From
 	message.Userid = j.Userid
 	message.Chatid = j.Chatid
@@ -502,62 +500,4 @@ func MsgParser(cfg *config.MyConfig, ctx context.Context, msg string) {
 	}
 }
 
-// ValidateRmsg валидирует входящее сообщение.
-func ValidateRmsg(cfg *config.MyConfig, j types.RMsg, msg string) (types.RMsg, error) {
-	if exist := j.From; exist == "" {
-		emsg := fmt.Errorf("incorrect msg from redis, no from field: %s", msg)
-
-		return j, emsg
-	}
-
-	if exist := j.Chatid; exist == "" {
-		emsg := fmt.Errorf("incorrect msg from redis, no chatid field: %s", msg)
-
-		return j, emsg
-	}
-
-	if exist := j.Userid; exist == "" {
-		emsg := fmt.Errorf("incorrect msg from redis, no userid field: %s", msg)
-
-		return j, emsg
-	}
-
-	// j.Threadid может быть пустым, значит либо нам его не дали, либо дали пустым. Это нормально.
-
-	if exist := j.Message; exist == "" {
-		emsg := fmt.Errorf("incorrect msg from redis, no message field: %s", msg)
-
-		return j, emsg
-	}
-
-	if exist := j.Plugin; exist == "" {
-		emsg := fmt.Errorf("incorrect msg from redis, no plugin field: %s", msg)
-
-		return j, emsg
-	}
-
-	if exist := j.Mode; exist == "" {
-		emsg := fmt.Errorf("incorrect msg from redis, no mode field: %s", msg)
-
-		return j, emsg
-	}
-
-	// j.Misc.Answer может и не быть, тогда ответа на такое сообщение не будет
-	// j.Misc.Botnick тоже можно не передавать, тогда будет записана пустая строка
-	// j.Misc.Csign если нам его не передали, возьмём значение из конфига
-	if exist := j.Misc.Csign; exist == "" {
-		j.Misc.Csign = cfg.Csign
-	}
-
-	// j.Misc.Fwdcnt если нам его не передали, то будет 0
-	if exist := j.Misc.Fwdcnt; exist == 0 {
-		j.Misc.Fwdcnt = 1
-	}
-
-	// j.Misc.GoodMorning может быть быть 1 или 0, по-умолчанию 0
-	// j.Misc.Msgformat может быть быть 1 или 0, по-умолчанию 0
-	// j.Misc.Username можно не передавать, тогда будет пустая строка
-	return j, nil
-}
-
 /* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

internal/webapp/ReadConfig.go

@@ -1,54 +1,24 @@
-package config
+package webapp
 
 import (
-	"aleesa-webapp-go/internal/defaults"
-	"bufio"
+	"aleesa-webapp-go/internal/config"
+	"aleesa-webapp-go/internal/log"
 	"encoding/json"
 	"fmt"
 	"os"
 	"unsafe"
 
 	"github.com/cockroachdb/pebble"
-	"github.com/go-redis/redis/v8"
 	"github.com/hjson/hjson-go"
-	log "github.com/sirupsen/logrus"
 )
 
-// MyConfig структурка, описывающая конфиг.
-type MyConfig struct {
-	Server         string `json:"server,omitempty"`
-	Port           int    `json:"port,omitempty"`
-	Timeout        int    `json:"timeout,omitempty"`
-	Loglevel       string `json:"loglevel,omitempty"`
-	Log            string `json:"log,omitempty"`
-	Channel        string `json:"channel,omitempty"`
-	DataDir        string `json:"datadir,omitempty"`
-	Csign          string `json:"csign,omitempty"`
-	ForwardsMax    int64  `json:"forwards_max,omitempty"`
-	OpenWeatherMap struct {
-		Enabled bool   `json:"enabled,omitempty"`
-		Country bool   `json:"country,omitempty"`
-		Appid   string `json:"appid,omitempty"`
-	} `json:"openweathermap,omitempty"`
-	Flickr struct {
-		Enabled          bool   `json:"enabled,omitempty"`
-		Key              string `json:"key,omitempty"`
-		Secret           string `json:"secret,omitempty"`
-		OAuthToken       string `json:"oauth_token,omitempty"`
-		OAuthTokenSecret string `json:"oauth_token_secret,omitempty"`
-	} `json:"flickr,omitempty"`
-	UserAgents  []string `json:"user_agents,omitempty"`
-	PcacheDB    map[string]*pebble.DB
-	RedisClient *redis.Client
-}
-
 // ReadConfig читает и валидирует конфиг, а также выставляет некоторые default-ы, если значений для параметров в конфиге
 // нет.
-func ReadConfig() (*MyConfig, error) {
-	locations, err := defaults.DefaultConfigFileLocations()
+func ReadConfig() error {
+	locations, err := DefaultConfigFileLocations()
 
 	if err != nil {
-		return &MyConfig{}, err
+		return err
 	}
 
 	for _, location := range locations {
@@ -60,7 +30,7 @@ func ReadConfig() (*MyConfig, error) {
 		}
 
 		// Конфиг-файл длинноват для конфига, попробуем следующего кандидата
-		if fileInfo.Size() > defaults.ConfigFileSize {
+		if fileInfo.Size() > ConfigFileSize {
 			log.Warnf("Config file %s is too long for config, skipping", location)
 
 			continue
@@ -79,7 +49,7 @@ func ReadConfig() (*MyConfig, error) {
 		// Интереснее на выходе получить структурку: то есть мы вначале конфиг преобразуем в map-ку, затем эту map-ку
 		// сериализуем в json, а потом json преврщааем в стркутурку. Не очень эффективно, но он и не часто требуется.
 		var (
-			sampleConfig *MyConfig
+			sampleConfig *config.MyConfig
 			tmp          map[string]interface{}
 		)
 
@@ -109,19 +79,19 @@ func ReadConfig() (*MyConfig, error) {
 
 		// Валидируем значения из конфига
 		if sampleConfig.Server == "" {
-			sampleConfig.Server = defaults.Host
+			sampleConfig.Server = Host
 		}
 
 		if sampleConfig.Port == 0 {
-			sampleConfig.Port = defaults.RedisPort
+			sampleConfig.Port = RedisPort
 		}
 
 		if sampleConfig.Timeout == 0 {
-			sampleConfig.Timeout = defaults.NetworkTimeout
+			sampleConfig.Timeout = NetworkTimeout
 		}
 
 		if sampleConfig.Loglevel == "" {
-			sampleConfig.Loglevel = defaults.Loglevel
+			sampleConfig.Loglevel = Loglevel
 		}
 
 		// sampleConfig.Log = "" if not set
@@ -133,7 +103,7 @@ func ReadConfig() (*MyConfig, error) {
 		}
 
 		if sampleConfig.DataDir == "" {
-			sampleConfig.DataDir = defaults.DataDir
+			sampleConfig.DataDir = DataDir
 		}
 
 		if sampleConfig.Csign == "" {
@@ -143,7 +113,7 @@ func ReadConfig() (*MyConfig, error) {
 		}
 
 		if sampleConfig.ForwardsMax == 0 {
-			sampleConfig.ForwardsMax = defaults.ForwardMax
+			sampleConfig.ForwardsMax = ForwardMax
 		}
 
 		if unsafe.Sizeof(sampleConfig.OpenWeatherMap) == 0 {
@@ -165,7 +135,7 @@ func ReadConfig() (*MyConfig, error) {
 		uaList, err := ReadLines(uaFile)
 
 		if err != nil {
-			return &MyConfig{}, fmt.Errorf("unable to read %s: %w", uaFile, err)
+			return fmt.Errorf("unable to read %s: %w", uaFile, err)
 		}
 
 		sampleConfig.UserAgents = uaList
@@ -174,37 +144,12 @@ func ReadConfig() (*MyConfig, error) {
 
 		log.Infof("Using %s as config file", location)
 
-		return sampleConfig, nil
+		Config = sampleConfig
+
+		return nil
 	}
 
-	return &MyConfig{}, fmt.Errorf("config was not loaded! Refusing to start")
-}
-
-// ReadLines читает даденный файл построчно в массив строк.
-func ReadLines(path string) ([]string, error) {
-	file, err := os.Open(path)
-
-	if err != nil {
-		return nil, err
-	}
-
-	defer func(file *os.File) {
-		err := file.Close()
-		if err != nil {
-			log.Errorf("Unable to close file %s:%s", path, err)
-		}
-	}(file)
-
-	var (
-		lines   []string
-		scanner = bufio.NewScanner(file)
-	)
-
-	for scanner.Scan() {
-		lines = append(lines, scanner.Text())
-	}
-
-	return lines, scanner.Err()
+	return fmt.Errorf("config was not loaded! Refusing to start")
 }
 
 /* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

internal/webapp/ReadLines.go

@@ -0,0 +1,36 @@
+package webapp
+
+import (
+	"aleesa-webapp-go/internal/log"
+	"bufio"
+	"os"
+)
+
+// ReadLines читает даденный файл построчно в массив строк.
+func ReadLines(path string) ([]string, error) {
+	file, err := os.Open(path)
+
+	if err != nil {
+		return nil, err
+	}
+
+	defer func(file *os.File) {
+		err := file.Close()
+		if err != nil {
+			log.Errorf("Unable to close file %s:%s", path, err)
+		}
+	}(file)
+
+	var (
+		lines   []string
+		scanner = bufio.NewScanner(file)
+	)
+
+	for scanner.Scan() {
+		lines = append(lines, scanner.Text())
+	}
+
+	return lines, scanner.Err()
+}
+
+/* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

internal/webapp/ValidateRmsg.go

@@ -0,0 +1,66 @@
+package webapp
+
+import (
+	"aleesa-webapp-go/internal/config"
+	"fmt"
+)
+
+// ValidateRmsg валидирует входящее сообщение.
+func ValidateRmsg(cfg *config.MyConfig, j RMsg, msg string) (RMsg, error) {
+	if exist := j.From; exist == "" {
+		emsg := fmt.Errorf("incorrect msg from redis, no from field: %s", msg)
+
+		return j, emsg
+	}
+
+	if exist := j.Chatid; exist == "" {
+		emsg := fmt.Errorf("incorrect msg from redis, no chatid field: %s", msg)
+
+		return j, emsg
+	}
+
+	if exist := j.Userid; exist == "" {
+		emsg := fmt.Errorf("incorrect msg from redis, no userid field: %s", msg)
+
+		return j, emsg
+	}
+
+	// j.Threadid может быть пустым, значит либо нам его не дали, либо дали пустым. Это нормально.
+
+	if exist := j.Message; exist == "" {
+		emsg := fmt.Errorf("incorrect msg from redis, no message field: %s", msg)
+
+		return j, emsg
+	}
+
+	if exist := j.Plugin; exist == "" {
+		emsg := fmt.Errorf("incorrect msg from redis, no plugin field: %s", msg)
+
+		return j, emsg
+	}
+
+	if exist := j.Mode; exist == "" {
+		emsg := fmt.Errorf("incorrect msg from redis, no mode field: %s", msg)
+
+		return j, emsg
+	}
+
+	// j.Misc.Answer может и не быть, тогда ответа на такое сообщение не будет
+	// j.Misc.Botnick тоже можно не передавать, тогда будет записана пустая строка
+	// j.Misc.Csign если нам его не передали, возьмём значение из конфига
+	if exist := j.Misc.Csign; exist == "" {
+		j.Misc.Csign = cfg.Csign
+	}
+
+	// j.Misc.Fwdcnt если нам его не передали, то будет 0
+	if exist := j.Misc.Fwdcnt; exist == 0 {
+		j.Misc.Fwdcnt = 1
+	}
+
+	// j.Misc.GoodMorning может быть быть 1 или 0, по-умолчанию 0
+	// j.Misc.Msgformat может быть быть 1 или 0, по-умолчанию 0
+	// j.Misc.Username можно не передавать, тогда будет пустая строка
+	return j, nil
+}
+
+/* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

internal/webapp/defaults.go

@@ -1,4 +1,4 @@
-package defaults
+package webapp
 
 import (
 	"fmt"
@@ -6,14 +6,6 @@ import (
 	"path/filepath"
 )
 
-var ForwardMax int64 = 5
-var ConfigFileSize int64 = 65535
-var DataDir = "data"
-var Host = "localhost"
-var Loglevel = "info"
-var RedisPort = 6379
-var NetworkTimeout = 10
-
 func DefaultConfigFileLocations() ([]string, error) {
 	var locations []string
 

internal/webapp/globals.go

@@ -0,0 +1,18 @@
+package webapp
+
+import (
+	"aleesa-webapp-go/internal/config"
+)
+
+var (
+	ForwardMax     int64 = 5
+	ConfigFileSize int64 = 65535
+	DataDir              = "data"
+	Host                 = "localhost"
+	Loglevel             = "info"
+	RedisPort            = 6379
+	NetworkTimeout       = 10
+	Config         *config.MyConfig
+)
+
+/* vim: set ft=go noet ai ts=4 sw=4 sts=4: */

internal/webapp/types.go

@@ -1,4 +1,4 @@
-package types
+package webapp
 
 // RMsg структурка, описывающая входящее сообщение из pubsub-канала redis-ки.
 type RMsg struct {

internal/xkcdru/xkcdru.go

@@ -2,13 +2,12 @@ package xkcdru
 
 import (
 	"aleesa-webapp-go/internal/config"
+	"aleesa-webapp-go/internal/log"
 	"fmt"
 	"io"
 	"math/rand/v2"
 	"net/http"
 	"time"
-
-	log "github.com/sirupsen/logrus"
 )
 
 func APIClient(cfg *config.MyConfig) (string, error) {

vendor/github.com/DataDog/zstd/README.md

@@ -6,8 +6,8 @@
 
 [C Zstd Homepage](https://github.com/facebook/zstd)
 
-The current headers and C files are from *v1.5.6* (Commit
-[794ea1b](https://github.com/facebook/zstd/releases/tag/v1.5.6)).
+The current headers and C files are from *v1.4.4* (Commit
+[10f0e699](https://github.com/facebook/zstd/releases/tag/v1.4.4)).
 
 ## Usage
 
@@ -19,21 +19,6 @@ There are two main APIs:
 The compress/decompress APIs mirror that of lz4, while the streaming API was
 designed to be a drop-in replacement for zlib.
 
-### Building against an external libzstd
-
-By default, zstd source code is vendored in this repository and the binding will be built with
-the vendored source code bundled.
-
-If you want to build this binding against an external static or shared libzstd library, you can
-use the `external_libzstd` build tag. This will look for the libzstd pkg-config file and extract
-build and linking parameters from that pkg-config file.
-
-Note that it requires at least libzstd 1.4.0.
-
-```bash
-go build -tags external_libzstd
-```
-
 ### Simple `Compress/Decompress`
 
 
@@ -75,9 +60,6 @@ NewWriterLevelDict(w io.Writer, level int, dict []byte) *Writer
 // Write compresses the input data and write it to the underlying writer
 (w *Writer) Write(p []byte) (int, error)
 
-// Flush writes any unwritten data to the underlying writer
-(w *Writer) Flush() error
-
 // Close flushes the buffer and frees C zstd objects
 (w *Writer) Close() error
 ```

vendor/github.com/DataDog/zstd/ZSTD_LICENSE

@@ -2,7 +2,7 @@ BSD License
 
 For Zstandard software
 
-Copyright (c) Meta Platforms, Inc. and affiliates. All rights reserved.
+Copyright (c) 2016-present, Facebook, Inc. All rights reserved.
 
 Redistribution and use in source and binary forms, with or without modification,
 are permitted provided that the following conditions are met:
@@ -14,9 +14,9 @@ are permitted provided that the following conditions are met:
    this list of conditions and the following disclaimer in the documentation
    and/or other materials provided with the distribution.
 
- * Neither the name Facebook, nor Meta, nor the names of its contributors may
-   be used to endorse or promote products derived from this software without
-   specific prior written permission.
+ * Neither the name Facebook nor the names of its contributors may be used to
+   endorse or promote products derived from this software without specific
+   prior written permission.
 
 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
 ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED

vendor/github.com/DataDog/zstd/allocations.h

@@ -1,58 +0,0 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
-/* This file provides custom allocation primitives
- */
-
-#define ZSTD_DEPS_NEED_MALLOC
-#include "zstd_deps.h"   /* ZSTD_malloc, ZSTD_calloc, ZSTD_free, ZSTD_memset */
-
-#include "compiler.h" /* MEM_STATIC */
-#define ZSTD_STATIC_LINKING_ONLY
-#include "zstd.h" /* ZSTD_customMem */
-
-#ifndef ZSTD_ALLOCATIONS_H
-#define ZSTD_ALLOCATIONS_H
-
-/* custom memory allocation functions */
-
-MEM_STATIC void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem)
-{
-    if (customMem.customAlloc)
-        return customMem.customAlloc(customMem.opaque, size);
-    return ZSTD_malloc(size);
-}
-
-MEM_STATIC void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem)
-{
-    if (customMem.customAlloc) {
-        /* calloc implemented as malloc+memset;
-         * not as efficient as calloc, but next best guess for custom malloc */
-        void* const ptr = customMem.customAlloc(customMem.opaque, size);
-        ZSTD_memset(ptr, 0, size);
-        return ptr;
-    }
-    return ZSTD_calloc(1, size);
-}
-
-MEM_STATIC void ZSTD_customFree(void* ptr, ZSTD_customMem customMem)
-{
-    if (ptr!=NULL) {
-        if (customMem.customFree)
-            customMem.customFree(customMem.opaque, ptr);
-        else
-            ZSTD_free(ptr);
-    }
-}
-
-#endif /* ZSTD_ALLOCATIONS_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/bits.h

@@ -1,203 +0,0 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
-#ifndef ZSTD_BITS_H
-#define ZSTD_BITS_H
-
-#include "mem.h"
-
-MEM_STATIC unsigned ZSTD_countTrailingZeros32_fallback(U32 val)
-{
-    assert(val != 0);
-    {
-        static const U32 DeBruijnBytePos[32] = {0, 1, 28, 2, 29, 14, 24, 3,
-                                                30, 22, 20, 15, 25, 17, 4, 8,
-                                                31, 27, 13, 23, 21, 19, 16, 7,
-                                                26, 12, 18, 6, 11, 5, 10, 9};
-        return DeBruijnBytePos[((U32) ((val & -(S32) val) * 0x077CB531U)) >> 27];
-    }
-}
-
-MEM_STATIC unsigned ZSTD_countTrailingZeros32(U32 val)
-{
-    assert(val != 0);
-#   if defined(_MSC_VER)
-#       if STATIC_BMI2 == 1
-            return (unsigned)_tzcnt_u32(val);
-#       else
-            if (val != 0) {
-                unsigned long r;
-                _BitScanForward(&r, val);
-                return (unsigned)r;
-            } else {
-                /* Should not reach this code path */
-                __assume(0);
-            }
-#       endif
-#   elif defined(__GNUC__) && (__GNUC__ >= 4)
-        return (unsigned)__builtin_ctz(val);
-#   else
-        return ZSTD_countTrailingZeros32_fallback(val);
-#   endif
-}
-
-MEM_STATIC unsigned ZSTD_countLeadingZeros32_fallback(U32 val) {
-    assert(val != 0);
-    {
-        static const U32 DeBruijnClz[32] = {0, 9, 1, 10, 13, 21, 2, 29,
-                                            11, 14, 16, 18, 22, 25, 3, 30,
-                                            8, 12, 20, 28, 15, 17, 24, 7,
-                                            19, 27, 23, 6, 26, 5, 4, 31};
-        val |= val >> 1;
-        val |= val >> 2;
-        val |= val >> 4;
-        val |= val >> 8;
-        val |= val >> 16;
-        return 31 - DeBruijnClz[(val * 0x07C4ACDDU) >> 27];
-    }
-}
-
-MEM_STATIC unsigned ZSTD_countLeadingZeros32(U32 val)
-{
-    assert(val != 0);
-#   if defined(_MSC_VER)
-#       if STATIC_BMI2 == 1
-            return (unsigned)_lzcnt_u32(val);
-#       else
-            if (val != 0) {
-                unsigned long r;
-                _BitScanReverse(&r, val);
-                return (unsigned)(31 - r);
-            } else {
-                /* Should not reach this code path */
-                __assume(0);
-            }
-#       endif
-#   elif defined(__GNUC__) && (__GNUC__ >= 4)
-        return (unsigned)__builtin_clz(val);
-#   else
-        return ZSTD_countLeadingZeros32_fallback(val);
-#   endif
-}
-
-MEM_STATIC unsigned ZSTD_countTrailingZeros64(U64 val)
-{
-    assert(val != 0);
-#   if defined(_MSC_VER) && defined(_WIN64)
-#       if STATIC_BMI2 == 1
-            return (unsigned)_tzcnt_u64(val);
-#       else
-            if (val != 0) {
-                unsigned long r;
-                _BitScanForward64(&r, val);
-                return (unsigned)r;
-            } else {
-                /* Should not reach this code path */
-                __assume(0);
-            }
-#       endif
-#   elif defined(__GNUC__) && (__GNUC__ >= 4) && defined(__LP64__)
-        return (unsigned)__builtin_ctzll(val);
-#   else
-        {
-            U32 mostSignificantWord = (U32)(val >> 32);
-            U32 leastSignificantWord = (U32)val;
-            if (leastSignificantWord == 0) {
-                return 32 + ZSTD_countTrailingZeros32(mostSignificantWord);
-            } else {
-                return ZSTD_countTrailingZeros32(leastSignificantWord);
-            }
-        }
-#   endif
-}
-
-MEM_STATIC unsigned ZSTD_countLeadingZeros64(U64 val)
-{
-    assert(val != 0);
-#   if defined(_MSC_VER) && defined(_WIN64)
-#       if STATIC_BMI2 == 1
-            return (unsigned)_lzcnt_u64(val);
-#       else
-            if (val != 0) {
-                unsigned long r;
-                _BitScanReverse64(&r, val);
-                return (unsigned)(63 - r);
-            } else {
-                /* Should not reach this code path */
-                __assume(0);
-            }
-#       endif
-#   elif defined(__GNUC__) && (__GNUC__ >= 4)
-        return (unsigned)(__builtin_clzll(val));
-#   else
-        {
-            U32 mostSignificantWord = (U32)(val >> 32);
-            U32 leastSignificantWord = (U32)val;
-            if (mostSignificantWord == 0) {
-                return 32 + ZSTD_countLeadingZeros32(leastSignificantWord);
-            } else {
-                return ZSTD_countLeadingZeros32(mostSignificantWord);
-            }
-        }
-#   endif
-}
-
-MEM_STATIC unsigned ZSTD_NbCommonBytes(size_t val)
-{
-    if (MEM_isLittleEndian()) {
-        if (MEM_64bits()) {
-            return ZSTD_countTrailingZeros64((U64)val) >> 3;
-        } else {
-            return ZSTD_countTrailingZeros32((U32)val) >> 3;
-        }
-    } else {  /* Big Endian CPU */
-        if (MEM_64bits()) {
-            return ZSTD_countLeadingZeros64((U64)val) >> 3;
-        } else {
-            return ZSTD_countLeadingZeros32((U32)val) >> 3;
-        }
-    }
-}
-
-MEM_STATIC unsigned ZSTD_highbit32(U32 val)   /* compress, dictBuilder, decodeCorpus */
-{
-    assert(val != 0);
-    return 31 - ZSTD_countLeadingZeros32(val);
-}
-
-/* ZSTD_rotateRight_*():
- * Rotates a bitfield to the right by "count" bits.
- * https://en.wikipedia.org/w/index.php?title=Circular_shift&oldid=991635599#Implementing_circular_shifts
- */
-MEM_STATIC
-U64 ZSTD_rotateRight_U64(U64 const value, U32 count) {
-    assert(count < 64);
-    count &= 0x3F; /* for fickle pattern recognition */
-    return (value >> count) | (U64)(value << ((0U - count) & 0x3F));
-}
-
-MEM_STATIC
-U32 ZSTD_rotateRight_U32(U32 const value, U32 count) {
-    assert(count < 32);
-    count &= 0x1F; /* for fickle pattern recognition */
-    return (value >> count) | (U32)(value << ((0U - count) & 0x1F));
-}
-
-MEM_STATIC
-U16 ZSTD_rotateRight_U16(U16 const value, U32 count) {
-    assert(count < 16);
-    count &= 0x0F; /* for fickle pattern recognition */
-    return (value >> count) | (U16)(value << ((0U - count) & 0x0F));
-}
-
-#endif /* ZSTD_BITS_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/bitstream.h

@@ -1,16 +1,35 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * bitstream
- * Part of FSE library
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- * You can contact the author at :
- * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   bitstream
+   Part of FSE library
+   Copyright (C) 2013-present, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+   You can contact the author at :
+   - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
 ****************************************************************** */
 #ifndef BITSTREAM_H_MODULE
 #define BITSTREAM_H_MODULE
@@ -18,6 +37,7 @@
 #if defined (__cplusplus)
 extern "C" {
 #endif
+
 /*
 *  This API consists of small unitary functions, which must be inlined for best performance.
 *  Since link-time-optimization is not available for all compilers,
@@ -28,21 +48,17 @@ extern "C" {
 *  Dependencies
 ******************************************/
 #include "mem.h"            /* unaligned access routines */
-#include "compiler.h"       /* UNLIKELY() */
 #include "debug.h"          /* assert(), DEBUGLOG(), RAWLOG() */
 #include "error_private.h"  /* error codes and messages */
-#include "bits.h"           /* ZSTD_highbit32 */
 
 
 /*=========================================
 *  Target specific
 =========================================*/
-#ifndef ZSTD_NO_INTRINSICS
-#  if (defined(__BMI__) || defined(__BMI2__)) && defined(__GNUC__)
-#    include <immintrin.h>   /* support for bextr (experimental)/bzhi */
-#  elif defined(__ICCARM__)
-#    include <intrinsics.h>
-#  endif
+#if defined(__BMI__) && defined(__GNUC__)
+#  include <immintrin.h>   /* support for bextr (experimental) */
+#elif defined(__ICCARM__)
+#  include <intrinsics.h>
 #endif
 
 #define STREAM_ACCUMULATOR_MIN_32  25
@@ -91,20 +107,19 @@ MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC);
 /*-********************************************
 *  bitStream decoding API (read backward)
 **********************************************/
-typedef size_t BitContainerType;
 typedef struct {
-    BitContainerType bitContainer;
+    size_t   bitContainer;
     unsigned bitsConsumed;
     const char* ptr;
     const char* start;
     const char* limitPtr;
 } BIT_DStream_t;
 
-typedef enum { BIT_DStream_unfinished = 0,  /* fully refilled */
-               BIT_DStream_endOfBuffer = 1, /* still some bits left in bitstream */
-               BIT_DStream_completed = 2,   /* bitstream entirely consumed, bit-exact */
-               BIT_DStream_overflow = 3     /* user requested more bits than present in bitstream */
-    } BIT_DStream_status;  /* result of BIT_reloadDStream() */
+typedef enum { BIT_DStream_unfinished = 0,
+               BIT_DStream_endOfBuffer = 1,
+               BIT_DStream_completed = 2,
+               BIT_DStream_overflow = 3 } BIT_DStream_status;  /* result of BIT_reloadDStream() */
+               /* 1,2,4,8 would be better for bitmap combinations, but slows down performance a bit ... :( */
 
 MEM_STATIC size_t   BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize);
 MEM_STATIC size_t   BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits);
@@ -114,7 +129,7 @@ MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* bitD);
 
 /* Start by invoking BIT_initDStream().
 *  A chunk of the bitStream is then stored into a local register.
-*  Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (BitContainerType).
+*  Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
 *  You can then retrieve bitFields stored into the local register, **in reverse order**.
 *  Local register is explicitly reloaded from memory by the BIT_reloadDStream() method.
 *  A reload guarantee a minimum of ((8*sizeof(bitD->bitContainer))-7) bits when its result is BIT_DStream_unfinished.
@@ -135,6 +150,39 @@ MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC);
 MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits);
 /* faster, but works only if nbBits >= 1 */
 
+
+
+/*-**************************************************************
+*  Internal functions
+****************************************************************/
+MEM_STATIC unsigned BIT_highbit32 (U32 val)
+{
+    assert(val != 0);
+    {
+#   if defined(_MSC_VER)   /* Visual */
+        unsigned long r=0;
+        _BitScanReverse ( &r, val );
+        return (unsigned) r;
+#   elif defined(__GNUC__) && (__GNUC__ >= 3)   /* Use GCC Intrinsic */
+        return __builtin_clz (val) ^ 31;
+#   elif defined(__ICCARM__)    /* IAR Intrinsic */
+        return 31 - __CLZ(val);
+#   else   /* Software version */
+        static const unsigned DeBruijnClz[32] = { 0,  9,  1, 10, 13, 21,  2, 29,
+                                                 11, 14, 16, 18, 22, 25,  3, 30,
+                                                  8, 12, 20, 28, 15, 17, 24,  7,
+                                                 19, 27, 23,  6, 26,  5,  4, 31 };
+        U32 v = val;
+        v |= v >> 1;
+        v |= v >> 2;
+        v |= v >> 4;
+        v |= v >> 8;
+        v |= v >> 16;
+        return DeBruijnClz[ (U32) (v * 0x07C4ACDDU) >> 27];
+#   endif
+    }
+}
+
 /*=====    Local Constants   =====*/
 static const unsigned BIT_mask[] = {
     0,          1,         3,         7,         0xF,       0x1F,
@@ -164,26 +212,16 @@ MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC,
     return 0;
 }
 
-FORCE_INLINE_TEMPLATE size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
-{
-#if defined(STATIC_BMI2) && STATIC_BMI2 == 1 && !defined(ZSTD_NO_INTRINSICS)
-    return  _bzhi_u64(bitContainer, nbBits);
-#else
-    assert(nbBits < BIT_MASK_SIZE);
-    return bitContainer & BIT_mask[nbBits];
-#endif
-}
-
 /*! BIT_addBits() :
  *  can add up to 31 bits into `bitC`.
  *  Note : does not check for register overflow ! */
 MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC,
                             size_t value, unsigned nbBits)
 {
-    DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32);
+    MEM_STATIC_ASSERT(BIT_MASK_SIZE == 32);
     assert(nbBits < BIT_MASK_SIZE);
     assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
-    bitC->bitContainer |= BIT_getLowerBits(value, nbBits) << bitC->bitPos;
+    bitC->bitContainer |= (value & BIT_mask[nbBits]) << bitC->bitPos;
     bitC->bitPos += nbBits;
 }
 
@@ -253,7 +291,7 @@ MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC)
  */
 MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize)
 {
-    if (srcSize < 1) { ZSTD_memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
+    if (srcSize < 1) { memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
 
     bitD->start = (const char*)srcBuffer;
     bitD->limitPtr = bitD->start + sizeof(bitD->bitContainer);
@@ -262,35 +300,35 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si
         bitD->ptr   = (const char*)srcBuffer + srcSize - sizeof(bitD->bitContainer);
         bitD->bitContainer = MEM_readLEST(bitD->ptr);
         { BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
-          bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0;  /* ensures bitsConsumed is always set */
+          bitD->bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0;  /* ensures bitsConsumed is always set */
           if (lastByte == 0) return ERROR(GENERIC); /* endMark not present */ }
     } else {
         bitD->ptr   = bitD->start;
         bitD->bitContainer = *(const BYTE*)(bitD->start);
         switch(srcSize)
         {
-        case 7: bitD->bitContainer += (BitContainerType)(((const BYTE*)(srcBuffer))[6]) << (sizeof(bitD->bitContainer)*8 - 16);
-                ZSTD_FALLTHROUGH;
+        case 7: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[6]) << (sizeof(bitD->bitContainer)*8 - 16);
+                /* fall-through */
 
-        case 6: bitD->bitContainer += (BitContainerType)(((const BYTE*)(srcBuffer))[5]) << (sizeof(bitD->bitContainer)*8 - 24);
-                ZSTD_FALLTHROUGH;
+        case 6: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[5]) << (sizeof(bitD->bitContainer)*8 - 24);
+                /* fall-through */
 
-        case 5: bitD->bitContainer += (BitContainerType)(((const BYTE*)(srcBuffer))[4]) << (sizeof(bitD->bitContainer)*8 - 32);
-                ZSTD_FALLTHROUGH;
+        case 5: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[4]) << (sizeof(bitD->bitContainer)*8 - 32);
+                /* fall-through */
 
-        case 4: bitD->bitContainer += (BitContainerType)(((const BYTE*)(srcBuffer))[3]) << 24;
-                ZSTD_FALLTHROUGH;
+        case 4: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[3]) << 24;
+                /* fall-through */
 
-        case 3: bitD->bitContainer += (BitContainerType)(((const BYTE*)(srcBuffer))[2]) << 16;
-                ZSTD_FALLTHROUGH;
+        case 3: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[2]) << 16;
+                /* fall-through */
 
-        case 2: bitD->bitContainer += (BitContainerType)(((const BYTE*)(srcBuffer))[1]) <<  8;
-                ZSTD_FALLTHROUGH;
+        case 2: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[1]) <<  8;
+                /* fall-through */
 
         default: break;
         }
         {   BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
-            bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0;
+            bitD->bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0;
             if (lastByte == 0) return ERROR(corruption_detected);  /* endMark not present */
         }
         bitD->bitsConsumed += (U32)(sizeof(bitD->bitContainer) - srcSize)*8;
@@ -299,26 +337,23 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si
     return srcSize;
 }
 
-FORCE_INLINE_TEMPLATE size_t BIT_getUpperBits(BitContainerType bitContainer, U32 const start)
+MEM_STATIC size_t BIT_getUpperBits(size_t bitContainer, U32 const start)
 {
     return bitContainer >> start;
 }
 
-FORCE_INLINE_TEMPLATE size_t BIT_getMiddleBits(BitContainerType bitContainer, U32 const start, U32 const nbBits)
+MEM_STATIC size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits)
 {
     U32 const regMask = sizeof(bitContainer)*8 - 1;
     /* if start > regMask, bitstream is corrupted, and result is undefined */
     assert(nbBits < BIT_MASK_SIZE);
-    /* x86 transform & ((1 << nbBits) - 1) to bzhi instruction, it is better
-     * than accessing memory. When bmi2 instruction is not present, we consider
-     * such cpus old (pre-Haswell, 2013) and their performance is not of that
-     * importance.
-     */
-#if defined(__x86_64__) || defined(_M_X86)
-    return (bitContainer >> (start & regMask)) & ((((U64)1) << nbBits) - 1);
-#else
     return (bitContainer >> (start & regMask)) & BIT_mask[nbBits];
-#endif
+}
+
+MEM_STATIC size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
+{
+    assert(nbBits < BIT_MASK_SIZE);
+    return bitContainer & BIT_mask[nbBits];
 }
 
 /*! BIT_lookBits() :
@@ -327,7 +362,7 @@ FORCE_INLINE_TEMPLATE size_t BIT_getMiddleBits(BitContainerType bitContainer, U3
  *  On 32-bits, maxNbBits==24.
  *  On 64-bits, maxNbBits==56.
  * @return : value extracted */
-FORCE_INLINE_TEMPLATE size_t BIT_lookBits(const BIT_DStream_t*  bitD, U32 nbBits)
+MEM_STATIC size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits)
 {
     /* arbitrate between double-shift and shift+mask */
 #if 1
@@ -350,7 +385,7 @@ MEM_STATIC size_t BIT_lookBitsFast(const BIT_DStream_t* bitD, U32 nbBits)
     return (bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> (((regMask+1)-nbBits) & regMask);
 }
 
-FORCE_INLINE_TEMPLATE void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
+MEM_STATIC void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
 {
     bitD->bitsConsumed += nbBits;
 }
@@ -359,7 +394,7 @@ FORCE_INLINE_TEMPLATE void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
  *  Read (consume) next n bits from local register and update.
  *  Pay attention to not read more than nbBits contained into local register.
  * @return : extracted value. */
-FORCE_INLINE_TEMPLATE size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
+MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
 {
     size_t const value = BIT_lookBits(bitD, nbBits);
     BIT_skipBits(bitD, nbBits);
@@ -367,7 +402,7 @@ FORCE_INLINE_TEMPLATE size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
 }
 
 /*! BIT_readBitsFast() :
- *  unsafe version; only works if nbBits >= 1 */
+ *  unsafe version; only works only if nbBits >= 1 */
 MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits)
 {
     size_t const value = BIT_lookBitsFast(bitD, nbBits);
@@ -376,60 +411,27 @@ MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits)
     return value;
 }
 
-/*! BIT_reloadDStream_internal() :
- *  Simple variant of BIT_reloadDStream(), with two conditions:
- *  1. bitstream is valid : bitsConsumed <= sizeof(bitD->bitContainer)*8
- *  2. look window is valid after shifted down : bitD->ptr >= bitD->start
- */
-MEM_STATIC BIT_DStream_status BIT_reloadDStream_internal(BIT_DStream_t* bitD)
-{
-    assert(bitD->bitsConsumed <= sizeof(bitD->bitContainer)*8);
-    bitD->ptr -= bitD->bitsConsumed >> 3;
-    assert(bitD->ptr >= bitD->start);
-    bitD->bitsConsumed &= 7;
-    bitD->bitContainer = MEM_readLEST(bitD->ptr);
-    return BIT_DStream_unfinished;
-}
-
-/*! BIT_reloadDStreamFast() :
- *  Similar to BIT_reloadDStream(), but with two differences:
- *  1. bitsConsumed <= sizeof(bitD->bitContainer)*8 must hold!
- *  2. Returns BIT_DStream_overflow when bitD->ptr < bitD->limitPtr, at this
- *     point you must use BIT_reloadDStream() to reload.
- */
-MEM_STATIC BIT_DStream_status BIT_reloadDStreamFast(BIT_DStream_t* bitD)
-{
-    if (UNLIKELY(bitD->ptr < bitD->limitPtr))
-        return BIT_DStream_overflow;
-    return BIT_reloadDStream_internal(bitD);
-}
-
 /*! BIT_reloadDStream() :
  *  Refill `bitD` from buffer previously set in BIT_initDStream() .
- *  This function is safe, it guarantees it will not never beyond src buffer.
+ *  This function is safe, it guarantees it will not read beyond src buffer.
  * @return : status of `BIT_DStream_t` internal register.
  *           when status == BIT_DStream_unfinished, internal register is filled with at least 25 or 57 bits */
-FORCE_INLINE_TEMPLATE BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD)
+MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD)
 {
-    /* note : once in overflow mode, a bitstream remains in this mode until it's reset */
-    if (UNLIKELY(bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8))) {
-        static const BitContainerType zeroFilled = 0;
-        bitD->ptr = (const char*)&zeroFilled; /* aliasing is allowed for char */
-        /* overflow detected, erroneous scenario or end of stream: no update */
+    if (bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8))  /* overflow detected, like end of stream */
         return BIT_DStream_overflow;
-    }
-
-    assert(bitD->ptr >= bitD->start);
 
     if (bitD->ptr >= bitD->limitPtr) {
-        return BIT_reloadDStream_internal(bitD);
+        bitD->ptr -= bitD->bitsConsumed >> 3;
+        bitD->bitsConsumed &= 7;
+        bitD->bitContainer = MEM_readLEST(bitD->ptr);
+        return BIT_DStream_unfinished;
     }
     if (bitD->ptr == bitD->start) {
-        /* reached end of bitStream => no update */
         if (bitD->bitsConsumed < sizeof(bitD->bitContainer)*8) return BIT_DStream_endOfBuffer;
         return BIT_DStream_completed;
     }
-    /* start < ptr < limitPtr => cautious update */
+    /* start < ptr < limitPtr */
     {   U32 nbBytes = bitD->bitsConsumed >> 3;
         BIT_DStream_status result = BIT_DStream_unfinished;
         if (bitD->ptr - nbBytes < bitD->start) {
@@ -456,5 +458,3 @@ MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* DStream)
 #endif
 
 #endif /* BITSTREAM_H_MODULE */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/clevels.h

@@ -1,137 +0,0 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
-#ifndef ZSTD_CLEVELS_H
-#define ZSTD_CLEVELS_H
-
-#define ZSTD_STATIC_LINKING_ONLY  /* ZSTD_compressionParameters  */
-#include "zstd.h"
-
-/*-=====  Pre-defined compression levels  =====-*/
-
-#define ZSTD_MAX_CLEVEL     22
-
-#ifdef __GNUC__
-__attribute__((__unused__))
-#endif
-
-static const ZSTD_compressionParameters ZSTD_defaultCParameters[4][ZSTD_MAX_CLEVEL+1] = {
-{   /* "default" - for any srcSize > 256 KB */
-    /* W,  C,  H,  S,  L, TL, strat */
-    { 19, 12, 13,  1,  6,  1, ZSTD_fast    },  /* base for negative levels */
-    { 19, 13, 14,  1,  7,  0, ZSTD_fast    },  /* level  1 */
-    { 20, 15, 16,  1,  6,  0, ZSTD_fast    },  /* level  2 */
-    { 21, 16, 17,  1,  5,  0, ZSTD_dfast   },  /* level  3 */
-    { 21, 18, 18,  1,  5,  0, ZSTD_dfast   },  /* level  4 */
-    { 21, 18, 19,  3,  5,  2, ZSTD_greedy  },  /* level  5 */
-    { 21, 18, 19,  3,  5,  4, ZSTD_lazy    },  /* level  6 */
-    { 21, 19, 20,  4,  5,  8, ZSTD_lazy    },  /* level  7 */
-    { 21, 19, 20,  4,  5, 16, ZSTD_lazy2   },  /* level  8 */
-    { 22, 20, 21,  4,  5, 16, ZSTD_lazy2   },  /* level  9 */
-    { 22, 21, 22,  5,  5, 16, ZSTD_lazy2   },  /* level 10 */
-    { 22, 21, 22,  6,  5, 16, ZSTD_lazy2   },  /* level 11 */
-    { 22, 22, 23,  6,  5, 32, ZSTD_lazy2   },  /* level 12 */
-    { 22, 22, 22,  4,  5, 32, ZSTD_btlazy2 },  /* level 13 */
-    { 22, 22, 23,  5,  5, 32, ZSTD_btlazy2 },  /* level 14 */
-    { 22, 23, 23,  6,  5, 32, ZSTD_btlazy2 },  /* level 15 */
-    { 22, 22, 22,  5,  5, 48, ZSTD_btopt   },  /* level 16 */
-    { 23, 23, 22,  5,  4, 64, ZSTD_btopt   },  /* level 17 */
-    { 23, 23, 22,  6,  3, 64, ZSTD_btultra },  /* level 18 */
-    { 23, 24, 22,  7,  3,256, ZSTD_btultra2},  /* level 19 */
-    { 25, 25, 23,  7,  3,256, ZSTD_btultra2},  /* level 20 */
-    { 26, 26, 24,  7,  3,512, ZSTD_btultra2},  /* level 21 */
-    { 27, 27, 25,  9,  3,999, ZSTD_btultra2},  /* level 22 */
-},
-{   /* for srcSize <= 256 KB */
-    /* W,  C,  H,  S,  L,  T, strat */
-    { 18, 12, 13,  1,  5,  1, ZSTD_fast    },  /* base for negative levels */
-    { 18, 13, 14,  1,  6,  0, ZSTD_fast    },  /* level  1 */
-    { 18, 14, 14,  1,  5,  0, ZSTD_dfast   },  /* level  2 */
-    { 18, 16, 16,  1,  4,  0, ZSTD_dfast   },  /* level  3 */
-    { 18, 16, 17,  3,  5,  2, ZSTD_greedy  },  /* level  4.*/
-    { 18, 17, 18,  5,  5,  2, ZSTD_greedy  },  /* level  5.*/
-    { 18, 18, 19,  3,  5,  4, ZSTD_lazy    },  /* level  6.*/
-    { 18, 18, 19,  4,  4,  4, ZSTD_lazy    },  /* level  7 */
-    { 18, 18, 19,  4,  4,  8, ZSTD_lazy2   },  /* level  8 */
-    { 18, 18, 19,  5,  4,  8, ZSTD_lazy2   },  /* level  9 */
-    { 18, 18, 19,  6,  4,  8, ZSTD_lazy2   },  /* level 10 */
-    { 18, 18, 19,  5,  4, 12, ZSTD_btlazy2 },  /* level 11.*/
-    { 18, 19, 19,  7,  4, 12, ZSTD_btlazy2 },  /* level 12.*/
-    { 18, 18, 19,  4,  4, 16, ZSTD_btopt   },  /* level 13 */
-    { 18, 18, 19,  4,  3, 32, ZSTD_btopt   },  /* level 14.*/
-    { 18, 18, 19,  6,  3,128, ZSTD_btopt   },  /* level 15.*/
-    { 18, 19, 19,  6,  3,128, ZSTD_btultra },  /* level 16.*/
-    { 18, 19, 19,  8,  3,256, ZSTD_btultra },  /* level 17.*/
-    { 18, 19, 19,  6,  3,128, ZSTD_btultra2},  /* level 18.*/
-    { 18, 19, 19,  8,  3,256, ZSTD_btultra2},  /* level 19.*/
-    { 18, 19, 19, 10,  3,512, ZSTD_btultra2},  /* level 20.*/
-    { 18, 19, 19, 12,  3,512, ZSTD_btultra2},  /* level 21.*/
-    { 18, 19, 19, 13,  3,999, ZSTD_btultra2},  /* level 22.*/
-},
-{   /* for srcSize <= 128 KB */
-    /* W,  C,  H,  S,  L,  T, strat */
-    { 17, 12, 12,  1,  5,  1, ZSTD_fast    },  /* base for negative levels */
-    { 17, 12, 13,  1,  6,  0, ZSTD_fast    },  /* level  1 */
-    { 17, 13, 15,  1,  5,  0, ZSTD_fast    },  /* level  2 */
-    { 17, 15, 16,  2,  5,  0, ZSTD_dfast   },  /* level  3 */
-    { 17, 17, 17,  2,  4,  0, ZSTD_dfast   },  /* level  4 */
-    { 17, 16, 17,  3,  4,  2, ZSTD_greedy  },  /* level  5 */
-    { 17, 16, 17,  3,  4,  4, ZSTD_lazy    },  /* level  6 */
-    { 17, 16, 17,  3,  4,  8, ZSTD_lazy2   },  /* level  7 */
-    { 17, 16, 17,  4,  4,  8, ZSTD_lazy2   },  /* level  8 */
-    { 17, 16, 17,  5,  4,  8, ZSTD_lazy2   },  /* level  9 */
-    { 17, 16, 17,  6,  4,  8, ZSTD_lazy2   },  /* level 10 */
-    { 17, 17, 17,  5,  4,  8, ZSTD_btlazy2 },  /* level 11 */
-    { 17, 18, 17,  7,  4, 12, ZSTD_btlazy2 },  /* level 12 */
-    { 17, 18, 17,  3,  4, 12, ZSTD_btopt   },  /* level 13.*/
-    { 17, 18, 17,  4,  3, 32, ZSTD_btopt   },  /* level 14.*/
-    { 17, 18, 17,  6,  3,256, ZSTD_btopt   },  /* level 15.*/
-    { 17, 18, 17,  6,  3,128, ZSTD_btultra },  /* level 16.*/
-    { 17, 18, 17,  8,  3,256, ZSTD_btultra },  /* level 17.*/
-    { 17, 18, 17, 10,  3,512, ZSTD_btultra },  /* level 18.*/
-    { 17, 18, 17,  5,  3,256, ZSTD_btultra2},  /* level 19.*/
-    { 17, 18, 17,  7,  3,512, ZSTD_btultra2},  /* level 20.*/
-    { 17, 18, 17,  9,  3,512, ZSTD_btultra2},  /* level 21.*/
-    { 17, 18, 17, 11,  3,999, ZSTD_btultra2},  /* level 22.*/
-},
-{   /* for srcSize <= 16 KB */
-    /* W,  C,  H,  S,  L,  T, strat */
-    { 14, 12, 13,  1,  5,  1, ZSTD_fast    },  /* base for negative levels */
-    { 14, 14, 15,  1,  5,  0, ZSTD_fast    },  /* level  1 */
-    { 14, 14, 15,  1,  4,  0, ZSTD_fast    },  /* level  2 */
-    { 14, 14, 15,  2,  4,  0, ZSTD_dfast   },  /* level  3 */
-    { 14, 14, 14,  4,  4,  2, ZSTD_greedy  },  /* level  4 */
-    { 14, 14, 14,  3,  4,  4, ZSTD_lazy    },  /* level  5.*/
-    { 14, 14, 14,  4,  4,  8, ZSTD_lazy2   },  /* level  6 */
-    { 14, 14, 14,  6,  4,  8, ZSTD_lazy2   },  /* level  7 */
-    { 14, 14, 14,  8,  4,  8, ZSTD_lazy2   },  /* level  8.*/
-    { 14, 15, 14,  5,  4,  8, ZSTD_btlazy2 },  /* level  9.*/
-    { 14, 15, 14,  9,  4,  8, ZSTD_btlazy2 },  /* level 10.*/
-    { 14, 15, 14,  3,  4, 12, ZSTD_btopt   },  /* level 11.*/
-    { 14, 15, 14,  4,  3, 24, ZSTD_btopt   },  /* level 12.*/
-    { 14, 15, 14,  5,  3, 32, ZSTD_btultra },  /* level 13.*/
-    { 14, 15, 15,  6,  3, 64, ZSTD_btultra },  /* level 14.*/
-    { 14, 15, 15,  7,  3,256, ZSTD_btultra },  /* level 15.*/
-    { 14, 15, 15,  5,  3, 48, ZSTD_btultra2},  /* level 16.*/
-    { 14, 15, 15,  6,  3,128, ZSTD_btultra2},  /* level 17.*/
-    { 14, 15, 15,  7,  3,256, ZSTD_btultra2},  /* level 18.*/
-    { 14, 15, 15,  8,  3,256, ZSTD_btultra2},  /* level 19.*/
-    { 14, 15, 15,  8,  3,512, ZSTD_btultra2},  /* level 20.*/
-    { 14, 15, 15,  9,  3,512, ZSTD_btultra2},  /* level 21.*/
-    { 14, 15, 15, 10,  3,999, ZSTD_btultra2},  /* level 22.*/
-},
-};
-
-
-
-#endif  /* ZSTD_CLEVELS_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/compiler.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -12,17 +11,13 @@
 #ifndef ZSTD_COMPILER_H
 #define ZSTD_COMPILER_H
 
-#include <stddef.h>
-
-#include "portability_macros.h"
-
 /*-*******************************************************
 *  Compiler specifics
 *********************************************************/
 /* force inlining */
 
 #if !defined(ZSTD_NO_INLINE)
-#if (defined(__GNUC__) && !defined(__STRICT_ANSI__)) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   /* C99 */
+#if defined (__GNUC__) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   /* C99 */
 #  define INLINE_KEYWORD inline
 #else
 #  define INLINE_KEYWORD
@@ -43,30 +38,12 @@
 
 #endif
 
-/**
-  On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC).
-  This explicitly marks such functions as __cdecl so that the code will still compile
-  if a CC other than __cdecl has been made the default.
-*/
-#if  defined(_MSC_VER)
-#  define WIN_CDECL __cdecl
-#else
-#  define WIN_CDECL
-#endif
-
-/* UNUSED_ATTR tells the compiler it is okay if the function is unused. */
-#if defined(__GNUC__)
-#  define UNUSED_ATTR __attribute__((unused))
-#else
-#  define UNUSED_ATTR
-#endif
-
 /**
  * FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant
  * parameters. They must be inlined for the compiler to eliminate the constant
  * branches.
  */
-#define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR UNUSED_ATTR
+#define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR
 /**
  * HINT_INLINE is used to help the compiler generate better code. It is *not*
  * used for "templates", so it can be tweaked based on the compilers
@@ -81,28 +58,14 @@
 #if !defined(__clang__) && defined(__GNUC__) && __GNUC__ >= 4 && __GNUC_MINOR__ >= 8 && __GNUC__ < 5
 #  define HINT_INLINE static INLINE_KEYWORD
 #else
-#  define HINT_INLINE FORCE_INLINE_TEMPLATE
+#  define HINT_INLINE static INLINE_KEYWORD FORCE_INLINE_ATTR
 #endif
 
-/* "soft" inline :
- * The compiler is free to select if it's a good idea to inline or not.
- * The main objective is to silence compiler warnings
- * when a defined function in included but not used.
- *
- * Note : this macro is prefixed `MEM_` because it used to be provided by `mem.h` unit.
- * Updating the prefix is probably preferable, but requires a fairly large codemod,
- * since this name is used everywhere.
- */
-#ifndef MEM_STATIC  /* already defined in Linux Kernel mem.h */
+/* UNUSED_ATTR tells the compiler it is okay if the function is unused. */
 #if defined(__GNUC__)
-#  define MEM_STATIC static __inline UNUSED_ATTR
-#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
-#  define MEM_STATIC static inline
-#elif defined(_MSC_VER)
-#  define MEM_STATIC static __inline
+#  define UNUSED_ATTR __attribute__((unused))
 #else
-#  define MEM_STATIC static  /* this version may generate warnings for unused static functions; disable the relevant warning */
-#endif
+#  define UNUSED_ATTR
 #endif
 
 /* force no inlining */
@@ -116,58 +79,64 @@
 #  endif
 #endif
 
-
 /* target attribute */
+#ifndef __has_attribute
+  #define __has_attribute(x) 0  /* Compatibility with non-clang compilers. */
+#endif
 #if defined(__GNUC__) || defined(__ICCARM__)
 #  define TARGET_ATTRIBUTE(target) __attribute__((__target__(target)))
 #else
 #  define TARGET_ATTRIBUTE(target)
 #endif
 
-/* Target attribute for BMI2 dynamic dispatch.
- * Enable lzcnt, bmi, and bmi2.
- * We test for bmi1 & bmi2. lzcnt is included in bmi1.
+/* Enable runtime BMI2 dispatch based on the CPU.
+ * Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default.
  */
-#define BMI2_TARGET_ATTRIBUTE TARGET_ATTRIBUTE("lzcnt,bmi,bmi2")
+#ifndef DYNAMIC_BMI2
+  #if ((defined(__clang__) && __has_attribute(__target__)) \
+      || (defined(__GNUC__) \
+          && (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8)))) \
+      && (defined(__x86_64__) || defined(_M_X86)) \
+      && !defined(__BMI2__)
+  #  define DYNAMIC_BMI2 1
+  #else
+  #  define DYNAMIC_BMI2 0
+  #endif
+#endif
 
 /* prefetch
  * can be disabled, by declaring NO_PREFETCH build macro */
 #if defined(NO_PREFETCH)
-#  define PREFETCH_L1(ptr)  do { (void)(ptr); } while (0)  /* disabled */
-#  define PREFETCH_L2(ptr)  do { (void)(ptr); } while (0)  /* disabled */
+#  define PREFETCH_L1(ptr)  (void)(ptr)  /* disabled */
+#  define PREFETCH_L2(ptr)  (void)(ptr)  /* disabled */
 #else
-#  if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) && !defined(_M_ARM64EC)  /* _mm_prefetch() is not defined outside of x86/x64 */
+#  if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))  /* _mm_prefetch() is not defined outside of x86/x64 */
 #    include <mmintrin.h>   /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */
 #    define PREFETCH_L1(ptr)  _mm_prefetch((const char*)(ptr), _MM_HINT_T0)
 #    define PREFETCH_L2(ptr)  _mm_prefetch((const char*)(ptr), _MM_HINT_T1)
 #  elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) )
 #    define PREFETCH_L1(ptr)  __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */)
 #    define PREFETCH_L2(ptr)  __builtin_prefetch((ptr), 0 /* rw==read */, 2 /* locality */)
-#  elif defined(__aarch64__)
-#    define PREFETCH_L1(ptr)  do { __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr))); } while (0)
-#    define PREFETCH_L2(ptr)  do { __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr))); } while (0)
 #  else
-#    define PREFETCH_L1(ptr) do { (void)(ptr); } while (0)  /* disabled */
-#    define PREFETCH_L2(ptr) do { (void)(ptr); } while (0)  /* disabled */
+#    define PREFETCH_L1(ptr) (void)(ptr)  /* disabled */
+#    define PREFETCH_L2(ptr) (void)(ptr)  /* disabled */
 #  endif
 #endif  /* NO_PREFETCH */
 
 #define CACHELINE_SIZE 64
 
-#define PREFETCH_AREA(p, s)                              \
-    do {                                                 \
-        const char* const _ptr = (const char*)(p);       \
-        size_t const _size = (size_t)(s);                \
-        size_t _pos;                                     \
-        for (_pos=0; _pos<_size; _pos+=CACHELINE_SIZE) { \
-            PREFETCH_L2(_ptr + _pos);                    \
-        }                                                \
-    } while (0)
+#define PREFETCH_AREA(p, s)  {            \
+    const char* const _ptr = (const char*)(p);  \
+    size_t const _size = (size_t)(s);     \
+    size_t _pos;                          \
+    for (_pos=0; _pos<_size; _pos+=CACHELINE_SIZE) {  \
+        PREFETCH_L2(_ptr + _pos);         \
+    }                                     \
+}
 
 /* vectorization
- * older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax,
- * and some compilers, like Intel ICC and MCST LCC, do not support it at all. */
-#if !defined(__INTEL_COMPILER) && !defined(__clang__) && defined(__GNUC__) && !defined(__LCC__)
+ * older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax */
+#if !defined(__clang__) && defined(__GNUC__)
 #  if (__GNUC__ == 4 && __GNUC_MINOR__ > 3) || (__GNUC__ >= 5)
 #    define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize")))
 #  else
@@ -177,25 +146,6 @@
 #  define DONT_VECTORIZE
 #endif
 
-/* Tell the compiler that a branch is likely or unlikely.
- * Only use these macros if it causes the compiler to generate better code.
- * If you can remove a LIKELY/UNLIKELY annotation without speed changes in gcc
- * and clang, please do.
- */
-#if defined(__GNUC__)
-#define LIKELY(x) (__builtin_expect((x), 1))
-#define UNLIKELY(x) (__builtin_expect((x), 0))
-#else
-#define LIKELY(x) (x)
-#define UNLIKELY(x) (x)
-#endif
-
-#if __has_builtin(__builtin_unreachable) || (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 5)))
-#  define ZSTD_UNREACHABLE do { assert(0), __builtin_unreachable(); } while (0)
-#else
-#  define ZSTD_UNREACHABLE do { assert(0); } while (0)
-#endif
-
 /* disable warnings */
 #ifdef _MSC_VER    /* Visual Studio */
 #  include <intrin.h>                    /* For Visual 2005 */
@@ -206,248 +156,4 @@
 #  pragma warning(disable : 4324)        /* disable: C4324: padded structure */
 #endif
 
-/*Like DYNAMIC_BMI2 but for compile time determination of BMI2 support*/
-#ifndef STATIC_BMI2
-#  if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))
-#    ifdef __AVX2__  //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2
-#       define STATIC_BMI2 1
-#    endif
-#  elif defined(__BMI2__) && defined(__x86_64__) && defined(__GNUC__)
-#    define STATIC_BMI2 1
-#  endif
-#endif
-
-#ifndef STATIC_BMI2
-    #define STATIC_BMI2 0
-#endif
-
-/* compile time determination of SIMD support */
-#if !defined(ZSTD_NO_INTRINSICS)
-#  if defined(__SSE2__) || defined(_M_AMD64) || (defined (_M_IX86) && defined(_M_IX86_FP) && (_M_IX86_FP >= 2))
-#    define ZSTD_ARCH_X86_SSE2
-#  endif
-#  if defined(__ARM_NEON) || defined(_M_ARM64)
-#    define ZSTD_ARCH_ARM_NEON
-#  endif
-#
-#  if defined(ZSTD_ARCH_X86_SSE2)
-#    include <emmintrin.h>
-#  elif defined(ZSTD_ARCH_ARM_NEON)
-#    include <arm_neon.h>
-#  endif
-#endif
-
-/* C-language Attributes are added in C23. */
-#if defined(__STDC_VERSION__) && (__STDC_VERSION__ > 201710L) && defined(__has_c_attribute)
-# define ZSTD_HAS_C_ATTRIBUTE(x) __has_c_attribute(x)
-#else
-# define ZSTD_HAS_C_ATTRIBUTE(x) 0
-#endif
-
-/* Only use C++ attributes in C++. Some compilers report support for C++
- * attributes when compiling with C.
- */
-#if defined(__cplusplus) && defined(__has_cpp_attribute)
-# define ZSTD_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x)
-#else
-# define ZSTD_HAS_CPP_ATTRIBUTE(x) 0
-#endif
-
-/* Define ZSTD_FALLTHROUGH macro for annotating switch case with the 'fallthrough' attribute.
- * - C23: https://en.cppreference.com/w/c/language/attributes/fallthrough
- * - CPP17: https://en.cppreference.com/w/cpp/language/attributes/fallthrough
- * - Else: __attribute__((__fallthrough__))
- */
-#ifndef ZSTD_FALLTHROUGH
-# if ZSTD_HAS_C_ATTRIBUTE(fallthrough)
-#  define ZSTD_FALLTHROUGH [[fallthrough]]
-# elif ZSTD_HAS_CPP_ATTRIBUTE(fallthrough)
-#  define ZSTD_FALLTHROUGH [[fallthrough]]
-# elif __has_attribute(__fallthrough__)
-/* Leading semicolon is to satisfy gcc-11 with -pedantic. Without the semicolon
- * gcc complains about: a label can only be part of a statement and a declaration is not a statement.
- */
-#  define ZSTD_FALLTHROUGH ; __attribute__((__fallthrough__))
-# else
-#  define ZSTD_FALLTHROUGH
-# endif
-#endif
-
-/*-**************************************************************
-*  Alignment check
-*****************************************************************/
-
-/* this test was initially positioned in mem.h,
- * but this file is removed (or replaced) for linux kernel
- * so it's now hosted in compiler.h,
- * which remains valid for both user & kernel spaces.
- */
-
-#ifndef ZSTD_ALIGNOF
-# if defined(__GNUC__) || defined(_MSC_VER)
-/* covers gcc, clang & MSVC */
-/* note : this section must come first, before C11,
- * due to a limitation in the kernel source generator */
-#  define ZSTD_ALIGNOF(T) __alignof(T)
-
-# elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L)
-/* C11 support */
-#  include <stdalign.h>
-#  define ZSTD_ALIGNOF(T) alignof(T)
-
-# else
-/* No known support for alignof() - imperfect backup */
-#  define ZSTD_ALIGNOF(T) (sizeof(void*) < sizeof(T) ? sizeof(void*) : sizeof(T))
-
-# endif
-#endif /* ZSTD_ALIGNOF */
-
-/*-**************************************************************
-*  Sanitizer
-*****************************************************************/
-
-/**
- * Zstd relies on pointer overflow in its decompressor.
- * We add this attribute to functions that rely on pointer overflow.
- */
-#ifndef ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-#  if __has_attribute(no_sanitize)
-#    if !defined(__clang__) && defined(__GNUC__) && __GNUC__ < 8
-       /* gcc < 8 only has signed-integer-overlow which triggers on pointer overflow */
-#      define ZSTD_ALLOW_POINTER_OVERFLOW_ATTR __attribute__((no_sanitize("signed-integer-overflow")))
-#    else
-       /* older versions of clang [3.7, 5.0) will warn that pointer-overflow is ignored. */
-#      define ZSTD_ALLOW_POINTER_OVERFLOW_ATTR __attribute__((no_sanitize("pointer-overflow")))
-#    endif
-#  else
-#    define ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-#  endif
-#endif
-
-/**
- * Helper function to perform a wrapped pointer difference without trigging
- * UBSAN.
- *
- * @returns lhs - rhs with wrapping
- */
-MEM_STATIC
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-ptrdiff_t ZSTD_wrappedPtrDiff(unsigned char const* lhs, unsigned char const* rhs)
-{
-    return lhs - rhs;
-}
-
-/**
- * Helper function to perform a wrapped pointer add without triggering UBSAN.
- *
- * @return ptr + add with wrapping
- */
-MEM_STATIC
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-unsigned char const* ZSTD_wrappedPtrAdd(unsigned char const* ptr, ptrdiff_t add)
-{
-    return ptr + add;
-}
-
-/**
- * Helper function to perform a wrapped pointer subtraction without triggering
- * UBSAN.
- *
- * @return ptr - sub with wrapping
- */
-MEM_STATIC
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-unsigned char const* ZSTD_wrappedPtrSub(unsigned char const* ptr, ptrdiff_t sub)
-{
-    return ptr - sub;
-}
-
-/**
- * Helper function to add to a pointer that works around C's undefined behavior
- * of adding 0 to NULL.
- *
- * @returns `ptr + add` except it defines `NULL + 0 == NULL`.
- */
-MEM_STATIC
-unsigned char* ZSTD_maybeNullPtrAdd(unsigned char* ptr, ptrdiff_t add)
-{
-    return add > 0 ? ptr + add : ptr;
-}
-
-/* Issue #3240 reports an ASAN failure on an llvm-mingw build. Out of an
- * abundance of caution, disable our custom poisoning on mingw. */
-#ifdef __MINGW32__
-#ifndef ZSTD_ASAN_DONT_POISON_WORKSPACE
-#define ZSTD_ASAN_DONT_POISON_WORKSPACE 1
-#endif
-#ifndef ZSTD_MSAN_DONT_POISON_WORKSPACE
-#define ZSTD_MSAN_DONT_POISON_WORKSPACE 1
-#endif
-#endif
-
-#if ZSTD_MEMORY_SANITIZER && !defined(ZSTD_MSAN_DONT_POISON_WORKSPACE)
-/* Not all platforms that support msan provide sanitizers/msan_interface.h.
- * We therefore declare the functions we need ourselves, rather than trying to
- * include the header file... */
-#include <stddef.h>  /* size_t */
-#define ZSTD_DEPS_NEED_STDINT
-#include "zstd_deps.h"  /* intptr_t */
-
-/* Make memory region fully initialized (without changing its contents). */
-void __msan_unpoison(const volatile void *a, size_t size);
-
-/* Make memory region fully uninitialized (without changing its contents).
-   This is a legacy interface that does not update origin information. Use
-   __msan_allocated_memory() instead. */
-void __msan_poison(const volatile void *a, size_t size);
-
-/* Returns the offset of the first (at least partially) poisoned byte in the
-   memory range, or -1 if the whole range is good. */
-intptr_t __msan_test_shadow(const volatile void *x, size_t size);
-
-/* Print shadow and origin for the memory range to stderr in a human-readable
-   format. */
-void __msan_print_shadow(const volatile void *x, size_t size);
-#endif
-
-#if ZSTD_ADDRESS_SANITIZER && !defined(ZSTD_ASAN_DONT_POISON_WORKSPACE)
-/* Not all platforms that support asan provide sanitizers/asan_interface.h.
- * We therefore declare the functions we need ourselves, rather than trying to
- * include the header file... */
-#include <stddef.h>  /* size_t */
-
-/**
- * Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable.
- *
- * This memory must be previously allocated by your program. Instrumented
- * code is forbidden from accessing addresses in this region until it is
- * unpoisoned. This function is not guaranteed to poison the entire region -
- * it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan
- * alignment restrictions.
- *
- * \note This function is not thread-safe because no two threads can poison or
- * unpoison memory in the same memory region simultaneously.
- *
- * \param addr Start of memory region.
- * \param size Size of memory region. */
-void __asan_poison_memory_region(void const volatile *addr, size_t size);
-
-/**
- * Marks a memory region (<c>[addr, addr+size)</c>) as addressable.
- *
- * This memory must be previously allocated by your program. Accessing
- * addresses in this region is allowed until this region is poisoned again.
- * This function could unpoison a super-region of <c>[addr, addr+size)</c> due
- * to ASan alignment restrictions.
- *
- * \note This function is not thread-safe because no two threads can
- * poison or unpoison memory in the same memory region simultaneously.
- *
- * \param addr Start of memory region.
- * \param size Size of memory region. */
-void __asan_unpoison_memory_region(void const volatile *addr, size_t size);
-#endif
-
 #endif /* ZSTD_COMPILER_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/cover.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -27,65 +26,47 @@
 #include <string.h> /* memset */
 #include <time.h>   /* clock */
 
-#ifndef ZDICT_STATIC_LINKING_ONLY
-#  define ZDICT_STATIC_LINKING_ONLY
-#endif
-
 #include "mem.h" /* read */
-#include "pool.h" /* POOL_ctx */
-#include "threading.h" /* ZSTD_pthread_mutex_t */
-#include "zstd_internal.h" /* includes zstd.h */
-#include "bits.h" /* ZSTD_highbit32 */
-#include "zdict.h"
+#include "pool.h"
+#include "threading.h"
 #include "cover.h"
+#include "zstd_internal.h" /* includes zstd.h */
+#ifndef ZDICT_STATIC_LINKING_ONLY
+#define ZDICT_STATIC_LINKING_ONLY
+#endif
+#include "zdict.h"
 
 /*-*************************************
 *  Constants
 ***************************************/
-/**
-* There are 32bit indexes used to ref samples, so limit samples size to 4GB
-* on 64bit builds.
-* For 32bit builds we choose 1 GB.
-* Most 32bit platforms have 2GB user-mode addressable space and we allocate a large
-* contiguous buffer, so 1GB is already a high limit.
-*/
 #define COVER_MAX_SAMPLES_SIZE (sizeof(size_t) == 8 ? ((unsigned)-1) : ((unsigned)1 GB))
-#define COVER_DEFAULT_SPLITPOINT 1.0
+#define DEFAULT_SPLITPOINT 1.0
 
 /*-*************************************
 *  Console display
 ***************************************/
-#ifndef LOCALDISPLAYLEVEL
-static int g_displayLevel = 0;
-#endif
-#undef  DISPLAY
+static int g_displayLevel = 2;
 #define DISPLAY(...)                                                           \
   {                                                                            \
     fprintf(stderr, __VA_ARGS__);                                              \
     fflush(stderr);                                                            \
   }
-#undef  LOCALDISPLAYLEVEL
 #define LOCALDISPLAYLEVEL(displayLevel, l, ...)                                \
   if (displayLevel >= l) {                                                     \
     DISPLAY(__VA_ARGS__);                                                      \
   } /* 0 : no display;   1: errors;   2: default;  3: details;  4: debug */
-#undef  DISPLAYLEVEL
 #define DISPLAYLEVEL(l, ...) LOCALDISPLAYLEVEL(g_displayLevel, l, __VA_ARGS__)
 
-#ifndef LOCALDISPLAYUPDATE
-static const clock_t g_refreshRate = CLOCKS_PER_SEC * 15 / 100;
-static clock_t g_time = 0;
-#endif
-#undef  LOCALDISPLAYUPDATE
 #define LOCALDISPLAYUPDATE(displayLevel, l, ...)                               \
   if (displayLevel >= l) {                                                     \
-    if ((clock() - g_time > g_refreshRate) || (displayLevel >= 4)) {           \
+    if ((clock() - g_time > refreshRate) || (displayLevel >= 4)) {             \
       g_time = clock();                                                        \
       DISPLAY(__VA_ARGS__);                                                    \
     }                                                                          \
   }
-#undef  DISPLAYUPDATE
 #define DISPLAYUPDATE(l, ...) LOCALDISPLAYUPDATE(g_displayLevel, l, __VA_ARGS__)
+static const clock_t refreshRate = CLOCKS_PER_SEC * 15 / 100;
+static clock_t g_time = 0;
 
 /*-*************************************
 * Hash table
@@ -139,9 +120,9 @@ static int COVER_map_init(COVER_map_t *map, U32 size) {
 /**
  * Internal hash function
  */
-static const U32 COVER_prime4bytes = 2654435761U;
+static const U32 prime4bytes = 2654435761U;
 static U32 COVER_map_hash(COVER_map_t *map, U32 key) {
-  return (key * COVER_prime4bytes) >> (32 - map->sizeLog);
+  return (key * prime4bytes) >> (32 - map->sizeLog);
 }
 
 /**
@@ -234,7 +215,7 @@ typedef struct {
 } COVER_ctx_t;
 
 /* We need a global context for qsort... */
-static COVER_ctx_t *g_coverCtx = NULL;
+static COVER_ctx_t *g_ctx = NULL;
 
 /*-*************************************
 *  Helper functions
@@ -277,11 +258,11 @@ static int COVER_cmp8(COVER_ctx_t *ctx, const void *lp, const void *rp) {
 
 /**
  * Same as COVER_cmp() except ties are broken by pointer value
- * NOTE: g_coverCtx must be set to call this function.  A global is required because
+ * NOTE: g_ctx must be set to call this function.  A global is required because
  * qsort doesn't take an opaque pointer.
  */
-static int WIN_CDECL COVER_strict_cmp(const void *lp, const void *rp) {
-  int result = COVER_cmp(g_coverCtx, lp, rp);
+static int COVER_strict_cmp(const void *lp, const void *rp) {
+  int result = COVER_cmp(g_ctx, lp, rp);
   if (result == 0) {
     result = lp < rp ? -1 : 1;
   }
@@ -290,8 +271,8 @@ static int WIN_CDECL COVER_strict_cmp(const void *lp, const void *rp) {
 /**
  * Faster version for d <= 8.
  */
-static int WIN_CDECL COVER_strict_cmp8(const void *lp, const void *rp) {
-  int result = COVER_cmp8(g_coverCtx, lp, rp);
+static int COVER_strict_cmp8(const void *lp, const void *rp) {
+  int result = COVER_cmp8(g_ctx, lp, rp);
   if (result == 0) {
     result = lp < rp ? -1 : 1;
   }
@@ -302,10 +283,9 @@ static int WIN_CDECL COVER_strict_cmp8(const void *lp, const void *rp) {
  * Returns the first pointer in [first, last) whose element does not compare
  * less than value.  If no such element exists it returns last.
  */
-static const size_t *COVER_lower_bound(const size_t* first, const size_t* last,
+static const size_t *COVER_lower_bound(const size_t *first, const size_t *last,
                                        size_t value) {
-  size_t count = (size_t)(last - first);
-  assert(last >= first);
+  size_t count = last - first;
   while (count != 0) {
     size_t step = count / 2;
     const size_t *ptr = first;
@@ -544,15 +524,14 @@ static void COVER_ctx_destroy(COVER_ctx_t *ctx) {
 
 /**
  * Prepare a context for dictionary building.
- * The context is only dependent on the parameter `d` and can be used multiple
+ * The context is only dependent on the parameter `d` and can used multiple
  * times.
  * Returns 0 on success or error code on error.
  * The context must be destroyed with `COVER_ctx_destroy()`.
  */
 static size_t COVER_ctx_init(COVER_ctx_t *ctx, const void *samplesBuffer,
                           const size_t *samplesSizes, unsigned nbSamples,
-                          unsigned d, double splitPoint)
-{
+                          unsigned d, double splitPoint) {
   const BYTE *const samples = (const BYTE *)samplesBuffer;
   const size_t totalSamplesSize = COVER_sum(samplesSizes, nbSamples);
   /* Split samples into testing and training sets */
@@ -624,7 +603,7 @@ static size_t COVER_ctx_init(COVER_ctx_t *ctx, const void *samplesBuffer,
     /* qsort doesn't take an opaque pointer, so pass as a global.
      * On OpenBSD qsort() is not guaranteed to be stable, their mergesort() is.
      */
-    g_coverCtx = ctx;
+    g_ctx = ctx;
 #if defined(__OpenBSD__)
     mergesort(ctx->suffix, ctx->suffixSize, sizeof(U32),
           (ctx->d <= 8 ? &COVER_strict_cmp8 : &COVER_strict_cmp));
@@ -650,7 +629,7 @@ static size_t COVER_ctx_init(COVER_ctx_t *ctx, const void *samplesBuffer,
 
 void COVER_warnOnSmallCorpus(size_t maxDictSize, size_t nbDmers, int displayLevel)
 {
-  const double ratio = (double)nbDmers / (double)maxDictSize;
+  const double ratio = (double)nbDmers / maxDictSize;
   if (ratio >= 10) {
       return;
   }
@@ -736,7 +715,7 @@ static size_t COVER_buildDictionary(const COVER_ctx_t *ctx, U32 *freqs,
   return tail;
 }
 
-ZDICTLIB_STATIC_API size_t ZDICT_trainFromBuffer_cover(
+ZDICTLIB_API size_t ZDICT_trainFromBuffer_cover(
     void *dictBuffer, size_t dictBufferCapacity,
     const void *samplesBuffer, const size_t *samplesSizes, unsigned nbSamples,
     ZDICT_cover_params_t parameters)
@@ -746,7 +725,7 @@ ZDICTLIB_STATIC_API size_t ZDICT_trainFromBuffer_cover(
   COVER_map_t activeDmers;
   parameters.splitPoint = 1.0;
   /* Initialize global data */
-  g_displayLevel = (int)parameters.zParams.notificationLevel;
+  g_displayLevel = parameters.zParams.notificationLevel;
   /* Checks */
   if (!COVER_checkParameters(parameters, dictBufferCapacity)) {
     DISPLAYLEVEL(1, "Cover parameters incorrect\n");
@@ -910,10 +889,8 @@ void COVER_best_start(COVER_best_t *best) {
  * Decrements liveJobs and signals any waiting threads if liveJobs == 0.
  * If this dictionary is the best so far save it and its parameters.
  */
-void COVER_best_finish(COVER_best_t* best,
-                      ZDICT_cover_params_t parameters,
-                      COVER_dictSelection_t selection)
-{
+void COVER_best_finish(COVER_best_t *best, ZDICT_cover_params_t parameters,
+                              COVER_dictSelection_t selection) {
   void* dict = selection.dictContent;
   size_t compressedSize = selection.totalCompressedSize;
   size_t dictSize = selection.dictSize;
@@ -956,17 +933,9 @@ void COVER_best_finish(COVER_best_t* best,
   }
 }
 
-static COVER_dictSelection_t setDictSelection(BYTE* buf, size_t s, size_t csz)
-{
-    COVER_dictSelection_t ds;
-    ds.dictContent = buf;
-    ds.dictSize = s;
-    ds.totalCompressedSize = csz;
-    return ds;
-}
-
 COVER_dictSelection_t COVER_dictSelectionError(size_t error) {
-    return setDictSelection(NULL, 0, error);
+    COVER_dictSelection_t selection = { NULL, 0, error };
+    return selection;
 }
 
 unsigned COVER_dictSelectionIsError(COVER_dictSelection_t selection) {
@@ -977,7 +946,7 @@ void COVER_dictSelectionFree(COVER_dictSelection_t selection){
   free(selection.dictContent);
 }
 
-COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBufferCapacity,
+COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent,
         size_t dictContentSize, const BYTE* samplesBuffer, const size_t* samplesSizes, unsigned nbFinalizeSamples,
         size_t nbCheckSamples, size_t nbSamples, ZDICT_cover_params_t params, size_t* offsets, size_t totalCompressedSize) {
 
@@ -985,8 +954,8 @@ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBuffe
   size_t largestCompressed = 0;
   BYTE* customDictContentEnd = customDictContent + dictContentSize;
 
-  BYTE* largestDictbuffer = (BYTE*)malloc(dictBufferCapacity);
-  BYTE* candidateDictBuffer = (BYTE*)malloc(dictBufferCapacity);
+  BYTE * largestDictbuffer = (BYTE *)malloc(dictContentSize);
+  BYTE * candidateDictBuffer = (BYTE *)malloc(dictContentSize);
   double regressionTolerance = ((double)params.shrinkDictMaxRegression / 100.0) + 1.00;
 
   if (!largestDictbuffer || !candidateDictBuffer) {
@@ -998,7 +967,7 @@ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBuffe
   /* Initial dictionary size and compressed size */
   memcpy(largestDictbuffer, customDictContent, dictContentSize);
   dictContentSize = ZDICT_finalizeDictionary(
-    largestDictbuffer, dictBufferCapacity, customDictContent, dictContentSize,
+    largestDictbuffer, dictContentSize, customDictContent, dictContentSize,
     samplesBuffer, samplesSizes, nbFinalizeSamples, params.zParams);
 
   if (ZDICT_isError(dictContentSize)) {
@@ -1019,8 +988,9 @@ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBuffe
   }
 
   if (params.shrinkDict == 0) {
+    COVER_dictSelection_t selection = { largestDictbuffer, dictContentSize, totalCompressedSize };
     free(candidateDictBuffer);
-    return setDictSelection(largestDictbuffer, dictContentSize, totalCompressedSize);
+    return selection;
   }
 
   largestDict = dictContentSize;
@@ -1031,7 +1001,7 @@ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBuffe
   while (dictContentSize < largestDict) {
     memcpy(candidateDictBuffer, largestDictbuffer, largestDict);
     dictContentSize = ZDICT_finalizeDictionary(
-      candidateDictBuffer, dictBufferCapacity, customDictContentEnd - dictContentSize, dictContentSize,
+      candidateDictBuffer, dictContentSize, customDictContentEnd - dictContentSize, dictContentSize,
       samplesBuffer, samplesSizes, nbFinalizeSamples, params.zParams);
 
     if (ZDICT_isError(dictContentSize)) {
@@ -1052,16 +1022,20 @@ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBuffe
       return COVER_dictSelectionError(totalCompressedSize);
     }
 
-    if ((double)totalCompressedSize <= (double)largestCompressed * regressionTolerance) {
+    if (totalCompressedSize <= largestCompressed * regressionTolerance) {
+      COVER_dictSelection_t selection = { candidateDictBuffer, dictContentSize, totalCompressedSize };
       free(largestDictbuffer);
-      return setDictSelection( candidateDictBuffer, dictContentSize, totalCompressedSize );
+      return selection;
     }
     dictContentSize *= 2;
   }
   dictContentSize = largestDict;
   totalCompressedSize = largestCompressed;
-  free(candidateDictBuffer);
-  return setDictSelection( largestDictbuffer, dictContentSize, totalCompressedSize );
+  {
+    COVER_dictSelection_t selection = { largestDictbuffer, dictContentSize, totalCompressedSize };
+    free(candidateDictBuffer);
+    return selection;
+  }
 }
 
 /**
@@ -1079,19 +1053,18 @@ typedef struct COVER_tryParameters_data_s {
  * This function is thread safe if zstd is compiled with multithreaded support.
  * It takes its parameters as an *OWNING* opaque pointer to support threading.
  */
-static void COVER_tryParameters(void *opaque)
-{
+static void COVER_tryParameters(void *opaque) {
   /* Save parameters as local variables */
-  COVER_tryParameters_data_t *const data = (COVER_tryParameters_data_t*)opaque;
+  COVER_tryParameters_data_t *const data = (COVER_tryParameters_data_t *)opaque;
   const COVER_ctx_t *const ctx = data->ctx;
   const ZDICT_cover_params_t parameters = data->parameters;
   size_t dictBufferCapacity = data->dictBufferCapacity;
   size_t totalCompressedSize = ERROR(GENERIC);
   /* Allocate space for hash table, dict, and freqs */
   COVER_map_t activeDmers;
-  BYTE* const dict = (BYTE*)malloc(dictBufferCapacity);
+  BYTE *const dict = (BYTE * const)malloc(dictBufferCapacity);
   COVER_dictSelection_t selection = COVER_dictSelectionError(ERROR(GENERIC));
-  U32* const freqs = (U32*)malloc(ctx->suffixSize * sizeof(U32));
+  U32 *freqs = (U32 *)malloc(ctx->suffixSize * sizeof(U32));
   if (!COVER_map_init(&activeDmers, parameters.k - parameters.d + 1)) {
     DISPLAYLEVEL(1, "Failed to allocate dmer map: out of memory\n");
     goto _cleanup;
@@ -1106,7 +1079,7 @@ static void COVER_tryParameters(void *opaque)
   {
     const size_t tail = COVER_buildDictionary(ctx, freqs, &activeDmers, dict,
                                               dictBufferCapacity, parameters);
-    selection = COVER_selectDict(dict + tail, dictBufferCapacity, dictBufferCapacity - tail,
+    selection = COVER_selectDict(dict + tail, dictBufferCapacity - tail,
         ctx->samples, ctx->samplesSizes, (unsigned)ctx->nbTrainSamples, ctx->nbTrainSamples, ctx->nbSamples, parameters, ctx->offsets,
         totalCompressedSize);
 
@@ -1121,18 +1094,19 @@ _cleanup:
   free(data);
   COVER_map_destroy(&activeDmers);
   COVER_dictSelectionFree(selection);
-  free(freqs);
+  if (freqs) {
+    free(freqs);
+  }
 }
 
-ZDICTLIB_STATIC_API size_t ZDICT_optimizeTrainFromBuffer_cover(
-    void* dictBuffer, size_t dictBufferCapacity, const void* samplesBuffer,
-    const size_t* samplesSizes, unsigned nbSamples,
-    ZDICT_cover_params_t* parameters)
-{
+ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_cover(
+    void *dictBuffer, size_t dictBufferCapacity, const void *samplesBuffer,
+    const size_t *samplesSizes, unsigned nbSamples,
+    ZDICT_cover_params_t *parameters) {
   /* constants */
   const unsigned nbThreads = parameters->nbThreads;
   const double splitPoint =
-      parameters->splitPoint <= 0.0 ? COVER_DEFAULT_SPLITPOINT : parameters->splitPoint;
+      parameters->splitPoint <= 0.0 ? DEFAULT_SPLITPOINT : parameters->splitPoint;
   const unsigned kMinD = parameters->d == 0 ? 6 : parameters->d;
   const unsigned kMaxD = parameters->d == 0 ? 8 : parameters->d;
   const unsigned kMinK = parameters->k == 0 ? 50 : parameters->k;
@@ -1260,5 +1234,3 @@ ZDICTLIB_STATIC_API size_t ZDICT_optimizeTrainFromBuffer_cover(
     return dictSize;
   }
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/cover.h

@@ -1,20 +1,14 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
+#include <stdio.h>  /* fprintf */
+#include <stdlib.h> /* malloc, free, qsort */
+#include <string.h> /* memset */
+#include <time.h>   /* clock */
+#include "mem.h" /* read */
+#include "pool.h"
+#include "threading.h"
+#include "zstd_internal.h" /* includes zstd.h */
 #ifndef ZDICT_STATIC_LINKING_ONLY
-#  define ZDICT_STATIC_LINKING_ONLY
+#define ZDICT_STATIC_LINKING_ONLY
 #endif
-
-#include "threading.h" /* ZSTD_pthread_mutex_t */
-#include "mem.h"   /* U32, BYTE */
 #include "zdict.h"
 
 /**
@@ -148,8 +142,6 @@ void COVER_dictSelectionFree(COVER_dictSelection_t selection);
  * smallest dictionary within a specified regression of the compressed size
  * from the largest dictionary.
  */
- COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBufferCapacity,
+ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent,
                        size_t dictContentSize, const BYTE* samplesBuffer, const size_t* samplesSizes, unsigned nbFinalizeSamples,
                        size_t nbCheckSamples, size_t nbSamples, ZDICT_cover_params_t params, size_t* offsets, size_t totalCompressedSize);
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/cpu.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2018-present, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -17,6 +16,8 @@
  * https://github.com/facebook/folly/blob/master/folly/CpuId.h
  */
 
+#include <string.h>
+
 #include "mem.h"
 
 #ifdef _MSC_VER
@@ -36,7 +37,6 @@ MEM_STATIC ZSTD_cpuid_t ZSTD_cpuid(void) {
     U32 f7b = 0;
     U32 f7c = 0;
 #if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_IX86))
-#if !defined(__clang__)
     int reg[4];
     __cpuid((int*)reg, 0);
     {
@@ -52,41 +52,6 @@ MEM_STATIC ZSTD_cpuid_t ZSTD_cpuid(void) {
             f7c = (U32)reg[2];
         }
     }
-#else
-    /* Clang compiler has a bug (fixed in https://reviews.llvm.org/D101338) in
-     * which the `__cpuid` intrinsic does not save and restore `rbx` as it needs
-     * to due to being a reserved register. So in that case, do the `cpuid`
-     * ourselves. Clang supports inline assembly anyway.
-     */
-    U32 n;
-    __asm__(
-        "pushq %%rbx\n\t"
-        "cpuid\n\t"
-        "popq %%rbx\n\t"
-        : "=a"(n)
-        : "a"(0)
-        : "rcx", "rdx");
-    if (n >= 1) {
-      U32 f1a;
-      __asm__(
-          "pushq %%rbx\n\t"
-          "cpuid\n\t"
-          "popq %%rbx\n\t"
-          : "=a"(f1a), "=c"(f1c), "=d"(f1d)
-          : "a"(1)
-          :);
-    }
-    if (n >= 7) {
-      __asm__(
-          "pushq %%rbx\n\t"
-          "cpuid\n\t"
-          "movq %%rbx, %%rax\n\t"
-          "popq %%rbx"
-          : "=a"(f7b), "=c"(f7c)
-          : "a"(7), "c"(0)
-          : "rdx");
-    }
-#endif
 #elif defined(__i386__) && defined(__PIC__) && !defined(__clang__) && defined(__GNUC__)
     /* The following block like the normal cpuid branch below, but gcc
      * reserves ebx for use of its pic register so we must specially
@@ -248,5 +213,3 @@ MEM_STATIC ZSTD_cpuid_t ZSTD_cpuid(void) {
 #undef X
 
 #endif /* ZSTD_COMMON_CPU_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/debug.c

@@ -1,16 +1,35 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * debug
- * Part of FSE library
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- * You can contact the author at :
- * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   debug
+   Part of FSE library
+   Copyright (C) 2013-present, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+   You can contact the author at :
+   - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
 ****************************************************************** */
 
 
@@ -22,12 +41,4 @@
 
 #include "debug.h"
 
-#if !defined(ZSTD_LINUX_KERNEL) || (DEBUGLEVEL>=2)
-/* We only use this when DEBUGLEVEL>=2, but we get -Werror=pedantic errors if a
- * translation unit is empty. So remove this from Linux kernel builds, but
- * otherwise just leave it in.
- */
 int g_debuglevel = DEBUGLEVEL;
-#endif
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/debug.h

@@ -1,16 +1,35 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * debug
- * Part of FSE library
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- * You can contact the author at :
- * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   debug
+   Part of FSE library
+   Copyright (C) 2013-present, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+   You can contact the author at :
+   - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
 ****************************************************************** */
 
 
@@ -52,6 +71,15 @@ extern "C" {
 #endif
 
 
+/* DEBUGFILE can be defined externally,
+ * typically through compiler command line.
+ * note : currently useless.
+ * Value must be stderr or stdout */
+#ifndef DEBUGFILE
+#  define DEBUGFILE stderr
+#endif
+
+
 /* recommended values for DEBUGLEVEL :
  * 0 : release mode, no debug, all run-time checks disabled
  * 1 : enables assert() only, no display
@@ -68,8 +96,7 @@ extern "C" {
  */
 
 #if (DEBUGLEVEL>=1)
-#  define ZSTD_DEPS_NEED_ASSERT
-#  include "zstd_deps.h"
+#  include <assert.h>
 #else
 #  ifndef assert   /* assert may be already defined, due to prior #include <assert.h> */
 #    define assert(condition) ((void)0)   /* disable assert (default) */
@@ -77,8 +104,7 @@ extern "C" {
 #endif
 
 #if (DEBUGLEVEL>=2)
-#  define ZSTD_DEPS_NEED_IO
-#  include "zstd_deps.h"
+#  include <stdio.h>
 extern int g_debuglevel; /* the variable is only declared,
                             it actually lives in debug.c,
                             and is shared by the whole process.
@@ -86,27 +112,18 @@ extern int g_debuglevel; /* the variable is only declared,
                             It's useful when enabling very verbose levels
                             on selective conditions (such as position in src) */
 
-#  define RAWLOG(l, ...)                   \
-    do {                                   \
-        if (l<=g_debuglevel) {             \
-            ZSTD_DEBUG_PRINT(__VA_ARGS__); \
-        }                                  \
-    } while (0)
-
-#define STRINGIFY(x) #x
-#define TOSTRING(x) STRINGIFY(x)
-#define LINE_AS_STRING TOSTRING(__LINE__)
-
-#  define DEBUGLOG(l, ...)                               \
-    do {                                                 \
-        if (l<=g_debuglevel) {                           \
-            ZSTD_DEBUG_PRINT(__FILE__ ":" LINE_AS_STRING ": " __VA_ARGS__); \
-            ZSTD_DEBUG_PRINT(" \n");                     \
-        }                                                \
-    } while (0)
+#  define RAWLOG(l, ...) {                                      \
+                if (l<=g_debuglevel) {                          \
+                    fprintf(stderr, __VA_ARGS__);               \
+            }   }
+#  define DEBUGLOG(l, ...) {                                    \
+                if (l<=g_debuglevel) {                          \
+                    fprintf(stderr, __FILE__ ": " __VA_ARGS__); \
+                    fprintf(stderr, " \n");                     \
+            }   }
 #else
-#  define RAWLOG(l, ...)   do { } while (0)    /* disabled */
-#  define DEBUGLOG(l, ...) do { } while (0)    /* disabled */
+#  define RAWLOG(l, ...)      {}    /* disabled */
+#  define DEBUGLOG(l, ...)    {}    /* disabled */
 #endif
 
 
@@ -115,5 +132,3 @@ extern int g_debuglevel; /* the variable is only declared,
 #endif
 
 #endif /* DEBUG_H_12987983217 */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/divsufsort.c

@@ -1,4 +1,3 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
  * divsufsort.c for libdivsufsort-lite
  * Copyright (c) 2003-2008 Yuta Mori All Rights Reserved.
@@ -1577,7 +1576,7 @@ note:
     /* Construct the inverse suffix array of type B* suffixes using trsort. */
     trsort(ISAb, SA, m, 1);
 
-    /* Set the sorted order of type B* suffixes. */
+    /* Set the sorted order of tyoe B* suffixes. */
     for(i = n - 1, j = m, c0 = T[n - 1]; 0 <= i;) {
       for(--i, c1 = c0; (0 <= i) && ((c0 = T[i]) >= c1); --i, c1 = c0) { }
       if(0 <= i) {
@@ -1912,5 +1911,3 @@ divbwt(const unsigned char *T, unsigned char *U, int *A, int n, unsigned char *
 
   return pidx;
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/divsufsort.h

@@ -1,4 +1,3 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
  * divsufsort.h for libdivsufsort-lite
  * Copyright (c) 2003-2008 Yuta Mori All Rights Reserved.
@@ -66,5 +65,3 @@ divbwt(const unsigned char *T, unsigned char *U, int *A, int n, unsigned char *
 #endif /* __cplusplus */
 
 #endif /* _DIVSUFSORT_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/entropy_common.c

@@ -1,17 +1,36 @@
-#ifndef USE_EXTERNAL_ZSTD
-/* ******************************************************************
- * Common functions of New Generation Entropy library
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- *  You can contact the author at :
- *  - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *  - Public forum : https://groups.google.com/forum/#!forum/lz4c
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
-****************************************************************** */
+/*
+   Common functions of New Generation Entropy library
+   Copyright (C) 2016, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+    You can contact the author at :
+    - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy
+    - Public forum : https://groups.google.com/forum/#!forum/lz4c
+*************************************************************************** */
 
 /* *************************************
 *  Dependencies
@@ -20,8 +39,8 @@
 #include "error_private.h"       /* ERR_*, ERROR */
 #define FSE_STATIC_LINKING_ONLY  /* FSE_MIN_TABLELOG */
 #include "fse.h"
+#define HUF_STATIC_LINKING_ONLY  /* HUF_TABLELOG_ABSOLUTEMAX */
 #include "huf.h"
-#include "bits.h"                /* ZSDT_highbit32, ZSTD_countTrailingZeros32 */
 
 
 /*===   Version   ===*/
@@ -39,9 +58,8 @@ const char* HUF_getErrorName(size_t code) { return ERR_getErrorName(code); }
 /*-**************************************************************
 *  FSE NCount encoding-decoding
 ****************************************************************/
-FORCE_INLINE_TEMPLATE
-size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
-                           const void* headerBuffer, size_t hbSize)
+size_t FSE_readNCount (short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
+                 const void* headerBuffer, size_t hbSize)
 {
     const BYTE* const istart = (const BYTE*) headerBuffer;
     const BYTE* const iend = istart + hbSize;
@@ -52,23 +70,23 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne
     U32 bitStream;
     int bitCount;
     unsigned charnum = 0;
-    unsigned const maxSV1 = *maxSVPtr + 1;
     int previous0 = 0;
 
-    if (hbSize < 8) {
-        /* This function only works when hbSize >= 8 */
-        char buffer[8] = {0};
-        ZSTD_memcpy(buffer, headerBuffer, hbSize);
+    if (hbSize < 4) {
+        /* This function only works when hbSize >= 4 */
+        char buffer[4];
+        memset(buffer, 0, sizeof(buffer));
+        memcpy(buffer, headerBuffer, hbSize);
         {   size_t const countSize = FSE_readNCount(normalizedCounter, maxSVPtr, tableLogPtr,
                                                     buffer, sizeof(buffer));
             if (FSE_isError(countSize)) return countSize;
             if (countSize > hbSize) return ERROR(corruption_detected);
             return countSize;
     }   }
-    assert(hbSize >= 8);
+    assert(hbSize >= 4);
 
     /* init */
-    ZSTD_memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0]));   /* all symbols not present in NCount have a frequency of 0 */
+    memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0]));   /* all symbols not present in NCount have a frequency of 0 */
     bitStream = MEM_readLE32(ip);
     nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG;   /* extract tableLog */
     if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX) return ERROR(tableLog_tooLarge);
@@ -79,58 +97,36 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne
     threshold = 1<<nbBits;
     nbBits++;
 
-    for (;;) {
+    while ((remaining>1) & (charnum<=*maxSVPtr)) {
         if (previous0) {
-            /* Count the number of repeats. Each time the
-             * 2-bit repeat code is 0b11 there is another
-             * repeat.
-             * Avoid UB by setting the high bit to 1.
-             */
-            int repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
-            while (repeats >= 12) {
-                charnum += 3 * 12;
-                if (LIKELY(ip <= iend-7)) {
-                    ip += 3;
+            unsigned n0 = charnum;
+            while ((bitStream & 0xFFFF) == 0xFFFF) {
+                n0 += 24;
+                if (ip < iend-5) {
+                    ip += 2;
+                    bitStream = MEM_readLE32(ip) >> bitCount;
                 } else {
-                    bitCount -= (int)(8 * (iend - 7 - ip));
-                    bitCount &= 31;
-                    ip = iend - 4;
-                }
-                bitStream = MEM_readLE32(ip) >> bitCount;
-                repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
+                    bitStream >>= 16;
+                    bitCount   += 16;
+            }   }
+            while ((bitStream & 3) == 3) {
+                n0 += 3;
+                bitStream >>= 2;
+                bitCount += 2;
             }
-            charnum += 3 * repeats;
-            bitStream >>= 2 * repeats;
-            bitCount += 2 * repeats;
-
-            /* Add the final repeat which isn't 0b11. */
-            assert((bitStream & 3) < 3);
-            charnum += bitStream & 3;
+            n0 += bitStream & 3;
             bitCount += 2;
-
-            /* This is an error, but break and return an error
-             * at the end, because returning out of a loop makes
-             * it harder for the compiler to optimize.
-             */
-            if (charnum >= maxSV1) break;
-
-            /* We don't need to set the normalized count to 0
-             * because we already memset the whole buffer to 0.
-             */
-
-            if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
+            if (n0 > *maxSVPtr) return ERROR(maxSymbolValue_tooSmall);
+            while (charnum < n0) normalizedCounter[charnum++] = 0;
+            if ((ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
                 assert((bitCount >> 3) <= 3); /* For first condition to work */
                 ip += bitCount>>3;
                 bitCount &= 7;
+                bitStream = MEM_readLE32(ip) >> bitCount;
             } else {
-                bitCount -= (int)(8 * (iend - 4 - ip));
-                bitCount &= 31;
-                ip = iend - 4;
-            }
-            bitStream = MEM_readLE32(ip) >> bitCount;
-        }
-        {
-            int const max = (2*threshold-1) - remaining;
+                bitStream >>= 2;
+        }   }
+        {   int const max = (2*threshold-1) - remaining;
             int count;
 
             if ((bitStream & (threshold-1)) < (U32)max) {
@@ -143,43 +139,24 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne
             }
 
             count--;   /* extra accuracy */
-            /* When it matters (small blocks), this is a
-             * predictable branch, because we don't use -1.
-             */
-            if (count >= 0) {
-                remaining -= count;
-            } else {
-                assert(count == -1);
-                remaining += count;
-            }
+            remaining -= count < 0 ? -count : count;   /* -1 means +1 */
             normalizedCounter[charnum++] = (short)count;
             previous0 = !count;
-
-            assert(threshold > 1);
-            if (remaining < threshold) {
-                /* This branch can be folded into the
-                 * threshold update condition because we
-                 * know that threshold > 1.
-                 */
-                if (remaining <= 1) break;
-                nbBits = ZSTD_highbit32(remaining) + 1;
-                threshold = 1 << (nbBits - 1);
+            while (remaining < threshold) {
+                nbBits--;
+                threshold >>= 1;
             }
-            if (charnum >= maxSV1) break;
 
-            if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
+            if ((ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
                 ip += bitCount>>3;
                 bitCount &= 7;
             } else {
                 bitCount -= (int)(8 * (iend - 4 - ip));
-                bitCount &= 31;
                 ip = iend - 4;
             }
-            bitStream = MEM_readLE32(ip) >> bitCount;
-    }   }
+            bitStream = MEM_readLE32(ip) >> (bitCount & 31);
+    }   }   /* while ((remaining>1) & (charnum<=*maxSVPtr)) */
     if (remaining != 1) return ERROR(corruption_detected);
-    /* Only possible when there are too many zeros. */
-    if (charnum > maxSV1) return ERROR(maxSymbolValue_tooSmall);
     if (bitCount > 32) return ERROR(corruption_detected);
     *maxSVPtr = charnum-1;
 
@@ -187,43 +164,6 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne
     return ip-istart;
 }
 
-/* Avoids the FORCE_INLINE of the _body() function. */
-static size_t FSE_readNCount_body_default(
-        short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
-        const void* headerBuffer, size_t hbSize)
-{
-    return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
-}
-
-#if DYNAMIC_BMI2
-BMI2_TARGET_ATTRIBUTE static size_t FSE_readNCount_body_bmi2(
-        short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
-        const void* headerBuffer, size_t hbSize)
-{
-    return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
-}
-#endif
-
-size_t FSE_readNCount_bmi2(
-        short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
-        const void* headerBuffer, size_t hbSize, int bmi2)
-{
-#if DYNAMIC_BMI2
-    if (bmi2) {
-        return FSE_readNCount_body_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
-    }
-#endif
-    (void)bmi2;
-    return FSE_readNCount_body_default(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
-}
-
-size_t FSE_readNCount(
-        short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
-        const void* headerBuffer, size_t hbSize)
-{
-    return FSE_readNCount_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize, /* bmi2 */ 0);
-}
-
 
 /*! HUF_readStats() :
     Read compact Huffman tree, saved by HUF_writeCTable().
@@ -235,17 +175,6 @@ size_t FSE_readNCount(
 size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats,
                      U32* nbSymbolsPtr, U32* tableLogPtr,
                      const void* src, size_t srcSize)
-{
-    U32 wksp[HUF_READ_STATS_WORKSPACE_SIZE_U32];
-    return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* flags */ 0);
-}
-
-FORCE_INLINE_TEMPLATE size_t
-HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats,
-                   U32* nbSymbolsPtr, U32* tableLogPtr,
-                   const void* src, size_t srcSize,
-                   void* workSpace, size_t wkspSize,
-                   int bmi2)
 {
     U32 weightTotal;
     const BYTE* ip = (const BYTE*) src;
@@ -254,7 +183,7 @@ HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats,
 
     if (!srcSize) return ERROR(srcSize_wrong);
     iSize = ip[0];
-    /* ZSTD_memset(huffWeight, 0, hwSize);   *//* is not necessary, even though some analyzer complain ... */
+    /* memset(huffWeight, 0, hwSize);   *//* is not necessary, even though some analyzer complain ... */
 
     if (iSize >= 128) {  /* special header */
         oSize = iSize - 127;
@@ -268,31 +197,31 @@ HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats,
                 huffWeight[n+1] = ip[n/2] & 15;
     }   }   }
     else  {   /* header compressed with FSE (normal case) */
+        FSE_DTable fseWorkspace[FSE_DTABLE_SIZE_U32(6)];  /* 6 is max possible tableLog for HUF header (maybe even 5, to be tested) */
         if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
-        /* max (hwSize-1) values decoded, as last one is implied */
-        oSize = FSE_decompress_wksp_bmi2(huffWeight, hwSize-1, ip+1, iSize, 6, workSpace, wkspSize, bmi2);
+        oSize = FSE_decompress_wksp(huffWeight, hwSize-1, ip+1, iSize, fseWorkspace, 6);   /* max (hwSize-1) values decoded, as last one is implied */
         if (FSE_isError(oSize)) return oSize;
     }
 
     /* collect weight stats */
-    ZSTD_memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32));
+    memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32));
     weightTotal = 0;
     {   U32 n; for (n=0; n<oSize; n++) {
-            if (huffWeight[n] > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
+            if (huffWeight[n] >= HUF_TABLELOG_MAX) return ERROR(corruption_detected);
             rankStats[huffWeight[n]]++;
             weightTotal += (1 << huffWeight[n]) >> 1;
     }   }
     if (weightTotal == 0) return ERROR(corruption_detected);
 
     /* get last non-null symbol weight (implied, total must be 2^n) */
-    {   U32 const tableLog = ZSTD_highbit32(weightTotal) + 1;
+    {   U32 const tableLog = BIT_highbit32(weightTotal) + 1;
         if (tableLog > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
         *tableLogPtr = tableLog;
         /* determine last weight */
         {   U32 const total = 1 << tableLog;
             U32 const rest = total - weightTotal;
-            U32 const verif = 1 << ZSTD_highbit32(rest);
-            U32 const lastWeight = ZSTD_highbit32(rest) + 1;
+            U32 const verif = 1 << BIT_highbit32(rest);
+            U32 const lastWeight = BIT_highbit32(rest) + 1;
             if (verif != rest) return ERROR(corruption_detected);    /* last value must be a clean power of 2 */
             huffWeight[oSize] = (BYTE)lastWeight;
             rankStats[lastWeight]++;
@@ -305,39 +234,3 @@ HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats,
     *nbSymbolsPtr = (U32)(oSize+1);
     return iSize+1;
 }
-
-/* Avoids the FORCE_INLINE of the _body() function. */
-static size_t HUF_readStats_body_default(BYTE* huffWeight, size_t hwSize, U32* rankStats,
-                     U32* nbSymbolsPtr, U32* tableLogPtr,
-                     const void* src, size_t srcSize,
-                     void* workSpace, size_t wkspSize)
-{
-    return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 0);
-}
-
-#if DYNAMIC_BMI2
-static BMI2_TARGET_ATTRIBUTE size_t HUF_readStats_body_bmi2(BYTE* huffWeight, size_t hwSize, U32* rankStats,
-                     U32* nbSymbolsPtr, U32* tableLogPtr,
-                     const void* src, size_t srcSize,
-                     void* workSpace, size_t wkspSize)
-{
-    return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 1);
-}
-#endif
-
-size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, U32* rankStats,
-                     U32* nbSymbolsPtr, U32* tableLogPtr,
-                     const void* src, size_t srcSize,
-                     void* workSpace, size_t wkspSize,
-                     int flags)
-{
-#if DYNAMIC_BMI2
-    if (flags & HUF_flags_bmi2) {
-        return HUF_readStats_body_bmi2(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
-    }
-#endif
-    (void)flags;
-    return HUF_readStats_body_default(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
-}
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/error_private.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -28,11 +27,9 @@ const char* ERR_getErrorString(ERR_enum code)
     case PREFIX(version_unsupported): return "Version not supported";
     case PREFIX(frameParameter_unsupported): return "Unsupported frame parameter";
     case PREFIX(frameParameter_windowTooLarge): return "Frame requires too much memory for decoding";
-    case PREFIX(corruption_detected): return "Data corruption detected";
+    case PREFIX(corruption_detected): return "Corrupted block detected";
     case PREFIX(checksum_wrong): return "Restored data doesn't match checksum";
-    case PREFIX(literals_headerWrong): return "Header of Literals' block doesn't respect format specification";
     case PREFIX(parameter_unsupported): return "Unsupported parameter";
-    case PREFIX(parameter_combination_unsupported): return "Unsupported combination of parameters";
     case PREFIX(parameter_outOfBound): return "Parameter is out of bound";
     case PREFIX(init_missing): return "Context should be init first";
     case PREFIX(memory_allocation): return "Allocation error : not enough memory";
@@ -41,26 +38,17 @@ const char* ERR_getErrorString(ERR_enum code)
     case PREFIX(tableLog_tooLarge): return "tableLog requires too much memory : unsupported";
     case PREFIX(maxSymbolValue_tooLarge): return "Unsupported max Symbol Value : too large";
     case PREFIX(maxSymbolValue_tooSmall): return "Specified maxSymbolValue is too small";
-    case PREFIX(stabilityCondition_notRespected): return "pledged buffer stability condition is not respected";
     case PREFIX(dictionary_corrupted): return "Dictionary is corrupted";
     case PREFIX(dictionary_wrong): return "Dictionary mismatch";
     case PREFIX(dictionaryCreation_failed): return "Cannot create Dictionary from provided samples";
     case PREFIX(dstSize_tooSmall): return "Destination buffer is too small";
     case PREFIX(srcSize_wrong): return "Src size is incorrect";
     case PREFIX(dstBuffer_null): return "Operation on NULL destination buffer";
-    case PREFIX(noForwardProgress_destFull): return "Operation made no progress over multiple calls, due to output buffer being full";
-    case PREFIX(noForwardProgress_inputEmpty): return "Operation made no progress over multiple calls, due to input being empty";
         /* following error codes are not stable and may be removed or changed in a future version */
     case PREFIX(frameIndex_tooLarge): return "Frame index is too large";
     case PREFIX(seekableIO): return "An I/O error occurred when reading/seeking";
-    case PREFIX(dstBuffer_wrong): return "Destination buffer is wrong";
-    case PREFIX(srcBuffer_wrong): return "Source buffer is wrong";
-    case PREFIX(sequenceProducer_failed): return "Block-level external sequence producer returned an error code";
-    case PREFIX(externalSequences_invalid): return "External sequences are not valid";
     case PREFIX(maxCode):
     default: return notErrorCode;
     }
 #endif
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/error_private.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -22,10 +21,8 @@ extern "C" {
 /* ****************************************
 *  Dependencies
 ******************************************/
+#include <stddef.h>        /* size_t */
 #include "zstd_errors.h"  /* enum list */
-#include "compiler.h"
-#include "debug.h"
-#include "zstd_deps.h"       /* size_t */
 
 
 /* ****************************************
@@ -52,7 +49,7 @@ typedef ZSTD_ErrorCode ERR_enum;
 /*-****************************************
 *  Error codes handling
 ******************************************/
-#undef ERROR   /* already defined on Visual Studio */
+#undef ERROR   /* reported already defined on VS 2015 (Rich Geldreich) */
 #define ERROR(name) ZSTD_ERROR(name)
 #define ZSTD_ERROR(name) ((size_t)-PREFIX(name))
 
@@ -60,15 +57,6 @@ ERR_STATIC unsigned ERR_isError(size_t code) { return (code > ERROR(maxCode)); }
 
 ERR_STATIC ERR_enum ERR_getErrorCode(size_t code) { if (!ERR_isError(code)) return (ERR_enum)0; return (ERR_enum) (0-code); }
 
-/* check and forward error code */
-#define CHECK_V_F(e, f)     \
-    size_t const e = f;     \
-    do {                    \
-        if (ERR_isError(e)) \
-            return e;       \
-    } while (0)
-#define CHECK_F(f)   do { CHECK_V_F(_var_err__, f); } while (0)
-
 
 /*-****************************************
 *  Error Strings
@@ -81,91 +69,8 @@ ERR_STATIC const char* ERR_getErrorName(size_t code)
     return ERR_getErrorString(ERR_getErrorCode(code));
 }
 
-/**
- * Ignore: this is an internal helper.
- *
- * This is a helper function to help force C99-correctness during compilation.
- * Under strict compilation modes, variadic macro arguments can't be empty.
- * However, variadic function arguments can be. Using a function therefore lets
- * us statically check that at least one (string) argument was passed,
- * independent of the compilation flags.
- */
-static INLINE_KEYWORD UNUSED_ATTR
-void _force_has_format_string(const char *format, ...) {
-  (void)format;
-}
-
-/**
- * Ignore: this is an internal helper.
- *
- * We want to force this function invocation to be syntactically correct, but
- * we don't want to force runtime evaluation of its arguments.
- */
-#define _FORCE_HAS_FORMAT_STRING(...)              \
-    do {                                           \
-        if (0) {                                   \
-            _force_has_format_string(__VA_ARGS__); \
-        }                                          \
-    } while (0)
-
-#define ERR_QUOTE(str) #str
-
-/**
- * Return the specified error if the condition evaluates to true.
- *
- * In debug modes, prints additional information.
- * In order to do that (particularly, printing the conditional that failed),
- * this can't just wrap RETURN_ERROR().
- */
-#define RETURN_ERROR_IF(cond, err, ...)                                        \
-    do {                                                                       \
-        if (cond) {                                                            \
-            RAWLOG(3, "%s:%d: ERROR!: check %s failed, returning %s",          \
-                  __FILE__, __LINE__, ERR_QUOTE(cond), ERR_QUOTE(ERROR(err))); \
-            _FORCE_HAS_FORMAT_STRING(__VA_ARGS__);                             \
-            RAWLOG(3, ": " __VA_ARGS__);                                       \
-            RAWLOG(3, "\n");                                                   \
-            return ERROR(err);                                                 \
-        }                                                                      \
-    } while (0)
-
-/**
- * Unconditionally return the specified error.
- *
- * In debug modes, prints additional information.
- */
-#define RETURN_ERROR(err, ...)                                               \
-    do {                                                                     \
-        RAWLOG(3, "%s:%d: ERROR!: unconditional check failed, returning %s", \
-              __FILE__, __LINE__, ERR_QUOTE(ERROR(err)));                    \
-        _FORCE_HAS_FORMAT_STRING(__VA_ARGS__);                               \
-        RAWLOG(3, ": " __VA_ARGS__);                                         \
-        RAWLOG(3, "\n");                                                     \
-        return ERROR(err);                                                   \
-    } while(0)
-
-/**
- * If the provided expression evaluates to an error code, returns that error code.
- *
- * In debug modes, prints additional information.
- */
-#define FORWARD_IF_ERROR(err, ...)                                                 \
-    do {                                                                           \
-        size_t const err_code = (err);                                             \
-        if (ERR_isError(err_code)) {                                               \
-            RAWLOG(3, "%s:%d: ERROR!: forwarding error in %s: %s",                 \
-                  __FILE__, __LINE__, ERR_QUOTE(err), ERR_getErrorName(err_code)); \
-            _FORCE_HAS_FORMAT_STRING(__VA_ARGS__);                                 \
-            RAWLOG(3, ": " __VA_ARGS__);                                           \
-            RAWLOG(3, "\n");                                                       \
-            return err_code;                                                       \
-        }                                                                          \
-    } while(0)
-
 #if defined (__cplusplus)
 }
 #endif
 
 #endif /* ERROR_H_MODULE */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/errors.go

@@ -1,6 +1,7 @@
 package zstd
 
 /*
+#define ZSTD_STATIC_LINKING_ONLY
 #include "zstd.h"
 */
 import "C"

vendor/github.com/DataDog/zstd/external_zstd.go

@@ -1,14 +0,0 @@
-//go:build external_libzstd
-// +build external_libzstd
-
-package zstd
-
-// #cgo CFLAGS: -DUSE_EXTERNAL_ZSTD
-// #cgo pkg-config: libzstd
-/*
-#include<zstd.h>
-#if ZSTD_VERSION_NUMBER < 10400
-#error "ZSTD version >= 1.4 is required"
-#endif
-*/
-import "C"

vendor/github.com/DataDog/zstd/fastcover.c

@@ -1,14 +1,3 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
 /*-*************************************
 *  Dependencies
 ***************************************/
@@ -17,33 +6,24 @@
 #include <string.h> /* memset */
 #include <time.h>   /* clock */
 
-#ifndef ZDICT_STATIC_LINKING_ONLY
-#  define ZDICT_STATIC_LINKING_ONLY
-#endif
-
 #include "mem.h" /* read */
 #include "pool.h"
 #include "threading.h"
-#include "zstd_internal.h" /* includes zstd.h */
-#include "zstd_compress_internal.h" /* ZSTD_hash*() */
-#include "zdict.h"
 #include "cover.h"
+#include "zstd_internal.h" /* includes zstd.h */
+#ifndef ZDICT_STATIC_LINKING_ONLY
+#define ZDICT_STATIC_LINKING_ONLY
+#endif
+#include "zdict.h"
 
 
 /*-*************************************
 *  Constants
 ***************************************/
-/**
-* There are 32bit indexes used to ref samples, so limit samples size to 4GB
-* on 64bit builds.
-* For 32bit builds we choose 1 GB.
-* Most 32bit platforms have 2GB user-mode addressable space and we allocate a large
-* contiguous buffer, so 1GB is already a high limit.
-*/
 #define FASTCOVER_MAX_SAMPLES_SIZE (sizeof(size_t) == 8 ? ((unsigned)-1) : ((unsigned)1 GB))
 #define FASTCOVER_MAX_F 31
 #define FASTCOVER_MAX_ACCEL 10
-#define FASTCOVER_DEFAULT_SPLITPOINT 0.75
+#define DEFAULT_SPLITPOINT 0.75
 #define DEFAULT_F 20
 #define DEFAULT_ACCEL 1
 
@@ -51,50 +31,50 @@
 /*-*************************************
 *  Console display
 ***************************************/
-#ifndef LOCALDISPLAYLEVEL
-static int g_displayLevel = 0;
-#endif
-#undef  DISPLAY
+static int g_displayLevel = 2;
 #define DISPLAY(...)                                                           \
   {                                                                            \
     fprintf(stderr, __VA_ARGS__);                                              \
     fflush(stderr);                                                            \
   }
-#undef  LOCALDISPLAYLEVEL
 #define LOCALDISPLAYLEVEL(displayLevel, l, ...)                                \
   if (displayLevel >= l) {                                                     \
     DISPLAY(__VA_ARGS__);                                                      \
   } /* 0 : no display;   1: errors;   2: default;  3: details;  4: debug */
-#undef  DISPLAYLEVEL
 #define DISPLAYLEVEL(l, ...) LOCALDISPLAYLEVEL(g_displayLevel, l, __VA_ARGS__)
 
-#ifndef LOCALDISPLAYUPDATE
-static const clock_t g_refreshRate = CLOCKS_PER_SEC * 15 / 100;
-static clock_t g_time = 0;
-#endif
-#undef  LOCALDISPLAYUPDATE
 #define LOCALDISPLAYUPDATE(displayLevel, l, ...)                               \
   if (displayLevel >= l) {                                                     \
-    if ((clock() - g_time > g_refreshRate) || (displayLevel >= 4)) {             \
+    if ((clock() - g_time > refreshRate) || (displayLevel >= 4)) {             \
       g_time = clock();                                                        \
       DISPLAY(__VA_ARGS__);                                                    \
     }                                                                          \
   }
-#undef  DISPLAYUPDATE
 #define DISPLAYUPDATE(l, ...) LOCALDISPLAYUPDATE(g_displayLevel, l, __VA_ARGS__)
+static const clock_t refreshRate = CLOCKS_PER_SEC * 15 / 100;
+static clock_t g_time = 0;
 
 
 /*-*************************************
 * Hash Functions
 ***************************************/
+static const U64 prime6bytes = 227718039650203ULL;
+static size_t ZSTD_hash6(U64 u, U32 h) { return (size_t)(((u  << (64-48)) * prime6bytes) >> (64-h)) ; }
+static size_t ZSTD_hash6Ptr(const void* p, U32 h) { return ZSTD_hash6(MEM_readLE64(p), h); }
+
+static const U64 prime8bytes = 0xCF1BBCDCB7A56463ULL;
+static size_t ZSTD_hash8(U64 u, U32 h) { return (size_t)(((u) * prime8bytes) >> (64-h)) ; }
+static size_t ZSTD_hash8Ptr(const void* p, U32 h) { return ZSTD_hash8(MEM_readLE64(p), h); }
+
+
 /**
- * Hash the d-byte value pointed to by p and mod 2^f into the frequency vector
+ * Hash the d-byte value pointed to by p and mod 2^f
  */
-static size_t FASTCOVER_hashPtrToIndex(const void* p, U32 f, unsigned d) {
+static size_t FASTCOVER_hashPtrToIndex(const void* p, U32 h, unsigned d) {
   if (d == 6) {
-    return ZSTD_hash6Ptr(p, f);
+    return ZSTD_hash6Ptr(p, h) & ((1 << h) - 1);
   }
-  return ZSTD_hash8Ptr(p, f);
+  return ZSTD_hash8Ptr(p, h) & ((1 << h) - 1);
 }
 
 
@@ -305,7 +285,7 @@ FASTCOVER_computeFrequency(U32* freqs, const FASTCOVER_ctx_t* ctx)
 
 /**
  * Prepare a context for dictionary building.
- * The context is only dependent on the parameter `d` and can be used multiple
+ * The context is only dependent on the parameter `d` and can used multiple
  * times.
  * Returns 0 on success or error code on error.
  * The context must be destroyed with `FASTCOVER_ctx_destroy()`.
@@ -471,20 +451,20 @@ typedef struct FASTCOVER_tryParameters_data_s {
  * This function is thread safe if zstd is compiled with multithreaded support.
  * It takes its parameters as an *OWNING* opaque pointer to support threading.
  */
-static void FASTCOVER_tryParameters(void* opaque)
+static void FASTCOVER_tryParameters(void *opaque)
 {
   /* Save parameters as local variables */
-  FASTCOVER_tryParameters_data_t *const data = (FASTCOVER_tryParameters_data_t*)opaque;
+  FASTCOVER_tryParameters_data_t *const data = (FASTCOVER_tryParameters_data_t *)opaque;
   const FASTCOVER_ctx_t *const ctx = data->ctx;
   const ZDICT_cover_params_t parameters = data->parameters;
   size_t dictBufferCapacity = data->dictBufferCapacity;
   size_t totalCompressedSize = ERROR(GENERIC);
   /* Initialize array to keep track of frequency of dmer within activeSegment */
-  U16* segmentFreqs = (U16*)calloc(((U64)1 << ctx->f), sizeof(U16));
+  U16* segmentFreqs = (U16 *)calloc(((U64)1 << ctx->f), sizeof(U16));
   /* Allocate space for hash table, dict, and freqs */
-  BYTE *const dict = (BYTE*)malloc(dictBufferCapacity);
+  BYTE *const dict = (BYTE * const)malloc(dictBufferCapacity);
   COVER_dictSelection_t selection = COVER_dictSelectionError(ERROR(GENERIC));
-  U32* freqs = (U32*) malloc(((U64)1 << ctx->f) * sizeof(U32));
+  U32 *freqs = (U32*) malloc(((U64)1 << ctx->f) * sizeof(U32));
   if (!segmentFreqs || !dict || !freqs) {
     DISPLAYLEVEL(1, "Failed to allocate buffers: out of memory\n");
     goto _cleanup;
@@ -496,7 +476,7 @@ static void FASTCOVER_tryParameters(void* opaque)
                                                     parameters, segmentFreqs);
 
     const unsigned nbFinalizeSamples = (unsigned)(ctx->nbTrainSamples * ctx->accelParams.finalize / 100);
-    selection = COVER_selectDict(dict + tail, dictBufferCapacity, dictBufferCapacity - tail,
+    selection = COVER_selectDict(dict + tail, dictBufferCapacity - tail,
          ctx->samples, ctx->samplesSizes, nbFinalizeSamples, ctx->nbTrainSamples, ctx->nbSamples, parameters, ctx->offsets,
          totalCompressedSize);
 
@@ -546,7 +526,7 @@ FASTCOVER_convertToFastCoverParams(ZDICT_cover_params_t coverParams,
 }
 
 
-ZDICTLIB_STATIC_API size_t
+ZDICTLIB_API size_t
 ZDICT_trainFromBuffer_fastCover(void* dictBuffer, size_t dictBufferCapacity,
                                 const void* samplesBuffer,
                                 const size_t* samplesSizes, unsigned nbSamples,
@@ -557,7 +537,7 @@ ZDICT_trainFromBuffer_fastCover(void* dictBuffer, size_t dictBufferCapacity,
     ZDICT_cover_params_t coverParams;
     FASTCOVER_accel_t accelParams;
     /* Initialize global data */
-    g_displayLevel = (int)parameters.zParams.notificationLevel;
+    g_displayLevel = parameters.zParams.notificationLevel;
     /* Assign splitPoint and f if not provided */
     parameters.splitPoint = 1.0;
     parameters.f = parameters.f == 0 ? DEFAULT_F : parameters.f;
@@ -615,7 +595,7 @@ ZDICT_trainFromBuffer_fastCover(void* dictBuffer, size_t dictBufferCapacity,
 }
 
 
-ZDICTLIB_STATIC_API size_t
+ZDICTLIB_API size_t
 ZDICT_optimizeTrainFromBuffer_fastCover(
                     void* dictBuffer, size_t dictBufferCapacity,
                     const void* samplesBuffer,
@@ -627,7 +607,7 @@ ZDICT_optimizeTrainFromBuffer_fastCover(
     /* constants */
     const unsigned nbThreads = parameters->nbThreads;
     const double splitPoint =
-        parameters->splitPoint <= 0.0 ? FASTCOVER_DEFAULT_SPLITPOINT : parameters->splitPoint;
+        parameters->splitPoint <= 0.0 ? DEFAULT_SPLITPOINT : parameters->splitPoint;
     const unsigned kMinD = parameters->d == 0 ? 6 : parameters->d;
     const unsigned kMaxD = parameters->d == 0 ? 8 : parameters->d;
     const unsigned kMinK = parameters->k == 0 ? 50 : parameters->k;
@@ -640,7 +620,7 @@ ZDICT_optimizeTrainFromBuffer_fastCover(
     const unsigned accel = parameters->accel == 0 ? DEFAULT_ACCEL : parameters->accel;
     const unsigned shrinkDict = 0;
     /* Local variables */
-    const int displayLevel = (int)parameters->zParams.notificationLevel;
+    const int displayLevel = parameters->zParams.notificationLevel;
     unsigned iteration = 1;
     unsigned d;
     unsigned k;
@@ -724,7 +704,7 @@ ZDICT_optimizeTrainFromBuffer_fastCover(
         data->parameters.splitPoint = splitPoint;
         data->parameters.steps = kSteps;
         data->parameters.shrinkDict = shrinkDict;
-        data->parameters.zParams.notificationLevel = (unsigned)g_displayLevel;
+        data->parameters.zParams.notificationLevel = g_displayLevel;
         /* Check the parameters */
         if (!FASTCOVER_checkParameters(data->parameters, dictBufferCapacity,
                                        data->ctx->f, accel)) {
@@ -765,5 +745,3 @@ ZDICT_optimizeTrainFromBuffer_fastCover(
     }
 
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/fse.h

@@ -1,16 +1,35 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * FSE : Finite State Entropy codec
- * Public Prototypes declaration
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- * You can contact the author at :
- * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   FSE : Finite State Entropy codec
+   Public Prototypes declaration
+   Copyright (C) 2013-2016, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+   You can contact the author at :
+   - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
 ****************************************************************** */
 
 #if defined (__cplusplus)
@@ -24,7 +43,7 @@ extern "C" {
 /*-*****************************************
 *  Dependencies
 ******************************************/
-#include "zstd_deps.h"    /* size_t, ptrdiff_t */
+#include <stddef.h>    /* size_t, ptrdiff_t */
 
 
 /*-*****************************************
@@ -54,6 +73,34 @@ extern "C" {
 FSE_PUBLIC_API unsigned FSE_versionNumber(void);   /**< library version number; to be used when checking dll version */
 
 
+/*-****************************************
+*  FSE simple functions
+******************************************/
+/*! FSE_compress() :
+    Compress content of buffer 'src', of size 'srcSize', into destination buffer 'dst'.
+    'dst' buffer must be already allocated. Compression runs faster is dstCapacity >= FSE_compressBound(srcSize).
+    @return : size of compressed data (<= dstCapacity).
+    Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
+                     if return == 1, srcData is a single byte symbol * srcSize times. Use RLE compression instead.
+                     if FSE_isError(return), compression failed (more details using FSE_getErrorName())
+*/
+FSE_PUBLIC_API size_t FSE_compress(void* dst, size_t dstCapacity,
+                             const void* src, size_t srcSize);
+
+/*! FSE_decompress():
+    Decompress FSE data from buffer 'cSrc', of size 'cSrcSize',
+    into already allocated destination buffer 'dst', of size 'dstCapacity'.
+    @return : size of regenerated data (<= maxDstSize),
+              or an error code, which can be tested using FSE_isError() .
+
+    ** Important ** : FSE_decompress() does not decompress non-compressible nor RLE data !!!
+    Why ? : making this distinction requires a header.
+    Header management is intentionally delegated to the user layer, which can better manage special cases.
+*/
+FSE_PUBLIC_API size_t FSE_decompress(void* dst,  size_t dstCapacity,
+                               const void* cSrc, size_t cSrcSize);
+
+
 /*-*****************************************
 *  Tool functions
 ******************************************/
@@ -64,6 +111,20 @@ FSE_PUBLIC_API unsigned    FSE_isError(size_t code);        /* tells if a return
 FSE_PUBLIC_API const char* FSE_getErrorName(size_t code);   /* provides error code string (useful for debugging) */
 
 
+/*-*****************************************
+*  FSE advanced functions
+******************************************/
+/*! FSE_compress2() :
+    Same as FSE_compress(), but allows the selection of 'maxSymbolValue' and 'tableLog'
+    Both parameters can be defined as '0' to mean : use default value
+    @return : size of compressed data
+    Special values : if return == 0, srcData is not compressible => Nothing is stored within cSrc !!!
+                     if return == 1, srcData is a single byte symbol * srcSize times. Use RLE compression.
+                     if FSE_isError(return), it's an error code.
+*/
+FSE_PUBLIC_API size_t FSE_compress2 (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
+
+
 /*-*****************************************
 *  FSE detailed API
 ******************************************/
@@ -96,16 +157,10 @@ FSE_PUBLIC_API unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize
 /*! FSE_normalizeCount():
     normalize counts so that sum(count[]) == Power_of_2 (2^tableLog)
     'normalizedCounter' is a table of short, of minimum size (maxSymbolValue+1).
-    useLowProbCount is a boolean parameter which trades off compressed size for
-    faster header decoding. When it is set to 1, the compressed data will be slightly
-    smaller. And when it is set to 0, FSE_readNCount() and FSE_buildDTable() will be
-    faster. If you are compressing a small amount of data (< 2 KB) then useLowProbCount=0
-    is a good default, since header deserialization makes a big speed difference.
-    Otherwise, useLowProbCount=1 is a good default, since the speed difference is small.
     @return : tableLog,
               or an errorCode, which can be tested using FSE_isError() */
 FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog,
-                    const unsigned* count, size_t srcSize, unsigned maxSymbolValue, unsigned useLowProbCount);
+                    const unsigned* count, size_t srcSize, unsigned maxSymbolValue);
 
 /*! FSE_NCountWriteBound():
     Provides the maximum possible size of an FSE normalized table, given 'maxSymbolValue' and 'tableLog'.
@@ -123,6 +178,8 @@ FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize,
 /*! Constructor and Destructor of FSE_CTable.
     Note that FSE_CTable size depends on 'tableLog' and 'maxSymbolValue' */
 typedef unsigned FSE_CTable;   /* don't allocate that. It's only meant to be more restrictive than void* */
+FSE_PUBLIC_API FSE_CTable* FSE_createCTable (unsigned maxSymbolValue, unsigned tableLog);
+FSE_PUBLIC_API void        FSE_freeCTable (FSE_CTable* ct);
 
 /*! FSE_buildCTable():
     Builds `ct`, which must be already allocated, using FSE_createCTable().
@@ -191,14 +248,23 @@ FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter,
                            unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
                            const void* rBuffer, size_t rBuffSize);
 
-/*! FSE_readNCount_bmi2():
- * Same as FSE_readNCount() but pass bmi2=1 when your CPU supports BMI2 and 0 otherwise.
- */
-FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter,
-                           unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
-                           const void* rBuffer, size_t rBuffSize, int bmi2);
-
+/*! Constructor and Destructor of FSE_DTable.
+    Note that its size depends on 'tableLog' */
 typedef unsigned FSE_DTable;   /* don't allocate that. It's just a way to be more restrictive than void* */
+FSE_PUBLIC_API FSE_DTable* FSE_createDTable(unsigned tableLog);
+FSE_PUBLIC_API void        FSE_freeDTable(FSE_DTable* dt);
+
+/*! FSE_buildDTable():
+    Builds 'dt', which must be already allocated, using FSE_createDTable().
+    return : 0, or an errorCode, which can be tested using FSE_isError() */
+FSE_PUBLIC_API size_t FSE_buildDTable (FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
+
+/*! FSE_decompress_usingDTable():
+    Decompress compressed source `cSrc` of size `cSrcSize` using `dt`
+    into `dst` which must be already allocated.
+    @return : size of regenerated data (necessarily <= `dstCapacity`),
+              or an errorCode, which can be tested using FSE_isError() */
+FSE_PUBLIC_API size_t FSE_decompress_usingDTable(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, const FSE_DTable* dt);
 
 /*!
 Tutorial :
@@ -230,7 +296,6 @@ If there is an error, the function will return an error code, which can be teste
 
 #endif  /* FSE_H */
 
-
 #if defined(FSE_STATIC_LINKING_ONLY) && !defined(FSE_H_FSE_STATIC_LINKING_ONLY)
 #define FSE_H_FSE_STATIC_LINKING_ONLY
 
@@ -243,12 +308,12 @@ If there is an error, the function will return an error code, which can be teste
 *******************************************/
 /* FSE buffer bounds */
 #define FSE_NCOUNTBOUND 512
-#define FSE_BLOCKBOUND(size) ((size) + ((size)>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */)
+#define FSE_BLOCKBOUND(size) (size + (size>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */)
 #define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size))   /* Macro version, useful for static allocation */
 
 /* It is possible to statically allocate FSE CTable/DTable as a table of FSE_CTable/FSE_DTable using below macros */
-#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue)   (1 + (1<<((maxTableLog)-1)) + (((maxSymbolValue)+1)*2))
-#define FSE_DTABLE_SIZE_U32(maxTableLog)                   (1 + (1<<(maxTableLog)))
+#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue)   (1 + (1<<(maxTableLog-1)) + ((maxSymbolValue+1)*2))
+#define FSE_DTABLE_SIZE_U32(maxTableLog)                   (1 + (1<<maxTableLog))
 
 /* or use the size to malloc() space directly. Pay attention to alignment restrictions though */
 #define FSE_CTABLE_SIZE(maxTableLog, maxSymbolValue)   (FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(FSE_CTable))
@@ -262,28 +327,33 @@ If there is an error, the function will return an error code, which can be teste
 unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus);
 /**< same as FSE_optimalTableLog(), which used `minus==2` */
 
+/* FSE_compress_wksp() :
+ * Same as FSE_compress2(), but using an externally allocated scratch buffer (`workSpace`).
+ * FSE_WKSP_SIZE_U32() provides the minimum size required for `workSpace` as a table of FSE_CTable.
+ */
+#define FSE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue)   ( FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) + ((maxTableLog > 12) ? (1 << (maxTableLog - 2)) : 1024) )
+size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
+
+size_t FSE_buildCTable_raw (FSE_CTable* ct, unsigned nbBits);
+/**< build a fake FSE_CTable, designed for a flat distribution, where each symbol uses nbBits */
+
 size_t FSE_buildCTable_rle (FSE_CTable* ct, unsigned char symbolValue);
 /**< build a fake FSE_CTable, designed to compress always the same symbolValue */
 
 /* FSE_buildCTable_wksp() :
  * Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`).
- * `wkspSize` must be >= `FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)` of `unsigned`.
- * See FSE_buildCTable_wksp() for breakdown of workspace usage.
+ * `wkspSize` must be >= `(1<<tableLog)`.
  */
-#define FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog) (((maxSymbolValue + 2) + (1ull << (tableLog)))/2 + sizeof(U64)/sizeof(U32) /* additional 8 bytes for potential table overwrite */)
-#define FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) (sizeof(unsigned) * FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog))
 size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
 
-#define FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) (sizeof(short) * (maxSymbolValue + 1) + (1ULL << maxTableLog) + 8)
-#define FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ((FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) + sizeof(unsigned) - 1) / sizeof(unsigned))
-FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
-/**< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */
+size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits);
+/**< build a fake FSE_DTable, designed to read a flat distribution where each symbol uses nbBits */
 
-#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + 1 + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1)
-#define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned))
-size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2);
-/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)`.
- * Set bmi2 to 1 if your CPU supports BMI2 or 0 if it doesn't */
+size_t FSE_buildDTable_rle (FSE_DTable* dt, unsigned char symbolValue);
+/**< build a fake FSE_DTable, designed to always generate the same symbolValue */
+
+size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, FSE_DTable* workSpace, unsigned maxLog);
+/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DTABLE_SIZE_U32(maxLog)` */
 
 typedef enum {
    FSE_repeat_none,  /**< Cannot use the previous table */
@@ -466,20 +536,20 @@ MEM_STATIC void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* statePtr, un
     FSE_symbolCompressionTransform const symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol];
     const U16* const stateTable = (const U16*)(statePtr->stateTable);
     U32 const nbBitsOut  = (U32)((statePtr->value + symbolTT.deltaNbBits) >> 16);
-    BIT_addBits(bitC,  (size_t)statePtr->value, nbBitsOut);
+    BIT_addBits(bitC, statePtr->value, nbBitsOut);
     statePtr->value = stateTable[ (statePtr->value >> nbBitsOut) + symbolTT.deltaFindState];
 }
 
 MEM_STATIC void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* statePtr)
 {
-    BIT_addBits(bitC, (size_t)statePtr->value, statePtr->stateLog);
+    BIT_addBits(bitC, statePtr->value, statePtr->stateLog);
     BIT_flushBits(bitC);
 }
 
 
 /* FSE_getMaxNbBits() :
  * Approximate maximum cost of a symbol, in bits.
- * Fractional get rounded up (i.e. a symbol with a normalized frequency of 3 gives the same result as a frequency of 2)
+ * Fractional get rounded up (i.e : a symbol with a normalized frequency of 3 gives the same result as a frequency of 2)
  * note 1 : assume symbolValue is valid (<= maxSymbolValue)
  * note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */
 MEM_STATIC U32 FSE_getMaxNbBits(const void* symbolTTPtr, U32 symbolValue)
@@ -594,9 +664,6 @@ MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr)
 #ifndef FSE_DEFAULT_MEMORY_USAGE
 #  define FSE_DEFAULT_MEMORY_USAGE 13
 #endif
-#if (FSE_DEFAULT_MEMORY_USAGE > FSE_MAX_MEMORY_USAGE)
-#  error "FSE_DEFAULT_MEMORY_USAGE must be <= FSE_MAX_MEMORY_USAGE"
-#endif
 
 /*!FSE_MAX_SYMBOL_VALUE :
 *  Maximum symbol value authorized.
@@ -630,7 +697,7 @@ MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr)
 #  error "FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX is not supported"
 #endif
 
-#define FSE_TABLESTEP(tableSize) (((tableSize)>>1) + ((tableSize)>>3) + 3)
+#define FSE_TABLESTEP(tableSize) ((tableSize>>1) + (tableSize>>3) + 3)
 
 
 #endif /* FSE_STATIC_LINKING_ONLY */
@@ -639,5 +706,3 @@ MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr)
 #if defined (__cplusplus)
 }
 #endif
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/fse_compress.c

@@ -1,21 +1,42 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * FSE : Finite State Entropy encoder
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- *  You can contact the author at :
- *  - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *  - Public forum : https://groups.google.com/forum/#!forum/lz4c
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   FSE : Finite State Entropy encoder
+   Copyright (C) 2013-present, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+    You can contact the author at :
+    - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
+    - Public forum : https://groups.google.com/forum/#!forum/lz4c
 ****************************************************************** */
 
 /* **************************************************************
 *  Includes
 ****************************************************************/
+#include <stdlib.h>     /* malloc, free, qsort */
+#include <string.h>     /* memcpy, memset */
 #include "compiler.h"
 #include "mem.h"        /* U32, U16, etc. */
 #include "debug.h"      /* assert, DEBUGLOG */
@@ -24,10 +45,6 @@
 #define FSE_STATIC_LINKING_ONLY
 #include "fse.h"
 #include "error_private.h"
-#define ZSTD_DEPS_NEED_MALLOC
-#define ZSTD_DEPS_NEED_MATH64
-#include "zstd_deps.h"  /* ZSTD_memset */
-#include "bits.h" /* ZSTD_highbit32 */
 
 
 /* **************************************************************
@@ -77,85 +94,41 @@ size_t FSE_buildCTable_wksp(FSE_CTable* ct,
     void* const FSCT = ((U32*)ptr) + 1 /* header */ + (tableLog ? tableSize>>1 : 1) ;
     FSE_symbolCompressionTransform* const symbolTT = (FSE_symbolCompressionTransform*) (FSCT);
     U32 const step = FSE_TABLESTEP(tableSize);
-    U32 const maxSV1 = maxSymbolValue+1;
-
-    U16* cumul = (U16*)workSpace;   /* size = maxSV1 */
-    FSE_FUNCTION_TYPE* const tableSymbol = (FSE_FUNCTION_TYPE*)(cumul + (maxSV1+1));  /* size = tableSize */
+    U32 cumul[FSE_MAX_SYMBOL_VALUE+2];
 
+    FSE_FUNCTION_TYPE* const tableSymbol = (FSE_FUNCTION_TYPE*)workSpace;
     U32 highThreshold = tableSize-1;
 
-    assert(((size_t)workSpace & 1) == 0);  /* Must be 2 bytes-aligned */
-    if (FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) > wkspSize) return ERROR(tableLog_tooLarge);
     /* CTable header */
+    if (((size_t)1 << tableLog) * sizeof(FSE_FUNCTION_TYPE) > wkspSize) return ERROR(tableLog_tooLarge);
     tableU16[-2] = (U16) tableLog;
     tableU16[-1] = (U16) maxSymbolValue;
     assert(tableLog < 16);   /* required for threshold strategy to work */
 
     /* For explanations on how to distribute symbol values over the table :
-     * https://fastcompression.blogspot.fr/2014/02/fse-distributing-symbol-values.html */
+     * http://fastcompression.blogspot.fr/2014/02/fse-distributing-symbol-values.html */
 
      #ifdef __clang_analyzer__
-     ZSTD_memset(tableSymbol, 0, sizeof(*tableSymbol) * tableSize);   /* useless initialization, just to keep scan-build happy */
+     memset(tableSymbol, 0, sizeof(*tableSymbol) * tableSize);   /* useless initialization, just to keep scan-build happy */
      #endif
 
     /* symbol start positions */
     {   U32 u;
         cumul[0] = 0;
-        for (u=1; u <= maxSV1; u++) {
+        for (u=1; u <= maxSymbolValue+1; u++) {
             if (normalizedCounter[u-1]==-1) {  /* Low proba symbol */
                 cumul[u] = cumul[u-1] + 1;
                 tableSymbol[highThreshold--] = (FSE_FUNCTION_TYPE)(u-1);
             } else {
-                assert(normalizedCounter[u-1] >= 0);
-                cumul[u] = cumul[u-1] + (U16)normalizedCounter[u-1];
-                assert(cumul[u] >= cumul[u-1]);  /* no overflow */
+                cumul[u] = cumul[u-1] + normalizedCounter[u-1];
         }   }
-        cumul[maxSV1] = (U16)(tableSize+1);
+        cumul[maxSymbolValue+1] = tableSize+1;
     }
 
     /* Spread symbols */
-    if (highThreshold == tableSize - 1) {
-        /* Case for no low prob count symbols. Lay down 8 bytes at a time
-         * to reduce branch misses since we are operating on a small block
-         */
-        BYTE* const spread = tableSymbol + tableSize; /* size = tableSize + 8 (may write beyond tableSize) */
-        {   U64 const add = 0x0101010101010101ull;
-            size_t pos = 0;
-            U64 sv = 0;
-            U32 s;
-            for (s=0; s<maxSV1; ++s, sv += add) {
-                int i;
-                int const n = normalizedCounter[s];
-                MEM_write64(spread + pos, sv);
-                for (i = 8; i < n; i += 8) {
-                    MEM_write64(spread + pos + i, sv);
-                }
-                assert(n>=0);
-                pos += (size_t)n;
-            }
-        }
-        /* Spread symbols across the table. Lack of lowprob symbols means that
-         * we don't need variable sized inner loop, so we can unroll the loop and
-         * reduce branch misses.
-         */
-        {   size_t position = 0;
-            size_t s;
-            size_t const unroll = 2; /* Experimentally determined optimal unroll */
-            assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */
-            for (s = 0; s < (size_t)tableSize; s += unroll) {
-                size_t u;
-                for (u = 0; u < unroll; ++u) {
-                    size_t const uPosition = (position + (u * step)) & tableMask;
-                    tableSymbol[uPosition] = spread[s + u];
-                }
-                position = (position + (unroll * step)) & tableMask;
-            }
-            assert(position == 0);   /* Must have initialized all positions */
-        }
-    } else {
-        U32 position = 0;
+    {   U32 position = 0;
         U32 symbol;
-        for (symbol=0; symbol<maxSV1; symbol++) {
+        for (symbol=0; symbol<=maxSymbolValue; symbol++) {
             int nbOccurrences;
             int const freq = normalizedCounter[symbol];
             for (nbOccurrences=0; nbOccurrences<freq; nbOccurrences++) {
@@ -164,6 +137,7 @@ size_t FSE_buildCTable_wksp(FSE_CTable* ct,
                 while (position > highThreshold)
                     position = (position + step) & tableMask;   /* Low proba area */
         }   }
+
         assert(position==0);  /* Must have initialized all positions */
     }
 
@@ -187,17 +161,16 @@ size_t FSE_buildCTable_wksp(FSE_CTable* ct,
             case -1:
             case  1:
                 symbolTT[s].deltaNbBits = (tableLog << 16) - (1<<tableLog);
-                assert(total <= INT_MAX);
-                symbolTT[s].deltaFindState = (int)(total - 1);
+                symbolTT[s].deltaFindState = total - 1;
                 total ++;
                 break;
             default :
-                assert(normalizedCounter[s] > 1);
-                {   U32 const maxBitsOut = tableLog - ZSTD_highbit32 ((U32)normalizedCounter[s]-1);
-                    U32 const minStatePlus = (U32)normalizedCounter[s] << maxBitsOut;
+                {
+                    U32 const maxBitsOut = tableLog - BIT_highbit32 (normalizedCounter[s]-1);
+                    U32 const minStatePlus = normalizedCounter[s] << maxBitsOut;
                     symbolTT[s].deltaNbBits = (maxBitsOut << 16) - minStatePlus;
-                    symbolTT[s].deltaFindState = (int)(total - (unsigned)normalizedCounter[s]);
-                    total +=  (unsigned)normalizedCounter[s];
+                    symbolTT[s].deltaFindState = total - normalizedCounter[s];
+                    total +=  normalizedCounter[s];
     }   }   }   }
 
 #if 0  /* debug : symbol costs */
@@ -208,26 +181,31 @@ size_t FSE_buildCTable_wksp(FSE_CTable* ct,
                 symbol, normalizedCounter[symbol],
                 FSE_getMaxNbBits(symbolTT, symbol),
                 (double)FSE_bitCost(symbolTT, tableLog, symbol, 8) / 256);
-    }   }
+        }
+    }
 #endif
 
     return 0;
 }
 
 
+size_t FSE_buildCTable(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog)
+{
+    FSE_FUNCTION_TYPE tableSymbol[FSE_MAX_TABLESIZE];   /* memset() is not necessary, even if static analyzer complain about it */
+    return FSE_buildCTable_wksp(ct, normalizedCounter, maxSymbolValue, tableLog, tableSymbol, sizeof(tableSymbol));
+}
+
+
 
 #ifndef FSE_COMMONDEFS_ONLY
 
+
 /*-**************************************************************
 *  FSE NCount encoding
 ****************************************************************/
 size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog)
 {
-    size_t const maxHeaderSize = (((maxSymbolValue+1) * tableLog
-                                   + 4 /* bitCount initialized at 4 */
-                                   + 2 /* first two symbols may use one additional bit each */) / 8)
-                                   + 1 /* round up to whole nb bytes */
-                                   + 2 /* additional two bytes for bitstream flush */;
+    size_t const maxHeaderSize = (((maxSymbolValue+1) * tableLog) >> 3) + 3;
     return maxSymbolValue ? maxHeaderSize : FSE_NCOUNTBOUND;  /* maxSymbolValue==0 ? use default */
 }
 
@@ -256,7 +234,7 @@ FSE_writeNCount_generic (void* header, size_t headerBufferSize,
     /* Init */
     remaining = tableSize+1;   /* +1 for extra accuracy */
     threshold = tableSize;
-    nbBits = (int)tableLog+1;
+    nbBits = tableLog+1;
 
     while ((symbol < alphabetSize) && (remaining>1)) {  /* stops at 1 */
         if (previousIs0) {
@@ -275,7 +253,7 @@ FSE_writeNCount_generic (void* header, size_t headerBufferSize,
             }
             while (symbol >= start+3) {
                 start+=3;
-                bitStream += 3U << bitCount;
+                bitStream += 3 << bitCount;
                 bitCount += 2;
             }
             bitStream += (symbol-start) << bitCount;
@@ -295,7 +273,7 @@ FSE_writeNCount_generic (void* header, size_t headerBufferSize,
             count++;   /* +1 for extra accuracy */
             if (count>=threshold)
                 count += max;   /* [0..max[ [max..threshold[ (...) [threshold+max 2*threshold[ */
-            bitStream += (U32)count << bitCount;
+            bitStream += count << bitCount;
             bitCount  += nbBits;
             bitCount  -= (count<max);
             previousIs0  = (count==1);
@@ -323,8 +301,7 @@ FSE_writeNCount_generic (void* header, size_t headerBufferSize,
     out[1] = (BYTE)(bitStream>>8);
     out+= (bitCount+7) /8;
 
-    assert(out >= ostart);
-    return (size_t)(out-ostart);
+    return (out-ostart);
 }
 
 
@@ -345,11 +322,21 @@ size_t FSE_writeNCount (void* buffer, size_t bufferSize,
 *  FSE Compression Code
 ****************************************************************/
 
+FSE_CTable* FSE_createCTable (unsigned maxSymbolValue, unsigned tableLog)
+{
+    size_t size;
+    if (tableLog > FSE_TABLELOG_ABSOLUTE_MAX) tableLog = FSE_TABLELOG_ABSOLUTE_MAX;
+    size = FSE_CTABLE_SIZE_U32 (tableLog, maxSymbolValue) * sizeof(U32);
+    return (FSE_CTable*)malloc(size);
+}
+
+void FSE_freeCTable (FSE_CTable* ct) { free(ct); }
+
 /* provides the minimum logSize to safely represent a distribution */
 static unsigned FSE_minTableLog(size_t srcSize, unsigned maxSymbolValue)
 {
-    U32 minBitsSrc = ZSTD_highbit32((U32)(srcSize)) + 1;
-    U32 minBitsSymbols = ZSTD_highbit32(maxSymbolValue) + 2;
+    U32 minBitsSrc = BIT_highbit32((U32)(srcSize)) + 1;
+    U32 minBitsSymbols = BIT_highbit32(maxSymbolValue) + 2;
     U32 minBits = minBitsSrc < minBitsSymbols ? minBitsSrc : minBitsSymbols;
     assert(srcSize > 1); /* Not supported, RLE should be used instead */
     return minBits;
@@ -357,7 +344,7 @@ static unsigned FSE_minTableLog(size_t srcSize, unsigned maxSymbolValue)
 
 unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus)
 {
-    U32 maxBitsSrc = ZSTD_highbit32((U32)(srcSize - 1)) - minus;
+    U32 maxBitsSrc = BIT_highbit32((U32)(srcSize - 1)) - minus;
     U32 tableLog = maxTableLog;
     U32 minBits = FSE_minTableLog(srcSize, maxSymbolValue);
     assert(srcSize > 1); /* Not supported, RLE should be used instead */
@@ -374,10 +361,11 @@ unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxS
     return FSE_optimalTableLog_internal(maxTableLog, srcSize, maxSymbolValue, 2);
 }
 
+
 /* Secondary normalization method.
    To be used when primary method fails. */
 
-static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, size_t total, U32 maxSymbolValue, short lowProbCount)
+static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, size_t total, U32 maxSymbolValue)
 {
     short const NOT_YET_ASSIGNED = -2;
     U32 s;
@@ -394,7 +382,7 @@ static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count,
             continue;
         }
         if (count[s] <= lowThreshold) {
-            norm[s] = lowProbCount;
+            norm[s] = -1;
             distributed++;
             total -= count[s];
             continue;
@@ -446,7 +434,7 @@ static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count,
 
     {   U64 const vStepLog = 62 - tableLog;
         U64 const mid = (1ULL << (vStepLog-1)) - 1;
-        U64 const rStep = ZSTD_div64((((U64)1<<vStepLog) * ToDistribute) + mid, (U32)total);   /* scale on remaining */
+        U64 const rStep = ((((U64)1<<vStepLog) * ToDistribute) + mid) / total;   /* scale on remaining */
         U64 tmpTotal = mid;
         for (s=0; s<=maxSymbolValue; s++) {
             if (norm[s]==NOT_YET_ASSIGNED) {
@@ -463,9 +451,10 @@ static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count,
     return 0;
 }
 
+
 size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog,
                            const unsigned* count, size_t total,
-                           unsigned maxSymbolValue, unsigned useLowProbCount)
+                           unsigned maxSymbolValue)
 {
     /* Sanity checks */
     if (tableLog==0) tableLog = FSE_DEFAULT_TABLELOG;
@@ -474,9 +463,8 @@ size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog,
     if (tableLog < FSE_minTableLog(total, maxSymbolValue)) return ERROR(GENERIC);   /* Too small tableLog, compression potentially impossible */
 
     {   static U32 const rtbTable[] = {     0, 473195, 504333, 520860, 550000, 700000, 750000, 830000 };
-        short const lowProbCount = useLowProbCount ? -1 : 1;
         U64 const scale = 62 - tableLog;
-        U64 const step = ZSTD_div64((U64)1<<62, (U32)total);   /* <== here, one division ! */
+        U64 const step = ((U64)1<<62) / total;   /* <== here, one division ! */
         U64 const vStep = 1ULL<<(scale-20);
         int stillToDistribute = 1<<tableLog;
         unsigned s;
@@ -488,7 +476,7 @@ size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog,
             if (count[s] == total) return 0;   /* rle special case */
             if (count[s] == 0) { normalizedCounter[s]=0; continue; }
             if (count[s] <= lowThreshold) {
-                normalizedCounter[s] = lowProbCount;
+                normalizedCounter[s] = -1;
                 stillToDistribute--;
             } else {
                 short proba = (short)((count[s]*step) >> scale);
@@ -502,7 +490,7 @@ size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog,
         }   }
         if (-stillToDistribute >= (normalizedCounter[largest] >> 1)) {
             /* corner case, need another normalization method */
-            size_t const errorCode = FSE_normalizeM2(normalizedCounter, tableLog, count, total, maxSymbolValue, lowProbCount);
+            size_t const errorCode = FSE_normalizeM2(normalizedCounter, tableLog, count, total, maxSymbolValue);
             if (FSE_isError(errorCode)) return errorCode;
         }
         else normalizedCounter[largest] += (short)stillToDistribute;
@@ -525,6 +513,40 @@ size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog,
     return tableLog;
 }
 
+
+/* fake FSE_CTable, for raw (uncompressed) input */
+size_t FSE_buildCTable_raw (FSE_CTable* ct, unsigned nbBits)
+{
+    const unsigned tableSize = 1 << nbBits;
+    const unsigned tableMask = tableSize - 1;
+    const unsigned maxSymbolValue = tableMask;
+    void* const ptr = ct;
+    U16* const tableU16 = ( (U16*) ptr) + 2;
+    void* const FSCT = ((U32*)ptr) + 1 /* header */ + (tableSize>>1);   /* assumption : tableLog >= 1 */
+    FSE_symbolCompressionTransform* const symbolTT = (FSE_symbolCompressionTransform*) (FSCT);
+    unsigned s;
+
+    /* Sanity checks */
+    if (nbBits < 1) return ERROR(GENERIC);             /* min size */
+
+    /* header */
+    tableU16[-2] = (U16) nbBits;
+    tableU16[-1] = (U16) maxSymbolValue;
+
+    /* Build table */
+    for (s=0; s<tableSize; s++)
+        tableU16[s] = (U16)(tableSize + s);
+
+    /* Build Symbol Transformation Table */
+    {   const U32 deltaNbBits = (nbBits << 16) - (1 << nbBits);
+        for (s=0; s<=maxSymbolValue; s++) {
+            symbolTT[s].deltaNbBits = deltaNbBits;
+            symbolTT[s].deltaFindState = s-1;
+    }   }
+
+    return 0;
+}
+
 /* fake FSE_CTable, for rle input (always same symbol) */
 size_t FSE_buildCTable_rle (FSE_CTable* ct, BYTE symbolValue)
 {
@@ -623,6 +645,77 @@ size_t FSE_compress_usingCTable (void* dst, size_t dstSize,
 
 size_t FSE_compressBound(size_t size) { return FSE_COMPRESSBOUND(size); }
 
-#endif   /* FSE_COMMONDEFS_ONLY */
+#define CHECK_V_F(e, f) size_t const e = f; if (ERR_isError(e)) return e
+#define CHECK_F(f)   { CHECK_V_F(_var_err__, f); }
 
-#endif /* USE_EXTERNAL_ZSTD */
+/* FSE_compress_wksp() :
+ * Same as FSE_compress2(), but using an externally allocated scratch buffer (`workSpace`).
+ * `wkspSize` size must be `(1<<tableLog)`.
+ */
+size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
+{
+    BYTE* const ostart = (BYTE*) dst;
+    BYTE* op = ostart;
+    BYTE* const oend = ostart + dstSize;
+
+    unsigned count[FSE_MAX_SYMBOL_VALUE+1];
+    S16   norm[FSE_MAX_SYMBOL_VALUE+1];
+    FSE_CTable* CTable = (FSE_CTable*)workSpace;
+    size_t const CTableSize = FSE_CTABLE_SIZE_U32(tableLog, maxSymbolValue);
+    void* scratchBuffer = (void*)(CTable + CTableSize);
+    size_t const scratchBufferSize = wkspSize - (CTableSize * sizeof(FSE_CTable));
+
+    /* init conditions */
+    if (wkspSize < FSE_WKSP_SIZE_U32(tableLog, maxSymbolValue)) return ERROR(tableLog_tooLarge);
+    if (srcSize <= 1) return 0;  /* Not compressible */
+    if (!maxSymbolValue) maxSymbolValue = FSE_MAX_SYMBOL_VALUE;
+    if (!tableLog) tableLog = FSE_DEFAULT_TABLELOG;
+
+    /* Scan input and build symbol stats */
+    {   CHECK_V_F(maxCount, HIST_count_wksp(count, &maxSymbolValue, src, srcSize, scratchBuffer, scratchBufferSize) );
+        if (maxCount == srcSize) return 1;   /* only a single symbol in src : rle */
+        if (maxCount == 1) return 0;         /* each symbol present maximum once => not compressible */
+        if (maxCount < (srcSize >> 7)) return 0;   /* Heuristic : not compressible enough */
+    }
+
+    tableLog = FSE_optimalTableLog(tableLog, srcSize, maxSymbolValue);
+    CHECK_F( FSE_normalizeCount(norm, tableLog, count, srcSize, maxSymbolValue) );
+
+    /* Write table description header */
+    {   CHECK_V_F(nc_err, FSE_writeNCount(op, oend-op, norm, maxSymbolValue, tableLog) );
+        op += nc_err;
+    }
+
+    /* Compress */
+    CHECK_F( FSE_buildCTable_wksp(CTable, norm, maxSymbolValue, tableLog, scratchBuffer, scratchBufferSize) );
+    {   CHECK_V_F(cSize, FSE_compress_usingCTable(op, oend - op, src, srcSize, CTable) );
+        if (cSize == 0) return 0;   /* not enough space for compressed data */
+        op += cSize;
+    }
+
+    /* check compressibility */
+    if ( (size_t)(op-ostart) >= srcSize-1 ) return 0;
+
+    return op-ostart;
+}
+
+typedef struct {
+    FSE_CTable CTable_max[FSE_CTABLE_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)];
+    BYTE scratchBuffer[1 << FSE_MAX_TABLELOG];
+} fseWkspMax_t;
+
+size_t FSE_compress2 (void* dst, size_t dstCapacity, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog)
+{
+    fseWkspMax_t scratchBuffer;
+    DEBUG_STATIC_ASSERT(sizeof(scratchBuffer) >= FSE_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE));   /* compilation failures here means scratchBuffer is not large enough */
+    if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
+    return FSE_compress_wksp(dst, dstCapacity, src, srcSize, maxSymbolValue, tableLog, &scratchBuffer, sizeof(scratchBuffer));
+}
+
+size_t FSE_compress (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
+{
+    return FSE_compress2(dst, dstCapacity, src, srcSize, FSE_MAX_SYMBOL_VALUE, FSE_DEFAULT_TABLELOG);
+}
+
+
+#endif   /* FSE_COMMONDEFS_ONLY */

vendor/github.com/DataDog/zstd/fse_decompress.c

@@ -1,30 +1,48 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * FSE : Finite State Entropy decoder
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- *  You can contact the author at :
- *  - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *  - Public forum : https://groups.google.com/forum/#!forum/lz4c
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   FSE : Finite State Entropy decoder
+   Copyright (C) 2013-2015, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+    You can contact the author at :
+    - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
+    - Public forum : https://groups.google.com/forum/#!forum/lz4c
 ****************************************************************** */
 
 
 /* **************************************************************
 *  Includes
 ****************************************************************/
-#include "debug.h"      /* assert */
+#include <stdlib.h>     /* malloc, free, qsort */
+#include <string.h>     /* memcpy, memset */
 #include "bitstream.h"
 #include "compiler.h"
 #define FSE_STATIC_LINKING_ONLY
 #include "fse.h"
 #include "error_private.h"
-#include "zstd_deps.h"  /* ZSTD_memcpy */
-#include "bits.h"       /* ZSTD_highbit32 */
 
 
 /* **************************************************************
@@ -33,6 +51,11 @@
 #define FSE_isError ERR_isError
 #define FSE_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c)   /* use only *after* variable declarations */
 
+/* check and forward error code */
+#ifndef CHECK_F
+#define CHECK_F(f) { size_t const e = f; if (FSE_isError(e)) return e; }
+#endif
+
 
 /* **************************************************************
 *  Templates
@@ -56,19 +79,30 @@
 #define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y)
 #define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y)
 
-static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
+
+/* Function templates */
+FSE_DTable* FSE_createDTable (unsigned tableLog)
+{
+    if (tableLog > FSE_TABLELOG_ABSOLUTE_MAX) tableLog = FSE_TABLELOG_ABSOLUTE_MAX;
+    return (FSE_DTable*)malloc( FSE_DTABLE_SIZE_U32(tableLog) * sizeof (U32) );
+}
+
+void FSE_freeDTable (FSE_DTable* dt)
+{
+    free(dt);
+}
+
+size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog)
 {
     void* const tdPtr = dt+1;   /* because *dt is unsigned, 32-bits aligned on 32-bits */
     FSE_DECODE_TYPE* const tableDecode = (FSE_DECODE_TYPE*) (tdPtr);
-    U16* symbolNext = (U16*)workSpace;
-    BYTE* spread = (BYTE*)(symbolNext + maxSymbolValue + 1);
+    U16 symbolNext[FSE_MAX_SYMBOL_VALUE+1];
 
     U32 const maxSV1 = maxSymbolValue + 1;
     U32 const tableSize = 1 << tableLog;
     U32 highThreshold = tableSize-1;
 
     /* Sanity Checks */
-    if (FSE_BUILD_DTABLE_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(maxSymbolValue_tooLarge);
     if (maxSymbolValue > FSE_MAX_SYMBOL_VALUE) return ERROR(maxSymbolValue_tooLarge);
     if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
 
@@ -84,57 +118,13 @@ static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCo
                     symbolNext[s] = 1;
                 } else {
                     if (normalizedCounter[s] >= largeLimit) DTableH.fastMode=0;
-                    symbolNext[s] = (U16)normalizedCounter[s];
+                    symbolNext[s] = normalizedCounter[s];
         }   }   }
-        ZSTD_memcpy(dt, &DTableH, sizeof(DTableH));
+        memcpy(dt, &DTableH, sizeof(DTableH));
     }
 
     /* Spread symbols */
-    if (highThreshold == tableSize - 1) {
-        size_t const tableMask = tableSize-1;
-        size_t const step = FSE_TABLESTEP(tableSize);
-        /* First lay down the symbols in order.
-         * We use a uint64_t to lay down 8 bytes at a time. This reduces branch
-         * misses since small blocks generally have small table logs, so nearly
-         * all symbols have counts <= 8. We ensure we have 8 bytes at the end of
-         * our buffer to handle the over-write.
-         */
-        {   U64 const add = 0x0101010101010101ull;
-            size_t pos = 0;
-            U64 sv = 0;
-            U32 s;
-            for (s=0; s<maxSV1; ++s, sv += add) {
-                int i;
-                int const n = normalizedCounter[s];
-                MEM_write64(spread + pos, sv);
-                for (i = 8; i < n; i += 8) {
-                    MEM_write64(spread + pos + i, sv);
-                }
-                pos += (size_t)n;
-        }   }
-        /* Now we spread those positions across the table.
-         * The benefit of doing it in two stages is that we avoid the
-         * variable size inner loop, which caused lots of branch misses.
-         * Now we can run through all the positions without any branch misses.
-         * We unroll the loop twice, since that is what empirically worked best.
-         */
-        {
-            size_t position = 0;
-            size_t s;
-            size_t const unroll = 2;
-            assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */
-            for (s = 0; s < (size_t)tableSize; s += unroll) {
-                size_t u;
-                for (u = 0; u < unroll; ++u) {
-                    size_t const uPosition = (position + (u * step)) & tableMask;
-                    tableDecode[uPosition].symbol = spread[s + u];
-                }
-                position = (position + (unroll * step)) & tableMask;
-            }
-            assert(position == 0);
-        }
-    } else {
-        U32 const tableMask = tableSize-1;
+    {   U32 const tableMask = tableSize-1;
         U32 const step = FSE_TABLESTEP(tableSize);
         U32 s, position = 0;
         for (s=0; s<maxSV1; s++) {
@@ -152,24 +142,62 @@ static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCo
         for (u=0; u<tableSize; u++) {
             FSE_FUNCTION_TYPE const symbol = (FSE_FUNCTION_TYPE)(tableDecode[u].symbol);
             U32 const nextState = symbolNext[symbol]++;
-            tableDecode[u].nbBits = (BYTE) (tableLog - ZSTD_highbit32(nextState) );
+            tableDecode[u].nbBits = (BYTE) (tableLog - BIT_highbit32(nextState) );
             tableDecode[u].newState = (U16) ( (nextState << tableDecode[u].nbBits) - tableSize);
     }   }
 
     return 0;
 }
 
-size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
-{
-    return FSE_buildDTable_internal(dt, normalizedCounter, maxSymbolValue, tableLog, workSpace, wkspSize);
-}
-
 
 #ifndef FSE_COMMONDEFS_ONLY
 
 /*-*******************************************************
 *  Decompression (Byte symbols)
 *********************************************************/
+size_t FSE_buildDTable_rle (FSE_DTable* dt, BYTE symbolValue)
+{
+    void* ptr = dt;
+    FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr;
+    void* dPtr = dt + 1;
+    FSE_decode_t* const cell = (FSE_decode_t*)dPtr;
+
+    DTableH->tableLog = 0;
+    DTableH->fastMode = 0;
+
+    cell->newState = 0;
+    cell->symbol = symbolValue;
+    cell->nbBits = 0;
+
+    return 0;
+}
+
+
+size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits)
+{
+    void* ptr = dt;
+    FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr;
+    void* dPtr = dt + 1;
+    FSE_decode_t* const dinfo = (FSE_decode_t*)dPtr;
+    const unsigned tableSize = 1 << nbBits;
+    const unsigned tableMask = tableSize - 1;
+    const unsigned maxSV1 = tableMask+1;
+    unsigned s;
+
+    /* Sanity checks */
+    if (nbBits < 1) return ERROR(GENERIC);         /* min size */
+
+    /* Build Decoding Table */
+    DTableH->tableLog = (U16)nbBits;
+    DTableH->fastMode = 1;
+    for (s=0; s<maxSV1; s++) {
+        dinfo[s].newState = 0;
+        dinfo[s].symbol = (BYTE)s;
+        dinfo[s].nbBits = (BYTE)nbBits;
+    }
+
+    return 0;
+}
 
 FORCE_INLINE_TEMPLATE size_t FSE_decompress_usingDTable_generic(
           void* dst, size_t maxDstSize,
@@ -230,87 +258,54 @@ FORCE_INLINE_TEMPLATE size_t FSE_decompress_usingDTable_generic(
             break;
     }   }
 
-    assert(op >= ostart);
-    return (size_t)(op-ostart);
+    return op-ostart;
 }
 
-typedef struct {
-    short ncount[FSE_MAX_SYMBOL_VALUE + 1];
-} FSE_DecompressWksp;
+
+size_t FSE_decompress_usingDTable(void* dst, size_t originalSize,
+                            const void* cSrc, size_t cSrcSize,
+                            const FSE_DTable* dt)
+{
+    const void* ptr = dt;
+    const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr;
+    const U32 fastMode = DTableH->fastMode;
+
+    /* select fast mode (static) */
+    if (fastMode) return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 1);
+    return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 0);
+}
 
 
-FORCE_INLINE_TEMPLATE size_t FSE_decompress_wksp_body(
-        void* dst, size_t dstCapacity,
-        const void* cSrc, size_t cSrcSize,
-        unsigned maxLog, void* workSpace, size_t wkspSize,
-        int bmi2)
+size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, FSE_DTable* workSpace, unsigned maxLog)
 {
     const BYTE* const istart = (const BYTE*)cSrc;
     const BYTE* ip = istart;
+    short counting[FSE_MAX_SYMBOL_VALUE+1];
     unsigned tableLog;
     unsigned maxSymbolValue = FSE_MAX_SYMBOL_VALUE;
-    FSE_DecompressWksp* const wksp = (FSE_DecompressWksp*)workSpace;
-    size_t const dtablePos = sizeof(FSE_DecompressWksp) / sizeof(FSE_DTable);
-    FSE_DTable* const dtable = (FSE_DTable*)workSpace + dtablePos;
-
-    FSE_STATIC_ASSERT((FSE_MAX_SYMBOL_VALUE + 1) % 2 == 0);
-    if (wkspSize < sizeof(*wksp)) return ERROR(GENERIC);
-
-    /* correct offset to dtable depends on this property */
-    FSE_STATIC_ASSERT(sizeof(FSE_DecompressWksp) % sizeof(FSE_DTable) == 0);
 
     /* normal FSE decoding mode */
-    {   size_t const NCountLength =
-            FSE_readNCount_bmi2(wksp->ncount, &maxSymbolValue, &tableLog, istart, cSrcSize, bmi2);
-        if (FSE_isError(NCountLength)) return NCountLength;
-        if (tableLog > maxLog) return ERROR(tableLog_tooLarge);
-        assert(NCountLength <= cSrcSize);
-        ip += NCountLength;
-        cSrcSize -= NCountLength;
-    }
+    size_t const NCountLength = FSE_readNCount (counting, &maxSymbolValue, &tableLog, istart, cSrcSize);
+    if (FSE_isError(NCountLength)) return NCountLength;
+    //if (NCountLength >= cSrcSize) return ERROR(srcSize_wrong);   /* too small input size; supposed to be already checked in NCountLength, only remaining case : NCountLength==cSrcSize */
+    if (tableLog > maxLog) return ERROR(tableLog_tooLarge);
+    ip += NCountLength;
+    cSrcSize -= NCountLength;
 
-    if (FSE_DECOMPRESS_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(tableLog_tooLarge);
-    assert(sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog) <= wkspSize);
-    workSpace = (BYTE*)workSpace + sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
-    wkspSize -= sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
+    CHECK_F( FSE_buildDTable (workSpace, counting, maxSymbolValue, tableLog) );
 
-    CHECK_F( FSE_buildDTable_internal(dtable, wksp->ncount, maxSymbolValue, tableLog, workSpace, wkspSize) );
-
-    {
-        const void* ptr = dtable;
-        const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr;
-        const U32 fastMode = DTableH->fastMode;
-
-        /* select fast mode (static) */
-        if (fastMode) return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, dtable, 1);
-        return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, dtable, 0);
-    }
+    return FSE_decompress_usingDTable (dst, dstCapacity, ip, cSrcSize, workSpace);   /* always return, even if it is an error code */
 }
 
-/* Avoids the FORCE_INLINE of the _body() function. */
-static size_t FSE_decompress_wksp_body_default(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
+
+typedef FSE_DTable DTable_max_t[FSE_DTABLE_SIZE_U32(FSE_MAX_TABLELOG)];
+
+size_t FSE_decompress(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize)
 {
-    return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 0);
+    DTable_max_t dt;   /* Static analyzer seems unable to understand this table will be properly initialized later */
+    return FSE_decompress_wksp(dst, dstCapacity, cSrc, cSrcSize, dt, FSE_MAX_TABLELOG);
 }
 
-#if DYNAMIC_BMI2
-BMI2_TARGET_ATTRIBUTE static size_t FSE_decompress_wksp_body_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
-{
-    return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 1);
-}
-#endif
 
-size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2)
-{
-#if DYNAMIC_BMI2
-    if (bmi2) {
-        return FSE_decompress_wksp_body_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
-    }
-#endif
-    (void)bmi2;
-    return FSE_decompress_wksp_body_default(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
-}
 
 #endif   /* FSE_COMMONDEFS_ONLY */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/hist.c

@@ -1,17 +1,36 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * hist : Histogram functions
- * part of Finite State Entropy project
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- *  You can contact the author at :
- *  - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *  - Public forum : https://groups.google.com/forum/#!forum/lz4c
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   hist : Histogram functions
+   part of Finite State Entropy project
+   Copyright (C) 2013-present, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+    You can contact the author at :
+    - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
+    - Public forum : https://groups.google.com/forum/#!forum/lz4c
 ****************************************************************** */
 
 /* --- dependencies --- */
@@ -35,7 +54,7 @@ unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
     unsigned maxSymbolValue = *maxSymbolValuePtr;
     unsigned largestCount=0;
 
-    ZSTD_memset(count, 0, (maxSymbolValue+1) * sizeof(*count));
+    memset(count, 0, (maxSymbolValue+1) * sizeof(*count));
     if (srcSize==0) { *maxSymbolValuePtr = 0; return 0; }
 
     while (ip<end) {
@@ -61,9 +80,9 @@ typedef enum { trustInput, checkMaxSymbolValue } HIST_checkInput_e;
  * this design makes better use of OoO cpus,
  * and is noticeably faster when some values are heavily repeated.
  * But it needs some additional workspace for intermediate tables.
- * `workSpace` must be a U32 table of size >= HIST_WKSP_SIZE_U32.
+ * `workSpace` size must be a table of size >= HIST_WKSP_SIZE_U32.
  * @return : largest histogram frequency,
- *           or an error code (notably when histogram's alphabet is larger than *maxSymbolValuePtr) */
+ *           or an error code (notably when histogram would be larger than *maxSymbolValuePtr). */
 static size_t HIST_count_parallel_wksp(
                                 unsigned* count, unsigned* maxSymbolValuePtr,
                                 const void* source, size_t sourceSize,
@@ -72,21 +91,22 @@ static size_t HIST_count_parallel_wksp(
 {
     const BYTE* ip = (const BYTE*)source;
     const BYTE* const iend = ip+sourceSize;
-    size_t const countSize = (*maxSymbolValuePtr + 1) * sizeof(*count);
+    unsigned maxSymbolValue = *maxSymbolValuePtr;
     unsigned max=0;
     U32* const Counting1 = workSpace;
     U32* const Counting2 = Counting1 + 256;
     U32* const Counting3 = Counting2 + 256;
     U32* const Counting4 = Counting3 + 256;
 
+    memset(workSpace, 0, 4*256*sizeof(unsigned));
+
     /* safety checks */
-    assert(*maxSymbolValuePtr <= 255);
     if (!sourceSize) {
-        ZSTD_memset(count, 0, countSize);
+        memset(count, 0, maxSymbolValue + 1);
         *maxSymbolValuePtr = 0;
         return 0;
     }
-    ZSTD_memset(workSpace, 0, 4*256*sizeof(unsigned));
+    if (!maxSymbolValue) maxSymbolValue = 255;            /* 0 == default */
 
     /* by stripes of 16 bytes */
     {   U32 cached = MEM_read32(ip); ip += 4;
@@ -118,18 +138,21 @@ static size_t HIST_count_parallel_wksp(
     /* finish last symbols */
     while (ip<iend) Counting1[*ip++]++;
 
-    {   U32 s;
-        for (s=0; s<256; s++) {
+    if (check) {   /* verify stats will fit into destination table */
+        U32 s; for (s=255; s>maxSymbolValue; s--) {
             Counting1[s] += Counting2[s] + Counting3[s] + Counting4[s];
-            if (Counting1[s] > max) max = Counting1[s];
+            if (Counting1[s]) return ERROR(maxSymbolValue_tooSmall);
     }   }
 
-    {   unsigned maxSymbolValue = 255;
-        while (!Counting1[maxSymbolValue]) maxSymbolValue--;
-        if (check && maxSymbolValue > *maxSymbolValuePtr) return ERROR(maxSymbolValue_tooSmall);
-        *maxSymbolValuePtr = maxSymbolValue;
-        ZSTD_memmove(count, Counting1, countSize);   /* in case count & Counting1 are overlapping */
-    }
+    {   U32 s;
+        if (maxSymbolValue > 255) maxSymbolValue = 255;
+        for (s=0; s<=maxSymbolValue; s++) {
+            count[s] = Counting1[s] + Counting2[s] + Counting3[s] + Counting4[s];
+            if (count[s] > max) max = count[s];
+    }   }
+
+    while (!count[maxSymbolValue]) maxSymbolValue--;
+    *maxSymbolValuePtr = maxSymbolValue;
     return (size_t)max;
 }
 
@@ -149,6 +172,14 @@ size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
     return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, trustInput, (U32*)workSpace);
 }
 
+/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */
+size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
+                     const void* source, size_t sourceSize)
+{
+    unsigned tmpCounters[HIST_WKSP_SIZE_U32];
+    return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters, sizeof(tmpCounters));
+}
+
 /* HIST_count_wksp() :
  * Same as HIST_count(), but using an externally provided scratch buffer.
  * `workSpace` size must be table of >= HIST_WKSP_SIZE_U32 unsigned */
@@ -164,21 +195,9 @@ size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
     return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, workSpace, workSpaceSize);
 }
 
-#ifndef ZSTD_NO_UNUSED_FUNCTIONS
-/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */
-size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
-                     const void* source, size_t sourceSize)
-{
-    unsigned tmpCounters[HIST_WKSP_SIZE_U32];
-    return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters, sizeof(tmpCounters));
-}
-
 size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr,
                  const void* src, size_t srcSize)
 {
     unsigned tmpCounters[HIST_WKSP_SIZE_U32];
     return HIST_count_wksp(count, maxSymbolValuePtr, src, srcSize, tmpCounters, sizeof(tmpCounters));
 }
-#endif
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/hist.h

@@ -1,21 +1,40 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * hist : Histogram functions
- * part of Finite State Entropy project
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- *  You can contact the author at :
- *  - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *  - Public forum : https://groups.google.com/forum/#!forum/lz4c
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   hist : Histogram functions
+   part of Finite State Entropy project
+   Copyright (C) 2013-present, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+    You can contact the author at :
+    - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
+    - Public forum : https://groups.google.com/forum/#!forum/lz4c
 ****************************************************************** */
 
 /* --- dependencies --- */
-#include "zstd_deps.h"   /* size_t */
+#include <stddef.h>   /* size_t */
 
 
 /* --- simple histogram functions --- */
@@ -74,5 +93,3 @@ size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
  */
 unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
                            const void* src, size_t srcSize);
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/huf.h

@@ -1,16 +1,35 @@
-#ifndef USE_EXTERNAL_ZSTD
 /* ******************************************************************
- * huff0 huffman codec,
- * part of Finite State Entropy library
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- *
- * You can contact the author at :
- * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+   huff0 huffman codec,
+   part of Finite State Entropy library
+   Copyright (C) 2013-present, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+   You can contact the author at :
+   - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
 ****************************************************************** */
 
 #if defined (__cplusplus)
@@ -21,30 +40,105 @@ extern "C" {
 #define HUF_H_298734234
 
 /* *** Dependencies *** */
-#include "zstd_deps.h"    /* size_t */
-#include "mem.h"          /* U32 */
-#define FSE_STATIC_LINKING_ONLY
-#include "fse.h"
+#include <stddef.h>    /* size_t */
+
+
+/* *** library symbols visibility *** */
+/* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual,
+ *        HUF symbols remain "private" (internal symbols for library only).
+ *        Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */
+#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
+#  define HUF_PUBLIC_API __attribute__ ((visibility ("default")))
+#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1)   /* Visual expected */
+#  define HUF_PUBLIC_API __declspec(dllexport)
+#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
+#  define HUF_PUBLIC_API __declspec(dllimport)  /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */
+#else
+#  define HUF_PUBLIC_API
+#endif
+
+
+/* ========================== */
+/* ***  simple functions  *** */
+/* ========================== */
+
+/** HUF_compress() :
+ *  Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'.
+ * 'dst' buffer must be already allocated.
+ *  Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize).
+ * `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB.
+ * @return : size of compressed data (<= `dstCapacity`).
+ *  Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
+ *                   if HUF_isError(return), compression failed (more details using HUF_getErrorName())
+ */
+HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity,
+                             const void* src, size_t srcSize);
+
+/** HUF_decompress() :
+ *  Decompress HUF data from buffer 'cSrc', of size 'cSrcSize',
+ *  into already allocated buffer 'dst', of minimum size 'dstSize'.
+ * `originalSize` : **must** be the ***exact*** size of original (uncompressed) data.
+ *  Note : in contrast with FSE, HUF_decompress can regenerate
+ *         RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data,
+ *         because it knows size to regenerate (originalSize).
+ * @return : size of regenerated data (== originalSize),
+ *           or an error code, which can be tested using HUF_isError()
+ */
+HUF_PUBLIC_API size_t HUF_decompress(void* dst,  size_t originalSize,
+                               const void* cSrc, size_t cSrcSize);
 
 
 /* ***   Tool functions *** */
-#define HUF_BLOCKSIZE_MAX (128 * 1024)   /**< maximum input size for a single block compressed with HUF_compress */
-size_t HUF_compressBound(size_t size);   /**< maximum compressed size (worst case) */
+#define HUF_BLOCKSIZE_MAX (128 * 1024)                  /**< maximum input size for a single block compressed with HUF_compress */
+HUF_PUBLIC_API size_t HUF_compressBound(size_t size);   /**< maximum compressed size (worst case) */
 
 /* Error Management */
-unsigned    HUF_isError(size_t code);       /**< tells if a return value is an error code */
-const char* HUF_getErrorName(size_t code);  /**< provides error code string (useful for debugging) */
+HUF_PUBLIC_API unsigned    HUF_isError(size_t code);       /**< tells if a return value is an error code */
+HUF_PUBLIC_API const char* HUF_getErrorName(size_t code);  /**< provides error code string (useful for debugging) */
 
 
-#define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */)
-#define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64))
+/* ***   Advanced function   *** */
+
+/** HUF_compress2() :
+ *  Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`.
+ * `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX .
+ * `tableLog` must be `<= HUF_TABLELOG_MAX` . */
+HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity,
+                               const void* src, size_t srcSize,
+                               unsigned maxSymbolValue, unsigned tableLog);
+
+/** HUF_compress4X_wksp() :
+ *  Same as HUF_compress2(), but uses externally allocated `workSpace`.
+ * `workspace` must have minimum alignment of 4, and be at least as large as HUF_WORKSPACE_SIZE */
+#define HUF_WORKSPACE_SIZE (6 << 10)
+#define HUF_WORKSPACE_SIZE_U32 (HUF_WORKSPACE_SIZE / sizeof(U32))
+HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity,
+                                     const void* src, size_t srcSize,
+                                     unsigned maxSymbolValue, unsigned tableLog,
+                                     void* workSpace, size_t wkspSize);
+
+#endif   /* HUF_H_298734234 */
+
+/* ******************************************************************
+ *  WARNING !!
+ *  The following section contains advanced and experimental definitions
+ *  which shall never be used in the context of a dynamic library,
+ *  because they are not guaranteed to remain stable in the future.
+ *  Only consider them in association with static linking.
+ * *****************************************************************/
+#if defined(HUF_STATIC_LINKING_ONLY) && !defined(HUF_H_HUF_STATIC_LINKING_ONLY)
+#define HUF_H_HUF_STATIC_LINKING_ONLY
+
+/* *** Dependencies *** */
+#include "mem.h"   /* U32 */
+
 
 /* *** Constants *** */
-#define HUF_TABLELOG_MAX      12      /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */
+#define HUF_TABLELOG_MAX      12      /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_ABSOLUTEMAX_TABLELOG */
 #define HUF_TABLELOG_DEFAULT  11      /* default tableLog value when none specified */
 #define HUF_SYMBOLVALUE_MAX  255
 
-#define HUF_TABLELOG_ABSOLUTEMAX  12  /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
+#define HUF_TABLELOG_ABSOLUTEMAX  15  /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
 #if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX)
 #  error "HUF_TABLELOG_MAX is too large !"
 #endif
@@ -59,12 +153,12 @@ const char* HUF_getErrorName(size_t code);  /**< provides error code string (use
 #define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size))   /* Macro version, useful for static allocation */
 
 /* static allocation of HUF's Compression Table */
-/* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */
-typedef size_t HUF_CElt;   /* consider it an incomplete type */
-#define HUF_CTABLE_SIZE_ST(maxSymbolValue)   ((maxSymbolValue)+2)   /* Use tables of size_t, for proper alignment */
-#define HUF_CTABLE_SIZE(maxSymbolValue)       (HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t))
+#define HUF_CTABLE_SIZE_U32(maxSymbolValue)   ((maxSymbolValue)+1)   /* Use tables of U32, for proper alignment */
+#define HUF_CTABLE_SIZE(maxSymbolValue)       (HUF_CTABLE_SIZE_U32(maxSymbolValue) * sizeof(U32))
 #define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \
-    HUF_CElt name[HUF_CTABLE_SIZE_ST(maxSymbolValue)] /* no final ; */
+    U32 name##hb[HUF_CTABLE_SIZE_U32(maxSymbolValue)]; \
+    void* name##hv = &(name##hb); \
+    HUF_CElt* name = (HUF_CElt*)(name##hv)   /* no final ; */
 
 /* static allocation of HUF's DTable */
 typedef U32 HUF_DTable;
@@ -78,49 +172,25 @@ typedef U32 HUF_DTable;
 /* ****************************************
 *  Advanced decompression functions
 ******************************************/
+size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
+#endif
 
-/**
- * Huffman flags bitset.
- * For all flags, 0 is the default value.
- */
-typedef enum {
-    /**
-     * If compiled with DYNAMIC_BMI2: Set flag only if the CPU supports BMI2 at runtime.
-     * Otherwise: Ignored.
-     */
-    HUF_flags_bmi2 = (1 << 0),
-    /**
-     * If set: Test possible table depths to find the one that produces the smallest header + encoded size.
-     * If unset: Use heuristic to find the table depth.
-     */
-    HUF_flags_optimalDepth = (1 << 1),
-    /**
-     * If set: If the previous table can encode the input, always reuse the previous table.
-     * If unset: If the previous table can encode the input, reuse the previous table if it results in a smaller output.
-     */
-    HUF_flags_preferRepeat = (1 << 2),
-    /**
-     * If set: Sample the input and check if the sample is uncompressible, if it is then don't attempt to compress.
-     * If unset: Always histogram the entire input.
-     */
-    HUF_flags_suspectUncompressible = (1 << 3),
-    /**
-     * If set: Don't use assembly implementations
-     * If unset: Allow using assembly implementations
-     */
-    HUF_flags_disableAsm = (1 << 4),
-    /**
-     * If set: Don't use the fast decoding loop, always use the fallback decoding loop.
-     * If unset: Use the fast decoding loop when possible.
-     */
-    HUF_flags_disableFast = (1 << 5)
-} HUF_flags_e;
+size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< decodes RLE and uncompressed */
+size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< considers RLE and uncompressed as errors */
+size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< considers RLE and uncompressed as errors */
+size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
+size_t HUF_decompress4X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< single-symbol decoder */
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
+size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< double-symbols decoder */
+#endif
 
 
 /* ****************************************
  *  HUF detailed API
  * ****************************************/
-#define HUF_OPTIMAL_DEPTH_THRESHOLD ZSTD_btultra
 
 /*! HUF_compress() does the following:
  *  1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
@@ -133,38 +203,33 @@ typedef enum {
  *  For example, it's possible to compress several blocks using the same 'CTable',
  *  or to save and regenerate 'CTable' using external methods.
  */
-unsigned HUF_minTableLog(unsigned symbolCardinality);
-unsigned HUF_cardinality(const unsigned* count, unsigned maxSymbolValue);
-unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, void* workSpace,
- size_t wkspSize, HUF_CElt* table, const unsigned* count, int flags); /* table is used as scratch space for building and testing tables, not a return value */
-size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize);
-size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
-size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
-int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
+unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
+typedef struct HUF_CElt_s HUF_CElt;   /* incomplete type */
+size_t HUF_buildCTable (HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits);   /* @return : maxNbBits; CTable and count can overlap. In which case, CTable will overwrite count content */
+size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog);
+size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
 
 typedef enum {
    HUF_repeat_none,  /**< Cannot use the previous table */
    HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */
    HUF_repeat_valid  /**< Can use the previous table and it is assumed to be valid */
  } HUF_repeat;
-
 /** HUF_compress4X_repeat() :
  *  Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
  *  If it uses hufTable it does not modify hufTable or repeat.
  *  If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
- *  If preferRepeat then the old table will always be used if valid.
- *  If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */
+ *  If preferRepeat then the old table will always be used if valid. */
 size_t HUF_compress4X_repeat(void* dst, size_t dstSize,
                        const void* src, size_t srcSize,
                        unsigned maxSymbolValue, unsigned tableLog,
                        void* workSpace, size_t wkspSize,    /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
-                       HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
+                       HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2);
 
 /** HUF_buildCTable_wksp() :
  *  Same as HUF_buildCTable(), but using externally allocated scratch buffer.
  * `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.
  */
-#define HUF_CTABLE_WORKSPACE_SIZE_U32 ((4 * (HUF_SYMBOLVALUE_MAX + 1)) + 192)
+#define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1)
 #define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
 size_t HUF_buildCTable_wksp (HUF_CElt* tree,
                        const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
@@ -179,40 +244,15 @@ size_t HUF_readStats(BYTE* huffWeight, size_t hwSize,
                      U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
                      const void* src, size_t srcSize);
 
-/*! HUF_readStats_wksp() :
- * Same as HUF_readStats() but takes an external workspace which must be
- * 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE.
- * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
- */
-#define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1)
-#define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned))
-size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize,
-                          U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
-                          const void* src, size_t srcSize,
-                          void* workspace, size_t wkspSize,
-                          int flags);
-
 /** HUF_readCTable() :
  *  Loading a CTable saved with HUF_writeCTable() */
-size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights);
+size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize);
 
-/** HUF_getNbBitsFromCTable() :
+/** HUF_getNbBits() :
  *  Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX
- *  Note 1 : If symbolValue > HUF_readCTableHeader(symbolTable).maxSymbolValue, returns 0
- *  Note 2 : is not inlined, as HUF_CElt definition is private
- */
-U32 HUF_getNbBitsFromCTable(const HUF_CElt* symbolTable, U32 symbolValue);
-
-typedef struct {
-    BYTE tableLog;
-    BYTE maxSymbolValue;
-    BYTE unused[sizeof(size_t) - 2];
-} HUF_CTableHeader;
-
-/** HUF_readCTableHeader() :
- *  @returns The header from the CTable specifying the tableLog and the maxSymbolValue.
- */
-HUF_CTableHeader HUF_readCTableHeader(HUF_CElt const* ctable);
+ *  Note 1 : is not inlined, as HUF_CElt definition is private
+ *  Note 2 : const void* used, so that it can provide a statically allocated table as argument (which uses type U32) */
+U32 HUF_getNbBits(const void* symbolTable, U32 symbolValue);
 
 /*
  * HUF_decompress() does the following:
@@ -238,52 +278,81 @@ U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
  *  a required workspace size greater than that specified in the following
  *  macro.
  */
-#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9))
+#define HUF_DECOMPRESS_WORKSPACE_SIZE (2 << 10)
 #define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
 
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize);
+size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize);
+size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
+#endif
+
+size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#endif
+
 
 /* ====================== */
 /* single stream variants */
 /* ====================== */
 
-size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
+size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
+size_t HUF_compress1X_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);  /**< `workSpace` must be a table of at least HUF_WORKSPACE_SIZE_U32 unsigned */
+size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
 /** HUF_compress1X_repeat() :
  *  Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
  *  If it uses hufTable it does not modify hufTable or repeat.
  *  If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
- *  If preferRepeat then the old table will always be used if valid.
- *  If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */
+ *  If preferRepeat then the old table will always be used if valid. */
 size_t HUF_compress1X_repeat(void* dst, size_t dstSize,
                        const void* src, size_t srcSize,
                        unsigned maxSymbolValue, unsigned tableLog,
                        void* workSpace, size_t wkspSize,   /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
-                       HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
+                       HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2);
 
-size_t HUF_decompress1X_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
+size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /* single-symbol decoder */
 #ifndef HUF_FORCE_DECOMPRESS_X1
-size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);   /**< double-symbols decoder */
+size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /* double-symbol decoder */
+#endif
+
+size_t HUF_decompress1X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);
+size_t HUF_decompress1X_DCtx_wksp (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_decompress1X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
+size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< single-symbol decoder */
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
+size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);   /**< double-symbols decoder */
+#endif
+
+size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);   /**< automatic selection of sing or double symbol decoder, based on DTable */
+#ifndef HUF_FORCE_DECOMPRESS_X2
+size_t HUF_decompress1X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
+#endif
+#ifndef HUF_FORCE_DECOMPRESS_X1
+size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
 #endif
 
 /* BMI2 variants.
  * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
  */
-size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
+size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
 #ifndef HUF_FORCE_DECOMPRESS_X2
-size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
-#endif
-size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
-size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
-#ifndef HUF_FORCE_DECOMPRESS_X2
-size_t HUF_readDTableX1_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
-#endif
-#ifndef HUF_FORCE_DECOMPRESS_X1
-size_t HUF_readDTableX2_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
+size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
 #endif
+size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
+size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);
 
-#endif   /* HUF_H_298734234 */
+#endif /* HUF_STATIC_LINKING_ONLY */
 
 #if defined (__cplusplus)
 }
 #endif
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/huf_decompress_amd64.S

@@ -1,595 +0,0 @@
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
-#include "portability_macros.h"
-
-#if defined(__ELF__) && defined(__GNUC__)
-/* Stack marking
- * ref: https://wiki.gentoo.org/wiki/Hardened/GNU_stack_quickstart
- */
-.section .note.GNU-stack,"",%progbits
-
-#if defined(__aarch64__)
-/* Mark that this assembly supports BTI & PAC, because it is empty for aarch64.
- * See: https://github.com/facebook/zstd/issues/3841
- * See: https://gcc.godbolt.org/z/sqr5T4ffK
- * See: https://lore.kernel.org/linux-arm-kernel/20200429211641.9279-8-broonie@kernel.org/
- * See: https://reviews.llvm.org/D62609
- */
-.pushsection .note.gnu.property, "a"
-.p2align 3
-.long 4                 /* size of the name - "GNU\0" */
-.long 0x10              /* size of descriptor */
-.long 0x5               /* NT_GNU_PROPERTY_TYPE_0 */
-.asciz "GNU"
-.long 0xc0000000        /* pr_type - GNU_PROPERTY_AARCH64_FEATURE_1_AND */
-.long 4                 /* pr_datasz - 4 bytes */
-.long 3                 /* pr_data - GNU_PROPERTY_AARCH64_FEATURE_1_BTI | GNU_PROPERTY_AARCH64_FEATURE_1_PAC */
-.p2align 3              /* pr_padding - bring everything to 8 byte alignment */
-.popsection
-#endif
-
-#endif
-
-#if ZSTD_ENABLE_ASM_X86_64_BMI2
-
-/* Calling convention:
- *
- * %rdi contains the first argument: HUF_DecompressAsmArgs*.
- * %rbp isn't maintained (no frame pointer).
- * %rsp contains the stack pointer that grows down.
- *      No red-zone is assumed, only addresses >= %rsp are used.
- * All register contents are preserved.
- *
- * TODO: Support Windows calling convention.
- */
-
-ZSTD_HIDE_ASM_FUNCTION(HUF_decompress4X1_usingDTable_internal_fast_asm_loop)
-ZSTD_HIDE_ASM_FUNCTION(HUF_decompress4X2_usingDTable_internal_fast_asm_loop)
-ZSTD_HIDE_ASM_FUNCTION(_HUF_decompress4X2_usingDTable_internal_fast_asm_loop)
-ZSTD_HIDE_ASM_FUNCTION(_HUF_decompress4X1_usingDTable_internal_fast_asm_loop)
-.global HUF_decompress4X1_usingDTable_internal_fast_asm_loop
-.global HUF_decompress4X2_usingDTable_internal_fast_asm_loop
-.global _HUF_decompress4X1_usingDTable_internal_fast_asm_loop
-.global _HUF_decompress4X2_usingDTable_internal_fast_asm_loop
-.text
-
-/* Sets up register mappings for clarity.
- * op[], bits[], dtable & ip[0] each get their own register.
- * ip[1,2,3] & olimit alias var[].
- * %rax is a scratch register.
- */
-
-#define op0    rsi
-#define op1    rbx
-#define op2    rcx
-#define op3    rdi
-
-#define ip0    r8
-#define ip1    r9
-#define ip2    r10
-#define ip3    r11
-
-#define bits0  rbp
-#define bits1  rdx
-#define bits2  r12
-#define bits3  r13
-#define dtable r14
-#define olimit r15
-
-/* var[] aliases ip[1,2,3] & olimit
- * ip[1,2,3] are saved every iteration.
- * olimit is only used in compute_olimit.
- */
-#define var0   r15
-#define var1   r9
-#define var2   r10
-#define var3   r11
-
-/* 32-bit var registers */
-#define vard0  r15d
-#define vard1  r9d
-#define vard2  r10d
-#define vard3  r11d
-
-/* Calls X(N) for each stream 0, 1, 2, 3. */
-#define FOR_EACH_STREAM(X) \
-    X(0);                  \
-    X(1);                  \
-    X(2);                  \
-    X(3)
-
-/* Calls X(N, idx) for each stream 0, 1, 2, 3. */
-#define FOR_EACH_STREAM_WITH_INDEX(X, idx) \
-    X(0, idx);                             \
-    X(1, idx);                             \
-    X(2, idx);                             \
-    X(3, idx)
-
-/* Define both _HUF_* & HUF_* symbols because MacOS
- * C symbols are prefixed with '_' & Linux symbols aren't.
- */
-_HUF_decompress4X1_usingDTable_internal_fast_asm_loop:
-HUF_decompress4X1_usingDTable_internal_fast_asm_loop:
-    ZSTD_CET_ENDBRANCH
-    /* Save all registers - even if they are callee saved for simplicity. */
-    push %rax
-    push %rbx
-    push %rcx
-    push %rdx
-    push %rbp
-    push %rsi
-    push %rdi
-    push %r8
-    push %r9
-    push %r10
-    push %r11
-    push %r12
-    push %r13
-    push %r14
-    push %r15
-
-    /* Read HUF_DecompressAsmArgs* args from %rax */
-    movq %rdi, %rax
-    movq  0(%rax), %ip0
-    movq  8(%rax), %ip1
-    movq 16(%rax), %ip2
-    movq 24(%rax), %ip3
-    movq 32(%rax), %op0
-    movq 40(%rax), %op1
-    movq 48(%rax), %op2
-    movq 56(%rax), %op3
-    movq 64(%rax), %bits0
-    movq 72(%rax), %bits1
-    movq 80(%rax), %bits2
-    movq 88(%rax), %bits3
-    movq 96(%rax), %dtable
-    push %rax      /* argument */
-    push 104(%rax) /* ilowest */
-    push 112(%rax) /* oend */
-    push %olimit   /* olimit space */
-
-    subq $24, %rsp
-
-.L_4X1_compute_olimit:
-    /* Computes how many iterations we can do safely
-     * %r15, %rax may be clobbered
-     * rbx, rdx must be saved
-     * op3 & ip0 mustn't be clobbered
-     */
-    movq %rbx, 0(%rsp)
-    movq %rdx, 8(%rsp)
-
-    movq 32(%rsp), %rax /* rax = oend */
-    subq %op3,    %rax  /* rax = oend - op3 */
-
-    /* r15 = (oend - op3) / 5 */
-    movabsq $-3689348814741910323, %rdx
-    mulq %rdx
-    movq %rdx, %r15
-    shrq $2, %r15
-
-    movq %ip0,     %rax /* rax = ip0 */
-    movq 40(%rsp), %rdx /* rdx = ilowest */
-    subq %rdx,     %rax /* rax = ip0 - ilowest */
-    movq %rax,     %rbx /* rbx = ip0 - ilowest */
-
-    /* rdx = (ip0 - ilowest) / 7 */
-    movabsq $2635249153387078803, %rdx
-    mulq %rdx
-    subq %rdx, %rbx
-    shrq %rbx
-    addq %rbx, %rdx
-    shrq $2, %rdx
-
-    /* r15 = min(%rdx, %r15) */
-    cmpq %rdx, %r15
-    cmova %rdx, %r15
-
-    /* r15 = r15 * 5 */
-    leaq (%r15, %r15, 4), %r15
-
-    /* olimit = op3 + r15 */
-    addq %op3, %olimit
-
-    movq 8(%rsp), %rdx
-    movq 0(%rsp), %rbx
-
-    /* If (op3 + 20 > olimit) */
-    movq %op3, %rax    /* rax = op3 */
-    cmpq %rax, %olimit /* op3 == olimit */
-    je .L_4X1_exit
-
-    /* If (ip1 < ip0) go to exit */
-    cmpq %ip0, %ip1
-    jb .L_4X1_exit
-
-    /* If (ip2 < ip1) go to exit */
-    cmpq %ip1, %ip2
-    jb .L_4X1_exit
-
-    /* If (ip3 < ip2) go to exit */
-    cmpq %ip2, %ip3
-    jb .L_4X1_exit
-
-/* Reads top 11 bits from bits[n]
- * Loads dt[bits[n]] into var[n]
- */
-#define GET_NEXT_DELT(n)                \
-    movq $53, %var##n;                  \
-    shrxq %var##n, %bits##n, %var##n;   \
-    movzwl (%dtable,%var##n,2),%vard##n
-
-/* var[n] must contain the DTable entry computed with GET_NEXT_DELT
- * Moves var[n] to %rax
- * bits[n] <<= var[n] & 63
- * op[n][idx] = %rax >> 8
- * %ah is a way to access bits [8, 16) of %rax
- */
-#define DECODE_FROM_DELT(n, idx)       \
-    movq %var##n, %rax;                \
-    shlxq %var##n, %bits##n, %bits##n; \
-    movb %ah, idx(%op##n)
-
-/* Assumes GET_NEXT_DELT has been called.
- * Calls DECODE_FROM_DELT then GET_NEXT_DELT
- */
-#define DECODE_AND_GET_NEXT(n, idx) \
-    DECODE_FROM_DELT(n, idx);       \
-    GET_NEXT_DELT(n)                \
-
-/* // ctz & nbBytes is stored in bits[n]
- * // nbBits is stored in %rax
- * ctz  = CTZ[bits[n]]
- * nbBits  = ctz & 7
- * nbBytes = ctz >> 3
- * op[n]  += 5
- * ip[n]  -= nbBytes
- * // Note: x86-64 is little-endian ==> no bswap
- * bits[n] = MEM_readST(ip[n]) | 1
- * bits[n] <<= nbBits
- */
-#define RELOAD_BITS(n)             \
-    bsfq %bits##n, %bits##n;       \
-    movq %bits##n, %rax;           \
-    andq $7, %rax;                 \
-    shrq $3, %bits##n;             \
-    leaq 5(%op##n), %op##n;        \
-    subq %bits##n, %ip##n;         \
-    movq (%ip##n), %bits##n;       \
-    orq $1, %bits##n;              \
-    shlx %rax, %bits##n, %bits##n
-
-    /* Store clobbered variables on the stack */
-    movq %olimit, 24(%rsp)
-    movq %ip1, 0(%rsp)
-    movq %ip2, 8(%rsp)
-    movq %ip3, 16(%rsp)
-
-    /* Call GET_NEXT_DELT for each stream */
-    FOR_EACH_STREAM(GET_NEXT_DELT)
-
-    .p2align 6
-
-.L_4X1_loop_body:
-    /* Decode 5 symbols in each of the 4 streams (20 total)
-     * Must have called GET_NEXT_DELT for each stream
-     */
-    FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 0)
-    FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 1)
-    FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 2)
-    FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 3)
-    FOR_EACH_STREAM_WITH_INDEX(DECODE_FROM_DELT, 4)
-
-    /* Load ip[1,2,3] from stack (var[] aliases them)
-     * ip[] is needed for RELOAD_BITS
-     * Each will be stored back to the stack after RELOAD
-     */
-    movq 0(%rsp), %ip1
-    movq 8(%rsp), %ip2
-    movq 16(%rsp), %ip3
-
-    /* Reload each stream & fetch the next table entry
-     * to prepare for the next iteration
-     */
-    RELOAD_BITS(0)
-    GET_NEXT_DELT(0)
-
-    RELOAD_BITS(1)
-    movq %ip1, 0(%rsp)
-    GET_NEXT_DELT(1)
-
-    RELOAD_BITS(2)
-    movq %ip2, 8(%rsp)
-    GET_NEXT_DELT(2)
-
-    RELOAD_BITS(3)
-    movq %ip3, 16(%rsp)
-    GET_NEXT_DELT(3)
-
-    /* If op3 < olimit: continue the loop */
-    cmp %op3, 24(%rsp)
-    ja .L_4X1_loop_body
-
-    /* Reload ip[1,2,3] from stack */
-    movq 0(%rsp), %ip1
-    movq 8(%rsp), %ip2
-    movq 16(%rsp), %ip3
-
-    /* Re-compute olimit */
-    jmp .L_4X1_compute_olimit
-
-#undef GET_NEXT_DELT
-#undef DECODE_FROM_DELT
-#undef DECODE
-#undef RELOAD_BITS
-.L_4X1_exit:
-    addq $24, %rsp
-
-    /* Restore stack (oend & olimit) */
-    pop %rax /* olimit */
-    pop %rax /* oend */
-    pop %rax /* ilowest */
-    pop %rax /* arg */
-
-    /* Save ip / op / bits */
-    movq %ip0,  0(%rax)
-    movq %ip1,  8(%rax)
-    movq %ip2, 16(%rax)
-    movq %ip3, 24(%rax)
-    movq %op0, 32(%rax)
-    movq %op1, 40(%rax)
-    movq %op2, 48(%rax)
-    movq %op3, 56(%rax)
-    movq %bits0, 64(%rax)
-    movq %bits1, 72(%rax)
-    movq %bits2, 80(%rax)
-    movq %bits3, 88(%rax)
-
-    /* Restore registers */
-    pop %r15
-    pop %r14
-    pop %r13
-    pop %r12
-    pop %r11
-    pop %r10
-    pop %r9
-    pop %r8
-    pop %rdi
-    pop %rsi
-    pop %rbp
-    pop %rdx
-    pop %rcx
-    pop %rbx
-    pop %rax
-    ret
-
-_HUF_decompress4X2_usingDTable_internal_fast_asm_loop:
-HUF_decompress4X2_usingDTable_internal_fast_asm_loop:
-    ZSTD_CET_ENDBRANCH
-    /* Save all registers - even if they are callee saved for simplicity. */
-    push %rax
-    push %rbx
-    push %rcx
-    push %rdx
-    push %rbp
-    push %rsi
-    push %rdi
-    push %r8
-    push %r9
-    push %r10
-    push %r11
-    push %r12
-    push %r13
-    push %r14
-    push %r15
-
-    movq %rdi, %rax
-    movq  0(%rax), %ip0
-    movq  8(%rax), %ip1
-    movq 16(%rax), %ip2
-    movq 24(%rax), %ip3
-    movq 32(%rax), %op0
-    movq 40(%rax), %op1
-    movq 48(%rax), %op2
-    movq 56(%rax), %op3
-    movq 64(%rax), %bits0
-    movq 72(%rax), %bits1
-    movq 80(%rax), %bits2
-    movq 88(%rax), %bits3
-    movq 96(%rax), %dtable
-    push %rax      /* argument */
-    push %rax      /* olimit */
-    push 104(%rax) /* ilowest */
-
-    movq 112(%rax), %rax
-    push %rax /* oend3 */
-
-    movq %op3, %rax
-    push %rax /* oend2 */
-
-    movq %op2, %rax
-    push %rax /* oend1 */
-
-    movq %op1, %rax
-    push %rax /* oend0 */
-
-    /* Scratch space */
-    subq $8, %rsp
-
-.L_4X2_compute_olimit:
-    /* Computes how many iterations we can do safely
-     * %r15, %rax may be clobbered
-     * rdx must be saved
-     * op[1,2,3,4] & ip0 mustn't be clobbered
-     */
-    movq %rdx, 0(%rsp)
-
-    /* We can consume up to 7 input bytes each iteration. */
-    movq %ip0,     %rax  /* rax = ip0 */
-    movq 40(%rsp), %rdx  /* rdx = ilowest */
-    subq %rdx,     %rax  /* rax = ip0 - ilowest */
-    movq %rax,    %r15   /* r15 = ip0 - ilowest */
-
-    /* rdx = rax / 7 */
-    movabsq $2635249153387078803, %rdx
-    mulq %rdx
-    subq %rdx, %r15
-    shrq %r15
-    addq %r15, %rdx
-    shrq $2, %rdx
-
-    /* r15 = (ip0 - ilowest) / 7 */
-    movq %rdx, %r15
-
-    /* r15 = min(r15, min(oend0 - op0, oend1 - op1, oend2 - op2, oend3 - op3) / 10) */
-    movq 8(%rsp),  %rax /* rax = oend0 */
-    subq %op0,     %rax /* rax = oend0 - op0 */
-    movq 16(%rsp), %rdx /* rdx = oend1 */
-    subq %op1,     %rdx /* rdx = oend1 - op1 */
-
-    cmpq  %rax,    %rdx
-    cmova %rax,    %rdx /* rdx = min(%rdx, %rax) */
-
-    movq 24(%rsp), %rax /* rax = oend2 */
-    subq %op2,     %rax /* rax = oend2 - op2 */
-
-    cmpq  %rax,    %rdx
-    cmova %rax,    %rdx /* rdx = min(%rdx, %rax) */
-
-    movq 32(%rsp), %rax /* rax = oend3 */
-    subq %op3,     %rax /* rax = oend3 - op3 */
-
-    cmpq  %rax,    %rdx
-    cmova %rax,    %rdx /* rdx = min(%rdx, %rax) */
-
-    movabsq $-3689348814741910323, %rax
-    mulq %rdx
-    shrq $3,       %rdx /* rdx = rdx / 10 */
-
-    /* r15 = min(%rdx, %r15) */
-    cmpq  %rdx, %r15
-    cmova %rdx, %r15
-
-    /* olimit = op3 + 5 * r15 */
-    movq %r15, %rax
-    leaq (%op3, %rax, 4), %olimit
-    addq %rax, %olimit
-
-    movq 0(%rsp), %rdx
-
-    /* If (op3 + 10 > olimit) */
-    movq %op3, %rax    /* rax = op3 */
-    cmpq %rax, %olimit /* op3 == olimit */
-    je .L_4X2_exit
-
-    /* If (ip1 < ip0) go to exit */
-    cmpq %ip0, %ip1
-    jb .L_4X2_exit
-
-    /* If (ip2 < ip1) go to exit */
-    cmpq %ip1, %ip2
-    jb .L_4X2_exit
-
-    /* If (ip3 < ip2) go to exit */
-    cmpq %ip2, %ip3
-    jb .L_4X2_exit
-
-#define DECODE(n, idx)              \
-    movq %bits##n, %rax;            \
-    shrq $53, %rax;                 \
-    movzwl 0(%dtable,%rax,4),%r8d;  \
-    movzbl 2(%dtable,%rax,4),%r15d; \
-    movzbl 3(%dtable,%rax,4),%eax;  \
-    movw %r8w, (%op##n);            \
-    shlxq %r15, %bits##n, %bits##n; \
-    addq %rax, %op##n
-
-#define RELOAD_BITS(n)              \
-    bsfq %bits##n, %bits##n;        \
-    movq %bits##n, %rax;            \
-    shrq $3, %bits##n;              \
-    andq $7, %rax;                  \
-    subq %bits##n, %ip##n;          \
-    movq (%ip##n), %bits##n;        \
-    orq $1, %bits##n;               \
-    shlxq %rax, %bits##n, %bits##n
-
-
-    movq %olimit, 48(%rsp)
-
-    .p2align 6
-
-.L_4X2_loop_body:
-    /* We clobber r8, so store it on the stack */
-    movq %r8, 0(%rsp)
-
-    /* Decode 5 symbols from each of the 4 streams (20 symbols total). */
-    FOR_EACH_STREAM_WITH_INDEX(DECODE, 0)
-    FOR_EACH_STREAM_WITH_INDEX(DECODE, 1)
-    FOR_EACH_STREAM_WITH_INDEX(DECODE, 2)
-    FOR_EACH_STREAM_WITH_INDEX(DECODE, 3)
-    FOR_EACH_STREAM_WITH_INDEX(DECODE, 4)
-
-    /* Reload r8 */
-    movq 0(%rsp), %r8
-
-    FOR_EACH_STREAM(RELOAD_BITS)
-
-    cmp %op3, 48(%rsp)
-    ja .L_4X2_loop_body
-    jmp .L_4X2_compute_olimit
-
-#undef DECODE
-#undef RELOAD_BITS
-.L_4X2_exit:
-    addq $8, %rsp
-    /* Restore stack (oend & olimit) */
-    pop %rax /* oend0 */
-    pop %rax /* oend1 */
-    pop %rax /* oend2 */
-    pop %rax /* oend3 */
-    pop %rax /* ilowest */
-    pop %rax /* olimit */
-    pop %rax /* arg */
-
-    /* Save ip / op / bits */
-    movq %ip0,  0(%rax)
-    movq %ip1,  8(%rax)
-    movq %ip2, 16(%rax)
-    movq %ip3, 24(%rax)
-    movq %op0, 32(%rax)
-    movq %op1, 40(%rax)
-    movq %op2, 48(%rax)
-    movq %op3, 56(%rax)
-    movq %bits0, 64(%rax)
-    movq %bits1, 72(%rax)
-    movq %bits2, 80(%rax)
-    movq %bits3, 88(%rax)
-
-    /* Restore registers */
-    pop %r15
-    pop %r14
-    pop %r13
-    pop %r12
-    pop %r11
-    pop %r10
-    pop %r9
-    pop %r8
-    pop %rdi
-    pop %rsi
-    pop %rbp
-    pop %rdx
-    pop %rcx
-    pop %rbx
-    pop %rax
-    ret
-
-#endif

vendor/github.com/DataDog/zstd/mem.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -19,10 +18,8 @@ extern "C" {
 /*-****************************************
 *  Dependencies
 ******************************************/
-#include <stddef.h>  /* size_t, ptrdiff_t */
-#include "compiler.h"  /* __has_builtin */
-#include "debug.h"  /* DEBUG_STATIC_ASSERT */
-#include "zstd_deps.h"  /* ZSTD_memcpy */
+#include <stddef.h>     /* size_t, ptrdiff_t */
+#include <string.h>     /* memcpy */
 
 
 /*-****************************************
@@ -32,19 +29,104 @@ extern "C" {
 #   include <stdlib.h>  /* _byteswap_ulong */
 #   include <intrin.h>  /* _byteswap_* */
 #endif
+#if defined(__GNUC__)
+#  define MEM_STATIC static __inline __attribute__((unused))
+#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
+#  define MEM_STATIC static inline
+#elif defined(_MSC_VER)
+#  define MEM_STATIC static __inline
+#else
+#  define MEM_STATIC static  /* this version may generate warnings for unused static functions; disable the relevant warning */
+#endif
+
+#ifndef __has_builtin
+#  define __has_builtin(x) 0  /* compat. with non-clang compilers */
+#endif
+
+/* code only tested on 32 and 64 bits systems */
+#define MEM_STATIC_ASSERT(c)   { enum { MEM_static_assert = 1/(int)(!!(c)) }; }
+MEM_STATIC void MEM_check(void) { MEM_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); }
+
+/* detects whether we are being compiled under msan */
+#if defined (__has_feature)
+#  if __has_feature(memory_sanitizer)
+#    define MEMORY_SANITIZER 1
+#  endif
+#endif
+
+#if defined (MEMORY_SANITIZER)
+/* Not all platforms that support msan provide sanitizers/msan_interface.h.
+ * We therefore declare the functions we need ourselves, rather than trying to
+ * include the header file... */
+
+#include <stdint.h> /* intptr_t */
+
+/* Make memory region fully initialized (without changing its contents). */
+void __msan_unpoison(const volatile void *a, size_t size);
+
+/* Make memory region fully uninitialized (without changing its contents).
+   This is a legacy interface that does not update origin information. Use
+   __msan_allocated_memory() instead. */
+void __msan_poison(const volatile void *a, size_t size);
+
+/* Returns the offset of the first (at least partially) poisoned byte in the
+   memory range, or -1 if the whole range is good. */
+intptr_t __msan_test_shadow(const volatile void *x, size_t size);
+#endif
+
+/* detects whether we are being compiled under asan */
+#if defined (__has_feature)
+#  if __has_feature(address_sanitizer)
+#    define ADDRESS_SANITIZER 1
+#  endif
+#elif defined(__SANITIZE_ADDRESS__)
+#  define ADDRESS_SANITIZER 1
+#endif
+
+#if defined (ADDRESS_SANITIZER)
+/* Not all platforms that support asan provide sanitizers/asan_interface.h.
+ * We therefore declare the functions we need ourselves, rather than trying to
+ * include the header file... */
+
+/**
+ * Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable.
+ *
+ * This memory must be previously allocated by your program. Instrumented
+ * code is forbidden from accessing addresses in this region until it is
+ * unpoisoned. This function is not guaranteed to poison the entire region -
+ * it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan
+ * alignment restrictions.
+ *
+ * \note This function is not thread-safe because no two threads can poison or
+ * unpoison memory in the same memory region simultaneously.
+ *
+ * \param addr Start of memory region.
+ * \param size Size of memory region. */
+void __asan_poison_memory_region(void const volatile *addr, size_t size);
+
+/**
+ * Marks a memory region (<c>[addr, addr+size)</c>) as addressable.
+ *
+ * This memory must be previously allocated by your program. Accessing
+ * addresses in this region is allowed until this region is poisoned again.
+ * This function could unpoison a super-region of <c>[addr, addr+size)</c> due
+ * to ASan alignment restrictions.
+ *
+ * \note This function is not thread-safe because no two threads can
+ * poison or unpoison memory in the same memory region simultaneously.
+ *
+ * \param addr Start of memory region.
+ * \param size Size of memory region. */
+void __asan_unpoison_memory_region(void const volatile *addr, size_t size);
+#endif
+
 
 /*-**************************************************************
 *  Basic Types
 *****************************************************************/
 #if  !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
-#  if defined(_AIX)
-#    include <inttypes.h>
-#  else
-#    include <stdint.h> /* intptr_t */
-#  endif
+# include <stdint.h>
   typedef   uint8_t BYTE;
-  typedef   uint8_t U8;
-  typedef    int8_t S8;
   typedef  uint16_t U16;
   typedef   int16_t S16;
   typedef  uint32_t U32;
@@ -57,8 +139,6 @@ extern "C" {
 #  error "this implementation requires char to be exactly 8-bit type"
 #endif
   typedef unsigned char      BYTE;
-  typedef unsigned char      U8;
-  typedef   signed char      S8;
 #if USHRT_MAX != 65535
 #  error "this implementation requires short to be exactly 16-bit type"
 #endif
@@ -77,63 +157,25 @@ extern "C" {
 
 
 /*-**************************************************************
-*  Memory I/O API
+*  Memory I/O
 *****************************************************************/
-/*=== Static platform detection ===*/
-MEM_STATIC unsigned MEM_32bits(void);
-MEM_STATIC unsigned MEM_64bits(void);
-MEM_STATIC unsigned MEM_isLittleEndian(void);
-
-/*=== Native unaligned read/write ===*/
-MEM_STATIC U16 MEM_read16(const void* memPtr);
-MEM_STATIC U32 MEM_read32(const void* memPtr);
-MEM_STATIC U64 MEM_read64(const void* memPtr);
-MEM_STATIC size_t MEM_readST(const void* memPtr);
-
-MEM_STATIC void MEM_write16(void* memPtr, U16 value);
-MEM_STATIC void MEM_write32(void* memPtr, U32 value);
-MEM_STATIC void MEM_write64(void* memPtr, U64 value);
-
-/*=== Little endian unaligned read/write ===*/
-MEM_STATIC U16 MEM_readLE16(const void* memPtr);
-MEM_STATIC U32 MEM_readLE24(const void* memPtr);
-MEM_STATIC U32 MEM_readLE32(const void* memPtr);
-MEM_STATIC U64 MEM_readLE64(const void* memPtr);
-MEM_STATIC size_t MEM_readLEST(const void* memPtr);
-
-MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val);
-MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val);
-MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32);
-MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64);
-MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val);
-
-/*=== Big endian unaligned read/write ===*/
-MEM_STATIC U32 MEM_readBE32(const void* memPtr);
-MEM_STATIC U64 MEM_readBE64(const void* memPtr);
-MEM_STATIC size_t MEM_readBEST(const void* memPtr);
-
-MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32);
-MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64);
-MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val);
-
-/*=== Byteswap ===*/
-MEM_STATIC U32 MEM_swap32(U32 in);
-MEM_STATIC U64 MEM_swap64(U64 in);
-MEM_STATIC size_t MEM_swapST(size_t in);
-
-
-/*-**************************************************************
-*  Memory I/O Implementation
-*****************************************************************/
-/* MEM_FORCE_MEMORY_ACCESS : For accessing unaligned memory:
- * Method 0 : always use `memcpy()`. Safe and portable.
- * Method 1 : Use compiler extension to set unaligned access.
+/* MEM_FORCE_MEMORY_ACCESS :
+ * By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable.
+ * Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal.
+ * The below switch allow to select different access method for improved performance.
+ * Method 0 (default) : use `memcpy()`. Safe and portable.
+ * Method 1 : `__packed` statement. It depends on compiler extension (i.e., not portable).
+ *            This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`.
  * Method 2 : direct access. This method is portable but violate C standard.
  *            It can generate buggy code on targets depending on alignment.
- * Default  : method 1 if supported, else method 0
+ *            In some circumstances, it's the only known way to get the most performance (i.e. GCC + ARMv6)
+ * See http://fastcompression.blogspot.fr/2015/08/accessing-unaligned-memory.html for details.
+ * Prefer these methods in priority order (0 > 1 > 2)
  */
 #ifndef MEM_FORCE_MEMORY_ACCESS   /* can be defined externally, on command line for example */
-#  ifdef __GNUC__
+#  if defined(__GNUC__) && ( defined(__ARM_ARCH_6__) || defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__) || defined(__ARM_ARCH_6T2__) )
+#    define MEM_FORCE_MEMORY_ACCESS 2
+#  elif defined(__INTEL_COMPILER) || defined(__GNUC__) || defined(__ICCARM__)
 #    define MEM_FORCE_MEMORY_ACCESS 1
 #  endif
 #endif
@@ -143,22 +185,8 @@ MEM_STATIC unsigned MEM_64bits(void) { return sizeof(size_t)==8; }
 
 MEM_STATIC unsigned MEM_isLittleEndian(void)
 {
-#if defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
-    return 1;
-#elif defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)
-    return 0;
-#elif defined(__clang__) && __LITTLE_ENDIAN__
-    return 1;
-#elif defined(__clang__) && __BIG_ENDIAN__
-    return 0;
-#elif defined(_MSC_VER) && (_M_AMD64 || _M_IX86)
-    return 1;
-#elif defined(__DMC__) && defined(_M_IX86)
-    return 1;
-#else
     const union { U32 u; BYTE c[4]; } one = { 1 };   /* don't use static : performance detrimental  */
     return one.c[0];
-#endif
 }
 
 #if defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==2)
@@ -176,19 +204,30 @@ MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(U64*)memPtr = value; }
 
 #elif defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==1)
 
-typedef __attribute__((aligned(1))) U16 unalign16;
-typedef __attribute__((aligned(1))) U32 unalign32;
-typedef __attribute__((aligned(1))) U64 unalign64;
-typedef __attribute__((aligned(1))) size_t unalignArch;
+/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
+/* currently only defined for gcc and icc */
+#if defined(_MSC_VER) || (defined(__INTEL_COMPILER) && defined(WIN32))
+    __pragma( pack(push, 1) )
+    typedef struct { U16 v; } unalign16;
+    typedef struct { U32 v; } unalign32;
+    typedef struct { U64 v; } unalign64;
+    typedef struct { size_t v; } unalignArch;
+    __pragma( pack(pop) )
+#else
+    typedef struct { U16 v; } __attribute__((packed)) unalign16;
+    typedef struct { U32 v; } __attribute__((packed)) unalign32;
+    typedef struct { U64 v; } __attribute__((packed)) unalign64;
+    typedef struct { size_t v; } __attribute__((packed)) unalignArch;
+#endif
 
-MEM_STATIC U16 MEM_read16(const void* ptr) { return *(const unalign16*)ptr; }
-MEM_STATIC U32 MEM_read32(const void* ptr) { return *(const unalign32*)ptr; }
-MEM_STATIC U64 MEM_read64(const void* ptr) { return *(const unalign64*)ptr; }
-MEM_STATIC size_t MEM_readST(const void* ptr) { return *(const unalignArch*)ptr; }
+MEM_STATIC U16 MEM_read16(const void* ptr) { return ((const unalign16*)ptr)->v; }
+MEM_STATIC U32 MEM_read32(const void* ptr) { return ((const unalign32*)ptr)->v; }
+MEM_STATIC U64 MEM_read64(const void* ptr) { return ((const unalign64*)ptr)->v; }
+MEM_STATIC size_t MEM_readST(const void* ptr) { return ((const unalignArch*)ptr)->v; }
 
-MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(unalign16*)memPtr = value; }
-MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(unalign32*)memPtr = value; }
-MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(unalign64*)memPtr = value; }
+MEM_STATIC void MEM_write16(void* memPtr, U16 value) { ((unalign16*)memPtr)->v = value; }
+MEM_STATIC void MEM_write32(void* memPtr, U32 value) { ((unalign32*)memPtr)->v = value; }
+MEM_STATIC void MEM_write64(void* memPtr, U64 value) { ((unalign64*)memPtr)->v = value; }
 
 #else
 
@@ -197,49 +236,41 @@ MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(unalign64*)memPtr = val
 
 MEM_STATIC U16 MEM_read16(const void* memPtr)
 {
-    U16 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
+    U16 val; memcpy(&val, memPtr, sizeof(val)); return val;
 }
 
 MEM_STATIC U32 MEM_read32(const void* memPtr)
 {
-    U32 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
+    U32 val; memcpy(&val, memPtr, sizeof(val)); return val;
 }
 
 MEM_STATIC U64 MEM_read64(const void* memPtr)
 {
-    U64 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
+    U64 val; memcpy(&val, memPtr, sizeof(val)); return val;
 }
 
 MEM_STATIC size_t MEM_readST(const void* memPtr)
 {
-    size_t val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
+    size_t val; memcpy(&val, memPtr, sizeof(val)); return val;
 }
 
 MEM_STATIC void MEM_write16(void* memPtr, U16 value)
 {
-    ZSTD_memcpy(memPtr, &value, sizeof(value));
+    memcpy(memPtr, &value, sizeof(value));
 }
 
 MEM_STATIC void MEM_write32(void* memPtr, U32 value)
 {
-    ZSTD_memcpy(memPtr, &value, sizeof(value));
+    memcpy(memPtr, &value, sizeof(value));
 }
 
 MEM_STATIC void MEM_write64(void* memPtr, U64 value)
 {
-    ZSTD_memcpy(memPtr, &value, sizeof(value));
+    memcpy(memPtr, &value, sizeof(value));
 }
 
 #endif /* MEM_FORCE_MEMORY_ACCESS */
 
-MEM_STATIC U32 MEM_swap32_fallback(U32 in)
-{
-    return  ((in << 24) & 0xff000000 ) |
-            ((in <<  8) & 0x00ff0000 ) |
-            ((in >>  8) & 0x0000ff00 ) |
-            ((in >> 24) & 0x000000ff );
-}
-
 MEM_STATIC U32 MEM_swap32(U32 in)
 {
 #if defined(_MSC_VER)     /* Visual Studio */
@@ -248,22 +279,13 @@ MEM_STATIC U32 MEM_swap32(U32 in)
   || (defined(__clang__) && __has_builtin(__builtin_bswap32))
     return __builtin_bswap32(in);
 #else
-    return MEM_swap32_fallback(in);
+    return  ((in << 24) & 0xff000000 ) |
+            ((in <<  8) & 0x00ff0000 ) |
+            ((in >>  8) & 0x0000ff00 ) |
+            ((in >> 24) & 0x000000ff );
 #endif
 }
 
-MEM_STATIC U64 MEM_swap64_fallback(U64 in)
-{
-     return  ((in << 56) & 0xff00000000000000ULL) |
-            ((in << 40) & 0x00ff000000000000ULL) |
-            ((in << 24) & 0x0000ff0000000000ULL) |
-            ((in << 8)  & 0x000000ff00000000ULL) |
-            ((in >> 8)  & 0x00000000ff000000ULL) |
-            ((in >> 24) & 0x0000000000ff0000ULL) |
-            ((in >> 40) & 0x000000000000ff00ULL) |
-            ((in >> 56) & 0x00000000000000ffULL);
-}
-
 MEM_STATIC U64 MEM_swap64(U64 in)
 {
 #if defined(_MSC_VER)     /* Visual Studio */
@@ -272,7 +294,14 @@ MEM_STATIC U64 MEM_swap64(U64 in)
   || (defined(__clang__) && __has_builtin(__builtin_bswap64))
     return __builtin_bswap64(in);
 #else
-    return MEM_swap64_fallback(in);
+    return  ((in << 56) & 0xff00000000000000ULL) |
+            ((in << 40) & 0x00ff000000000000ULL) |
+            ((in << 24) & 0x0000ff0000000000ULL) |
+            ((in << 8)  & 0x000000ff00000000ULL) |
+            ((in >> 8)  & 0x00000000ff000000ULL) |
+            ((in >> 24) & 0x0000000000ff0000ULL) |
+            ((in >> 40) & 0x000000000000ff00ULL) |
+            ((in >> 56) & 0x00000000000000ffULL);
 #endif
 }
 
@@ -309,7 +338,7 @@ MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val)
 
 MEM_STATIC U32 MEM_readLE24(const void* memPtr)
 {
-    return (U32)MEM_readLE16(memPtr) + ((U32)(((const BYTE*)memPtr)[2]) << 16);
+    return MEM_readLE16(memPtr) + (((const BYTE*)memPtr)[2] << 16);
 }
 
 MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val)
@@ -416,14 +445,9 @@ MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val)
         MEM_writeBE64(memPtr, (U64)val);
 }
 
-/* code only tested on 32 and 64 bits systems */
-MEM_STATIC void MEM_check(void) { DEBUG_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); }
-
 
 #if defined (__cplusplus)
 }
 #endif
 
 #endif /* MEM_H_MODULE */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/pool.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -11,9 +10,9 @@
 
 
 /* ======   Dependencies   ======= */
-#include "allocations.h"  /* ZSTD_customCalloc, ZSTD_customFree */
-#include "zstd_deps.h" /* size_t */
+#include <stddef.h>    /* size_t */
 #include "debug.h"     /* assert */
+#include "zstd_internal.h"  /* ZSTD_malloc, ZSTD_free */
 #include "pool.h"
 
 /* ======   Compiler specifics   ====== */
@@ -87,7 +86,7 @@ static void* POOL_thread(void* opaque) {
         {   POOL_job const job = ctx->queue[ctx->queueHead];
             ctx->queueHead = (ctx->queueHead + 1) % ctx->queueSize;
             ctx->numThreadsBusy++;
-            ctx->queueEmpty = (ctx->queueHead == ctx->queueTail);
+            ctx->queueEmpty = ctx->queueHead == ctx->queueTail;
             /* Unlock the mutex, signal a pusher, and run the job */
             ZSTD_pthread_cond_signal(&ctx->queuePushCond);
             ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
@@ -97,37 +96,33 @@ static void* POOL_thread(void* opaque) {
             /* If the intended queue size was 0, signal after finishing job */
             ZSTD_pthread_mutex_lock(&ctx->queueMutex);
             ctx->numThreadsBusy--;
-            ZSTD_pthread_cond_signal(&ctx->queuePushCond);
+            if (ctx->queueSize == 1) {
+                ZSTD_pthread_cond_signal(&ctx->queuePushCond);
+            }
             ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
         }
     }  /* for (;;) */
     assert(0);  /* Unreachable */
 }
 
-/* ZSTD_createThreadPool() : public access point */
-POOL_ctx* ZSTD_createThreadPool(size_t numThreads) {
-    return POOL_create (numThreads, 0);
-}
-
 POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
     return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
 }
 
 POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
-                               ZSTD_customMem customMem)
-{
+                               ZSTD_customMem customMem) {
     POOL_ctx* ctx;
     /* Check parameters */
     if (!numThreads) { return NULL; }
     /* Allocate the context and zero initialize */
-    ctx = (POOL_ctx*)ZSTD_customCalloc(sizeof(POOL_ctx), customMem);
+    ctx = (POOL_ctx*)ZSTD_calloc(sizeof(POOL_ctx), customMem);
     if (!ctx) { return NULL; }
     /* Initialize the job queue.
      * It needs one extra space since one space is wasted to differentiate
      * empty and full queues.
      */
     ctx->queueSize = queueSize + 1;
-    ctx->queue = (POOL_job*)ZSTD_customCalloc(ctx->queueSize * sizeof(POOL_job), customMem);
+    ctx->queue = (POOL_job*)ZSTD_malloc(ctx->queueSize * sizeof(POOL_job), customMem);
     ctx->queueHead = 0;
     ctx->queueTail = 0;
     ctx->numThreadsBusy = 0;
@@ -141,7 +136,7 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
     }
     ctx->shutdown = 0;
     /* Allocate space for the thread handles */
-    ctx->threads = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
+    ctx->threads = (ZSTD_pthread_t*)ZSTD_malloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
     ctx->threadCapacity = 0;
     ctx->customMem = customMem;
     /* Check for errors */
@@ -174,7 +169,7 @@ static void POOL_join(POOL_ctx* ctx) {
     /* Join all of the threads */
     {   size_t i;
         for (i = 0; i < ctx->threadCapacity; ++i) {
-            ZSTD_pthread_join(ctx->threads[i]);  /* note : could fail */
+            ZSTD_pthread_join(ctx->threads[i], NULL);  /* note : could fail */
     }   }
 }
 
@@ -184,27 +179,14 @@ void POOL_free(POOL_ctx *ctx) {
     ZSTD_pthread_mutex_destroy(&ctx->queueMutex);
     ZSTD_pthread_cond_destroy(&ctx->queuePushCond);
     ZSTD_pthread_cond_destroy(&ctx->queuePopCond);
-    ZSTD_customFree(ctx->queue, ctx->customMem);
-    ZSTD_customFree(ctx->threads, ctx->customMem);
-    ZSTD_customFree(ctx, ctx->customMem);
+    ZSTD_free(ctx->queue, ctx->customMem);
+    ZSTD_free(ctx->threads, ctx->customMem);
+    ZSTD_free(ctx, ctx->customMem);
 }
 
-/*! POOL_joinJobs() :
- *  Waits for all queued jobs to finish executing.
- */
-void POOL_joinJobs(POOL_ctx* ctx) {
-    ZSTD_pthread_mutex_lock(&ctx->queueMutex);
-    while(!ctx->queueEmpty || ctx->numThreadsBusy > 0) {
-        ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
-    }
-    ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
-}
 
-void ZSTD_freeThreadPool (ZSTD_threadPool* pool) {
-  POOL_free (pool);
-}
 
-size_t POOL_sizeof(const POOL_ctx* ctx) {
+size_t POOL_sizeof(POOL_ctx *ctx) {
     if (ctx==NULL) return 0;  /* supports sizeof NULL */
     return sizeof(*ctx)
         + ctx->queueSize * sizeof(POOL_job)
@@ -221,11 +203,11 @@ static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads)
         return 0;
     }
     /* numThreads > threadCapacity */
-    {   ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
+    {   ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_malloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
         if (!threadPool) return 1;
         /* replace existing thread pool */
-        ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(ZSTD_pthread_t));
-        ZSTD_customFree(ctx->threads, ctx->customMem);
+        memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool));
+        ZSTD_free(ctx->threads, ctx->customMem);
         ctx->threads = threadPool;
         /* Initialize additional threads */
         {   size_t threadId;
@@ -269,12 +251,9 @@ static int isQueueFull(POOL_ctx const* ctx) {
 }
 
 
-static void
-POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
+static void POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
 {
-    POOL_job job;
-    job.function = function;
-    job.opaque = opaque;
+    POOL_job const job = {function, opaque};
     assert(ctx != NULL);
     if (ctx->shutdown) return;
 
@@ -322,28 +301,21 @@ int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque)
 struct POOL_ctx_s {
     int dummy;
 };
-static POOL_ctx g_poolCtx;
+static POOL_ctx g_ctx;
 
 POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
     return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
 }
 
-POOL_ctx*
-POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem)
-{
+POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem) {
     (void)numThreads;
     (void)queueSize;
     (void)customMem;
-    return &g_poolCtx;
+    return &g_ctx;
 }
 
 void POOL_free(POOL_ctx* ctx) {
-    assert(!ctx || ctx == &g_poolCtx);
-    (void)ctx;
-}
-
-void POOL_joinJobs(POOL_ctx* ctx){
-    assert(!ctx || ctx == &g_poolCtx);
+    assert(!ctx || ctx == &g_ctx);
     (void)ctx;
 }
 
@@ -363,12 +335,10 @@ int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) {
     return 1;
 }
 
-size_t POOL_sizeof(const POOL_ctx* ctx) {
+size_t POOL_sizeof(POOL_ctx* ctx) {
     if (ctx==NULL) return 0;  /* supports sizeof NULL */
-    assert(ctx == &g_poolCtx);
+    assert(ctx == &g_ctx);
     return sizeof(*ctx);
 }
 
 #endif  /* ZSTD_MULTITHREAD */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/pool.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -17,7 +16,7 @@ extern "C" {
 #endif
 
 
-#include "zstd_deps.h"
+#include <stddef.h>   /* size_t */
 #define ZSTD_STATIC_LINKING_ONLY   /* ZSTD_customMem */
 #include "zstd.h"
 
@@ -39,16 +38,10 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
  */
 void POOL_free(POOL_ctx* ctx);
 
-
-/*! POOL_joinJobs() :
- *  Waits for all queued jobs to finish executing.
- */
-void POOL_joinJobs(POOL_ctx* ctx);
-
 /*! POOL_resize() :
  *  Expands or shrinks pool's number of threads.
  *  This is more efficient than releasing + creating a new context,
- *  since it tries to preserve and reuse existing threads.
+ *  since it tries to preserve and re-use existing threads.
  * `numThreads` must be at least 1.
  * @return : 0 when resize was successful,
  *           !0 (typically 1) if there is an error.
@@ -60,7 +53,7 @@ int POOL_resize(POOL_ctx* ctx, size_t numThreads);
  * @return threadpool memory usage
  *  note : compatible with NULL (returns 0 in this case)
  */
-size_t POOL_sizeof(const POOL_ctx* ctx);
+size_t POOL_sizeof(POOL_ctx* ctx);
 
 /*! POOL_function :
  *  The function type that can be added to a thread pool.
@@ -77,7 +70,7 @@ void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque);
 
 
 /*! POOL_tryAdd() :
- *  Add the job `function(opaque)` to thread pool _if_ a queue slot is available.
+ *  Add the job `function(opaque)` to thread pool _if_ a worker is available.
  *  Returns immediately even if not (does not block).
  * @return : 1 if successful, 0 if not.
  */
@@ -89,5 +82,3 @@ int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque);
 #endif
 
 #endif
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/portability_macros.h

@@ -1,161 +0,0 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
-#ifndef ZSTD_PORTABILITY_MACROS_H
-#define ZSTD_PORTABILITY_MACROS_H
-
-/**
- * This header file contains macro definitions to support portability.
- * This header is shared between C and ASM code, so it MUST only
- * contain macro definitions. It MUST not contain any C code.
- *
- * This header ONLY defines macros to detect platforms/feature support.
- *
- */
-
-
-/* compat. with non-clang compilers */
-#ifndef __has_attribute
-  #define __has_attribute(x) 0
-#endif
-
-/* compat. with non-clang compilers */
-#ifndef __has_builtin
-#  define __has_builtin(x) 0
-#endif
-
-/* compat. with non-clang compilers */
-#ifndef __has_feature
-#  define __has_feature(x) 0
-#endif
-
-/* detects whether we are being compiled under msan */
-#ifndef ZSTD_MEMORY_SANITIZER
-#  if __has_feature(memory_sanitizer)
-#    define ZSTD_MEMORY_SANITIZER 1
-#  else
-#    define ZSTD_MEMORY_SANITIZER 0
-#  endif
-#endif
-
-/* detects whether we are being compiled under asan */
-#ifndef ZSTD_ADDRESS_SANITIZER
-#  if __has_feature(address_sanitizer)
-#    define ZSTD_ADDRESS_SANITIZER 1
-#  elif defined(__SANITIZE_ADDRESS__)
-#    define ZSTD_ADDRESS_SANITIZER 1
-#  else
-#    define ZSTD_ADDRESS_SANITIZER 0
-#  endif
-#endif
-
-/* detects whether we are being compiled under dfsan */
-#ifndef ZSTD_DATAFLOW_SANITIZER
-# if __has_feature(dataflow_sanitizer)
-#  define ZSTD_DATAFLOW_SANITIZER 1
-# else
-#  define ZSTD_DATAFLOW_SANITIZER 0
-# endif
-#endif
-
-/* Mark the internal assembly functions as hidden  */
-#ifdef __ELF__
-# define ZSTD_HIDE_ASM_FUNCTION(func) .hidden func
-#elif defined(__APPLE__)
-# define ZSTD_HIDE_ASM_FUNCTION(func) .private_extern func
-#else
-# define ZSTD_HIDE_ASM_FUNCTION(func)
-#endif
-
-/* Enable runtime BMI2 dispatch based on the CPU.
- * Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default.
- */
-#ifndef DYNAMIC_BMI2
-  #if ((defined(__clang__) && __has_attribute(__target__)) \
-      || (defined(__GNUC__) \
-          && (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8)))) \
-      && (defined(__x86_64__) || defined(_M_X64)) \
-      && !defined(__BMI2__)
-  #  define DYNAMIC_BMI2 1
-  #else
-  #  define DYNAMIC_BMI2 0
-  #endif
-#endif
-
-/**
- * Only enable assembly for GNUC compatible compilers,
- * because other platforms may not support GAS assembly syntax.
- *
- * Only enable assembly for Linux / MacOS, other platforms may
- * work, but they haven't been tested. This could likely be
- * extended to BSD systems.
- *
- * Disable assembly when MSAN is enabled, because MSAN requires
- * 100% of code to be instrumented to work.
- */
-#if defined(__GNUC__)
-#  if defined(__linux__) || defined(__linux) || defined(__APPLE__)
-#    if ZSTD_MEMORY_SANITIZER
-#      define ZSTD_ASM_SUPPORTED 0
-#    elif ZSTD_DATAFLOW_SANITIZER
-#      define ZSTD_ASM_SUPPORTED 0
-#    else
-#      define ZSTD_ASM_SUPPORTED 1
-#    endif
-#  else
-#    define ZSTD_ASM_SUPPORTED 0
-#  endif
-#else
-#  define ZSTD_ASM_SUPPORTED 0
-#endif
-
-/**
- * Determines whether we should enable assembly for x86-64
- * with BMI2.
- *
- * Enable if all of the following conditions hold:
- * - ASM hasn't been explicitly disabled by defining ZSTD_DISABLE_ASM
- * - Assembly is supported
- * - We are compiling for x86-64 and either:
- *   - DYNAMIC_BMI2 is enabled
- *   - BMI2 is supported at compile time
- */
-#if !defined(ZSTD_DISABLE_ASM) &&                                 \
-    ZSTD_ASM_SUPPORTED &&                                         \
-    defined(__x86_64__) &&                                        \
-    (DYNAMIC_BMI2 || defined(__BMI2__))
-# define ZSTD_ENABLE_ASM_X86_64_BMI2 1
-#else
-# define ZSTD_ENABLE_ASM_X86_64_BMI2 0
-#endif
-
-/*
- * For x86 ELF targets, add .note.gnu.property section for Intel CET in
- * assembly sources when CET is enabled.
- *
- * Additionally, any function that may be called indirectly must begin
- * with ZSTD_CET_ENDBRANCH.
- */
-#if defined(__ELF__) && (defined(__x86_64__) || defined(__i386__)) \
-    && defined(__has_include)
-# if __has_include(<cet.h>)
-#  include <cet.h>
-#  define ZSTD_CET_ENDBRANCH _CET_ENDBR
-# endif
-#endif
-
-#ifndef ZSTD_CET_ENDBRANCH
-# define ZSTD_CET_ENDBRANCH
-#endif
-
-#endif /* ZSTD_PORTABILITY_MACROS_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/threading.c

@@ -1,15 +1,13 @@
-#ifndef USE_EXTERNAL_ZSTD
 /**
  * Copyright (c) 2016 Tino Reichardt
  * All rights reserved.
  *
- * You can contact the author at:
- * - zstdmt source repository: https://github.com/mcmilk/zstdmt
- *
  * This source code is licensed under both the BSD-style license (found in the
  * LICENSE file in the root directory of this source tree) and the GPLv2 (found
  * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+ *
+ * You can contact the author at:
+ * - zstdmt source repository: https://github.com/mcmilk/zstdmt
  */
 
 /**
@@ -24,7 +22,8 @@ int g_ZSTD_threading_useless_symbol;
 #if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
 
 /**
- * Windows minimalist Pthread Wrapper
+ * Windows minimalist Pthread Wrapper, based on :
+ * http://www.cse.wustl.edu/~schmidt/win32-cv-1.html
  */
 
 
@@ -35,94 +34,37 @@ int g_ZSTD_threading_useless_symbol;
 
 /* ===  Implementation  === */
 
-typedef struct {
-    void* (*start_routine)(void*);
-    void* arg;
-    int initialized;
-    ZSTD_pthread_cond_t initialized_cond;
-    ZSTD_pthread_mutex_t initialized_mutex;
-} ZSTD_thread_params_t;
-
 static unsigned __stdcall worker(void *arg)
 {
-    void* (*start_routine)(void*);
-    void* thread_arg;
-
-    /* Initialized thread_arg and start_routine and signal main thread that we don't need it
-     * to wait any longer.
-     */
-    {
-        ZSTD_thread_params_t*  thread_param = (ZSTD_thread_params_t*)arg;
-        thread_arg = thread_param->arg;
-        start_routine = thread_param->start_routine;
-
-        /* Signal main thread that we are running and do not depend on its memory anymore */
-        ZSTD_pthread_mutex_lock(&thread_param->initialized_mutex);
-        thread_param->initialized = 1;
-        ZSTD_pthread_cond_signal(&thread_param->initialized_cond);
-        ZSTD_pthread_mutex_unlock(&thread_param->initialized_mutex);
-    }
-
-    start_routine(thread_arg);
-
+    ZSTD_pthread_t* const thread = (ZSTD_pthread_t*) arg;
+    thread->arg = thread->start_routine(thread->arg);
     return 0;
 }
 
 int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
             void* (*start_routine) (void*), void* arg)
 {
-    ZSTD_thread_params_t thread_param;
     (void)unused;
+    thread->arg = arg;
+    thread->start_routine = start_routine;
+    thread->handle = (HANDLE) _beginthreadex(NULL, 0, worker, thread, 0, NULL);
 
-    if (thread==NULL) return -1;
-    *thread = NULL;
-
-    thread_param.start_routine = start_routine;
-    thread_param.arg = arg;
-    thread_param.initialized = 0;
-
-    /* Setup thread initialization synchronization */
-    if(ZSTD_pthread_cond_init(&thread_param.initialized_cond, NULL)) {
-        /* Should never happen on Windows */
-        return -1;
-    }
-    if(ZSTD_pthread_mutex_init(&thread_param.initialized_mutex, NULL)) {
-        /* Should never happen on Windows */
-        ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
-        return -1;
-    }
-
-    /* Spawn thread */
-    *thread = (HANDLE)_beginthreadex(NULL, 0, worker, &thread_param, 0, NULL);
-    if (*thread==NULL) {
-        ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
-        ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
+    if (!thread->handle)
         return errno;
-    }
-
-    /* Wait for thread to be initialized */
-    ZSTD_pthread_mutex_lock(&thread_param.initialized_mutex);
-    while(!thread_param.initialized) {
-        ZSTD_pthread_cond_wait(&thread_param.initialized_cond, &thread_param.initialized_mutex);
-    }
-    ZSTD_pthread_mutex_unlock(&thread_param.initialized_mutex);
-    ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
-    ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
-
-    return 0;
+    else
+        return 0;
 }
 
-int ZSTD_pthread_join(ZSTD_pthread_t thread)
+int ZSTD_pthread_join(ZSTD_pthread_t thread, void **value_ptr)
 {
     DWORD result;
 
-    if (!thread) return 0;
-
-    result = WaitForSingleObject(thread, INFINITE);
-    CloseHandle(thread);
+    if (!thread.handle) return 0;
 
+    result = WaitForSingleObject(thread.handle, INFINITE);
     switch (result) {
     case WAIT_OBJECT_0:
+        if (value_ptr) *value_ptr = thread.arg;
         return 0;
     case WAIT_ABANDONED:
         return EINVAL;
@@ -135,13 +77,11 @@ int ZSTD_pthread_join(ZSTD_pthread_t thread)
 
 #if defined(ZSTD_MULTITHREAD) && DEBUGLEVEL >= 1 && !defined(_WIN32)
 
-#define ZSTD_DEPS_NEED_MALLOC
-#include "zstd_deps.h"
+#include <stdlib.h>
 
 int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr)
 {
-    assert(mutex != NULL);
-    *mutex = (pthread_mutex_t*)ZSTD_malloc(sizeof(pthread_mutex_t));
+    *mutex = (pthread_mutex_t*)malloc(sizeof(pthread_mutex_t));
     if (!*mutex)
         return 1;
     return pthread_mutex_init(*mutex, attr);
@@ -149,20 +89,18 @@ int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t con
 
 int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex)
 {
-    assert(mutex != NULL);
     if (!*mutex)
         return 0;
     {
         int const ret = pthread_mutex_destroy(*mutex);
-        ZSTD_free(*mutex);
+        free(*mutex);
         return ret;
     }
 }
 
 int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr)
 {
-    assert(cond != NULL);
-    *cond = (pthread_cond_t*)ZSTD_malloc(sizeof(pthread_cond_t));
+    *cond = (pthread_cond_t*)malloc(sizeof(pthread_cond_t));
     if (!*cond)
         return 1;
     return pthread_cond_init(*cond, attr);
@@ -170,16 +108,13 @@ int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const*
 
 int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond)
 {
-    assert(cond != NULL);
     if (!*cond)
         return 0;
     {
         int const ret = pthread_cond_destroy(*cond);
-        ZSTD_free(*cond);
+        free(*cond);
         return ret;
     }
 }
 
 #endif
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/threading.h

@@ -1,15 +1,13 @@
-#ifndef USE_EXTERNAL_ZSTD
 /**
  * Copyright (c) 2016 Tino Reichardt
  * All rights reserved.
  *
- * You can contact the author at:
- * - zstdmt source repository: https://github.com/mcmilk/zstdmt
- *
  * This source code is licensed under both the BSD-style license (found in the
  * LICENSE file in the root directory of this source tree) and the GPLv2 (found
  * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+ *
+ * You can contact the author at:
+ * - zstdmt source repository: https://github.com/mcmilk/zstdmt
  */
 
 #ifndef THREADING_H_938743
@@ -24,7 +22,8 @@ extern "C" {
 #if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
 
 /**
- * Windows minimalist Pthread Wrapper
+ * Windows minimalist Pthread Wrapper, based on :
+ * http://www.cse.wustl.edu/~schmidt/win32-cv-1.html
  */
 #ifdef WINVER
 #  undef WINVER
@@ -62,12 +61,16 @@ extern "C" {
 #define ZSTD_pthread_cond_broadcast(a)  WakeAllConditionVariable((a))
 
 /* ZSTD_pthread_create() and ZSTD_pthread_join() */
-typedef HANDLE ZSTD_pthread_t;
+typedef struct {
+    HANDLE handle;
+    void* (*start_routine)(void*);
+    void* arg;
+} ZSTD_pthread_t;
 
 int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
                    void* (*start_routine) (void*), void* arg);
 
-int ZSTD_pthread_join(ZSTD_pthread_t thread);
+int ZSTD_pthread_join(ZSTD_pthread_t thread, void** value_ptr);
 
 /**
  * add here more wrappers as required
@@ -95,7 +98,7 @@ int ZSTD_pthread_join(ZSTD_pthread_t thread);
 
 #define ZSTD_pthread_t                  pthread_t
 #define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
-#define ZSTD_pthread_join(a)         pthread_join((a),NULL)
+#define ZSTD_pthread_join(a, b)         pthread_join((a),(b))
 
 #else /* DEBUGLEVEL >= 1 */
 
@@ -120,7 +123,7 @@ int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond);
 
 #define ZSTD_pthread_t                  pthread_t
 #define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
-#define ZSTD_pthread_join(a)         pthread_join((a),NULL)
+#define ZSTD_pthread_join(a, b)         pthread_join((a),(b))
 
 #endif
 
@@ -149,5 +152,3 @@ typedef int ZSTD_pthread_cond_t;
 #endif
 
 #endif /* THREADING_H_938743 */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/travis_test_32.sh

@@ -1,8 +1,6 @@
 #!/bin/bash
 # Get utilities
-#yum -y -q -e 0 install wget tar unzip gcc
-apt-get update
-apt-get -y install wget tar unzip gcc
+yum -y -q -e 0 install wget tar unzip gcc
 
 # Get Go
 wget -q https://dl.google.com/go/go1.13.linux-386.tar.gz
@@ -15,5 +13,5 @@ unzip mr.zip
 
 # Build and run tests
 go build
-DISABLE_BIG_TESTS=1 PAYLOAD=$(pwd)/mr go test -v
-DISABLE_BIG_TESTS=1 PAYLOAD=$(pwd)/mr go test -bench .
+PAYLOAD=$(pwd)/mr go test -v
+PAYLOAD=$(pwd)/mr go test -bench .

vendor/github.com/DataDog/zstd/xxhash.c

@@ -1,21 +1,882 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * xxHash - Extremely Fast Hash algorithm
- * Copyright (c) Yann Collet - Meta Platforms, Inc
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
+*  xxHash - Fast Hash algorithm
+*  Copyright (C) 2012-2016, Yann Collet
+*
+*  BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+*
+*  Redistribution and use in source and binary forms, with or without
+*  modification, are permitted provided that the following conditions are
+*  met:
+*
+*  * Redistributions of source code must retain the above copyright
+*  notice, this list of conditions and the following disclaimer.
+*  * Redistributions in binary form must reproduce the above
+*  copyright notice, this list of conditions and the following disclaimer
+*  in the documentation and/or other materials provided with the
+*  distribution.
+*
+*  THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+*  "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+*  LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+*  A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+*  OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+*  SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+*  LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+*  DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+*  THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+*  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+*  OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+*
+*  You can contact the author at :
+*  - xxHash homepage: http://www.xxhash.com
+*  - xxHash source repository : https://github.com/Cyan4973/xxHash
+*/
+
+
+/* *************************************
+*  Tuning parameters
+***************************************/
+/*!XXH_FORCE_MEMORY_ACCESS :
+ * By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable.
+ * Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal.
+ * The below switch allow to select different access method for improved performance.
+ * Method 0 (default) : use `memcpy()`. Safe and portable.
+ * Method 1 : `__packed` statement. It depends on compiler extension (ie, not portable).
+ *            This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`.
+ * Method 2 : direct access. This method doesn't depend on compiler but violate C standard.
+ *            It can generate buggy code on targets which do not support unaligned memory accesses.
+ *            But in some circumstances, it's the only known way to get the most performance (ie GCC + ARMv6)
+ * See http://stackoverflow.com/a/32095106/646947 for details.
+ * Prefer these methods in priority order (0 > 1 > 2)
  */
+#ifndef XXH_FORCE_MEMORY_ACCESS   /* can be defined externally, on command line for example */
+#  if defined(__GNUC__) && ( defined(__ARM_ARCH_6__) || defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__) || defined(__ARM_ARCH_6T2__) )
+#    define XXH_FORCE_MEMORY_ACCESS 2
+#  elif (defined(__INTEL_COMPILER) && !defined(WIN32)) || \
+  (defined(__GNUC__) && ( defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A__) || defined(__ARM_ARCH_7R__) || defined(__ARM_ARCH_7M__) || defined(__ARM_ARCH_7S__) )) || \
+  defined(__ICCARM__)
+#    define XXH_FORCE_MEMORY_ACCESS 1
+#  endif
+#endif
 
-/*
- * xxhash.c instantiates functions defined in xxhash.h
+/*!XXH_ACCEPT_NULL_INPUT_POINTER :
+ * If the input pointer is a null pointer, xxHash default behavior is to trigger a memory access error, since it is a bad pointer.
+ * When this option is enabled, xxHash output for null input pointers will be the same as a null-length input.
+ * By default, this option is disabled. To enable it, uncomment below define :
  */
+/* #define XXH_ACCEPT_NULL_INPUT_POINTER 1 */
 
-#define XXH_STATIC_LINKING_ONLY /* access advanced declarations */
-#define XXH_IMPLEMENTATION      /* access definitions */
+/*!XXH_FORCE_NATIVE_FORMAT :
+ * By default, xxHash library provides endian-independent Hash values, based on little-endian convention.
+ * Results are therefore identical for little-endian and big-endian CPU.
+ * This comes at a performance cost for big-endian CPU, since some swapping is required to emulate little-endian format.
+ * Should endian-independence be of no importance for your application, you may set the #define below to 1,
+ * to improve speed for Big-endian CPU.
+ * This option has no impact on Little_Endian CPU.
+ */
+#ifndef XXH_FORCE_NATIVE_FORMAT   /* can be defined externally */
+#  define XXH_FORCE_NATIVE_FORMAT 0
+#endif
 
+/*!XXH_FORCE_ALIGN_CHECK :
+ * This is a minor performance trick, only useful with lots of very small keys.
+ * It means : check for aligned/unaligned input.
+ * The check costs one initial branch per hash; set to 0 when the input data
+ * is guaranteed to be aligned.
+ */
+#ifndef XXH_FORCE_ALIGN_CHECK /* can be defined externally */
+#  if defined(__i386) || defined(_M_IX86) || defined(__x86_64__) || defined(_M_X64)
+#    define XXH_FORCE_ALIGN_CHECK 0
+#  else
+#    define XXH_FORCE_ALIGN_CHECK 1
+#  endif
+#endif
+
+
+/* *************************************
+*  Includes & Memory related functions
+***************************************/
+/* Modify the local functions below should you wish to use some other memory routines */
+/* for malloc(), free() */
+#include <stdlib.h>
+#include <stddef.h>     /* size_t */
+static void* XXH_malloc(size_t s) { return malloc(s); }
+static void  XXH_free  (void* p)  { free(p); }
+/* for memcpy() */
+#include <string.h>
+static void* XXH_memcpy(void* dest, const void* src, size_t size) { return memcpy(dest,src,size); }
+
+#ifndef XXH_STATIC_LINKING_ONLY
+#  define XXH_STATIC_LINKING_ONLY
+#endif
 #include "xxhash.h"
 
-#endif /* USE_EXTERNAL_ZSTD */
+
+/* *************************************
+*  Compiler Specific Options
+***************************************/
+#if defined (__GNUC__) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   /* C99 */
+#  define INLINE_KEYWORD inline
+#else
+#  define INLINE_KEYWORD
+#endif
+
+#if defined(__GNUC__) || defined(__ICCARM__)
+#  define FORCE_INLINE_ATTR __attribute__((always_inline))
+#elif defined(_MSC_VER)
+#  define FORCE_INLINE_ATTR __forceinline
+#else
+#  define FORCE_INLINE_ATTR
+#endif
+
+#define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR
+
+
+#ifdef _MSC_VER
+#  pragma warning(disable : 4127)      /* disable: C4127: conditional expression is constant */
+#endif
+
+
+/* *************************************
+*  Basic Types
+***************************************/
+#ifndef MEM_MODULE
+# define MEM_MODULE
+# if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
+#   include <stdint.h>
+    typedef uint8_t  BYTE;
+    typedef uint16_t U16;
+    typedef uint32_t U32;
+    typedef  int32_t S32;
+    typedef uint64_t U64;
+#  else
+    typedef unsigned char      BYTE;
+    typedef unsigned short     U16;
+    typedef unsigned int       U32;
+    typedef   signed int       S32;
+    typedef unsigned long long U64;   /* if your compiler doesn't support unsigned long long, replace by another 64-bit type here. Note that xxhash.h will also need to be updated. */
+#  endif
+#endif
+
+
+#if (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==2))
+
+/* Force direct memory access. Only works on CPU which support unaligned memory access in hardware */
+static U32 XXH_read32(const void* memPtr) { return *(const U32*) memPtr; }
+static U64 XXH_read64(const void* memPtr) { return *(const U64*) memPtr; }
+
+#elif (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==1))
+
+/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
+/* currently only defined for gcc and icc */
+typedef union { U32 u32; U64 u64; } __attribute__((packed)) unalign;
+
+static U32 XXH_read32(const void* ptr) { return ((const unalign*)ptr)->u32; }
+static U64 XXH_read64(const void* ptr) { return ((const unalign*)ptr)->u64; }
+
+#else
+
+/* portable and safe solution. Generally efficient.
+ * see : http://stackoverflow.com/a/32095106/646947
+ */
+
+static U32 XXH_read32(const void* memPtr)
+{
+    U32 val;
+    memcpy(&val, memPtr, sizeof(val));
+    return val;
+}
+
+static U64 XXH_read64(const void* memPtr)
+{
+    U64 val;
+    memcpy(&val, memPtr, sizeof(val));
+    return val;
+}
+
+#endif   /* XXH_FORCE_DIRECT_MEMORY_ACCESS */
+
+
+/* ****************************************
+*  Compiler-specific Functions and Macros
+******************************************/
+#define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
+
+/* Note : although _rotl exists for minGW (GCC under windows), performance seems poor */
+#if defined(_MSC_VER)
+#  define XXH_rotl32(x,r) _rotl(x,r)
+#  define XXH_rotl64(x,r) _rotl64(x,r)
+#else
+#if defined(__ICCARM__)
+#  include <intrinsics.h>
+#  define XXH_rotl32(x,r) __ROR(x,(32 - r))
+#else
+#  define XXH_rotl32(x,r) ((x << r) | (x >> (32 - r)))
+#endif
+#  define XXH_rotl64(x,r) ((x << r) | (x >> (64 - r)))
+#endif
+
+#if defined(_MSC_VER)     /* Visual Studio */
+#  define XXH_swap32 _byteswap_ulong
+#  define XXH_swap64 _byteswap_uint64
+#elif GCC_VERSION >= 403
+#  define XXH_swap32 __builtin_bswap32
+#  define XXH_swap64 __builtin_bswap64
+#else
+static U32 XXH_swap32 (U32 x)
+{
+    return  ((x << 24) & 0xff000000 ) |
+            ((x <<  8) & 0x00ff0000 ) |
+            ((x >>  8) & 0x0000ff00 ) |
+            ((x >> 24) & 0x000000ff );
+}
+static U64 XXH_swap64 (U64 x)
+{
+    return  ((x << 56) & 0xff00000000000000ULL) |
+            ((x << 40) & 0x00ff000000000000ULL) |
+            ((x << 24) & 0x0000ff0000000000ULL) |
+            ((x << 8)  & 0x000000ff00000000ULL) |
+            ((x >> 8)  & 0x00000000ff000000ULL) |
+            ((x >> 24) & 0x0000000000ff0000ULL) |
+            ((x >> 40) & 0x000000000000ff00ULL) |
+            ((x >> 56) & 0x00000000000000ffULL);
+}
+#endif
+
+
+/* *************************************
+*  Architecture Macros
+***************************************/
+typedef enum { XXH_bigEndian=0, XXH_littleEndian=1 } XXH_endianess;
+
+/* XXH_CPU_LITTLE_ENDIAN can be defined externally, for example on the compiler command line */
+#ifndef XXH_CPU_LITTLE_ENDIAN
+    static const int g_one = 1;
+#   define XXH_CPU_LITTLE_ENDIAN   (*(const char*)(&g_one))
+#endif
+
+
+/* ***************************
+*  Memory reads
+*****************************/
+typedef enum { XXH_aligned, XXH_unaligned } XXH_alignment;
+
+FORCE_INLINE_TEMPLATE U32 XXH_readLE32_align(const void* ptr, XXH_endianess endian, XXH_alignment align)
+{
+    if (align==XXH_unaligned)
+        return endian==XXH_littleEndian ? XXH_read32(ptr) : XXH_swap32(XXH_read32(ptr));
+    else
+        return endian==XXH_littleEndian ? *(const U32*)ptr : XXH_swap32(*(const U32*)ptr);
+}
+
+FORCE_INLINE_TEMPLATE U32 XXH_readLE32(const void* ptr, XXH_endianess endian)
+{
+    return XXH_readLE32_align(ptr, endian, XXH_unaligned);
+}
+
+static U32 XXH_readBE32(const void* ptr)
+{
+    return XXH_CPU_LITTLE_ENDIAN ? XXH_swap32(XXH_read32(ptr)) : XXH_read32(ptr);
+}
+
+FORCE_INLINE_TEMPLATE U64 XXH_readLE64_align(const void* ptr, XXH_endianess endian, XXH_alignment align)
+{
+    if (align==XXH_unaligned)
+        return endian==XXH_littleEndian ? XXH_read64(ptr) : XXH_swap64(XXH_read64(ptr));
+    else
+        return endian==XXH_littleEndian ? *(const U64*)ptr : XXH_swap64(*(const U64*)ptr);
+}
+
+FORCE_INLINE_TEMPLATE U64 XXH_readLE64(const void* ptr, XXH_endianess endian)
+{
+    return XXH_readLE64_align(ptr, endian, XXH_unaligned);
+}
+
+static U64 XXH_readBE64(const void* ptr)
+{
+    return XXH_CPU_LITTLE_ENDIAN ? XXH_swap64(XXH_read64(ptr)) : XXH_read64(ptr);
+}
+
+
+/* *************************************
+*  Macros
+***************************************/
+#define XXH_STATIC_ASSERT(c)   { enum { XXH_static_assert = 1/(int)(!!(c)) }; }    /* use only *after* variable declarations */
+
+
+/* *************************************
+*  Constants
+***************************************/
+static const U32 PRIME32_1 = 2654435761U;
+static const U32 PRIME32_2 = 2246822519U;
+static const U32 PRIME32_3 = 3266489917U;
+static const U32 PRIME32_4 =  668265263U;
+static const U32 PRIME32_5 =  374761393U;
+
+static const U64 PRIME64_1 = 11400714785074694791ULL;
+static const U64 PRIME64_2 = 14029467366897019727ULL;
+static const U64 PRIME64_3 =  1609587929392839161ULL;
+static const U64 PRIME64_4 =  9650029242287828579ULL;
+static const U64 PRIME64_5 =  2870177450012600261ULL;
+
+XXH_PUBLIC_API unsigned XXH_versionNumber (void) { return XXH_VERSION_NUMBER; }
+
+
+/* **************************
+*  Utils
+****************************/
+XXH_PUBLIC_API void XXH32_copyState(XXH32_state_t* restrict dstState, const XXH32_state_t* restrict srcState)
+{
+    memcpy(dstState, srcState, sizeof(*dstState));
+}
+
+XXH_PUBLIC_API void XXH64_copyState(XXH64_state_t* restrict dstState, const XXH64_state_t* restrict srcState)
+{
+    memcpy(dstState, srcState, sizeof(*dstState));
+}
+
+
+/* ***************************
+*  Simple Hash Functions
+*****************************/
+
+static U32 XXH32_round(U32 seed, U32 input)
+{
+    seed += input * PRIME32_2;
+    seed  = XXH_rotl32(seed, 13);
+    seed *= PRIME32_1;
+    return seed;
+}
+
+FORCE_INLINE_TEMPLATE U32 XXH32_endian_align(const void* input, size_t len, U32 seed, XXH_endianess endian, XXH_alignment align)
+{
+    const BYTE* p = (const BYTE*)input;
+    const BYTE* bEnd = p + len;
+    U32 h32;
+#define XXH_get32bits(p) XXH_readLE32_align(p, endian, align)
+
+#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
+    if (p==NULL) {
+        len=0;
+        bEnd=p=(const BYTE*)(size_t)16;
+    }
+#endif
+
+    if (len>=16) {
+        const BYTE* const limit = bEnd - 16;
+        U32 v1 = seed + PRIME32_1 + PRIME32_2;
+        U32 v2 = seed + PRIME32_2;
+        U32 v3 = seed + 0;
+        U32 v4 = seed - PRIME32_1;
+
+        do {
+            v1 = XXH32_round(v1, XXH_get32bits(p)); p+=4;
+            v2 = XXH32_round(v2, XXH_get32bits(p)); p+=4;
+            v3 = XXH32_round(v3, XXH_get32bits(p)); p+=4;
+            v4 = XXH32_round(v4, XXH_get32bits(p)); p+=4;
+        } while (p<=limit);
+
+        h32 = XXH_rotl32(v1, 1) + XXH_rotl32(v2, 7) + XXH_rotl32(v3, 12) + XXH_rotl32(v4, 18);
+    } else {
+        h32  = seed + PRIME32_5;
+    }
+
+    h32 += (U32) len;
+
+    while (p+4<=bEnd) {
+        h32 += XXH_get32bits(p) * PRIME32_3;
+        h32  = XXH_rotl32(h32, 17) * PRIME32_4 ;
+        p+=4;
+    }
+
+    while (p<bEnd) {
+        h32 += (*p) * PRIME32_5;
+        h32 = XXH_rotl32(h32, 11) * PRIME32_1 ;
+        p++;
+    }
+
+    h32 ^= h32 >> 15;
+    h32 *= PRIME32_2;
+    h32 ^= h32 >> 13;
+    h32 *= PRIME32_3;
+    h32 ^= h32 >> 16;
+
+    return h32;
+}
+
+
+XXH_PUBLIC_API unsigned int XXH32 (const void* input, size_t len, unsigned int seed)
+{
+#if 0
+    /* Simple version, good for code maintenance, but unfortunately slow for small inputs */
+    XXH32_CREATESTATE_STATIC(state);
+    XXH32_reset(state, seed);
+    XXH32_update(state, input, len);
+    return XXH32_digest(state);
+#else
+    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
+
+    if (XXH_FORCE_ALIGN_CHECK) {
+        if ((((size_t)input) & 3) == 0) {   /* Input is 4-bytes aligned, leverage the speed benefit */
+            if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
+                return XXH32_endian_align(input, len, seed, XXH_littleEndian, XXH_aligned);
+            else
+                return XXH32_endian_align(input, len, seed, XXH_bigEndian, XXH_aligned);
+    }   }
+
+    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
+        return XXH32_endian_align(input, len, seed, XXH_littleEndian, XXH_unaligned);
+    else
+        return XXH32_endian_align(input, len, seed, XXH_bigEndian, XXH_unaligned);
+#endif
+}
+
+
+static U64 XXH64_round(U64 acc, U64 input)
+{
+    acc += input * PRIME64_2;
+    acc  = XXH_rotl64(acc, 31);
+    acc *= PRIME64_1;
+    return acc;
+}
+
+static U64 XXH64_mergeRound(U64 acc, U64 val)
+{
+    val  = XXH64_round(0, val);
+    acc ^= val;
+    acc  = acc * PRIME64_1 + PRIME64_4;
+    return acc;
+}
+
+FORCE_INLINE_TEMPLATE U64 XXH64_endian_align(const void* input, size_t len, U64 seed, XXH_endianess endian, XXH_alignment align)
+{
+    const BYTE* p = (const BYTE*)input;
+    const BYTE* const bEnd = p + len;
+    U64 h64;
+#define XXH_get64bits(p) XXH_readLE64_align(p, endian, align)
+
+#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
+    if (p==NULL) {
+        len=0;
+        bEnd=p=(const BYTE*)(size_t)32;
+    }
+#endif
+
+    if (len>=32) {
+        const BYTE* const limit = bEnd - 32;
+        U64 v1 = seed + PRIME64_1 + PRIME64_2;
+        U64 v2 = seed + PRIME64_2;
+        U64 v3 = seed + 0;
+        U64 v4 = seed - PRIME64_1;
+
+        do {
+            v1 = XXH64_round(v1, XXH_get64bits(p)); p+=8;
+            v2 = XXH64_round(v2, XXH_get64bits(p)); p+=8;
+            v3 = XXH64_round(v3, XXH_get64bits(p)); p+=8;
+            v4 = XXH64_round(v4, XXH_get64bits(p)); p+=8;
+        } while (p<=limit);
+
+        h64 = XXH_rotl64(v1, 1) + XXH_rotl64(v2, 7) + XXH_rotl64(v3, 12) + XXH_rotl64(v4, 18);
+        h64 = XXH64_mergeRound(h64, v1);
+        h64 = XXH64_mergeRound(h64, v2);
+        h64 = XXH64_mergeRound(h64, v3);
+        h64 = XXH64_mergeRound(h64, v4);
+
+    } else {
+        h64  = seed + PRIME64_5;
+    }
+
+    h64 += (U64) len;
+
+    while (p+8<=bEnd) {
+        U64 const k1 = XXH64_round(0, XXH_get64bits(p));
+        h64 ^= k1;
+        h64  = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4;
+        p+=8;
+    }
+
+    if (p+4<=bEnd) {
+        h64 ^= (U64)(XXH_get32bits(p)) * PRIME64_1;
+        h64 = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
+        p+=4;
+    }
+
+    while (p<bEnd) {
+        h64 ^= (*p) * PRIME64_5;
+        h64 = XXH_rotl64(h64, 11) * PRIME64_1;
+        p++;
+    }
+
+    h64 ^= h64 >> 33;
+    h64 *= PRIME64_2;
+    h64 ^= h64 >> 29;
+    h64 *= PRIME64_3;
+    h64 ^= h64 >> 32;
+
+    return h64;
+}
+
+
+XXH_PUBLIC_API unsigned long long XXH64 (const void* input, size_t len, unsigned long long seed)
+{
+#if 0
+    /* Simple version, good for code maintenance, but unfortunately slow for small inputs */
+    XXH64_CREATESTATE_STATIC(state);
+    XXH64_reset(state, seed);
+    XXH64_update(state, input, len);
+    return XXH64_digest(state);
+#else
+    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
+
+    if (XXH_FORCE_ALIGN_CHECK) {
+        if ((((size_t)input) & 7)==0) {  /* Input is aligned, let's leverage the speed advantage */
+            if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
+                return XXH64_endian_align(input, len, seed, XXH_littleEndian, XXH_aligned);
+            else
+                return XXH64_endian_align(input, len, seed, XXH_bigEndian, XXH_aligned);
+    }   }
+
+    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
+        return XXH64_endian_align(input, len, seed, XXH_littleEndian, XXH_unaligned);
+    else
+        return XXH64_endian_align(input, len, seed, XXH_bigEndian, XXH_unaligned);
+#endif
+}
+
+
+/* **************************************************
+*  Advanced Hash Functions
+****************************************************/
+
+XXH_PUBLIC_API XXH32_state_t* XXH32_createState(void)
+{
+    return (XXH32_state_t*)XXH_malloc(sizeof(XXH32_state_t));
+}
+XXH_PUBLIC_API XXH_errorcode XXH32_freeState(XXH32_state_t* statePtr)
+{
+    XXH_free(statePtr);
+    return XXH_OK;
+}
+
+XXH_PUBLIC_API XXH64_state_t* XXH64_createState(void)
+{
+    return (XXH64_state_t*)XXH_malloc(sizeof(XXH64_state_t));
+}
+XXH_PUBLIC_API XXH_errorcode XXH64_freeState(XXH64_state_t* statePtr)
+{
+    XXH_free(statePtr);
+    return XXH_OK;
+}
+
+
+/*** Hash feed ***/
+
+XXH_PUBLIC_API XXH_errorcode XXH32_reset(XXH32_state_t* statePtr, unsigned int seed)
+{
+    XXH32_state_t state;   /* using a local state to memcpy() in order to avoid strict-aliasing warnings */
+    memset(&state, 0, sizeof(state)-4);   /* do not write into reserved, for future removal */
+    state.v1 = seed + PRIME32_1 + PRIME32_2;
+    state.v2 = seed + PRIME32_2;
+    state.v3 = seed + 0;
+    state.v4 = seed - PRIME32_1;
+    memcpy(statePtr, &state, sizeof(state));
+    return XXH_OK;
+}
+
+
+XXH_PUBLIC_API XXH_errorcode XXH64_reset(XXH64_state_t* statePtr, unsigned long long seed)
+{
+    XXH64_state_t state;   /* using a local state to memcpy() in order to avoid strict-aliasing warnings */
+    memset(&state, 0, sizeof(state)-8);   /* do not write into reserved, for future removal */
+    state.v1 = seed + PRIME64_1 + PRIME64_2;
+    state.v2 = seed + PRIME64_2;
+    state.v3 = seed + 0;
+    state.v4 = seed - PRIME64_1;
+    memcpy(statePtr, &state, sizeof(state));
+    return XXH_OK;
+}
+
+
+FORCE_INLINE_TEMPLATE XXH_errorcode XXH32_update_endian (XXH32_state_t* state, const void* input, size_t len, XXH_endianess endian)
+{
+    const BYTE* p = (const BYTE*)input;
+    const BYTE* const bEnd = p + len;
+
+#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
+    if (input==NULL) return XXH_ERROR;
+#endif
+
+    state->total_len_32 += (unsigned)len;
+    state->large_len |= (len>=16) | (state->total_len_32>=16);
+
+    if (state->memsize + len < 16)  {   /* fill in tmp buffer */
+        XXH_memcpy((BYTE*)(state->mem32) + state->memsize, input, len);
+        state->memsize += (unsigned)len;
+        return XXH_OK;
+    }
+
+    if (state->memsize) {   /* some data left from previous update */
+        XXH_memcpy((BYTE*)(state->mem32) + state->memsize, input, 16-state->memsize);
+        {   const U32* p32 = state->mem32;
+            state->v1 = XXH32_round(state->v1, XXH_readLE32(p32, endian)); p32++;
+            state->v2 = XXH32_round(state->v2, XXH_readLE32(p32, endian)); p32++;
+            state->v3 = XXH32_round(state->v3, XXH_readLE32(p32, endian)); p32++;
+            state->v4 = XXH32_round(state->v4, XXH_readLE32(p32, endian)); p32++;
+        }
+        p += 16-state->memsize;
+        state->memsize = 0;
+    }
+
+    if (p <= bEnd-16) {
+        const BYTE* const limit = bEnd - 16;
+        U32 v1 = state->v1;
+        U32 v2 = state->v2;
+        U32 v3 = state->v3;
+        U32 v4 = state->v4;
+
+        do {
+            v1 = XXH32_round(v1, XXH_readLE32(p, endian)); p+=4;
+            v2 = XXH32_round(v2, XXH_readLE32(p, endian)); p+=4;
+            v3 = XXH32_round(v3, XXH_readLE32(p, endian)); p+=4;
+            v4 = XXH32_round(v4, XXH_readLE32(p, endian)); p+=4;
+        } while (p<=limit);
+
+        state->v1 = v1;
+        state->v2 = v2;
+        state->v3 = v3;
+        state->v4 = v4;
+    }
+
+    if (p < bEnd) {
+        XXH_memcpy(state->mem32, p, (size_t)(bEnd-p));
+        state->memsize = (unsigned)(bEnd-p);
+    }
+
+    return XXH_OK;
+}
+
+XXH_PUBLIC_API XXH_errorcode XXH32_update (XXH32_state_t* state_in, const void* input, size_t len)
+{
+    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
+
+    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
+        return XXH32_update_endian(state_in, input, len, XXH_littleEndian);
+    else
+        return XXH32_update_endian(state_in, input, len, XXH_bigEndian);
+}
+
+
+
+FORCE_INLINE_TEMPLATE U32 XXH32_digest_endian (const XXH32_state_t* state, XXH_endianess endian)
+{
+    const BYTE * p = (const BYTE*)state->mem32;
+    const BYTE* const bEnd = (const BYTE*)(state->mem32) + state->memsize;
+    U32 h32;
+
+    if (state->large_len) {
+        h32 = XXH_rotl32(state->v1, 1) + XXH_rotl32(state->v2, 7) + XXH_rotl32(state->v3, 12) + XXH_rotl32(state->v4, 18);
+    } else {
+        h32 = state->v3 /* == seed */ + PRIME32_5;
+    }
+
+    h32 += state->total_len_32;
+
+    while (p+4<=bEnd) {
+        h32 += XXH_readLE32(p, endian) * PRIME32_3;
+        h32  = XXH_rotl32(h32, 17) * PRIME32_4;
+        p+=4;
+    }
+
+    while (p<bEnd) {
+        h32 += (*p) * PRIME32_5;
+        h32  = XXH_rotl32(h32, 11) * PRIME32_1;
+        p++;
+    }
+
+    h32 ^= h32 >> 15;
+    h32 *= PRIME32_2;
+    h32 ^= h32 >> 13;
+    h32 *= PRIME32_3;
+    h32 ^= h32 >> 16;
+
+    return h32;
+}
+
+
+XXH_PUBLIC_API unsigned int XXH32_digest (const XXH32_state_t* state_in)
+{
+    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
+
+    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
+        return XXH32_digest_endian(state_in, XXH_littleEndian);
+    else
+        return XXH32_digest_endian(state_in, XXH_bigEndian);
+}
+
+
+
+/* **** XXH64 **** */
+
+FORCE_INLINE_TEMPLATE XXH_errorcode XXH64_update_endian (XXH64_state_t* state, const void* input, size_t len, XXH_endianess endian)
+{
+    const BYTE* p = (const BYTE*)input;
+    const BYTE* const bEnd = p + len;
+
+#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
+    if (input==NULL) return XXH_ERROR;
+#endif
+
+    state->total_len += len;
+
+    if (state->memsize + len < 32) {  /* fill in tmp buffer */
+        XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, len);
+        state->memsize += (U32)len;
+        return XXH_OK;
+    }
+
+    if (state->memsize) {   /* tmp buffer is full */
+        XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, 32-state->memsize);
+        state->v1 = XXH64_round(state->v1, XXH_readLE64(state->mem64+0, endian));
+        state->v2 = XXH64_round(state->v2, XXH_readLE64(state->mem64+1, endian));
+        state->v3 = XXH64_round(state->v3, XXH_readLE64(state->mem64+2, endian));
+        state->v4 = XXH64_round(state->v4, XXH_readLE64(state->mem64+3, endian));
+        p += 32-state->memsize;
+        state->memsize = 0;
+    }
+
+    if (p+32 <= bEnd) {
+        const BYTE* const limit = bEnd - 32;
+        U64 v1 = state->v1;
+        U64 v2 = state->v2;
+        U64 v3 = state->v3;
+        U64 v4 = state->v4;
+
+        do {
+            v1 = XXH64_round(v1, XXH_readLE64(p, endian)); p+=8;
+            v2 = XXH64_round(v2, XXH_readLE64(p, endian)); p+=8;
+            v3 = XXH64_round(v3, XXH_readLE64(p, endian)); p+=8;
+            v4 = XXH64_round(v4, XXH_readLE64(p, endian)); p+=8;
+        } while (p<=limit);
+
+        state->v1 = v1;
+        state->v2 = v2;
+        state->v3 = v3;
+        state->v4 = v4;
+    }
+
+    if (p < bEnd) {
+        XXH_memcpy(state->mem64, p, (size_t)(bEnd-p));
+        state->memsize = (unsigned)(bEnd-p);
+    }
+
+    return XXH_OK;
+}
+
+XXH_PUBLIC_API XXH_errorcode XXH64_update (XXH64_state_t* state_in, const void* input, size_t len)
+{
+    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
+
+    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
+        return XXH64_update_endian(state_in, input, len, XXH_littleEndian);
+    else
+        return XXH64_update_endian(state_in, input, len, XXH_bigEndian);
+}
+
+
+
+FORCE_INLINE_TEMPLATE U64 XXH64_digest_endian (const XXH64_state_t* state, XXH_endianess endian)
+{
+    const BYTE * p = (const BYTE*)state->mem64;
+    const BYTE* const bEnd = (const BYTE*)state->mem64 + state->memsize;
+    U64 h64;
+
+    if (state->total_len >= 32) {
+        U64 const v1 = state->v1;
+        U64 const v2 = state->v2;
+        U64 const v3 = state->v3;
+        U64 const v4 = state->v4;
+
+        h64 = XXH_rotl64(v1, 1) + XXH_rotl64(v2, 7) + XXH_rotl64(v3, 12) + XXH_rotl64(v4, 18);
+        h64 = XXH64_mergeRound(h64, v1);
+        h64 = XXH64_mergeRound(h64, v2);
+        h64 = XXH64_mergeRound(h64, v3);
+        h64 = XXH64_mergeRound(h64, v4);
+    } else {
+        h64  = state->v3 + PRIME64_5;
+    }
+
+    h64 += (U64) state->total_len;
+
+    while (p+8<=bEnd) {
+        U64 const k1 = XXH64_round(0, XXH_readLE64(p, endian));
+        h64 ^= k1;
+        h64  = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4;
+        p+=8;
+    }
+
+    if (p+4<=bEnd) {
+        h64 ^= (U64)(XXH_readLE32(p, endian)) * PRIME64_1;
+        h64  = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
+        p+=4;
+    }
+
+    while (p<bEnd) {
+        h64 ^= (*p) * PRIME64_5;
+        h64  = XXH_rotl64(h64, 11) * PRIME64_1;
+        p++;
+    }
+
+    h64 ^= h64 >> 33;
+    h64 *= PRIME64_2;
+    h64 ^= h64 >> 29;
+    h64 *= PRIME64_3;
+    h64 ^= h64 >> 32;
+
+    return h64;
+}
+
+
+XXH_PUBLIC_API unsigned long long XXH64_digest (const XXH64_state_t* state_in)
+{
+    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
+
+    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
+        return XXH64_digest_endian(state_in, XXH_littleEndian);
+    else
+        return XXH64_digest_endian(state_in, XXH_bigEndian);
+}
+
+
+/* **************************
+*  Canonical representation
+****************************/
+
+/*! Default XXH result types are basic unsigned 32 and 64 bits.
+*   The canonical representation follows human-readable write convention, aka big-endian (large digits first).
+*   These functions allow transformation of hash result into and from its canonical format.
+*   This way, hash values can be written into a file or buffer, and remain comparable across different systems and programs.
+*/
+
+XXH_PUBLIC_API void XXH32_canonicalFromHash(XXH32_canonical_t* dst, XXH32_hash_t hash)
+{
+    XXH_STATIC_ASSERT(sizeof(XXH32_canonical_t) == sizeof(XXH32_hash_t));
+    if (XXH_CPU_LITTLE_ENDIAN) hash = XXH_swap32(hash);
+    memcpy(dst, &hash, sizeof(*dst));
+}
+
+XXH_PUBLIC_API void XXH64_canonicalFromHash(XXH64_canonical_t* dst, XXH64_hash_t hash)
+{
+    XXH_STATIC_ASSERT(sizeof(XXH64_canonical_t) == sizeof(XXH64_hash_t));
+    if (XXH_CPU_LITTLE_ENDIAN) hash = XXH_swap64(hash);
+    memcpy(dst, &hash, sizeof(*dst));
+}
+
+XXH_PUBLIC_API XXH32_hash_t XXH32_hashFromCanonical(const XXH32_canonical_t* src)
+{
+    return XXH_readBE32(src);
+}
+
+XXH_PUBLIC_API XXH64_hash_t XXH64_hashFromCanonical(const XXH64_canonical_t* src)
+{
+    return XXH_readBE64(src);
+}

vendor/github.com/DataDog/zstd/zbuff.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -213,5 +212,3 @@ ZBUFF_DEPRECATED("use ZSTD_initDStream_usingDict") size_t ZBUFF_compressInit_adv
 #if defined (__cplusplus)
 }
 #endif
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zbuff_common.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -25,5 +24,3 @@ unsigned ZBUFF_isError(size_t errorCode) { return ERR_isError(errorCode); }
 /*! ZBUFF_getErrorName() :
 *   provides error code string from function result (useful for debugging) */
 const char* ZBUFF_getErrorName(size_t errorCode) { return ERR_getErrorName(errorCode); }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zbuff_compress.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -16,7 +15,6 @@
 ***************************************/
 #define ZBUFF_STATIC_LINKING_ONLY
 #include "zbuff.h"
-#include "error_private.h"
 
 
 /*-***********************************************************
@@ -75,32 +73,13 @@ size_t ZBUFF_compressInit_advanced(ZBUFF_CCtx* zbc,
                                    ZSTD_parameters params, unsigned long long pledgedSrcSize)
 {
     if (pledgedSrcSize==0) pledgedSrcSize = ZSTD_CONTENTSIZE_UNKNOWN;  /* preserve "0 == unknown" behavior */
-    FORWARD_IF_ERROR(ZSTD_CCtx_reset(zbc, ZSTD_reset_session_only), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setPledgedSrcSize(zbc, pledgedSrcSize), "");
-
-    FORWARD_IF_ERROR(ZSTD_checkCParams(params.cParams), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_windowLog, params.cParams.windowLog), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_hashLog, params.cParams.hashLog), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_chainLog, params.cParams.chainLog), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_searchLog, params.cParams.searchLog), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_minMatch, params.cParams.minMatch), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_targetLength, params.cParams.targetLength), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_strategy, params.cParams.strategy), "");
-
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_contentSizeFlag, params.fParams.contentSizeFlag), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_checksumFlag, params.fParams.checksumFlag), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_dictIDFlag, params.fParams.noDictIDFlag), "");
-
-    FORWARD_IF_ERROR(ZSTD_CCtx_loadDictionary(zbc, dict, dictSize), "");
-    return 0;
+    return ZSTD_initCStream_advanced(zbc, dict, dictSize, params, pledgedSrcSize);
 }
 
+
 size_t ZBUFF_compressInitDictionary(ZBUFF_CCtx* zbc, const void* dict, size_t dictSize, int compressionLevel)
 {
-    FORWARD_IF_ERROR(ZSTD_CCtx_reset(zbc, ZSTD_reset_session_only), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_setParameter(zbc, ZSTD_c_compressionLevel, compressionLevel), "");
-    FORWARD_IF_ERROR(ZSTD_CCtx_loadDictionary(zbc, dict, dictSize), "");
-    return 0;
+    return ZSTD_initCStream_usingDict(zbc, dict, dictSize, compressionLevel);
 }
 
 size_t ZBUFF_compressInit(ZBUFF_CCtx* zbc, int compressionLevel)
@@ -166,5 +145,3 @@ size_t ZBUFF_compressEnd(ZBUFF_CCtx* zbc, void* dst, size_t* dstCapacityPtr)
 ***************************************/
 size_t ZBUFF_recommendedCInSize(void)  { return ZSTD_CStreamInSize(); }
 size_t ZBUFF_recommendedCOutSize(void) { return ZSTD_CStreamOutSize(); }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zbuff_decompress.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -14,8 +13,6 @@
 /* *************************************
 *  Dependencies
 ***************************************/
-#define ZSTD_DISABLE_DEPRECATE_WARNINGS  /* suppress warning on ZSTD_initDStream_usingDict */
-#include "zstd.h"        /* ZSTD_CStream, ZSTD_DStream, ZSTDLIB_API */
 #define ZBUFF_STATIC_LINKING_ONLY
 #include "zbuff.h"
 
@@ -76,5 +73,3 @@ size_t ZBUFF_decompressContinue(ZBUFF_DCtx* zbd,
 ***************************************/
 size_t ZBUFF_recommendedDInSize(void)  { return ZSTD_DStreamInSize(); }
 size_t ZBUFF_recommendedDOutSize(void) { return ZSTD_DStreamOutSize(); }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zdict.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -24,13 +23,9 @@
 /* Unix Large Files support (>4GB) */
 #define _FILE_OFFSET_BITS 64
 #if (defined(__sun__) && (!defined(__LP64__)))   /* Sun Solaris 32-bits requires specific definitions */
-#  ifndef _LARGEFILE_SOURCE
 #  define _LARGEFILE_SOURCE
-#  endif
 #elif ! defined(__LP64__)                        /* No point defining Large file for 64 bit */
-#  ifndef _LARGEFILE64_SOURCE
 #  define _LARGEFILE64_SOURCE
-#  endif
 #endif
 
 
@@ -42,19 +37,17 @@
 #include <stdio.h>         /* fprintf, fopen, ftello64 */
 #include <time.h>          /* clock */
 
-#ifndef ZDICT_STATIC_LINKING_ONLY
-#  define ZDICT_STATIC_LINKING_ONLY
-#endif
-
 #include "mem.h"           /* read */
 #include "fse.h"           /* FSE_normalizeCount, FSE_writeNCount */
+#define HUF_STATIC_LINKING_ONLY
 #include "huf.h"           /* HUF_buildCTable, HUF_writeCTable */
 #include "zstd_internal.h" /* includes zstd.h */
 #include "xxhash.h"        /* XXH64 */
-#include "zstd_compress_internal.h" /* ZSTD_loadCEntropy() */
-#include "zdict.h"
 #include "divsufsort.h"
-#include "bits.h"          /* ZSTD_NbCommonBytes */
+#ifndef ZDICT_STATIC_LINKING_ONLY
+#  define ZDICT_STATIC_LINKING_ONLY
+#endif
+#include "zdict.h"
 
 
 /*-*************************************
@@ -68,16 +61,15 @@
 
 #define NOISELENGTH 32
 
+static const int g_compressionLevel_default = 3;
 static const U32 g_selectivity_default = 9;
 
 
 /*-*************************************
 *  Console display
 ***************************************/
-#undef  DISPLAY
-#define DISPLAY(...)         do { fprintf(stderr, __VA_ARGS__); fflush( stderr ); } while (0)
-#undef  DISPLAYLEVEL
-#define DISPLAYLEVEL(l, ...) do { if (notificationLevel>=l) { DISPLAY(__VA_ARGS__); } } while (0)    /* 0 : no display;   1: errors;   2: default;  3: details;  4: debug */
+#define DISPLAY(...)         { fprintf(stderr, __VA_ARGS__); fflush( stderr ); }
+#define DISPLAYLEVEL(l, ...) if (notificationLevel>=l) { DISPLAY(__VA_ARGS__); }    /* 0 : no display;   1: errors;   2: default;  3: details;  4: debug */
 
 static clock_t ZDICT_clockSpan(clock_t nPrevious) { return clock() - nPrevious; }
 
@@ -107,30 +99,69 @@ unsigned ZDICT_getDictID(const void* dictBuffer, size_t dictSize)
     return MEM_readLE32((const char*)dictBuffer + 4);
 }
 
-size_t ZDICT_getDictHeaderSize(const void* dictBuffer, size_t dictSize)
-{
-    size_t headerSize;
-    if (dictSize <= 8 || MEM_readLE32(dictBuffer) != ZSTD_MAGIC_DICTIONARY) return ERROR(dictionary_corrupted);
-
-    {   ZSTD_compressedBlockState_t* bs = (ZSTD_compressedBlockState_t*)malloc(sizeof(ZSTD_compressedBlockState_t));
-        U32* wksp = (U32*)malloc(HUF_WORKSPACE_SIZE);
-        if (!bs || !wksp) {
-            headerSize = ERROR(memory_allocation);
-        } else {
-            ZSTD_reset_compressedBlockState(bs);
-            headerSize = ZSTD_loadCEntropy(bs, wksp, dictBuffer, dictSize);
-        }
-
-        free(bs);
-        free(wksp);
-    }
-
-    return headerSize;
-}
 
 /*-********************************************************
 *  Dictionary training functions
 **********************************************************/
+static unsigned ZDICT_NbCommonBytes (size_t val)
+{
+    if (MEM_isLittleEndian()) {
+        if (MEM_64bits()) {
+#       if defined(_MSC_VER) && defined(_WIN64)
+            unsigned long r = 0;
+            _BitScanForward64( &r, (U64)val );
+            return (unsigned)(r>>3);
+#       elif defined(__GNUC__) && (__GNUC__ >= 3)
+            return (__builtin_ctzll((U64)val) >> 3);
+#       else
+            static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2, 0, 3, 1, 3, 1, 4, 2, 7, 0, 2, 3, 6, 1, 5, 3, 5, 1, 3, 4, 4, 2, 5, 6, 7, 7, 0, 1, 2, 3, 3, 4, 6, 2, 6, 5, 5, 3, 4, 5, 6, 7, 1, 2, 4, 6, 4, 4, 5, 7, 2, 6, 5, 7, 6, 7, 7 };
+            return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
+#       endif
+        } else { /* 32 bits */
+#       if defined(_MSC_VER)
+            unsigned long r=0;
+            _BitScanForward( &r, (U32)val );
+            return (unsigned)(r>>3);
+#       elif defined(__GNUC__) && (__GNUC__ >= 3)
+            return (__builtin_ctz((U32)val) >> 3);
+#       else
+            static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, 3, 2, 2, 1, 3, 2, 0, 1, 3, 3, 1, 2, 2, 2, 2, 0, 3, 1, 2, 0, 1, 0, 1, 1 };
+            return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
+#       endif
+        }
+    } else {  /* Big Endian CPU */
+        if (MEM_64bits()) {
+#       if defined(_MSC_VER) && defined(_WIN64)
+            unsigned long r = 0;
+            _BitScanReverse64( &r, val );
+            return (unsigned)(r>>3);
+#       elif defined(__GNUC__) && (__GNUC__ >= 3)
+            return (__builtin_clzll(val) >> 3);
+#       else
+            unsigned r;
+            const unsigned n32 = sizeof(size_t)*4;   /* calculate this way due to compiler complaining in 32-bits mode */
+            if (!(val>>n32)) { r=4; } else { r=0; val>>=n32; }
+            if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
+            r += (!val);
+            return r;
+#       endif
+        } else { /* 32 bits */
+#       if defined(_MSC_VER)
+            unsigned long r = 0;
+            _BitScanReverse( &r, (unsigned long)val );
+            return (unsigned)(r>>3);
+#       elif defined(__GNUC__) && (__GNUC__ >= 3)
+            return (__builtin_clz((U32)val) >> 3);
+#       else
+            unsigned r;
+            if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
+            r += (!val);
+            return r;
+#       endif
+    }   }
+}
+
+
 /*! ZDICT_count() :
     Count the nb of common bytes between 2 pointers.
     Note : this function presumes end of buffer followed by noisy guard band.
@@ -145,7 +176,7 @@ static size_t ZDICT_count(const void* pIn, const void* pMatch)
             pMatch = (const char*)pMatch+sizeof(size_t);
             continue;
         }
-        pIn = (const char*)pIn+ZSTD_NbCommonBytes(diff);
+        pIn = (const char*)pIn+ZDICT_NbCommonBytes(diff);
         return (size_t)((const char*)pIn - pStart);
     }
 }
@@ -177,7 +208,7 @@ static dictItem ZDICT_analyzePos(
     U32 savings[LLIMIT] = {0};
     const BYTE* b = (const BYTE*)buffer;
     size_t maxLength = LLIMIT;
-    size_t pos = (size_t)suffix[start];
+    size_t pos = suffix[start];
     U32 end = start;
     dictItem solution;
 
@@ -311,7 +342,7 @@ static dictItem ZDICT_analyzePos(
             savings[i] = savings[i-1] + (lengthList[i] * (i-3));
 
         DISPLAYLEVEL(4, "Selected dict at position %u, of length %u : saves %u (ratio: %.2f)  \n",
-                     (unsigned)pos, (unsigned)maxLength, (unsigned)savings[maxLength], (double)savings[maxLength] / (double)maxLength);
+                     (unsigned)pos, (unsigned)maxLength, (unsigned)savings[maxLength], (double)savings[maxLength] / maxLength);
 
         solution.pos = (U32)pos;
         solution.length = (U32)maxLength;
@@ -321,7 +352,7 @@ static dictItem ZDICT_analyzePos(
         {   U32 id;
             for (id=start; id<end; id++) {
                 U32 p, pEnd, length;
-                U32 const testedPos = (U32)suffix[id];
+                U32 const testedPos = suffix[id];
                 if (testedPos == pos)
                     length = solution.length;
                 else {
@@ -373,7 +404,7 @@ static U32 ZDICT_tryMerge(dictItem* table, dictItem elt, U32 eltNbToSkip, const
             elt = table[u];
             /* sort : improve rank */
             while ((u>1) && (table[u-1].savings < elt.savings))
-                table[u] = table[u-1], u--;
+            table[u] = table[u-1], u--;
             table[u] = elt;
             return u;
     }   }
@@ -384,7 +415,7 @@ static U32 ZDICT_tryMerge(dictItem* table, dictItem elt, U32 eltNbToSkip, const
 
         if ((table[u].pos + table[u].length >= elt.pos) && (table[u].pos < elt.pos)) {  /* overlap, existing < new */
             /* append */
-            int const addedLength = (int)eltEnd - (int)(table[u].pos + table[u].length);
+            int const addedLength = (int)eltEnd - (table[u].pos + table[u].length);
             table[u].savings += elt.length / 8;    /* rough approx bonus */
             if (addedLength > 0) {   /* otherwise, elt fully included into existing */
                 table[u].length += addedLength;
@@ -477,17 +508,10 @@ static size_t ZDICT_trainBuffer_legacy(dictItem* dictList, U32 dictListSize,
     clock_t displayClock = 0;
     clock_t const refreshRate = CLOCKS_PER_SEC * 3 / 10;
 
-#   undef  DISPLAYUPDATE
-#   define DISPLAYUPDATE(l, ...)                                   \
-        do {                                                       \
-            if (notificationLevel>=l) {                            \
-                if (ZDICT_clockSpan(displayClock) > refreshRate) { \
-                    displayClock = clock();                        \
-                    DISPLAY(__VA_ARGS__);                          \
-                }                                                  \
-                if (notificationLevel>=4) fflush(stderr);          \
-            }                                                      \
-        } while (0)
+#   define DISPLAYUPDATE(l, ...) if (notificationLevel>=l) { \
+            if (ZDICT_clockSpan(displayClock) > refreshRate)  \
+            { displayClock = clock(); DISPLAY(__VA_ARGS__); \
+            if (notificationLevel>=4) fflush(stderr); } }
 
     /* init */
     DISPLAYLEVEL(2, "\r%70s\r", "");   /* clean display line */
@@ -530,7 +554,7 @@ static size_t ZDICT_trainBuffer_legacy(dictItem* dictList, U32 dictListSize,
             if (solution.length==0) { cursor++; continue; }
             ZDICT_insertDictItem(dictList, dictListSize, solution, buffer);
             cursor += solution.length;
-            DISPLAYUPDATE(2, "\r%4.2f %% \r", (double)cursor / (double)bufferSize * 100.0);
+            DISPLAYUPDATE(2, "\r%4.2f %% \r", (double)cursor / bufferSize * 100);
     }   }
 
 _cleanup:
@@ -564,20 +588,20 @@ typedef struct
 
 #define MAXREPOFFSET 1024
 
-static void ZDICT_countEStats(EStats_ress_t esr, const ZSTD_parameters* params,
+static void ZDICT_countEStats(EStats_ress_t esr, ZSTD_parameters params,
                               unsigned* countLit, unsigned* offsetcodeCount, unsigned* matchlengthCount, unsigned* litlengthCount, U32* repOffsets,
                               const void* src, size_t srcSize,
                               U32 notificationLevel)
 {
-    size_t const blockSizeMax = MIN (ZSTD_BLOCKSIZE_MAX, 1 << params->cParams.windowLog);
+    size_t const blockSizeMax = MIN (ZSTD_BLOCKSIZE_MAX, 1 << params.cParams.windowLog);
     size_t cSize;
 
     if (srcSize > blockSizeMax) srcSize = blockSizeMax;   /* protection vs large samples */
-    {   size_t const errorCode = ZSTD_compressBegin_usingCDict_deprecated(esr.zc, esr.dict);
+    {   size_t const errorCode = ZSTD_compressBegin_usingCDict(esr.zc, esr.dict);
         if (ZSTD_isError(errorCode)) { DISPLAYLEVEL(1, "warning : ZSTD_compressBegin_usingCDict failed \n"); return; }
 
     }
-    cSize = ZSTD_compressBlock_deprecated(esr.zc, esr.workPlace, ZSTD_BLOCKSIZE_MAX, src, srcSize);
+    cSize = ZSTD_compressBlock(esr.zc, esr.workPlace, ZSTD_BLOCKSIZE_MAX, src, srcSize);
     if (ZSTD_isError(cSize)) { DISPLAYLEVEL(3, "warning : could not compress sample size %u \n", (unsigned)srcSize); return; }
 
     if (cSize) {  /* if == 0; block is not compressible */
@@ -610,8 +634,8 @@ static void ZDICT_countEStats(EStats_ress_t esr, const ZSTD_parameters* params,
 
             if (nbSeq >= 2) { /* rep offsets */
                 const seqDef* const seq = seqStorePtr->sequencesStart;
-                U32 offset1 = seq[0].offBase - ZSTD_REP_NUM;
-                U32 offset2 = seq[1].offBase - ZSTD_REP_NUM;
+                U32 offset1 = seq[0].offset - 3;
+                U32 offset2 = seq[1].offset - 3;
                 if (offset1 >= MAXREPOFFSET) offset1 = 0;
                 if (offset2 >= MAXREPOFFSET) offset2 = 0;
                 repOffsets[offset1] += 3;
@@ -658,7 +682,7 @@ static void ZDICT_flatLit(unsigned* countLit)
 
 #define OFFCODE_MAX 30  /* only applicable to first block */
 static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
-                                   int compressionLevel,
+                                   unsigned compressionLevel,
                              const void*  srcBuffer, const size_t* fileSizes, unsigned nbFiles,
                              const void* dictBuffer, size_t  dictBufferSize,
                                    unsigned notificationLevel)
@@ -682,7 +706,6 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     size_t const totalSrcSize = ZDICT_totalSampleSize(fileSizes, nbFiles);
     size_t const averageSampleSize = totalSrcSize / (nbFiles + !nbFiles);
     BYTE* dstPtr = (BYTE*)dstBuffer;
-    U32 wksp[HUF_CTABLE_WORKSPACE_SIZE_U32];
 
     /* init */
     DEBUGLOG(4, "ZDICT_analyzeEntropy");
@@ -694,7 +717,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     memset(repOffset, 0, sizeof(repOffset));
     repOffset[1] = repOffset[4] = repOffset[8] = 1;
     memset(bestRepOffset, 0, sizeof(bestRepOffset));
-    if (compressionLevel==0) compressionLevel = ZSTD_CLEVEL_DEFAULT;
+    if (compressionLevel==0) compressionLevel = g_compressionLevel_default;
     params = ZSTD_getParams(compressionLevel, averageSampleSize, dictBufferSize);
 
     esr.dict = ZSTD_createCDict_advanced(dictBuffer, dictBufferSize, ZSTD_dlm_byRef, ZSTD_dct_rawContent, params.cParams, ZSTD_defaultCMem);
@@ -708,22 +731,15 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
 
     /* collect stats on all samples */
     for (u=0; u<nbFiles; u++) {
-        ZDICT_countEStats(esr, &params,
+        ZDICT_countEStats(esr, params,
                           countLit, offcodeCount, matchLengthCount, litLengthCount, repOffset,
                          (const char*)srcBuffer + pos, fileSizes[u],
                           notificationLevel);
         pos += fileSizes[u];
     }
 
-    if (notificationLevel >= 4) {
-        /* writeStats */
-        DISPLAYLEVEL(4, "Offset Code Frequencies : \n");
-        for (u=0; u<=offcodeMax; u++) {
-            DISPLAYLEVEL(4, "%2u :%7u \n", u, offcodeCount[u]);
-    }   }
-
     /* analyze, build stats, starting with literals */
-    {   size_t maxNbBits = HUF_buildCTable_wksp(hufTable, countLit, 255, huffLog, wksp, sizeof(wksp));
+    {   size_t maxNbBits = HUF_buildCTable (hufTable, countLit, 255, huffLog);
         if (HUF_isError(maxNbBits)) {
             eSize = maxNbBits;
             DISPLAYLEVEL(1, " HUF_buildCTable error \n");
@@ -732,7 +748,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
         if (maxNbBits==8) {  /* not compressible : will fail on HUF_writeCTable() */
             DISPLAYLEVEL(2, "warning : pathological dataset : literals are not compressible : samples are noisy or too regular \n");
             ZDICT_flatLit(countLit);  /* replace distribution by a fake "mostly flat but still compressible" distribution, that HUF_writeCTable() can encode */
-            maxNbBits = HUF_buildCTable_wksp(hufTable, countLit, 255, huffLog, wksp, sizeof(wksp));
+            maxNbBits = HUF_buildCTable (hufTable, countLit, 255, huffLog);
             assert(maxNbBits==9);
         }
         huffLog = (U32)maxNbBits;
@@ -746,7 +762,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     /* note : the result of this phase should be used to better appreciate the impact on statistics */
 
     total=0; for (u=0; u<=offcodeMax; u++) total+=offcodeCount[u];
-    errorCode = FSE_normalizeCount(offcodeNCount, Offlog, offcodeCount, total, offcodeMax, /* useLowProbCount */ 1);
+    errorCode = FSE_normalizeCount(offcodeNCount, Offlog, offcodeCount, total, offcodeMax);
     if (FSE_isError(errorCode)) {
         eSize = errorCode;
         DISPLAYLEVEL(1, "FSE_normalizeCount error with offcodeCount \n");
@@ -755,7 +771,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     Offlog = (U32)errorCode;
 
     total=0; for (u=0; u<=MaxML; u++) total+=matchLengthCount[u];
-    errorCode = FSE_normalizeCount(matchLengthNCount, mlLog, matchLengthCount, total, MaxML, /* useLowProbCount */ 1);
+    errorCode = FSE_normalizeCount(matchLengthNCount, mlLog, matchLengthCount, total, MaxML);
     if (FSE_isError(errorCode)) {
         eSize = errorCode;
         DISPLAYLEVEL(1, "FSE_normalizeCount error with matchLengthCount \n");
@@ -764,7 +780,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     mlLog = (U32)errorCode;
 
     total=0; for (u=0; u<=MaxLL; u++) total+=litLengthCount[u];
-    errorCode = FSE_normalizeCount(litLengthNCount, llLog, litLengthCount, total, MaxLL, /* useLowProbCount */ 1);
+    errorCode = FSE_normalizeCount(litLengthNCount, llLog, litLengthCount, total, MaxLL);
     if (FSE_isError(errorCode)) {
         eSize = errorCode;
         DISPLAYLEVEL(1, "FSE_normalizeCount error with litLengthCount \n");
@@ -773,7 +789,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     llLog = (U32)errorCode;
 
     /* write result to buffer */
-    {   size_t const hhSize = HUF_writeCTable_wksp(dstPtr, maxDstSize, hufTable, 255, huffLog, wksp, sizeof(wksp));
+    {   size_t const hhSize = HUF_writeCTable(dstPtr, maxDstSize, hufTable, 255, huffLog);
         if (HUF_isError(hhSize)) {
             eSize = hhSize;
             DISPLAYLEVEL(1, "HUF_writeCTable error \n");
@@ -828,7 +844,7 @@ static size_t ZDICT_analyzeEntropy(void*  dstBuffer, size_t maxDstSize,
     MEM_writeLE32(dstPtr+8, bestRepOffset[2].offset);
 #else
     /* at this stage, we don't use the result of "most common first offset",
-     * as the impact of statistics is not properly evaluated */
+       as the impact of statistics is not properly evaluated */
     MEM_writeLE32(dstPtr+0, repStartValue[0]);
     MEM_writeLE32(dstPtr+4, repStartValue[1]);
     MEM_writeLE32(dstPtr+8, repStartValue[2]);
@@ -844,17 +860,6 @@ _cleanup:
 }
 
 
-/**
- * @returns the maximum repcode value
- */
-static U32 ZDICT_maxRep(U32 const reps[ZSTD_REP_NUM])
-{
-    U32 maxRep = reps[0];
-    int r;
-    for (r = 1; r < ZSTD_REP_NUM; ++r)
-        maxRep = MAX(maxRep, reps[r]);
-    return maxRep;
-}
 
 size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity,
                           const void* customDictContent, size_t dictContentSize,
@@ -864,15 +869,13 @@ size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity,
     size_t hSize;
 #define HBUFFSIZE 256   /* should prove large enough for all entropy headers */
     BYTE header[HBUFFSIZE];
-    int const compressionLevel = (params.compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : params.compressionLevel;
+    int const compressionLevel = (params.compressionLevel == 0) ? g_compressionLevel_default : params.compressionLevel;
     U32 const notificationLevel = params.notificationLevel;
-    /* The final dictionary content must be at least as large as the largest repcode */
-    size_t const minContentSize = (size_t)ZDICT_maxRep(repStartValue);
-    size_t paddingSize;
 
     /* check conditions */
     DEBUGLOG(4, "ZDICT_finalizeDictionary");
     if (dictBufferCapacity < dictContentSize) return ERROR(dstSize_tooSmall);
+    if (dictContentSize < ZDICT_CONTENTSIZE_MIN) return ERROR(srcSize_wrong);
     if (dictBufferCapacity < ZDICT_DICTSIZE_MIN) return ERROR(dstSize_tooSmall);
 
     /* dictionary header */
@@ -896,43 +899,12 @@ size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity,
         hSize += eSize;
     }
 
-    /* Shrink the content size if it doesn't fit in the buffer */
-    if (hSize + dictContentSize > dictBufferCapacity) {
-        dictContentSize = dictBufferCapacity - hSize;
-    }
-
-    /* Pad the dictionary content with zeros if it is too small */
-    if (dictContentSize < minContentSize) {
-        RETURN_ERROR_IF(hSize + minContentSize > dictBufferCapacity, dstSize_tooSmall,
-                        "dictBufferCapacity too small to fit max repcode");
-        paddingSize = minContentSize - dictContentSize;
-    } else {
-        paddingSize = 0;
-    }
-
-    {
-        size_t const dictSize = hSize + paddingSize + dictContentSize;
-
-        /* The dictionary consists of the header, optional padding, and the content.
-         * The padding comes before the content because the "best" position in the
-         * dictionary is the last byte.
-         */
-        BYTE* const outDictHeader = (BYTE*)dictBuffer;
-        BYTE* const outDictPadding = outDictHeader + hSize;
-        BYTE* const outDictContent = outDictPadding + paddingSize;
-
-        assert(dictSize <= dictBufferCapacity);
-        assert(outDictContent + dictContentSize == (BYTE*)dictBuffer + dictSize);
-
-        /* First copy the customDictContent into its final location.
-         * `customDictContent` and `dictBuffer` may overlap, so we must
-         * do this before any other writes into the output buffer.
-         * Then copy the header & padding into the output buffer.
-         */
-        memmove(outDictContent, customDictContent, dictContentSize);
-        memcpy(outDictHeader, header, hSize);
-        memset(outDictPadding, 0, paddingSize);
-
+    /* copy elements in final buffer ; note : src and dst buffer can overlap */
+    if (hSize + dictContentSize > dictBufferCapacity) dictContentSize = dictBufferCapacity - hSize;
+    {   size_t const dictSize = hSize + dictContentSize;
+        char* dictEnd = (char*)dictBuffer + dictSize;
+        memmove(dictEnd - dictContentSize, customDictContent, dictContentSize);
+        memcpy(dictBuffer, header, hSize);
         return dictSize;
     }
 }
@@ -943,7 +915,7 @@ static size_t ZDICT_addEntropyTablesFromBuffer_advanced(
         const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
         ZDICT_params_t params)
 {
-    int const compressionLevel = (params.compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : params.compressionLevel;
+    int const compressionLevel = (params.compressionLevel == 0) ? g_compressionLevel_default : params.compressionLevel;
     U32 const notificationLevel = params.notificationLevel;
     size_t hSize = 8;
 
@@ -972,11 +944,16 @@ static size_t ZDICT_addEntropyTablesFromBuffer_advanced(
     return MIN(dictBufferCapacity, hSize+dictContentSize);
 }
 
+/* Hidden declaration for dbio.c */
+size_t ZDICT_trainFromBuffer_unsafe_legacy(
+                            void* dictBuffer, size_t maxDictSize,
+                            const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
+                            ZDICT_legacy_params_t params);
 /*! ZDICT_trainFromBuffer_unsafe_legacy() :
-*   Warning : `samplesBuffer` must be followed by noisy guard band !!!
+*   Warning : `samplesBuffer` must be followed by noisy guard band.
 *   @return : size of dictionary, or an error code which can be tested with ZDICT_isError()
 */
-static size_t ZDICT_trainFromBuffer_unsafe_legacy(
+size_t ZDICT_trainFromBuffer_unsafe_legacy(
                             void* dictBuffer, size_t maxDictSize,
                             const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
                             ZDICT_legacy_params_t params)
@@ -1113,8 +1090,8 @@ size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity,
     memset(&params, 0, sizeof(params));
     params.d = 8;
     params.steps = 4;
-    /* Use default level since no compression level information is available */
-    params.zParams.compressionLevel = ZSTD_CLEVEL_DEFAULT;
+    /* Default to level 6 since no compression level information is available */
+    params.zParams.compressionLevel = 3;
 #if defined(DEBUGLEVEL) && (DEBUGLEVEL>=1)
     params.zParams.notificationLevel = DEBUGLEVEL;
 #endif
@@ -1132,5 +1109,3 @@ size_t ZDICT_addEntropyTablesFromBuffer(void* dictBuffer, size_t dictContentSize
                                                      samplesBuffer, samplesSizes, nbSamples,
                                                      params);
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zdict.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -9,184 +8,34 @@
  * You may select, at your option, one of the above-listed licenses.
  */
 
+#ifndef DICTBUILDER_H_001
+#define DICTBUILDER_H_001
+
 #if defined (__cplusplus)
 extern "C" {
 #endif
 
-#ifndef ZSTD_ZDICT_H
-#define ZSTD_ZDICT_H
 
 /*======  Dependencies  ======*/
 #include <stddef.h>  /* size_t */
 
 
 /* =====   ZDICTLIB_API : control library symbols visibility   ===== */
-#ifndef ZDICTLIB_VISIBLE
-   /* Backwards compatibility with old macro name */
-#  ifdef ZDICTLIB_VISIBILITY
-#    define ZDICTLIB_VISIBLE ZDICTLIB_VISIBILITY
-#  elif defined(__GNUC__) && (__GNUC__ >= 4) && !defined(__MINGW32__)
-#    define ZDICTLIB_VISIBLE __attribute__ ((visibility ("default")))
+#ifndef ZDICTLIB_VISIBILITY
+#  if defined(__GNUC__) && (__GNUC__ >= 4)
+#    define ZDICTLIB_VISIBILITY __attribute__ ((visibility ("default")))
 #  else
-#    define ZDICTLIB_VISIBLE
+#    define ZDICTLIB_VISIBILITY
 #  endif
 #endif
-
-#ifndef ZDICTLIB_HIDDEN
-#  if defined(__GNUC__) && (__GNUC__ >= 4) && !defined(__MINGW32__)
-#    define ZDICTLIB_HIDDEN __attribute__ ((visibility ("hidden")))
-#  else
-#    define ZDICTLIB_HIDDEN
-#  endif
-#endif
-
 #if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1)
-#  define ZDICTLIB_API __declspec(dllexport) ZDICTLIB_VISIBLE
+#  define ZDICTLIB_API __declspec(dllexport) ZDICTLIB_VISIBILITY
 #elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1)
-#  define ZDICTLIB_API __declspec(dllimport) ZDICTLIB_VISIBLE /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
+#  define ZDICTLIB_API __declspec(dllimport) ZDICTLIB_VISIBILITY /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
 #else
-#  define ZDICTLIB_API ZDICTLIB_VISIBLE
+#  define ZDICTLIB_API ZDICTLIB_VISIBILITY
 #endif
 
-/*******************************************************************************
- * Zstd dictionary builder
- *
- * FAQ
- * ===
- * Why should I use a dictionary?
- * ------------------------------
- *
- * Zstd can use dictionaries to improve compression ratio of small data.
- * Traditionally small files don't compress well because there is very little
- * repetition in a single sample, since it is small. But, if you are compressing
- * many similar files, like a bunch of JSON records that share the same
- * structure, you can train a dictionary on ahead of time on some samples of
- * these files. Then, zstd can use the dictionary to find repetitions that are
- * present across samples. This can vastly improve compression ratio.
- *
- * When is a dictionary useful?
- * ----------------------------
- *
- * Dictionaries are useful when compressing many small files that are similar.
- * The larger a file is, the less benefit a dictionary will have. Generally,
- * we don't expect dictionary compression to be effective past 100KB. And the
- * smaller a file is, the more we would expect the dictionary to help.
- *
- * How do I use a dictionary?
- * --------------------------
- *
- * Simply pass the dictionary to the zstd compressor with
- * `ZSTD_CCtx_loadDictionary()`. The same dictionary must then be passed to
- * the decompressor, using `ZSTD_DCtx_loadDictionary()`. There are other
- * more advanced functions that allow selecting some options, see zstd.h for
- * complete documentation.
- *
- * What is a zstd dictionary?
- * --------------------------
- *
- * A zstd dictionary has two pieces: Its header, and its content. The header
- * contains a magic number, the dictionary ID, and entropy tables. These
- * entropy tables allow zstd to save on header costs in the compressed file,
- * which really matters for small data. The content is just bytes, which are
- * repeated content that is common across many samples.
- *
- * What is a raw content dictionary?
- * ---------------------------------
- *
- * A raw content dictionary is just bytes. It doesn't have a zstd dictionary
- * header, a dictionary ID, or entropy tables. Any buffer is a valid raw
- * content dictionary.
- *
- * How do I train a dictionary?
- * ----------------------------
- *
- * Gather samples from your use case. These samples should be similar to each
- * other. If you have several use cases, you could try to train one dictionary
- * per use case.
- *
- * Pass those samples to `ZDICT_trainFromBuffer()` and that will train your
- * dictionary. There are a few advanced versions of this function, but this
- * is a great starting point. If you want to further tune your dictionary
- * you could try `ZDICT_optimizeTrainFromBuffer_cover()`. If that is too slow
- * you can try `ZDICT_optimizeTrainFromBuffer_fastCover()`.
- *
- * If the dictionary training function fails, that is likely because you
- * either passed too few samples, or a dictionary would not be effective
- * for your data. Look at the messages that the dictionary trainer printed,
- * if it doesn't say too few samples, then a dictionary would not be effective.
- *
- * How large should my dictionary be?
- * ----------------------------------
- *
- * A reasonable dictionary size, the `dictBufferCapacity`, is about 100KB.
- * The zstd CLI defaults to a 110KB dictionary. You likely don't need a
- * dictionary larger than that. But, most use cases can get away with a
- * smaller dictionary. The advanced dictionary builders can automatically
- * shrink the dictionary for you, and select the smallest size that doesn't
- * hurt compression ratio too much. See the `shrinkDict` parameter.
- * A smaller dictionary can save memory, and potentially speed up
- * compression.
- *
- * How many samples should I provide to the dictionary builder?
- * ------------------------------------------------------------
- *
- * We generally recommend passing ~100x the size of the dictionary
- * in samples. A few thousand should suffice. Having too few samples
- * can hurt the dictionaries effectiveness. Having more samples will
- * only improve the dictionaries effectiveness. But having too many
- * samples can slow down the dictionary builder.
- *
- * How do I determine if a dictionary will be effective?
- * -----------------------------------------------------
- *
- * Simply train a dictionary and try it out. You can use zstd's built in
- * benchmarking tool to test the dictionary effectiveness.
- *
- *   # Benchmark levels 1-3 without a dictionary
- *   zstd -b1e3 -r /path/to/my/files
- *   # Benchmark levels 1-3 with a dictionary
- *   zstd -b1e3 -r /path/to/my/files -D /path/to/my/dictionary
- *
- * When should I retrain a dictionary?
- * -----------------------------------
- *
- * You should retrain a dictionary when its effectiveness drops. Dictionary
- * effectiveness drops as the data you are compressing changes. Generally, we do
- * expect dictionaries to "decay" over time, as your data changes, but the rate
- * at which they decay depends on your use case. Internally, we regularly
- * retrain dictionaries, and if the new dictionary performs significantly
- * better than the old dictionary, we will ship the new dictionary.
- *
- * I have a raw content dictionary, how do I turn it into a zstd dictionary?
- * -------------------------------------------------------------------------
- *
- * If you have a raw content dictionary, e.g. by manually constructing it, or
- * using a third-party dictionary builder, you can turn it into a zstd
- * dictionary by using `ZDICT_finalizeDictionary()`. You'll also have to
- * provide some samples of the data. It will add the zstd header to the
- * raw content, which contains a dictionary ID and entropy tables, which
- * will improve compression ratio, and allow zstd to write the dictionary ID
- * into the frame, if you so choose.
- *
- * Do I have to use zstd's dictionary builder?
- * -------------------------------------------
- *
- * No! You can construct dictionary content however you please, it is just
- * bytes. It will always be valid as a raw content dictionary. If you want
- * a zstd dictionary, which can improve compression ratio, use
- * `ZDICT_finalizeDictionary()`.
- *
- * What is the attack surface of a zstd dictionary?
- * ------------------------------------------------
- *
- * Zstd is heavily fuzz tested, including loading fuzzed dictionaries, so
- * zstd should never crash, or access out-of-bounds memory no matter what
- * the dictionary is. However, if an attacker can control the dictionary
- * during decompression, they can cause zstd to generate arbitrary bytes,
- * just like if they controlled the compressed data.
- *
- ******************************************************************************/
-
 
 /*! ZDICT_trainFromBuffer():
  *  Train a dictionary from an array of samples.
@@ -212,81 +61,15 @@ ZDICTLIB_API size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCap
                                     const void* samplesBuffer,
                                     const size_t* samplesSizes, unsigned nbSamples);
 
-typedef struct {
-    int      compressionLevel;   /**< optimize for a specific zstd compression level; 0 means default */
-    unsigned notificationLevel;  /**< Write log to stderr; 0 = none (default); 1 = errors; 2 = progression; 3 = details; 4 = debug; */
-    unsigned dictID;             /**< force dictID value; 0 means auto mode (32-bits random value)
-                                  *   NOTE: The zstd format reserves some dictionary IDs for future use.
-                                  *         You may use them in private settings, but be warned that they
-                                  *         may be used by zstd in a public dictionary registry in the future.
-                                  *         These dictionary IDs are:
-                                  *           - low range  : <= 32767
-                                  *           - high range : >= (2^31)
-                                  */
-} ZDICT_params_t;
-
-/*! ZDICT_finalizeDictionary():
- * Given a custom content as a basis for dictionary, and a set of samples,
- * finalize dictionary by adding headers and statistics according to the zstd
- * dictionary format.
- *
- * Samples must be stored concatenated in a flat buffer `samplesBuffer`,
- * supplied with an array of sizes `samplesSizes`, providing the size of each
- * sample in order. The samples are used to construct the statistics, so they
- * should be representative of what you will compress with this dictionary.
- *
- * The compression level can be set in `parameters`. You should pass the
- * compression level you expect to use in production. The statistics for each
- * compression level differ, so tuning the dictionary for the compression level
- * can help quite a bit.
- *
- * You can set an explicit dictionary ID in `parameters`, or allow us to pick
- * a random dictionary ID for you, but we can't guarantee no collisions.
- *
- * The dstDictBuffer and the dictContent may overlap, and the content will be
- * appended to the end of the header. If the header + the content doesn't fit in
- * maxDictSize the beginning of the content is truncated to make room, since it
- * is presumed that the most profitable content is at the end of the dictionary,
- * since that is the cheapest to reference.
- *
- * `maxDictSize` must be >= max(dictContentSize, ZSTD_DICTSIZE_MIN).
- *
- * @return: size of dictionary stored into `dstDictBuffer` (<= `maxDictSize`),
- *          or an error code, which can be tested by ZDICT_isError().
- * Note: ZDICT_finalizeDictionary() will push notifications into stderr if
- *       instructed to, using notificationLevel>0.
- * NOTE: This function currently may fail in several edge cases including:
- *         * Not enough samples
- *         * Samples are uncompressible
- *         * Samples are all exactly the same
- */
-ZDICTLIB_API size_t ZDICT_finalizeDictionary(void* dstDictBuffer, size_t maxDictSize,
-                                const void* dictContent, size_t dictContentSize,
-                                const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
-                                ZDICT_params_t parameters);
-
 
 /*======   Helper functions   ======*/
 ZDICTLIB_API unsigned ZDICT_getDictID(const void* dictBuffer, size_t dictSize);  /**< extracts dictID; @return zero if error (not a valid dictionary) */
-ZDICTLIB_API size_t ZDICT_getDictHeaderSize(const void* dictBuffer, size_t dictSize);  /* returns dict header size; returns a ZSTD error code on failure */
 ZDICTLIB_API unsigned ZDICT_isError(size_t errorCode);
 ZDICTLIB_API const char* ZDICT_getErrorName(size_t errorCode);
 
-#endif   /* ZSTD_ZDICT_H */
 
-#if defined(ZDICT_STATIC_LINKING_ONLY) && !defined(ZSTD_ZDICT_H_STATIC)
-#define ZSTD_ZDICT_H_STATIC
 
-/* This can be overridden externally to hide static symbols. */
-#ifndef ZDICTLIB_STATIC_API
-#  if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1)
-#    define ZDICTLIB_STATIC_API __declspec(dllexport) ZDICTLIB_VISIBLE
-#  elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1)
-#    define ZDICTLIB_STATIC_API __declspec(dllimport) ZDICTLIB_VISIBLE
-#  else
-#    define ZDICTLIB_STATIC_API ZDICTLIB_VISIBLE
-#  endif
-#endif
+#ifdef ZDICT_STATIC_LINKING_ONLY
 
 /* ====================================================================================
  * The definitions in this section are considered experimental.
@@ -295,9 +78,11 @@ ZDICTLIB_API const char* ZDICT_getErrorName(size_t errorCode);
  * Use them only in association with static linking.
  * ==================================================================================== */
 
-#define ZDICT_DICTSIZE_MIN    256
-/* Deprecated: Remove in v1.6.0 */
-#define ZDICT_CONTENTSIZE_MIN 128
+typedef struct {
+    int      compressionLevel;   /* optimize for a specific zstd compression level; 0 means default */
+    unsigned notificationLevel;  /* Write log to stderr; 0 = none (default); 1 = errors; 2 = progression; 3 = details; 4 = debug; */
+    unsigned dictID;             /* force dictID value; 0 means auto mode (32-bits random value) */
+} ZDICT_params_t;
 
 /*! ZDICT_cover_params_t:
  *  k and d are the only required parameters.
@@ -342,7 +127,7 @@ typedef struct {
  *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
  *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
  */
-ZDICTLIB_STATIC_API size_t ZDICT_trainFromBuffer_cover(
+ZDICTLIB_API size_t ZDICT_trainFromBuffer_cover(
           void *dictBuffer, size_t dictBufferCapacity,
     const void *samplesBuffer, const size_t *samplesSizes, unsigned nbSamples,
           ZDICT_cover_params_t parameters);
@@ -364,7 +149,7 @@ ZDICTLIB_STATIC_API size_t ZDICT_trainFromBuffer_cover(
  *          See ZDICT_trainFromBuffer() for details on failure modes.
  * Note: ZDICT_optimizeTrainFromBuffer_cover() requires about 8 bytes of memory for each input byte and additionally another 5 bytes of memory for each byte of memory for each thread.
  */
-ZDICTLIB_STATIC_API size_t ZDICT_optimizeTrainFromBuffer_cover(
+ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_cover(
           void* dictBuffer, size_t dictBufferCapacity,
     const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
           ZDICT_cover_params_t* parameters);
@@ -385,7 +170,7 @@ ZDICTLIB_STATIC_API size_t ZDICT_optimizeTrainFromBuffer_cover(
  *        In general, it's recommended to provide a few thousands samples, though this can vary a lot.
  *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
  */
-ZDICTLIB_STATIC_API size_t ZDICT_trainFromBuffer_fastCover(void *dictBuffer,
+ZDICTLIB_API size_t ZDICT_trainFromBuffer_fastCover(void *dictBuffer,
                     size_t dictBufferCapacity, const void *samplesBuffer,
                     const size_t *samplesSizes, unsigned nbSamples,
                     ZDICT_fastCover_params_t parameters);
@@ -408,11 +193,33 @@ ZDICTLIB_STATIC_API size_t ZDICT_trainFromBuffer_fastCover(void *dictBuffer,
  *          See ZDICT_trainFromBuffer() for details on failure modes.
  * Note: ZDICT_optimizeTrainFromBuffer_fastCover() requires about 6 * 2^f bytes of memory for each thread.
  */
-ZDICTLIB_STATIC_API size_t ZDICT_optimizeTrainFromBuffer_fastCover(void* dictBuffer,
+ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_fastCover(void* dictBuffer,
                     size_t dictBufferCapacity, const void* samplesBuffer,
                     const size_t* samplesSizes, unsigned nbSamples,
                     ZDICT_fastCover_params_t* parameters);
 
+/*! ZDICT_finalizeDictionary():
+ * Given a custom content as a basis for dictionary, and a set of samples,
+ * finalize dictionary by adding headers and statistics.
+ *
+ * Samples must be stored concatenated in a flat buffer `samplesBuffer`,
+ * supplied with an array of sizes `samplesSizes`, providing the size of each sample in order.
+ *
+ * dictContentSize must be >= ZDICT_CONTENTSIZE_MIN bytes.
+ * maxDictSize must be >= dictContentSize, and must be >= ZDICT_DICTSIZE_MIN bytes.
+ *
+ * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`),
+ *          or an error code, which can be tested by ZDICT_isError().
+ * Note: ZDICT_finalizeDictionary() will push notifications into stderr if instructed to, using notificationLevel>0.
+ * Note 2: dictBuffer and dictContent can overlap
+ */
+#define ZDICT_CONTENTSIZE_MIN 128
+#define ZDICT_DICTSIZE_MIN    256
+ZDICTLIB_API size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity,
+                                const void* dictContent, size_t dictContentSize,
+                                const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
+                                ZDICT_params_t parameters);
+
 typedef struct {
     unsigned selectivityLevel;   /* 0 means default; larger => select more => larger dictionary */
     ZDICT_params_t zParams;
@@ -433,45 +240,43 @@ typedef struct {
  *        It's recommended that total size of all samples be about ~x100 times the target size of dictionary.
  *  Note: ZDICT_trainFromBuffer_legacy() will send notifications into stderr if instructed to, using notificationLevel>0.
  */
-ZDICTLIB_STATIC_API size_t ZDICT_trainFromBuffer_legacy(
-    void* dictBuffer, size_t dictBufferCapacity,
-    const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples,
+ZDICTLIB_API size_t ZDICT_trainFromBuffer_legacy(
+    void *dictBuffer, size_t dictBufferCapacity,
+    const void *samplesBuffer, const size_t *samplesSizes, unsigned nbSamples,
     ZDICT_legacy_params_t parameters);
 
-
 /* Deprecation warnings */
 /* It is generally possible to disable deprecation warnings from compiler,
    for example with -Wno-deprecated-declarations for gcc
    or _CRT_SECURE_NO_WARNINGS in Visual.
    Otherwise, it's also possible to manually define ZDICT_DISABLE_DEPRECATE_WARNINGS */
 #ifdef ZDICT_DISABLE_DEPRECATE_WARNINGS
-#  define ZDICT_DEPRECATED(message) /* disable deprecation warnings */
+#  define ZDICT_DEPRECATED(message) ZDICTLIB_API   /* disable deprecation warnings */
 #else
 #  define ZDICT_GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
 #  if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */
-#    define ZDICT_DEPRECATED(message) [[deprecated(message)]]
-#  elif defined(__clang__) || (ZDICT_GCC_VERSION >= 405)
-#    define ZDICT_DEPRECATED(message) __attribute__((deprecated(message)))
+#    define ZDICT_DEPRECATED(message) [[deprecated(message)]] ZDICTLIB_API
+#  elif (ZDICT_GCC_VERSION >= 405) || defined(__clang__)
+#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated(message)))
 #  elif (ZDICT_GCC_VERSION >= 301)
-#    define ZDICT_DEPRECATED(message) __attribute__((deprecated))
+#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated))
 #  elif defined(_MSC_VER)
-#    define ZDICT_DEPRECATED(message) __declspec(deprecated(message))
+#    define ZDICT_DEPRECATED(message) ZDICTLIB_API __declspec(deprecated(message))
 #  else
 #    pragma message("WARNING: You need to implement ZDICT_DEPRECATED for this compiler")
-#    define ZDICT_DEPRECATED(message)
+#    define ZDICT_DEPRECATED(message) ZDICTLIB_API
 #  endif
 #endif /* ZDICT_DISABLE_DEPRECATE_WARNINGS */
 
 ZDICT_DEPRECATED("use ZDICT_finalizeDictionary() instead")
-ZDICTLIB_STATIC_API
 size_t ZDICT_addEntropyTablesFromBuffer(void* dictBuffer, size_t dictContentSize, size_t dictBufferCapacity,
                                   const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples);
 
 
-#endif   /* ZSTD_ZDICT_H_STATIC */
+#endif   /* ZDICT_STATIC_LINKING_ONLY */
 
 #if defined (__cplusplus)
 }
 #endif
 
-#endif /* USE_EXTERNAL_ZSTD */
+#endif   /* DICTBUILDER_H_001 */

vendor/github.com/DataDog/zstd/zstd.go

@@ -1,18 +1,29 @@
 package zstd
 
 /*
-// support decoding of "legacy" zstd payloads from versions [0.4, 0.8], matching the
-// default configuration of the zstd command line tool:
-// https://github.com/facebook/zstd/blob/dev/programs/README.md
-#cgo CFLAGS: -DZSTD_LEGACY_SUPPORT=4 -DZSTD_MULTITHREAD=1
-
+#define ZSTD_STATIC_LINKING_ONLY
 #include "zstd.h"
+#include "stdint.h"  // for uintptr_t
+
+// The following *_wrapper function are used for removing superflouos
+// memory allocations when calling the wrapped functions from Go code.
+// See https://github.com/golang/go/issues/24450 for details.
+
+static size_t ZSTD_compress_wrapper(uintptr_t dst, size_t maxDstSize, const uintptr_t src, size_t srcSize, int compressionLevel) {
+	return ZSTD_compress((void*)dst, maxDstSize, (const void*)src, srcSize, compressionLevel);
+}
+
+static size_t ZSTD_decompress_wrapper(uintptr_t dst, size_t maxDstSize, uintptr_t src, size_t srcSize) {
+	return ZSTD_decompress((void*)dst, maxDstSize, (const void *)src, srcSize);
+}
+
 */
 import "C"
 import (
 	"bytes"
 	"errors"
 	"io/ioutil"
+	"runtime"
 	"unsafe"
 )
 
@@ -28,17 +39,6 @@ var (
 	ErrEmptySlice = errors.New("Bytes slice is empty")
 )
 
-const (
-	// decompressSizeBufferLimit is the limit we set on creating a decompression buffer for the Decompress API
-	// This is made to prevent DOS from maliciously-created payloads (aka zipbomb).
-	// For large payloads with a compression ratio > 10, you can do your own allocation and pass it to the method:
-	// dst := make([]byte, 1GB)
-	// decompressed, err := zstd.Decompress(dst, src)
-	decompressSizeBufferLimit = 1000 * 1000
-
-	zstdFrameHeaderSizeMin = 2 // From zstd.h. Since it's experimental API, hardcoding it
-)
-
 // CompressBound returns the worst case size needed for a destination buffer,
 // which can be used to preallocate a destination buffer or select a previously
 // allocated buffer from a pool.
@@ -57,33 +57,6 @@ func cCompressBound(srcSize int) int {
 	return int(C.ZSTD_compressBound(C.size_t(srcSize)))
 }
 
-// decompressSizeHint tries to give a hint on how much of the output buffer size we should have
-// based on zstd frame descriptors. To prevent DOS from maliciously-created payloads, limit the size
-func decompressSizeHint(src []byte) int {
-	// 1 MB or 50x input size
-	upperBound := 50 * len(src)
-	if upperBound < decompressSizeBufferLimit {
-		upperBound = decompressSizeBufferLimit
-	}
-
-	hint := upperBound
-	if len(src) >= zstdFrameHeaderSizeMin {
-		hint = int(C.ZSTD_getFrameContentSize(unsafe.Pointer(&src[0]), C.size_t(len(src))))
-		if hint < 0 { // On error, just use upperBound
-			hint = upperBound
-		}
-		if hint == 0 { // When compressing the empty slice, we need an output of at least 1 to pass down to the C lib
-			hint = 1
-		}
-	}
-
-	// Take the minimum of both
-	if hint > upperBound {
-		return upperBound
-	}
-	return hint
-}
-
 // Compress src into dst.  If you have a buffer to use, you can pass it to
 // prevent allocation.  If it is too small, or if nil is passed, a new buffer
 // will be allocated and returned.
@@ -100,26 +73,19 @@ func CompressLevel(dst, src []byte, level int) ([]byte, error) {
 		dst = make([]byte, bound)
 	}
 
-	// We need unsafe.Pointer(&src[0]) in the Cgo call to avoid "Go pointer to Go pointer" panics.
-	// This means we need to special case empty input. See:
-	// https://github.com/golang/go/issues/14210#issuecomment-346402945
-	var cWritten C.size_t
-	if len(src) == 0 {
-		cWritten = C.ZSTD_compress(
-			unsafe.Pointer(&dst[0]),
-			C.size_t(len(dst)),
-			unsafe.Pointer(nil),
-			C.size_t(0),
-			C.int(level))
-	} else {
-		cWritten = C.ZSTD_compress(
-			unsafe.Pointer(&dst[0]),
-			C.size_t(len(dst)),
-			unsafe.Pointer(&src[0]),
-			C.size_t(len(src)),
-			C.int(level))
+	srcPtr := C.uintptr_t(uintptr(0)) // Do not point anywhere, if src is empty
+	if len(src) > 0 {
+		srcPtr = C.uintptr_t(uintptr(unsafe.Pointer(&src[0])))
 	}
 
+	cWritten := C.ZSTD_compress_wrapper(
+		C.uintptr_t(uintptr(unsafe.Pointer(&dst[0]))),
+		C.size_t(len(dst)),
+		srcPtr,
+		C.size_t(len(src)),
+		C.int(level))
+
+	runtime.KeepAlive(src)
 	written := int(cWritten)
 	// Check if the return is an Error code
 	if err := getError(written); err != nil {
@@ -135,20 +101,43 @@ func Decompress(dst, src []byte) ([]byte, error) {
 	if len(src) == 0 {
 		return []byte{}, ErrEmptySlice
 	}
+	decompress := func(dst, src []byte) ([]byte, error) {
 
-	bound := decompressSizeHint(src)
-	if cap(dst) >= bound {
-		dst = dst[0:cap(dst)]
-	} else {
-		dst = make([]byte, bound)
-	}
+		cWritten := C.ZSTD_decompress_wrapper(
+			C.uintptr_t(uintptr(unsafe.Pointer(&dst[0]))),
+			C.size_t(len(dst)),
+			C.uintptr_t(uintptr(unsafe.Pointer(&src[0]))),
+			C.size_t(len(src)))
 
-	written, err := DecompressInto(dst, src)
-	if err == nil {
+		runtime.KeepAlive(src)
+		written := int(cWritten)
+		// Check error
+		if err := getError(written); err != nil {
+			return nil, err
+		}
 		return dst[:written], nil
 	}
-	if !IsDstSizeTooSmallError(err) {
-		return nil, err
+
+	if len(dst) == 0 {
+		// Attempt to use zStd to determine decompressed size (may result in error or 0)
+		size := int(C.size_t(C.ZSTD_getDecompressedSize(unsafe.Pointer(&src[0]), C.size_t(len(src)))))
+
+		if err := getError(size); err != nil {
+			return nil, err
+		}
+
+		if size > 0 {
+			dst = make([]byte, size)
+		} else {
+			dst = make([]byte, len(src)*3) // starting guess
+		}
+	}
+	for i := 0; i < 3; i++ { // 3 tries to allocate a bigger buffer
+		result, err := decompress(dst, src)
+		if !IsDstSizeTooSmallError(err) {
+			return result, err
+		}
+		dst = make([]byte, len(dst)*2) // Grow buffer by 2
 	}
 
 	// We failed getting a dst buffer of correct size, use stream API
@@ -156,19 +145,3 @@ func Decompress(dst, src []byte) ([]byte, error) {
 	defer r.Close()
 	return ioutil.ReadAll(r)
 }
-
-// DecompressInto decompresses src into dst. Unlike Decompress, DecompressInto
-// requires that dst be sufficiently large to hold the decompressed payload.
-// DecompressInto may be used when the caller knows the size of the decompressed
-// payload before attempting decompression.
-//
-// It returns the number of bytes copied and an error if any is encountered. If
-// dst is too small, DecompressInto errors.
-func DecompressInto(dst, src []byte) (int, error) {
-	written := int(C.ZSTD_decompress(
-		unsafe.Pointer(&dst[0]),
-		C.size_t(len(dst)),
-		unsafe.Pointer(&src[0]),
-		C.size_t(len(src))))
-	return written, getError(written)
-}

vendor/github.com/DataDog/zstd/zstd_bulk.go

@@ -1,151 +0,0 @@
-package zstd
-
-/*
-#include "zstd.h"
-*/
-import "C"
-import (
-	"errors"
-	"runtime"
-	"unsafe"
-)
-
-var (
-	// ErrEmptyDictionary is returned when the given dictionary is empty
-	ErrEmptyDictionary = errors.New("Dictionary is empty")
-	// ErrBadDictionary is returned when cannot load the given dictionary
-	ErrBadDictionary = errors.New("Cannot load dictionary")
-)
-
-// BulkProcessor implements Bulk processing dictionary API.
-// When compressing multiple messages or blocks using the same dictionary,
-// it's recommended to digest the dictionary only once, since it's a costly operation.
-// NewBulkProcessor() will create a state from digesting a dictionary.
-// The resulting state can be used for future compression/decompression operations with very limited startup cost.
-// BulkProcessor can be created once and shared by multiple threads concurrently, since its usage is read-only.
-// The state will be freed when gc cleans up BulkProcessor.
-type BulkProcessor struct {
-	cDict *C.struct_ZSTD_CDict_s
-	dDict *C.struct_ZSTD_DDict_s
-}
-
-// NewBulkProcessor creates a new BulkProcessor with a pre-trained dictionary and compression level
-func NewBulkProcessor(dictionary []byte, compressionLevel int) (*BulkProcessor, error) {
-	if len(dictionary) < 1 {
-		return nil, ErrEmptyDictionary
-	}
-
-	p := &BulkProcessor{}
-	runtime.SetFinalizer(p, finalizeBulkProcessor)
-
-	p.cDict = C.ZSTD_createCDict(
-		unsafe.Pointer(&dictionary[0]),
-		C.size_t(len(dictionary)),
-		C.int(compressionLevel),
-	)
-	if p.cDict == nil {
-		return nil, ErrBadDictionary
-	}
-	p.dDict = C.ZSTD_createDDict(
-		unsafe.Pointer(&dictionary[0]),
-		C.size_t(len(dictionary)),
-	)
-	if p.dDict == nil {
-		return nil, ErrBadDictionary
-	}
-
-	return p, nil
-}
-
-// Compress compresses `src` into `dst` with the dictionary given when creating the BulkProcessor.
-// If you have a buffer to use, you can pass it to prevent allocation.
-// If it is too small, or if nil is passed, a new buffer will be allocated and returned.
-func (p *BulkProcessor) Compress(dst, src []byte) ([]byte, error) {
-	bound := CompressBound(len(src))
-	if cap(dst) >= bound {
-		dst = dst[0:bound]
-	} else {
-		dst = make([]byte, bound)
-	}
-
-	cctx := C.ZSTD_createCCtx()
-	// We need unsafe.Pointer(&src[0]) in the Cgo call to avoid "Go pointer to Go pointer" panics.
-	// This means we need to special case empty input. See:
-	// https://github.com/golang/go/issues/14210#issuecomment-346402945
-	var cWritten C.size_t
-	if len(src) == 0 {
-		cWritten = C.ZSTD_compress_usingCDict(
-			cctx,
-			unsafe.Pointer(&dst[0]),
-			C.size_t(len(dst)),
-			unsafe.Pointer(nil),
-			C.size_t(len(src)),
-			p.cDict,
-		)
-	} else {
-		cWritten = C.ZSTD_compress_usingCDict(
-			cctx,
-			unsafe.Pointer(&dst[0]),
-			C.size_t(len(dst)),
-			unsafe.Pointer(&src[0]),
-			C.size_t(len(src)),
-			p.cDict,
-		)
-	}
-
-	C.ZSTD_freeCCtx(cctx)
-
-	written := int(cWritten)
-	if err := getError(written); err != nil {
-		return nil, err
-	}
-	return dst[:written], nil
-}
-
-// Decompress decompresses `src` into `dst` with the dictionary given when creating the BulkProcessor.
-// If you have a buffer to use, you can pass it to prevent allocation.
-// If it is too small, or if nil is passed, a new buffer will be allocated and returned.
-func (p *BulkProcessor) Decompress(dst, src []byte) ([]byte, error) {
-	if len(src) == 0 {
-		return nil, ErrEmptySlice
-	}
-
-	contentSize := decompressSizeHint(src)
-	if cap(dst) >= contentSize {
-		dst = dst[0:cap(dst)]
-	} else {
-		dst = make([]byte, contentSize)
-	}
-
-	if len(dst) == 0 {
-		return dst, nil
-	}
-
-	dctx := C.ZSTD_createDCtx()
-	cWritten := C.ZSTD_decompress_usingDDict(
-		dctx,
-		unsafe.Pointer(&dst[0]),
-		C.size_t(len(dst)),
-		unsafe.Pointer(&src[0]),
-		C.size_t(len(src)),
-		p.dDict,
-	)
-	C.ZSTD_freeDCtx(dctx)
-
-	written := int(cWritten)
-	if err := getError(written); err != nil {
-		return nil, err
-	}
-
-	return dst[:written], nil
-}
-
-// finalizeBulkProcessor frees compression and decompression dictionaries from memory
-func finalizeBulkProcessor(p *BulkProcessor) {
-	if p.cDict != nil {
-		C.ZSTD_freeCDict(p.cDict)
-	}
-	if p.dDict != nil {
-		C.ZSTD_freeDDict(p.dDict)
-	}
-}

vendor/github.com/DataDog/zstd/zstd_common.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -14,7 +13,8 @@
 /*-*************************************
 *  Dependencies
 ***************************************/
-#define ZSTD_DEPS_NEED_MALLOC
+#include <stdlib.h>      /* malloc, calloc, free */
+#include <string.h>      /* memset */
 #include "error_private.h"
 #include "zstd_internal.h"
 
@@ -48,4 +48,36 @@ ZSTD_ErrorCode ZSTD_getErrorCode(size_t code) { return ERR_getErrorCode(code); }
  *  provides error code string from enum */
 const char* ZSTD_getErrorString(ZSTD_ErrorCode code) { return ERR_getErrorString(code); }
 
-#endif /* USE_EXTERNAL_ZSTD */
+
+
+/*=**************************************************************
+*  Custom allocator
+****************************************************************/
+void* ZSTD_malloc(size_t size, ZSTD_customMem customMem)
+{
+    if (customMem.customAlloc)
+        return customMem.customAlloc(customMem.opaque, size);
+    return malloc(size);
+}
+
+void* ZSTD_calloc(size_t size, ZSTD_customMem customMem)
+{
+    if (customMem.customAlloc) {
+        /* calloc implemented as malloc+memset;
+         * not as efficient as calloc, but next best guess for custom malloc */
+        void* const ptr = customMem.customAlloc(customMem.opaque, size);
+        memset(ptr, 0, size);
+        return ptr;
+    }
+    return calloc(1, size);
+}
+
+void ZSTD_free(void* ptr, ZSTD_customMem customMem)
+{
+    if (ptr!=NULL) {
+        if (customMem.customFree)
+            customMem.customFree(customMem.opaque, ptr);
+        else
+            free(ptr);
+    }
+}

vendor/github.com/DataDog/zstd/zstd_compress_literals.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -14,37 +13,12 @@
  ***************************************/
 #include "zstd_compress_literals.h"
 
-
-/* **************************************************************
-*  Debug Traces
-****************************************************************/
-#if DEBUGLEVEL >= 2
-
-static size_t showHexa(const void* src, size_t srcSize)
-{
-    const BYTE* const ip = (const BYTE*)src;
-    size_t u;
-    for (u=0; u<srcSize; u++) {
-        RAWLOG(5, " %02X", ip[u]); (void)ip;
-    }
-    RAWLOG(5, " \n");
-    return srcSize;
-}
-
-#endif
-
-
-/* **************************************************************
-*  Literals compression - special cases
-****************************************************************/
 size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
 {
-    BYTE* const ostart = (BYTE*)dst;
+    BYTE* const ostart = (BYTE* const)dst;
     U32   const flSize = 1 + (srcSize>31) + (srcSize>4095);
 
-    DEBUGLOG(5, "ZSTD_noCompressLiterals: srcSize=%zu, dstCapacity=%zu", srcSize, dstCapacity);
-
-    RETURN_ERROR_IF(srcSize + flSize > dstCapacity, dstSize_tooSmall, "");
+    RETURN_ERROR_IF(srcSize + flSize > dstCapacity, dstSize_tooSmall);
 
     switch(flSize)
     {
@@ -61,31 +35,16 @@ size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src,
             assert(0);
     }
 
-    ZSTD_memcpy(ostart + flSize, src, srcSize);
-    DEBUGLOG(5, "Raw (uncompressed) literals: %u -> %u", (U32)srcSize, (U32)(srcSize + flSize));
+    memcpy(ostart + flSize, src, srcSize);
     return srcSize + flSize;
 }
 
-static int allBytesIdentical(const void* src, size_t srcSize)
-{
-    assert(srcSize >= 1);
-    assert(src != NULL);
-    {   const BYTE b = ((const BYTE*)src)[0];
-        size_t p;
-        for (p=1; p<srcSize; p++) {
-            if (((const BYTE*)src)[p] != b) return 0;
-        }
-        return 1;
-    }
-}
-
 size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
 {
-    BYTE* const ostart = (BYTE*)dst;
+    BYTE* const ostart = (BYTE* const)dst;
     U32   const flSize = 1 + (srcSize>31) + (srcSize>4095);
 
-    assert(dstCapacity >= 4); (void)dstCapacity;
-    assert(allBytesIdentical(src, srcSize));
+    (void)dstCapacity;  /* dstCapacity already guaranteed to be >=4, hence large enough */
 
     switch(flSize)
     {
@@ -103,103 +62,66 @@ size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void*
     }
 
     ostart[flSize] = *(const BYTE*)src;
-    DEBUGLOG(5, "RLE : Repeated Literal (%02X: %u times) -> %u bytes encoded", ((const BYTE*)src)[0], (U32)srcSize, (U32)flSize + 1);
     return flSize+1;
 }
 
-/* ZSTD_minLiteralsToCompress() :
- * returns minimal amount of literals
- * for literal compression to even be attempted.
- * Minimum is made tighter as compression strategy increases.
- */
-static size_t
-ZSTD_minLiteralsToCompress(ZSTD_strategy strategy, HUF_repeat huf_repeat)
-{
-    assert((int)strategy >= 0);
-    assert((int)strategy <= 9);
-    /* btultra2 : min 8 bytes;
-     * then 2x larger for each successive compression strategy
-     * max threshold 64 bytes */
-    {   int const shift = MIN(9-(int)strategy, 3);
-        size_t const mintc = (huf_repeat == HUF_repeat_valid) ? 6 : (size_t)8 << shift;
-        DEBUGLOG(7, "minLiteralsToCompress = %zu", mintc);
-        return mintc;
-    }
-}
-
-size_t ZSTD_compressLiterals (
-                  void* dst, size_t dstCapacity,
-            const void* src, size_t srcSize,
-                  void* entropyWorkspace, size_t entropyWorkspaceSize,
-            const ZSTD_hufCTables_t* prevHuf,
-                  ZSTD_hufCTables_t* nextHuf,
-                  ZSTD_strategy strategy,
-                  int disableLiteralCompression,
-                  int suspectUncompressible,
-                  int bmi2)
+size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf,
+                              ZSTD_hufCTables_t* nextHuf,
+                              ZSTD_strategy strategy, int disableLiteralCompression,
+                              void* dst, size_t dstCapacity,
+                        const void* src, size_t srcSize,
+                              void* entropyWorkspace, size_t entropyWorkspaceSize,
+                        const int bmi2)
 {
+    size_t const minGain = ZSTD_minGain(srcSize, strategy);
     size_t const lhSize = 3 + (srcSize >= 1 KB) + (srcSize >= 16 KB);
     BYTE*  const ostart = (BYTE*)dst;
     U32 singleStream = srcSize < 256;
     symbolEncodingType_e hType = set_compressed;
     size_t cLitSize;
 
-    DEBUGLOG(5,"ZSTD_compressLiterals (disableLiteralCompression=%i, srcSize=%u, dstCapacity=%zu)",
-                disableLiteralCompression, (U32)srcSize, dstCapacity);
-
-    DEBUGLOG(6, "Completed literals listing (%zu bytes)", showHexa(src, srcSize));
+    DEBUGLOG(5,"ZSTD_compressLiterals (disableLiteralCompression=%i)",
+                disableLiteralCompression);
 
     /* Prepare nextEntropy assuming reusing the existing table */
-    ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
+    memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
 
     if (disableLiteralCompression)
         return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
 
-    /* if too small, don't even attempt compression (speed opt) */
-    if (srcSize < ZSTD_minLiteralsToCompress(strategy, prevHuf->repeatMode))
-        return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
+    /* small ? don't even attempt compression (speed opt) */
+#   define COMPRESS_LITERALS_SIZE_MIN 63
+    {   size_t const minLitSize = (prevHuf->repeatMode == HUF_repeat_valid) ? 6 : COMPRESS_LITERALS_SIZE_MIN;
+        if (srcSize <= minLitSize) return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
+    }
 
     RETURN_ERROR_IF(dstCapacity < lhSize+1, dstSize_tooSmall, "not enough space for compression");
     {   HUF_repeat repeat = prevHuf->repeatMode;
-        int const flags = 0
-            | (bmi2 ? HUF_flags_bmi2 : 0)
-            | (strategy < ZSTD_lazy && srcSize <= 1024 ? HUF_flags_preferRepeat : 0)
-            | (strategy >= HUF_OPTIMAL_DEPTH_THRESHOLD ? HUF_flags_optimalDepth : 0)
-            | (suspectUncompressible ? HUF_flags_suspectUncompressible : 0);
-
-        typedef size_t (*huf_compress_f)(void*, size_t, const void*, size_t, unsigned, unsigned, void*, size_t, HUF_CElt*, HUF_repeat*, int);
-        huf_compress_f huf_compress;
+        int const preferRepeat = strategy < ZSTD_lazy ? srcSize <= 1024 : 0;
         if (repeat == HUF_repeat_valid && lhSize == 3) singleStream = 1;
-        huf_compress = singleStream ? HUF_compress1X_repeat : HUF_compress4X_repeat;
-        cLitSize = huf_compress(ostart+lhSize, dstCapacity-lhSize,
-                                src, srcSize,
-                                HUF_SYMBOLVALUE_MAX, LitHufLog,
-                                entropyWorkspace, entropyWorkspaceSize,
-                                (HUF_CElt*)nextHuf->CTable,
-                                &repeat, flags);
-        DEBUGLOG(5, "%zu literals compressed into %zu bytes (before header)", srcSize, cLitSize);
+        cLitSize = singleStream ?
+            HUF_compress1X_repeat(
+                ostart+lhSize, dstCapacity-lhSize, src, srcSize,
+                255, 11, entropyWorkspace, entropyWorkspaceSize,
+                (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2) :
+            HUF_compress4X_repeat(
+                ostart+lhSize, dstCapacity-lhSize, src, srcSize,
+                255, 11, entropyWorkspace, entropyWorkspaceSize,
+                (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2);
         if (repeat != HUF_repeat_none) {
             /* reused the existing table */
-            DEBUGLOG(5, "reusing statistics from previous huffman block");
             hType = set_repeat;
         }
     }
 
-    {   size_t const minGain = ZSTD_minGain(srcSize, strategy);
-        if ((cLitSize==0) || (cLitSize >= srcSize - minGain) || ERR_isError(cLitSize)) {
-            ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
-            return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
-    }   }
+    if ((cLitSize==0) | (cLitSize >= srcSize - minGain) | ERR_isError(cLitSize)) {
+        memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
+        return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
+    }
     if (cLitSize==1) {
-        /* A return value of 1 signals that the alphabet consists of a single symbol.
-         * However, in some rare circumstances, it could be the compressed size (a single byte).
-         * For that outcome to have a chance to happen, it's necessary that `srcSize < 8`.
-         * (it's also necessary to not generate statistics).
-         * Therefore, in such a case, actively check that all bytes are identical. */
-        if ((srcSize >= 8) || allBytesIdentical(src, srcSize)) {
-            ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
-            return ZSTD_compressRleLiteralsBlock(dst, dstCapacity, src, srcSize);
-    }   }
+        memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
+        return ZSTD_compressRleLiteralsBlock(dst, dstCapacity, src, srcSize);
+    }
 
     if (hType == set_compressed) {
         /* using a newly constructed table */
@@ -210,19 +132,16 @@ size_t ZSTD_compressLiterals (
     switch(lhSize)
     {
     case 3: /* 2 - 2 - 10 - 10 */
-        if (!singleStream) assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
-        {   U32 const lhc = hType + ((U32)(!singleStream) << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<14);
+        {   U32 const lhc = hType + ((!singleStream) << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<14);
             MEM_writeLE24(ostart, lhc);
             break;
         }
     case 4: /* 2 - 2 - 14 - 14 */
-        assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
         {   U32 const lhc = hType + (2 << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<18);
             MEM_writeLE32(ostart, lhc);
             break;
         }
     case 5: /* 2 - 2 - 18 - 18 */
-        assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
         {   U32 const lhc = hType + (3 << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<22);
             MEM_writeLE32(ostart, lhc);
             ostart[4] = (BYTE)(cLitSize >> 10);
@@ -231,8 +150,5 @@ size_t ZSTD_compressLiterals (
     default:  /* not possible : lhSize is {3,4,5} */
         assert(0);
     }
-    DEBUGLOG(5, "Compressed literals: %u -> %u", (U32)srcSize, (U32)(lhSize+cLitSize));
     return lhSize+cLitSize;
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_compress_literals.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -17,26 +16,14 @@
 
 size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize);
 
-/* ZSTD_compressRleLiteralsBlock() :
- * Conditions :
- * - All bytes in @src are identical
- * - dstCapacity >= 4 */
 size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize);
 
-/* ZSTD_compressLiterals():
- * @entropyWorkspace: must be aligned on 4-bytes boundaries
- * @entropyWorkspaceSize : must be >= HUF_WORKSPACE_SIZE
- * @suspectUncompressible: sampling checks, to potentially skip huffman coding
- */
-size_t ZSTD_compressLiterals (void* dst, size_t dstCapacity,
-                        const void* src, size_t srcSize,
-                              void* entropyWorkspace, size_t entropyWorkspaceSize,
-                        const ZSTD_hufCTables_t* prevHuf,
+size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf,
                               ZSTD_hufCTables_t* nextHuf,
                               ZSTD_strategy strategy, int disableLiteralCompression,
-                              int suspectUncompressible,
-                              int bmi2);
+                              void* dst, size_t dstCapacity,
+                        const void* src, size_t srcSize,
+                              void* entropyWorkspace, size_t entropyWorkspaceSize,
+                        const int bmi2);
 
 #endif /* ZSTD_COMPRESS_LITERALS_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_compress_sequences.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -51,19 +50,6 @@ static unsigned ZSTD_getFSEMaxSymbolValue(FSE_CTable const* ctable) {
   return maxSymbolValue;
 }
 
-/**
- * Returns true if we should use ncount=-1 else we should
- * use ncount=1 for low probability symbols instead.
- */
-static unsigned ZSTD_useLowProbCount(size_t const nbSeq)
-{
-    /* Heuristic: This should cover most blocks <= 16K and
-     * start to fade out after 16K to about 32K depending on
-     * compressibility.
-     */
-    return nbSeq >= 2048;
-}
-
 /**
  * Returns the cost in bytes of encoding the normalized count header.
  * Returns an error if any of the helper functions return an error.
@@ -74,7 +60,7 @@ static size_t ZSTD_NCountCost(unsigned const* count, unsigned const max,
     BYTE wksp[FSE_NCOUNTBOUND];
     S16 norm[MaxSeq + 1];
     const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max);
-    FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq, max, ZSTD_useLowProbCount(nbSeq)), "");
+    FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq, max));
     return FSE_writeNCount(wksp, sizeof(wksp), norm, max, tableLog);
 }
 
@@ -86,8 +72,6 @@ static size_t ZSTD_entropyCost(unsigned const* count, unsigned const max, size_t
 {
     unsigned cost = 0;
     unsigned s;
-
-    assert(total > 0);
     for (s = 0; s <= max; ++s) {
         unsigned norm = (unsigned)((256 * count[s]) / total);
         if (count[s] != 0 && norm == 0)
@@ -102,7 +86,7 @@ static size_t ZSTD_entropyCost(unsigned const* count, unsigned const max, size_t
  * Returns the cost in bits of encoding the distribution in count using ctable.
  * Returns an error if ctable cannot represent all the symbols in count.
  */
-size_t ZSTD_fseBitCost(
+static size_t ZSTD_fseBitCost(
     FSE_CTable const* ctable,
     unsigned const* count,
     unsigned const max)
@@ -112,22 +96,18 @@ size_t ZSTD_fseBitCost(
     unsigned s;
     FSE_CState_t cstate;
     FSE_initCState(&cstate, ctable);
-    if (ZSTD_getFSEMaxSymbolValue(ctable) < max) {
-        DEBUGLOG(5, "Repeat FSE_CTable has maxSymbolValue %u < %u",
+    RETURN_ERROR_IF(ZSTD_getFSEMaxSymbolValue(ctable) < max, GENERIC,
+                    "Repeat FSE_CTable has maxSymbolValue %u < %u",
                     ZSTD_getFSEMaxSymbolValue(ctable), max);
-        return ERROR(GENERIC);
-    }
     for (s = 0; s <= max; ++s) {
         unsigned const tableLog = cstate.stateLog;
         unsigned const badCost = (tableLog + 1) << kAccuracyLog;
         unsigned const bitCost = FSE_bitCost(cstate.symbolTT, tableLog, s, kAccuracyLog);
         if (count[s] == 0)
             continue;
-        if (bitCost >= badCost) {
-            DEBUGLOG(5, "Repeat FSE_CTable has Prob[%u] == 0", s);
-            return ERROR(GENERIC);
-        }
-        cost += (size_t)count[s] * bitCost;
+        RETURN_ERROR_IF(bitCost >= badCost, GENERIC,
+                        "Repeat FSE_CTable has Prob[%u] == 0", s);
+        cost += count[s] * bitCost;
     }
     return cost >> kAccuracyLog;
 }
@@ -137,15 +117,15 @@ size_t ZSTD_fseBitCost(
  * table described by norm. The max symbol support by norm is assumed >= max.
  * norm must be valid for every symbol with non-zero probability in count.
  */
-size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
-                             unsigned const* count, unsigned const max)
+static size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
+                                    unsigned const* count, unsigned const max)
 {
     unsigned const shift = 8 - accuracyLog;
     size_t cost = 0;
     unsigned s;
     assert(accuracyLog <= 8);
     for (s = 0; s <= max; ++s) {
-        unsigned const normAcc = (norm[s] != -1) ? (unsigned)norm[s] : 1;
+        unsigned const normAcc = norm[s] != -1 ? norm[s] : 1;
         unsigned const norm256 = normAcc << shift;
         assert(norm256 > 0);
         assert(norm256 < 256);
@@ -167,7 +147,7 @@ ZSTD_selectEncodingType(
     if (mostFrequent == nbSeq) {
         *repeatMode = FSE_repeat_none;
         if (isDefaultAllowed && nbSeq <= 2) {
-            /* Prefer set_basic over set_rle when there are 2 or fewer symbols,
+            /* Prefer set_basic over set_rle when there are 2 or less symbols,
              * since RLE uses 1 byte, but set_basic uses 5-6 bits per symbol.
              * If basic encoding isn't possible, always choose RLE.
              */
@@ -235,11 +215,6 @@ ZSTD_selectEncodingType(
     return set_compressed;
 }
 
-typedef struct {
-    S16 norm[MaxSeq + 1];
-    U32 wksp[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(MaxSeq, MaxFSELog)];
-} ZSTD_BuildCTableWksp;
-
 size_t
 ZSTD_buildCTable(void* dst, size_t dstCapacity,
                 FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
@@ -255,18 +230,18 @@ ZSTD_buildCTable(void* dst, size_t dstCapacity,
 
     switch (type) {
     case set_rle:
-        FORWARD_IF_ERROR(FSE_buildCTable_rle(nextCTable, (BYTE)max), "");
-        RETURN_ERROR_IF(dstCapacity==0, dstSize_tooSmall, "not enough space");
+        FORWARD_IF_ERROR(FSE_buildCTable_rle(nextCTable, (BYTE)max));
+        RETURN_ERROR_IF(dstCapacity==0, dstSize_tooSmall);
         *op = codeTable[0];
         return 1;
     case set_repeat:
-        ZSTD_memcpy(nextCTable, prevCTable, prevCTableSize);
+        memcpy(nextCTable, prevCTable, prevCTableSize);
         return 0;
     case set_basic:
-        FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, entropyWorkspace, entropyWorkspaceSize), "");  /* note : could be pre-calculated */
+        FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, entropyWorkspace, entropyWorkspaceSize));  /* note : could be pre-calculated */
         return 0;
     case set_compressed: {
-        ZSTD_BuildCTableWksp* wksp = (ZSTD_BuildCTableWksp*)entropyWorkspace;
+        S16 norm[MaxSeq + 1];
         size_t nbSeq_1 = nbSeq;
         const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max);
         if (count[codeTable[nbSeq-1]] > 1) {
@@ -274,17 +249,14 @@ ZSTD_buildCTable(void* dst, size_t dstCapacity,
             nbSeq_1--;
         }
         assert(nbSeq_1 > 1);
-        assert(entropyWorkspaceSize >= sizeof(ZSTD_BuildCTableWksp));
-        (void)entropyWorkspaceSize;
-        FORWARD_IF_ERROR(FSE_normalizeCount(wksp->norm, tableLog, count, nbSeq_1, max, ZSTD_useLowProbCount(nbSeq_1)), "FSE_normalizeCount failed");
-        assert(oend >= op);
-        {   size_t const NCountSize = FSE_writeNCount(op, (size_t)(oend - op), wksp->norm, max, tableLog);   /* overflow protected */
-            FORWARD_IF_ERROR(NCountSize, "FSE_writeNCount failed");
-            FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, wksp->norm, max, tableLog, wksp->wksp, sizeof(wksp->wksp)), "FSE_buildCTable_wksp failed");
+        FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq_1, max));
+        {   size_t const NCountSize = FSE_writeNCount(op, oend - op, norm, max, tableLog);   /* overflow protected */
+            FORWARD_IF_ERROR(NCountSize);
+            FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, norm, max, tableLog, entropyWorkspace, entropyWorkspaceSize));
             return NCountSize;
         }
     }
-    default: assert(0); RETURN_ERROR(GENERIC, "impossible to reach");
+    default: assert(0); RETURN_ERROR(GENERIC);
     }
 }
 
@@ -314,19 +286,19 @@ ZSTD_encodeSequences_body(
     FSE_initCState2(&stateLitLength,   CTable_LitLength,   llCodeTable[nbSeq-1]);
     BIT_addBits(&blockStream, sequences[nbSeq-1].litLength, LL_bits[llCodeTable[nbSeq-1]]);
     if (MEM_32bits()) BIT_flushBits(&blockStream);
-    BIT_addBits(&blockStream, sequences[nbSeq-1].mlBase, ML_bits[mlCodeTable[nbSeq-1]]);
+    BIT_addBits(&blockStream, sequences[nbSeq-1].matchLength, ML_bits[mlCodeTable[nbSeq-1]]);
     if (MEM_32bits()) BIT_flushBits(&blockStream);
     if (longOffsets) {
         U32 const ofBits = ofCodeTable[nbSeq-1];
-        unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
+        int const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
         if (extraBits) {
-            BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, extraBits);
+            BIT_addBits(&blockStream, sequences[nbSeq-1].offset, extraBits);
             BIT_flushBits(&blockStream);
         }
-        BIT_addBits(&blockStream, sequences[nbSeq-1].offBase >> extraBits,
+        BIT_addBits(&blockStream, sequences[nbSeq-1].offset >> extraBits,
                     ofBits - extraBits);
     } else {
-        BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, ofCodeTable[nbSeq-1]);
+        BIT_addBits(&blockStream, sequences[nbSeq-1].offset, ofCodeTable[nbSeq-1]);
     }
     BIT_flushBits(&blockStream);
 
@@ -340,8 +312,8 @@ ZSTD_encodeSequences_body(
             U32  const mlBits = ML_bits[mlCode];
             DEBUGLOG(6, "encoding: litlen:%2u - matchlen:%2u - offCode:%7u",
                         (unsigned)sequences[n].litLength,
-                        (unsigned)sequences[n].mlBase + MINMATCH,
-                        (unsigned)sequences[n].offBase);
+                        (unsigned)sequences[n].matchLength + MINMATCH,
+                        (unsigned)sequences[n].offset);
                                                                             /* 32b*/  /* 64b*/
                                                                             /* (7)*/  /* (7)*/
             FSE_encodeSymbol(&blockStream, &stateOffsetBits, ofCode);       /* 15 */  /* 15 */
@@ -352,18 +324,18 @@ ZSTD_encodeSequences_body(
                 BIT_flushBits(&blockStream);                                /* (7)*/
             BIT_addBits(&blockStream, sequences[n].litLength, llBits);
             if (MEM_32bits() && ((llBits+mlBits)>24)) BIT_flushBits(&blockStream);
-            BIT_addBits(&blockStream, sequences[n].mlBase, mlBits);
+            BIT_addBits(&blockStream, sequences[n].matchLength, mlBits);
             if (MEM_32bits() || (ofBits+mlBits+llBits > 56)) BIT_flushBits(&blockStream);
             if (longOffsets) {
-                unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
+                int const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
                 if (extraBits) {
-                    BIT_addBits(&blockStream, sequences[n].offBase, extraBits);
+                    BIT_addBits(&blockStream, sequences[n].offset, extraBits);
                     BIT_flushBits(&blockStream);                            /* (7)*/
                 }
-                BIT_addBits(&blockStream, sequences[n].offBase >> extraBits,
+                BIT_addBits(&blockStream, sequences[n].offset >> extraBits,
                             ofBits - extraBits);                            /* 31 */
             } else {
-                BIT_addBits(&blockStream, sequences[n].offBase, ofBits);     /* 31 */
+                BIT_addBits(&blockStream, sequences[n].offset, ofBits);     /* 31 */
             }
             BIT_flushBits(&blockStream);                                    /* (7)*/
             DEBUGLOG(7, "remaining space : %i", (int)(blockStream.endPtr - blockStream.ptr));
@@ -400,7 +372,7 @@ ZSTD_encodeSequences_default(
 
 #if DYNAMIC_BMI2
 
-static BMI2_TARGET_ATTRIBUTE size_t
+static TARGET_ATTRIBUTE("bmi2") size_t
 ZSTD_encodeSequences_bmi2(
             void* dst, size_t dstCapacity,
             FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
@@ -441,5 +413,3 @@ size_t ZSTD_encodeSequences(
                                         CTable_LitLength, llCodeTable,
                                         sequences, nbSeq, longOffsets);
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_compress_sequences.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -45,13 +44,4 @@ size_t ZSTD_encodeSequences(
             FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
             seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2);
 
-size_t ZSTD_fseBitCost(
-    FSE_CTable const* ctable,
-    unsigned const* count,
-    unsigned const max);
-
-size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
-                             unsigned const* count, unsigned const max);
 #endif /* ZSTD_COMPRESS_SEQUENCES_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_compress_superblock.c

@@ -1,691 +0,0 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
- /*-*************************************
- *  Dependencies
- ***************************************/
-#include "zstd_compress_superblock.h"
-
-#include "zstd_internal.h"  /* ZSTD_getSequenceLength */
-#include "hist.h"                     /* HIST_countFast_wksp */
-#include "zstd_compress_internal.h"   /* ZSTD_[huf|fse|entropy]CTablesMetadata_t */
-#include "zstd_compress_sequences.h"
-#include "zstd_compress_literals.h"
-
-/** ZSTD_compressSubBlock_literal() :
- *  Compresses literals section for a sub-block.
- *  When we have to write the Huffman table we will sometimes choose a header
- *  size larger than necessary. This is because we have to pick the header size
- *  before we know the table size + compressed size, so we have a bound on the
- *  table size. If we guessed incorrectly, we fall back to uncompressed literals.
- *
- *  We write the header when writeEntropy=1 and set entropyWritten=1 when we succeeded
- *  in writing the header, otherwise it is set to 0.
- *
- *  hufMetadata->hType has literals block type info.
- *      If it is set_basic, all sub-blocks literals section will be Raw_Literals_Block.
- *      If it is set_rle, all sub-blocks literals section will be RLE_Literals_Block.
- *      If it is set_compressed, first sub-block's literals section will be Compressed_Literals_Block
- *      If it is set_compressed, first sub-block's literals section will be Treeless_Literals_Block
- *      and the following sub-blocks' literals sections will be Treeless_Literals_Block.
- *  @return : compressed size of literals section of a sub-block
- *            Or 0 if unable to compress.
- *            Or error code */
-static size_t
-ZSTD_compressSubBlock_literal(const HUF_CElt* hufTable,
-                              const ZSTD_hufCTablesMetadata_t* hufMetadata,
-                              const BYTE* literals, size_t litSize,
-                              void* dst, size_t dstSize,
-                              const int bmi2, int writeEntropy, int* entropyWritten)
-{
-    size_t const header = writeEntropy ? 200 : 0;
-    size_t const lhSize = 3 + (litSize >= (1 KB - header)) + (litSize >= (16 KB - header));
-    BYTE* const ostart = (BYTE*)dst;
-    BYTE* const oend = ostart + dstSize;
-    BYTE* op = ostart + lhSize;
-    U32 const singleStream = lhSize == 3;
-    symbolEncodingType_e hType = writeEntropy ? hufMetadata->hType : set_repeat;
-    size_t cLitSize = 0;
-
-    DEBUGLOG(5, "ZSTD_compressSubBlock_literal (litSize=%zu, lhSize=%zu, writeEntropy=%d)", litSize, lhSize, writeEntropy);
-
-    *entropyWritten = 0;
-    if (litSize == 0 || hufMetadata->hType == set_basic) {
-      DEBUGLOG(5, "ZSTD_compressSubBlock_literal using raw literal");
-      return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
-    } else if (hufMetadata->hType == set_rle) {
-      DEBUGLOG(5, "ZSTD_compressSubBlock_literal using rle literal");
-      return ZSTD_compressRleLiteralsBlock(dst, dstSize, literals, litSize);
-    }
-
-    assert(litSize > 0);
-    assert(hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat);
-
-    if (writeEntropy && hufMetadata->hType == set_compressed) {
-        ZSTD_memcpy(op, hufMetadata->hufDesBuffer, hufMetadata->hufDesSize);
-        op += hufMetadata->hufDesSize;
-        cLitSize += hufMetadata->hufDesSize;
-        DEBUGLOG(5, "ZSTD_compressSubBlock_literal (hSize=%zu)", hufMetadata->hufDesSize);
-    }
-
-    {   int const flags = bmi2 ? HUF_flags_bmi2 : 0;
-        const size_t cSize = singleStream ? HUF_compress1X_usingCTable(op, (size_t)(oend-op), literals, litSize, hufTable, flags)
-                                          : HUF_compress4X_usingCTable(op, (size_t)(oend-op), literals, litSize, hufTable, flags);
-        op += cSize;
-        cLitSize += cSize;
-        if (cSize == 0 || ERR_isError(cSize)) {
-            DEBUGLOG(5, "Failed to write entropy tables %s", ZSTD_getErrorName(cSize));
-            return 0;
-        }
-        /* If we expand and we aren't writing a header then emit uncompressed */
-        if (!writeEntropy && cLitSize >= litSize) {
-            DEBUGLOG(5, "ZSTD_compressSubBlock_literal using raw literal because uncompressible");
-            return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
-        }
-        /* If we are writing headers then allow expansion that doesn't change our header size. */
-        if (lhSize < (size_t)(3 + (cLitSize >= 1 KB) + (cLitSize >= 16 KB))) {
-            assert(cLitSize > litSize);
-            DEBUGLOG(5, "Literals expanded beyond allowed header size");
-            return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
-        }
-        DEBUGLOG(5, "ZSTD_compressSubBlock_literal (cSize=%zu)", cSize);
-    }
-
-    /* Build header */
-    switch(lhSize)
-    {
-    case 3: /* 2 - 2 - 10 - 10 */
-        {   U32 const lhc = hType + ((U32)(!singleStream) << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<14);
-            MEM_writeLE24(ostart, lhc);
-            break;
-        }
-    case 4: /* 2 - 2 - 14 - 14 */
-        {   U32 const lhc = hType + (2 << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<18);
-            MEM_writeLE32(ostart, lhc);
-            break;
-        }
-    case 5: /* 2 - 2 - 18 - 18 */
-        {   U32 const lhc = hType + (3 << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<22);
-            MEM_writeLE32(ostart, lhc);
-            ostart[4] = (BYTE)(cLitSize >> 10);
-            break;
-        }
-    default:  /* not possible : lhSize is {3,4,5} */
-        assert(0);
-    }
-    *entropyWritten = 1;
-    DEBUGLOG(5, "Compressed literals: %u -> %u", (U32)litSize, (U32)(op-ostart));
-    return (size_t)(op-ostart);
-}
-
-static size_t
-ZSTD_seqDecompressedSize(seqStore_t const* seqStore,
-                   const seqDef* sequences, size_t nbSeqs,
-                         size_t litSize, int lastSubBlock)
-{
-    size_t matchLengthSum = 0;
-    size_t litLengthSum = 0;
-    size_t n;
-    for (n=0; n<nbSeqs; n++) {
-        const ZSTD_sequenceLength seqLen = ZSTD_getSequenceLength(seqStore, sequences+n);
-        litLengthSum += seqLen.litLength;
-        matchLengthSum += seqLen.matchLength;
-    }
-    DEBUGLOG(5, "ZSTD_seqDecompressedSize: %u sequences from %p: %u literals + %u matchlength",
-                (unsigned)nbSeqs, (const void*)sequences,
-                (unsigned)litLengthSum, (unsigned)matchLengthSum);
-    if (!lastSubBlock)
-        assert(litLengthSum == litSize);
-    else
-        assert(litLengthSum <= litSize);
-    (void)litLengthSum;
-    return matchLengthSum + litSize;
-}
-
-/** ZSTD_compressSubBlock_sequences() :
- *  Compresses sequences section for a sub-block.
- *  fseMetadata->llType, fseMetadata->ofType, and fseMetadata->mlType have
- *  symbol compression modes for the super-block.
- *  The first successfully compressed block will have these in its header.
- *  We set entropyWritten=1 when we succeed in compressing the sequences.
- *  The following sub-blocks will always have repeat mode.
- *  @return : compressed size of sequences section of a sub-block
- *            Or 0 if it is unable to compress
- *            Or error code. */
-static size_t
-ZSTD_compressSubBlock_sequences(const ZSTD_fseCTables_t* fseTables,
-                                const ZSTD_fseCTablesMetadata_t* fseMetadata,
-                                const seqDef* sequences, size_t nbSeq,
-                                const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode,
-                                const ZSTD_CCtx_params* cctxParams,
-                                void* dst, size_t dstCapacity,
-                                const int bmi2, int writeEntropy, int* entropyWritten)
-{
-    const int longOffsets = cctxParams->cParams.windowLog > STREAM_ACCUMULATOR_MIN;
-    BYTE* const ostart = (BYTE*)dst;
-    BYTE* const oend = ostart + dstCapacity;
-    BYTE* op = ostart;
-    BYTE* seqHead;
-
-    DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (nbSeq=%zu, writeEntropy=%d, longOffsets=%d)", nbSeq, writeEntropy, longOffsets);
-
-    *entropyWritten = 0;
-    /* Sequences Header */
-    RETURN_ERROR_IF((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead*/,
-                    dstSize_tooSmall, "");
-    if (nbSeq < 128)
-        *op++ = (BYTE)nbSeq;
-    else if (nbSeq < LONGNBSEQ)
-        op[0] = (BYTE)((nbSeq>>8) + 0x80), op[1] = (BYTE)nbSeq, op+=2;
-    else
-        op[0]=0xFF, MEM_writeLE16(op+1, (U16)(nbSeq - LONGNBSEQ)), op+=3;
-    if (nbSeq==0) {
-        return (size_t)(op - ostart);
-    }
-
-    /* seqHead : flags for FSE encoding type */
-    seqHead = op++;
-
-    DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (seqHeadSize=%u)", (unsigned)(op-ostart));
-
-    if (writeEntropy) {
-        const U32 LLtype = fseMetadata->llType;
-        const U32 Offtype = fseMetadata->ofType;
-        const U32 MLtype = fseMetadata->mlType;
-        DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (fseTablesSize=%zu)", fseMetadata->fseTablesSize);
-        *seqHead = (BYTE)((LLtype<<6) + (Offtype<<4) + (MLtype<<2));
-        ZSTD_memcpy(op, fseMetadata->fseTablesBuffer, fseMetadata->fseTablesSize);
-        op += fseMetadata->fseTablesSize;
-    } else {
-        const U32 repeat = set_repeat;
-        *seqHead = (BYTE)((repeat<<6) + (repeat<<4) + (repeat<<2));
-    }
-
-    {   size_t const bitstreamSize = ZSTD_encodeSequences(
-                                        op, (size_t)(oend - op),
-                                        fseTables->matchlengthCTable, mlCode,
-                                        fseTables->offcodeCTable, ofCode,
-                                        fseTables->litlengthCTable, llCode,
-                                        sequences, nbSeq,
-                                        longOffsets, bmi2);
-        FORWARD_IF_ERROR(bitstreamSize, "ZSTD_encodeSequences failed");
-        op += bitstreamSize;
-        /* zstd versions <= 1.3.4 mistakenly report corruption when
-         * FSE_readNCount() receives a buffer < 4 bytes.
-         * Fixed by https://github.com/facebook/zstd/pull/1146.
-         * This can happen when the last set_compressed table present is 2
-         * bytes and the bitstream is only one byte.
-         * In this exceedingly rare case, we will simply emit an uncompressed
-         * block, since it isn't worth optimizing.
-         */
-#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
-        if (writeEntropy && fseMetadata->lastCountSize && fseMetadata->lastCountSize + bitstreamSize < 4) {
-            /* NCountSize >= 2 && bitstreamSize > 0 ==> lastCountSize == 3 */
-            assert(fseMetadata->lastCountSize + bitstreamSize == 3);
-            DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.3.4 by "
-                        "emitting an uncompressed block.");
-            return 0;
-        }
-#endif
-        DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (bitstreamSize=%zu)", bitstreamSize);
-    }
-
-    /* zstd versions <= 1.4.0 mistakenly report error when
-     * sequences section body size is less than 3 bytes.
-     * Fixed by https://github.com/facebook/zstd/pull/1664.
-     * This can happen when the previous sequences section block is compressed
-     * with rle mode and the current block's sequences section is compressed
-     * with repeat mode where sequences section body size can be 1 byte.
-     */
-#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
-    if (op-seqHead < 4) {
-        DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.4.0 by emitting "
-                    "an uncompressed block when sequences are < 4 bytes");
-        return 0;
-    }
-#endif
-
-    *entropyWritten = 1;
-    return (size_t)(op - ostart);
-}
-
-/** ZSTD_compressSubBlock() :
- *  Compresses a single sub-block.
- *  @return : compressed size of the sub-block
- *            Or 0 if it failed to compress. */
-static size_t ZSTD_compressSubBlock(const ZSTD_entropyCTables_t* entropy,
-                                    const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
-                                    const seqDef* sequences, size_t nbSeq,
-                                    const BYTE* literals, size_t litSize,
-                                    const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode,
-                                    const ZSTD_CCtx_params* cctxParams,
-                                    void* dst, size_t dstCapacity,
-                                    const int bmi2,
-                                    int writeLitEntropy, int writeSeqEntropy,
-                                    int* litEntropyWritten, int* seqEntropyWritten,
-                                    U32 lastBlock)
-{
-    BYTE* const ostart = (BYTE*)dst;
-    BYTE* const oend = ostart + dstCapacity;
-    BYTE* op = ostart + ZSTD_blockHeaderSize;
-    DEBUGLOG(5, "ZSTD_compressSubBlock (litSize=%zu, nbSeq=%zu, writeLitEntropy=%d, writeSeqEntropy=%d, lastBlock=%d)",
-                litSize, nbSeq, writeLitEntropy, writeSeqEntropy, lastBlock);
-    {   size_t cLitSize = ZSTD_compressSubBlock_literal((const HUF_CElt*)entropy->huf.CTable,
-                                                        &entropyMetadata->hufMetadata, literals, litSize,
-                                                        op, (size_t)(oend-op),
-                                                        bmi2, writeLitEntropy, litEntropyWritten);
-        FORWARD_IF_ERROR(cLitSize, "ZSTD_compressSubBlock_literal failed");
-        if (cLitSize == 0) return 0;
-        op += cLitSize;
-    }
-    {   size_t cSeqSize = ZSTD_compressSubBlock_sequences(&entropy->fse,
-                                                  &entropyMetadata->fseMetadata,
-                                                  sequences, nbSeq,
-                                                  llCode, mlCode, ofCode,
-                                                  cctxParams,
-                                                  op, (size_t)(oend-op),
-                                                  bmi2, writeSeqEntropy, seqEntropyWritten);
-        FORWARD_IF_ERROR(cSeqSize, "ZSTD_compressSubBlock_sequences failed");
-        if (cSeqSize == 0) return 0;
-        op += cSeqSize;
-    }
-    /* Write block header */
-    {   size_t cSize = (size_t)(op-ostart) - ZSTD_blockHeaderSize;
-        U32 const cBlockHeader24 = lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3);
-        MEM_writeLE24(ostart, cBlockHeader24);
-    }
-    return (size_t)(op-ostart);
-}
-
-static size_t ZSTD_estimateSubBlockSize_literal(const BYTE* literals, size_t litSize,
-                                                const ZSTD_hufCTables_t* huf,
-                                                const ZSTD_hufCTablesMetadata_t* hufMetadata,
-                                                void* workspace, size_t wkspSize,
-                                                int writeEntropy)
-{
-    unsigned* const countWksp = (unsigned*)workspace;
-    unsigned maxSymbolValue = 255;
-    size_t literalSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */
-
-    if (hufMetadata->hType == set_basic) return litSize;
-    else if (hufMetadata->hType == set_rle) return 1;
-    else if (hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat) {
-        size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)literals, litSize, workspace, wkspSize);
-        if (ZSTD_isError(largest)) return litSize;
-        {   size_t cLitSizeEstimate = HUF_estimateCompressedSize((const HUF_CElt*)huf->CTable, countWksp, maxSymbolValue);
-            if (writeEntropy) cLitSizeEstimate += hufMetadata->hufDesSize;
-            return cLitSizeEstimate + literalSectionHeaderSize;
-    }   }
-    assert(0); /* impossible */
-    return 0;
-}
-
-static size_t ZSTD_estimateSubBlockSize_symbolType(symbolEncodingType_e type,
-                        const BYTE* codeTable, unsigned maxCode,
-                        size_t nbSeq, const FSE_CTable* fseCTable,
-                        const U8* additionalBits,
-                        short const* defaultNorm, U32 defaultNormLog, U32 defaultMax,
-                        void* workspace, size_t wkspSize)
-{
-    unsigned* const countWksp = (unsigned*)workspace;
-    const BYTE* ctp = codeTable;
-    const BYTE* const ctStart = ctp;
-    const BYTE* const ctEnd = ctStart + nbSeq;
-    size_t cSymbolTypeSizeEstimateInBits = 0;
-    unsigned max = maxCode;
-
-    HIST_countFast_wksp(countWksp, &max, codeTable, nbSeq, workspace, wkspSize);  /* can't fail */
-    if (type == set_basic) {
-        /* We selected this encoding type, so it must be valid. */
-        assert(max <= defaultMax);
-        cSymbolTypeSizeEstimateInBits = max <= defaultMax
-                ? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max)
-                : ERROR(GENERIC);
-    } else if (type == set_rle) {
-        cSymbolTypeSizeEstimateInBits = 0;
-    } else if (type == set_compressed || type == set_repeat) {
-        cSymbolTypeSizeEstimateInBits = ZSTD_fseBitCost(fseCTable, countWksp, max);
-    }
-    if (ZSTD_isError(cSymbolTypeSizeEstimateInBits)) return nbSeq * 10;
-    while (ctp < ctEnd) {
-        if (additionalBits) cSymbolTypeSizeEstimateInBits += additionalBits[*ctp];
-        else cSymbolTypeSizeEstimateInBits += *ctp; /* for offset, offset code is also the number of additional bits */
-        ctp++;
-    }
-    return cSymbolTypeSizeEstimateInBits / 8;
-}
-
-static size_t ZSTD_estimateSubBlockSize_sequences(const BYTE* ofCodeTable,
-                                                  const BYTE* llCodeTable,
-                                                  const BYTE* mlCodeTable,
-                                                  size_t nbSeq,
-                                                  const ZSTD_fseCTables_t* fseTables,
-                                                  const ZSTD_fseCTablesMetadata_t* fseMetadata,
-                                                  void* workspace, size_t wkspSize,
-                                                  int writeEntropy)
-{
-    size_t const sequencesSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */
-    size_t cSeqSizeEstimate = 0;
-    if (nbSeq == 0) return sequencesSectionHeaderSize;
-    cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->ofType, ofCodeTable, MaxOff,
-                                         nbSeq, fseTables->offcodeCTable, NULL,
-                                         OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff,
-                                         workspace, wkspSize);
-    cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->llType, llCodeTable, MaxLL,
-                                         nbSeq, fseTables->litlengthCTable, LL_bits,
-                                         LL_defaultNorm, LL_defaultNormLog, MaxLL,
-                                         workspace, wkspSize);
-    cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->mlType, mlCodeTable, MaxML,
-                                         nbSeq, fseTables->matchlengthCTable, ML_bits,
-                                         ML_defaultNorm, ML_defaultNormLog, MaxML,
-                                         workspace, wkspSize);
-    if (writeEntropy) cSeqSizeEstimate += fseMetadata->fseTablesSize;
-    return cSeqSizeEstimate + sequencesSectionHeaderSize;
-}
-
-typedef struct {
-    size_t estLitSize;
-    size_t estBlockSize;
-} EstimatedBlockSize;
-static EstimatedBlockSize ZSTD_estimateSubBlockSize(const BYTE* literals, size_t litSize,
-                                        const BYTE* ofCodeTable,
-                                        const BYTE* llCodeTable,
-                                        const BYTE* mlCodeTable,
-                                        size_t nbSeq,
-                                        const ZSTD_entropyCTables_t* entropy,
-                                        const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
-                                        void* workspace, size_t wkspSize,
-                                        int writeLitEntropy, int writeSeqEntropy)
-{
-    EstimatedBlockSize ebs;
-    ebs.estLitSize = ZSTD_estimateSubBlockSize_literal(literals, litSize,
-                                                        &entropy->huf, &entropyMetadata->hufMetadata,
-                                                        workspace, wkspSize, writeLitEntropy);
-    ebs.estBlockSize = ZSTD_estimateSubBlockSize_sequences(ofCodeTable, llCodeTable, mlCodeTable,
-                                                         nbSeq, &entropy->fse, &entropyMetadata->fseMetadata,
-                                                         workspace, wkspSize, writeSeqEntropy);
-    ebs.estBlockSize += ebs.estLitSize + ZSTD_blockHeaderSize;
-    return ebs;
-}
-
-static int ZSTD_needSequenceEntropyTables(ZSTD_fseCTablesMetadata_t const* fseMetadata)
-{
-    if (fseMetadata->llType == set_compressed || fseMetadata->llType == set_rle)
-        return 1;
-    if (fseMetadata->mlType == set_compressed || fseMetadata->mlType == set_rle)
-        return 1;
-    if (fseMetadata->ofType == set_compressed || fseMetadata->ofType == set_rle)
-        return 1;
-    return 0;
-}
-
-static size_t countLiterals(seqStore_t const* seqStore, const seqDef* sp, size_t seqCount)
-{
-    size_t n, total = 0;
-    assert(sp != NULL);
-    for (n=0; n<seqCount; n++) {
-        total += ZSTD_getSequenceLength(seqStore, sp+n).litLength;
-    }
-    DEBUGLOG(6, "countLiterals for %zu sequences from %p => %zu bytes", seqCount, (const void*)sp, total);
-    return total;
-}
-
-#define BYTESCALE 256
-
-static size_t sizeBlockSequences(const seqDef* sp, size_t nbSeqs,
-                size_t targetBudget, size_t avgLitCost, size_t avgSeqCost,
-                int firstSubBlock)
-{
-    size_t n, budget = 0, inSize=0;
-    /* entropy headers */
-    size_t const headerSize = (size_t)firstSubBlock * 120 * BYTESCALE; /* generous estimate */
-    assert(firstSubBlock==0 || firstSubBlock==1);
-    budget += headerSize;
-
-    /* first sequence => at least one sequence*/
-    budget += sp[0].litLength * avgLitCost + avgSeqCost;
-    if (budget > targetBudget) return 1;
-    inSize = sp[0].litLength + (sp[0].mlBase+MINMATCH);
-
-    /* loop over sequences */
-    for (n=1; n<nbSeqs; n++) {
-        size_t currentCost = sp[n].litLength * avgLitCost + avgSeqCost;
-        budget += currentCost;
-        inSize += sp[n].litLength + (sp[n].mlBase+MINMATCH);
-        /* stop when sub-block budget is reached */
-        if ( (budget > targetBudget)
-            /* though continue to expand until the sub-block is deemed compressible */
-          && (budget < inSize * BYTESCALE) )
-            break;
-    }
-
-    return n;
-}
-
-/** ZSTD_compressSubBlock_multi() :
- *  Breaks super-block into multiple sub-blocks and compresses them.
- *  Entropy will be written into the first block.
- *  The following blocks use repeat_mode to compress.
- *  Sub-blocks are all compressed, except the last one when beneficial.
- *  @return : compressed size of the super block (which features multiple ZSTD blocks)
- *            or 0 if it failed to compress. */
-static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr,
-                            const ZSTD_compressedBlockState_t* prevCBlock,
-                            ZSTD_compressedBlockState_t* nextCBlock,
-                            const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
-                            const ZSTD_CCtx_params* cctxParams,
-                                  void* dst, size_t dstCapacity,
-                            const void* src, size_t srcSize,
-                            const int bmi2, U32 lastBlock,
-                            void* workspace, size_t wkspSize)
-{
-    const seqDef* const sstart = seqStorePtr->sequencesStart;
-    const seqDef* const send = seqStorePtr->sequences;
-    const seqDef* sp = sstart; /* tracks progresses within seqStorePtr->sequences */
-    size_t const nbSeqs = (size_t)(send - sstart);
-    const BYTE* const lstart = seqStorePtr->litStart;
-    const BYTE* const lend = seqStorePtr->lit;
-    const BYTE* lp = lstart;
-    size_t const nbLiterals = (size_t)(lend - lstart);
-    BYTE const* ip = (BYTE const*)src;
-    BYTE const* const iend = ip + srcSize;
-    BYTE* const ostart = (BYTE*)dst;
-    BYTE* const oend = ostart + dstCapacity;
-    BYTE* op = ostart;
-    const BYTE* llCodePtr = seqStorePtr->llCode;
-    const BYTE* mlCodePtr = seqStorePtr->mlCode;
-    const BYTE* ofCodePtr = seqStorePtr->ofCode;
-    size_t const minTarget = ZSTD_TARGETCBLOCKSIZE_MIN; /* enforce minimum size, to reduce undesirable side effects */
-    size_t const targetCBlockSize = MAX(minTarget, cctxParams->targetCBlockSize);
-    int writeLitEntropy = (entropyMetadata->hufMetadata.hType == set_compressed);
-    int writeSeqEntropy = 1;
-
-    DEBUGLOG(5, "ZSTD_compressSubBlock_multi (srcSize=%u, litSize=%u, nbSeq=%u)",
-               (unsigned)srcSize, (unsigned)(lend-lstart), (unsigned)(send-sstart));
-
-        /* let's start by a general estimation for the full block */
-    if (nbSeqs > 0) {
-        EstimatedBlockSize const ebs =
-                ZSTD_estimateSubBlockSize(lp, nbLiterals,
-                                        ofCodePtr, llCodePtr, mlCodePtr, nbSeqs,
-                                        &nextCBlock->entropy, entropyMetadata,
-                                        workspace, wkspSize,
-                                        writeLitEntropy, writeSeqEntropy);
-        /* quick estimation */
-        size_t const avgLitCost = nbLiterals ? (ebs.estLitSize * BYTESCALE) / nbLiterals : BYTESCALE;
-        size_t const avgSeqCost = ((ebs.estBlockSize - ebs.estLitSize) * BYTESCALE) / nbSeqs;
-        const size_t nbSubBlocks = MAX((ebs.estBlockSize + (targetCBlockSize/2)) / targetCBlockSize, 1);
-        size_t n, avgBlockBudget, blockBudgetSupp=0;
-        avgBlockBudget = (ebs.estBlockSize * BYTESCALE) / nbSubBlocks;
-        DEBUGLOG(5, "estimated fullblock size=%u bytes ; avgLitCost=%.2f ; avgSeqCost=%.2f ; targetCBlockSize=%u, nbSubBlocks=%u ; avgBlockBudget=%.0f bytes",
-                    (unsigned)ebs.estBlockSize, (double)avgLitCost/BYTESCALE, (double)avgSeqCost/BYTESCALE,
-                    (unsigned)targetCBlockSize, (unsigned)nbSubBlocks, (double)avgBlockBudget/BYTESCALE);
-        /* simplification: if estimates states that the full superblock doesn't compress, just bail out immediately
-         * this will result in the production of a single uncompressed block covering @srcSize.*/
-        if (ebs.estBlockSize > srcSize) return 0;
-
-        /* compress and write sub-blocks */
-        assert(nbSubBlocks>0);
-        for (n=0; n < nbSubBlocks-1; n++) {
-            /* determine nb of sequences for current sub-block + nbLiterals from next sequence */
-            size_t const seqCount = sizeBlockSequences(sp, (size_t)(send-sp),
-                                        avgBlockBudget + blockBudgetSupp, avgLitCost, avgSeqCost, n==0);
-            /* if reached last sequence : break to last sub-block (simplification) */
-            assert(seqCount <= (size_t)(send-sp));
-            if (sp + seqCount == send) break;
-            assert(seqCount > 0);
-            /* compress sub-block */
-            {   int litEntropyWritten = 0;
-                int seqEntropyWritten = 0;
-                size_t litSize = countLiterals(seqStorePtr, sp, seqCount);
-                const size_t decompressedSize =
-                        ZSTD_seqDecompressedSize(seqStorePtr, sp, seqCount, litSize, 0);
-                size_t const cSize = ZSTD_compressSubBlock(&nextCBlock->entropy, entropyMetadata,
-                                                sp, seqCount,
-                                                lp, litSize,
-                                                llCodePtr, mlCodePtr, ofCodePtr,
-                                                cctxParams,
-                                                op, (size_t)(oend-op),
-                                                bmi2, writeLitEntropy, writeSeqEntropy,
-                                                &litEntropyWritten, &seqEntropyWritten,
-                                                0);
-                FORWARD_IF_ERROR(cSize, "ZSTD_compressSubBlock failed");
-
-                /* check compressibility, update state components */
-                if (cSize > 0 && cSize < decompressedSize) {
-                    DEBUGLOG(5, "Committed sub-block compressing %u bytes => %u bytes",
-                                (unsigned)decompressedSize, (unsigned)cSize);
-                    assert(ip + decompressedSize <= iend);
-                    ip += decompressedSize;
-                    lp += litSize;
-                    op += cSize;
-                    llCodePtr += seqCount;
-                    mlCodePtr += seqCount;
-                    ofCodePtr += seqCount;
-                    /* Entropy only needs to be written once */
-                    if (litEntropyWritten) {
-                        writeLitEntropy = 0;
-                    }
-                    if (seqEntropyWritten) {
-                        writeSeqEntropy = 0;
-                    }
-                    sp += seqCount;
-                    blockBudgetSupp = 0;
-            }   }
-            /* otherwise : do not compress yet, coalesce current sub-block with following one */
-        }
-    } /* if (nbSeqs > 0) */
-
-    /* write last block */
-    DEBUGLOG(5, "Generate last sub-block: %u sequences remaining", (unsigned)(send - sp));
-    {   int litEntropyWritten = 0;
-        int seqEntropyWritten = 0;
-        size_t litSize = (size_t)(lend - lp);
-        size_t seqCount = (size_t)(send - sp);
-        const size_t decompressedSize =
-                ZSTD_seqDecompressedSize(seqStorePtr, sp, seqCount, litSize, 1);
-        size_t const cSize = ZSTD_compressSubBlock(&nextCBlock->entropy, entropyMetadata,
-                                            sp, seqCount,
-                                            lp, litSize,
-                                            llCodePtr, mlCodePtr, ofCodePtr,
-                                            cctxParams,
-                                            op, (size_t)(oend-op),
-                                            bmi2, writeLitEntropy, writeSeqEntropy,
-                                            &litEntropyWritten, &seqEntropyWritten,
-                                            lastBlock);
-        FORWARD_IF_ERROR(cSize, "ZSTD_compressSubBlock failed");
-
-        /* update pointers, the nb of literals borrowed from next sequence must be preserved */
-        if (cSize > 0 && cSize < decompressedSize) {
-            DEBUGLOG(5, "Last sub-block compressed %u bytes => %u bytes",
-                        (unsigned)decompressedSize, (unsigned)cSize);
-            assert(ip + decompressedSize <= iend);
-            ip += decompressedSize;
-            lp += litSize;
-            op += cSize;
-            llCodePtr += seqCount;
-            mlCodePtr += seqCount;
-            ofCodePtr += seqCount;
-            /* Entropy only needs to be written once */
-            if (litEntropyWritten) {
-                writeLitEntropy = 0;
-            }
-            if (seqEntropyWritten) {
-                writeSeqEntropy = 0;
-            }
-            sp += seqCount;
-        }
-    }
-
-
-    if (writeLitEntropy) {
-        DEBUGLOG(5, "Literal entropy tables were never written");
-        ZSTD_memcpy(&nextCBlock->entropy.huf, &prevCBlock->entropy.huf, sizeof(prevCBlock->entropy.huf));
-    }
-    if (writeSeqEntropy && ZSTD_needSequenceEntropyTables(&entropyMetadata->fseMetadata)) {
-        /* If we haven't written our entropy tables, then we've violated our contract and
-         * must emit an uncompressed block.
-         */
-        DEBUGLOG(5, "Sequence entropy tables were never written => cancel, emit an uncompressed block");
-        return 0;
-    }
-
-    if (ip < iend) {
-        /* some data left : last part of the block sent uncompressed */
-        size_t const rSize = (size_t)((iend - ip));
-        size_t const cSize = ZSTD_noCompressBlock(op, (size_t)(oend - op), ip, rSize, lastBlock);
-        DEBUGLOG(5, "Generate last uncompressed sub-block of %u bytes", (unsigned)(rSize));
-        FORWARD_IF_ERROR(cSize, "ZSTD_noCompressBlock failed");
-        assert(cSize != 0);
-        op += cSize;
-        /* We have to regenerate the repcodes because we've skipped some sequences */
-        if (sp < send) {
-            const seqDef* seq;
-            repcodes_t rep;
-            ZSTD_memcpy(&rep, prevCBlock->rep, sizeof(rep));
-            for (seq = sstart; seq < sp; ++seq) {
-                ZSTD_updateRep(rep.rep, seq->offBase, ZSTD_getSequenceLength(seqStorePtr, seq).litLength == 0);
-            }
-            ZSTD_memcpy(nextCBlock->rep, &rep, sizeof(rep));
-        }
-    }
-
-    DEBUGLOG(5, "ZSTD_compressSubBlock_multi compressed all subBlocks: total compressed size = %u",
-                (unsigned)(op-ostart));
-    return (size_t)(op-ostart);
-}
-
-size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc,
-                               void* dst, size_t dstCapacity,
-                               const void* src, size_t srcSize,
-                               unsigned lastBlock)
-{
-    ZSTD_entropyCTablesMetadata_t entropyMetadata;
-
-    FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(&zc->seqStore,
-          &zc->blockState.prevCBlock->entropy,
-          &zc->blockState.nextCBlock->entropy,
-          &zc->appliedParams,
-          &entropyMetadata,
-          zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), "");
-
-    return ZSTD_compressSubBlock_multi(&zc->seqStore,
-            zc->blockState.prevCBlock,
-            zc->blockState.nextCBlock,
-            &entropyMetadata,
-            &zc->appliedParams,
-            dst, dstCapacity,
-            src, srcSize,
-            zc->bmi2, lastBlock,
-            zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */);
-}
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_compress_superblock.h

@@ -1,35 +0,0 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
-#ifndef ZSTD_COMPRESS_ADVANCED_H
-#define ZSTD_COMPRESS_ADVANCED_H
-
-/*-*************************************
-*  Dependencies
-***************************************/
-
-#include "zstd.h" /* ZSTD_CCtx */
-
-/*-*************************************
-*  Target Compressed Block Size
-***************************************/
-
-/* ZSTD_compressSuperBlock() :
- * Used to compress a super block when targetCBlockSize is being used.
- * The given block will be compressed into multiple sub blocks that are around targetCBlockSize. */
-size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc,
-                               void* dst, size_t dstCapacity,
-                               void const* src, size_t srcSize,
-                               unsigned lastBlock);
-
-#endif /* ZSTD_COMPRESS_ADVANCED_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_ctx.go

@@ -1,131 +0,0 @@
-package zstd
-
-/*
-#include "zstd.h"
-*/
-import "C"
-import (
-	"bytes"
-	"io/ioutil"
-	"runtime"
-	"unsafe"
-)
-
-type Ctx interface {
-	// Compress src into dst.  If you have a buffer to use, you can pass it to
-	// prevent allocation.  If it is too small, or if nil is passed, a new buffer
-	// will be allocated and returned.
-	Compress(dst, src []byte) ([]byte, error)
-
-	// CompressLevel is the same as Compress but you can pass a compression level
-	CompressLevel(dst, src []byte, level int) ([]byte, error)
-
-	// Decompress src into dst.  If you have a buffer to use, you can pass it to
-	// prevent allocation.  If it is too small, or if nil is passed, a new buffer
-	// will be allocated and returned.
-	Decompress(dst, src []byte) ([]byte, error)
-}
-
-type ctx struct {
-	cctx *C.ZSTD_CCtx
-	dctx *C.ZSTD_DCtx
-}
-
-// Create a new ZStd Context.
-//  When compressing/decompressing many times, it is recommended to allocate a
-//  context just once, and re-use it for each successive compression operation.
-//  This will make workload friendlier for system's memory.
-//  Note : re-using context is just a speed / resource optimization.
-//         It doesn't change the compression ratio, which remains identical.
-//  Note 2 : In multi-threaded environments,
-//         use one different context per thread for parallel execution.
-//
-func NewCtx() Ctx {
-	c := &ctx{
-		cctx: C.ZSTD_createCCtx(),
-		dctx: C.ZSTD_createDCtx(),
-	}
-
-	runtime.SetFinalizer(c, finalizeCtx)
-	return c
-}
-
-func (c *ctx) Compress(dst, src []byte) ([]byte, error) {
-	return c.CompressLevel(dst, src, DefaultCompression)
-}
-
-func (c *ctx) CompressLevel(dst, src []byte, level int) ([]byte, error) {
-	bound := CompressBound(len(src))
-	if cap(dst) >= bound {
-		dst = dst[0:bound] // Reuse dst buffer
-	} else {
-		dst = make([]byte, bound)
-	}
-
-	// We need unsafe.Pointer(&src[0]) in the Cgo call to avoid "Go pointer to Go pointer" panics.
-	// This means we need to special case empty input. See:
-	// https://github.com/golang/go/issues/14210#issuecomment-346402945
-	var cWritten C.size_t
-	if len(src) == 0 {
-		cWritten = C.ZSTD_compressCCtx(
-			c.cctx,
-			unsafe.Pointer(&dst[0]),
-			C.size_t(len(dst)),
-			unsafe.Pointer(nil),
-			C.size_t(0),
-			C.int(level))
-	} else {
-		cWritten = C.ZSTD_compressCCtx(
-			c.cctx,
-			unsafe.Pointer(&dst[0]),
-			C.size_t(len(dst)),
-			unsafe.Pointer(&src[0]),
-			C.size_t(len(src)),
-			C.int(level))
-	}
-
-	written := int(cWritten)
-	// Check if the return is an Error code
-	if err := getError(written); err != nil {
-		return nil, err
-	}
-	return dst[:written], nil
-}
-
-func (c *ctx) Decompress(dst, src []byte) ([]byte, error) {
-	if len(src) == 0 {
-		return []byte{}, ErrEmptySlice
-	}
-
-	bound := decompressSizeHint(src)
-	if cap(dst) >= bound {
-		dst = dst[0:cap(dst)]
-	} else {
-		dst = make([]byte, bound)
-	}
-
-	written := int(C.ZSTD_decompressDCtx(
-		c.dctx,
-		unsafe.Pointer(&dst[0]),
-		C.size_t(len(dst)),
-		unsafe.Pointer(&src[0]),
-		C.size_t(len(src))))
-
-	err := getError(written)
-	if err == nil {
-		return dst[:written], nil
-	}
-	if !IsDstSizeTooSmallError(err) {
-		return nil, err
-	}
-
-	// We failed getting a dst buffer of correct size, use stream API
-	r := NewReader(bytes.NewReader(src))
-	defer r.Close()
-	return ioutil.ReadAll(r)
-}
-
-func finalizeCtx(c *ctx) {
-	C.ZSTD_freeCCtx(c.cctx)
-	C.ZSTD_freeDCtx(c.dctx)
-}

vendor/github.com/DataDog/zstd/zstd_cwksp.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -15,9 +14,7 @@
 /*-*************************************
 *  Dependencies
 ***************************************/
-#include "allocations.h"  /* ZSTD_customMalloc, ZSTD_customFree */
 #include "zstd_internal.h"
-#include "portability_macros.h"
 
 #if defined (__cplusplus)
 extern "C" {
@@ -27,6 +24,16 @@ extern "C" {
 *  Constants
 ***************************************/
 
+/* define "workspace is too large" as this number of times larger than needed */
+#define ZSTD_WORKSPACETOOLARGE_FACTOR 3
+
+/* when workspace is continuously too large
+ * during at least this number of times,
+ * context's memory usage is considered wasteful,
+ * because it's sized to handle a worst case scenario which rarely happens.
+ * In which case, resize it down to free some memory */
+#define ZSTD_WORKSPACETOOLARGE_MAXDURATION 128
+
 /* Since the workspace is effectively its own little malloc implementation /
  * arena, when we run under ASAN, we should similarly insert redzones between
  * each internal element of the workspace, so ASAN will catch overruns that
@@ -38,30 +45,15 @@ extern "C" {
 #define ZSTD_CWKSP_ASAN_REDZONE_SIZE 128
 #endif
 
-
-/* Set our tables and aligneds to align by 64 bytes */
-#define ZSTD_CWKSP_ALIGNMENT_BYTES 64
-
 /*-*************************************
 *  Structures
 ***************************************/
 typedef enum {
     ZSTD_cwksp_alloc_objects,
-    ZSTD_cwksp_alloc_aligned_init_once,
-    ZSTD_cwksp_alloc_aligned,
-    ZSTD_cwksp_alloc_buffers
+    ZSTD_cwksp_alloc_buffers,
+    ZSTD_cwksp_alloc_aligned
 } ZSTD_cwksp_alloc_phase_e;
 
-/**
- * Used to describe whether the workspace is statically allocated (and will not
- * necessarily ever be freed), or if it's dynamically allocated and we can
- * expect a well-formed caller to free this.
- */
-typedef enum {
-    ZSTD_cwksp_dynamic_alloc,
-    ZSTD_cwksp_static_alloc
-} ZSTD_cwksp_static_alloc_e;
-
 /**
  * Zstd fits all its internal datastructures into a single continuous buffer,
  * so that it only needs to perform a single OS allocation (or so that a buffer
@@ -102,15 +94,15 @@ typedef enum {
  *
  * Workspace Layout:
  *
- * [                        ... workspace ...                           ]
- * [objects][tables ->] free space [<- buffers][<- aligned][<- init once]
+ * [                        ... workspace ...                         ]
+ * [objects][tables ... ->] free space [<- ... aligned][<- ... buffers]
  *
  * The various objects that live in the workspace are divided into the
  * following categories, and are allocated separately:
  *
  * - Static objects: this is optionally the enclosing ZSTD_CCtx or ZSTD_CDict,
  *   so that literally everything fits in a single buffer. Note: if present,
- *   this must be the first object in the workspace, since ZSTD_customFree{CCtx,
+ *   this must be the first object in the workspace, since ZSTD_free{CCtx,
  *   CDict}() rely on a pointer comparison to see whether one or two frees are
  *   required.
  *
@@ -125,20 +117,10 @@ typedef enum {
  * - Tables: these are any of several different datastructures (hash tables,
  *   chain tables, binary trees) that all respect a common format: they are
  *   uint32_t arrays, all of whose values are between 0 and (nextSrc - base).
- *   Their sizes depend on the cparams. These tables are 64-byte aligned.
+ *   Their sizes depend on the cparams.
  *
- * - Init once: these buffers require to be initialized at least once before
- *   use. They should be used when we want to skip memory initialization
- *   while not triggering memory checkers (like Valgrind) when reading from
- *   from this memory without writing to it first.
- *   These buffers should be used carefully as they might contain data
- *   from previous compressions.
- *   Buffers are aligned to 64 bytes.
- *
- * - Aligned: these buffers don't require any initialization before they're
- *   used. The user of the buffer should make sure they write into a buffer
- *   location before reading from it.
- *   Buffers are aligned to 64 bytes.
+ * - Aligned: these buffers are used for various purposes that require 4 byte
+ *   alignment, but don't require any initialization before they're used.
  *
  * - Buffers: these buffers are used for various purposes that don't require
  *   any alignment or initialization before they're used. This means they can
@@ -150,9 +132,9 @@ typedef enum {
  * correctly packed into the workspace buffer. That order is:
  *
  * 1. Objects
- * 2. Init once / Tables
- * 3. Aligned / Tables
- * 4. Buffers / Tables
+ * 2. Buffers
+ * 3. Aligned
+ * 4. Tables
  *
  * Attempts to reserve objects of different types out of order will fail.
  */
@@ -164,12 +146,10 @@ typedef struct {
     void* tableEnd;
     void* tableValidEnd;
     void* allocStart;
-    void* initOnceStart;
 
-    BYTE allocFailed;
+    int allocFailed;
     int workspaceOversizedDuration;
     ZSTD_cwksp_alloc_phase_e phase;
-    ZSTD_cwksp_static_alloc_e isStatic;
 } ZSTD_cwksp;
 
 /*-*************************************
@@ -177,7 +157,6 @@ typedef struct {
 ***************************************/
 
 MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws);
-MEM_STATIC void*  ZSTD_cwksp_initialAllocStart(ZSTD_cwksp* ws);
 
 MEM_STATIC void ZSTD_cwksp_assert_internal_consistency(ZSTD_cwksp* ws) {
     (void)ws;
@@ -187,21 +166,6 @@ MEM_STATIC void ZSTD_cwksp_assert_internal_consistency(ZSTD_cwksp* ws) {
     assert(ws->tableEnd <= ws->allocStart);
     assert(ws->tableValidEnd <= ws->allocStart);
     assert(ws->allocStart <= ws->workspaceEnd);
-    assert(ws->initOnceStart <= ZSTD_cwksp_initialAllocStart(ws));
-    assert(ws->workspace <= ws->initOnceStart);
-#if ZSTD_MEMORY_SANITIZER
-    {
-        intptr_t const offset = __msan_test_shadow(ws->initOnceStart,
-            (U8*)ZSTD_cwksp_initialAllocStart(ws) - (U8*)ws->initOnceStart);
-        (void)offset;
-#if defined(ZSTD_MSAN_PRINT)
-        if(offset!=-1) {
-            __msan_print_shadow((U8*)ws->initOnceStart + offset - 8, 32);
-        }
-#endif
-        assert(offset==-1);
-    };
-#endif
 }
 
 /**
@@ -222,72 +186,63 @@ MEM_STATIC size_t ZSTD_cwksp_align(size_t size, size_t const align) {
  * Since tables aren't currently redzoned, you don't need to call through this
  * to figure out how much space you need for the matchState tables. Everything
  * else is though.
- *
- * Do not use for sizing aligned buffers. Instead, use ZSTD_cwksp_aligned_alloc_size().
  */
 MEM_STATIC size_t ZSTD_cwksp_alloc_size(size_t size) {
-    if (size == 0)
-        return 0;
-#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
+#if defined (ADDRESS_SANITIZER) && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
     return size + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
 #else
     return size;
 #endif
 }
 
-/**
- * Returns an adjusted alloc size that is the nearest larger multiple of 64 bytes.
- * Used to determine the number of bytes required for a given "aligned".
- */
-MEM_STATIC size_t ZSTD_cwksp_aligned_alloc_size(size_t size) {
-    return ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(size, ZSTD_CWKSP_ALIGNMENT_BYTES));
+MEM_STATIC void ZSTD_cwksp_internal_advance_phase(
+        ZSTD_cwksp* ws, ZSTD_cwksp_alloc_phase_e phase) {
+    assert(phase >= ws->phase);
+    if (phase > ws->phase) {
+        if (ws->phase < ZSTD_cwksp_alloc_buffers &&
+                phase >= ZSTD_cwksp_alloc_buffers) {
+            ws->tableValidEnd = ws->objectEnd;
+        }
+        if (ws->phase < ZSTD_cwksp_alloc_aligned &&
+                phase >= ZSTD_cwksp_alloc_aligned) {
+            /* If unaligned allocations down from a too-large top have left us
+             * unaligned, we need to realign our alloc ptr. Technically, this
+             * can consume space that is unaccounted for in the neededSpace
+             * calculation. However, I believe this can only happen when the
+             * workspace is too large, and specifically when it is too large
+             * by a larger margin than the space that will be consumed. */
+            /* TODO: cleaner, compiler warning friendly way to do this??? */
+            ws->allocStart = (BYTE*)ws->allocStart - ((size_t)ws->allocStart & (sizeof(U32)-1));
+            if (ws->allocStart < ws->tableValidEnd) {
+                ws->tableValidEnd = ws->allocStart;
+            }
+        }
+        ws->phase = phase;
+    }
 }
 
 /**
- * Returns the amount of additional space the cwksp must allocate
- * for internal purposes (currently only alignment).
+ * Returns whether this object/buffer/etc was allocated in this workspace.
  */
-MEM_STATIC size_t ZSTD_cwksp_slack_space_required(void) {
-    /* For alignment, the wksp will always allocate an additional 2*ZSTD_CWKSP_ALIGNMENT_BYTES
-     * bytes to align the beginning of tables section and end of buffers;
-     */
-    size_t const slackSpace = ZSTD_CWKSP_ALIGNMENT_BYTES * 2;
-    return slackSpace;
-}
-
-
-/**
- * Return the number of additional bytes required to align a pointer to the given number of bytes.
- * alignBytes must be a power of two.
- */
-MEM_STATIC size_t ZSTD_cwksp_bytes_to_align_ptr(void* ptr, const size_t alignBytes) {
-    size_t const alignBytesMask = alignBytes - 1;
-    size_t const bytes = (alignBytes - ((size_t)ptr & (alignBytesMask))) & alignBytesMask;
-    assert((alignBytes & alignBytesMask) == 0);
-    assert(bytes < alignBytes);
-    return bytes;
-}
-
-/**
- * Returns the initial value for allocStart which is used to determine the position from
- * which we can allocate from the end of the workspace.
- */
-MEM_STATIC void*  ZSTD_cwksp_initialAllocStart(ZSTD_cwksp* ws) {
-    return (void*)((size_t)ws->workspaceEnd & ~(ZSTD_CWKSP_ALIGNMENT_BYTES-1));
+MEM_STATIC int ZSTD_cwksp_owns_buffer(const ZSTD_cwksp* ws, const void* ptr) {
+    return (ptr != NULL) && (ws->workspace <= ptr) && (ptr <= ws->workspaceEnd);
 }
 
 /**
  * Internal function. Do not use directly.
- * Reserves the given number of bytes within the aligned/buffer segment of the wksp,
- * which counts from the end of the wksp (as opposed to the object/table segment).
- *
- * Returns a pointer to the beginning of that space.
  */
-MEM_STATIC void*
-ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t const bytes)
-{
-    void* const alloc = (BYTE*)ws->allocStart - bytes;
-    void* const bottom = ws->tableEnd;
+MEM_STATIC void* ZSTD_cwksp_reserve_internal(
+        ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase_e phase) {
+    void* alloc;
+    void* bottom = ws->tableEnd;
+    ZSTD_cwksp_internal_advance_phase(ws, phase);
+    alloc = (BYTE *)ws->allocStart - bytes;
+
+#if defined (ADDRESS_SANITIZER) && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
+    /* over-reserve space */
+    alloc = (BYTE *)alloc - 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
+#endif
+
     DEBUGLOG(5, "cwksp: reserving %p %zd bytes, %zd bytes remaining",
         alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes);
     ZSTD_cwksp_assert_internal_consistency(ws);
@@ -297,88 +252,16 @@ ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t const bytes)
         ws->allocFailed = 1;
         return NULL;
     }
-    /* the area is reserved from the end of wksp.
-     * If it overlaps with tableValidEnd, it voids guarantees on values' range */
     if (alloc < ws->tableValidEnd) {
         ws->tableValidEnd = alloc;
     }
     ws->allocStart = alloc;
-    return alloc;
-}
 
-/**
- * Moves the cwksp to the next phase, and does any necessary allocations.
- * cwksp initialization must necessarily go through each phase in order.
- * Returns a 0 on success, or zstd error
- */
-MEM_STATIC size_t
-ZSTD_cwksp_internal_advance_phase(ZSTD_cwksp* ws, ZSTD_cwksp_alloc_phase_e phase)
-{
-    assert(phase >= ws->phase);
-    if (phase > ws->phase) {
-        /* Going from allocating objects to allocating initOnce / tables */
-        if (ws->phase < ZSTD_cwksp_alloc_aligned_init_once &&
-            phase >= ZSTD_cwksp_alloc_aligned_init_once) {
-            ws->tableValidEnd = ws->objectEnd;
-            ws->initOnceStart = ZSTD_cwksp_initialAllocStart(ws);
-
-            {   /* Align the start of the tables to 64 bytes. Use [0, 63] bytes */
-                void *const alloc = ws->objectEnd;
-                size_t const bytesToAlign = ZSTD_cwksp_bytes_to_align_ptr(alloc, ZSTD_CWKSP_ALIGNMENT_BYTES);
-                void *const objectEnd = (BYTE *) alloc + bytesToAlign;
-                DEBUGLOG(5, "reserving table alignment addtl space: %zu", bytesToAlign);
-                RETURN_ERROR_IF(objectEnd > ws->workspaceEnd, memory_allocation,
-                                "table phase - alignment initial allocation failed!");
-                ws->objectEnd = objectEnd;
-                ws->tableEnd = objectEnd;  /* table area starts being empty */
-                if (ws->tableValidEnd < ws->tableEnd) {
-                    ws->tableValidEnd = ws->tableEnd;
-                }
-            }
-        }
-        ws->phase = phase;
-        ZSTD_cwksp_assert_internal_consistency(ws);
-    }
-    return 0;
-}
-
-/**
- * Returns whether this object/buffer/etc was allocated in this workspace.
- */
-MEM_STATIC int ZSTD_cwksp_owns_buffer(const ZSTD_cwksp* ws, const void* ptr)
-{
-    return (ptr != NULL) && (ws->workspace <= ptr) && (ptr < ws->workspaceEnd);
-}
-
-/**
- * Internal function. Do not use directly.
- */
-MEM_STATIC void*
-ZSTD_cwksp_reserve_internal(ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase_e phase)
-{
-    void* alloc;
-    if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase)) || bytes == 0) {
-        return NULL;
-    }
-
-#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
-    /* over-reserve space */
-    bytes += 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
-#endif
-
-    alloc = ZSTD_cwksp_reserve_internal_buffer_space(ws, bytes);
-
-#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
+#if defined (ADDRESS_SANITIZER) && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
     /* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on
      * either size. */
-    if (alloc) {
-        alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
-        if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
-            /* We need to keep the redzone poisoned while unpoisoning the bytes that
-             * are actually allocated. */
-            __asan_unpoison_memory_region(alloc, bytes - 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE);
-        }
-    }
+    alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
+    __asan_unpoison_memory_region(alloc, bytes);
 #endif
 
     return alloc;
@@ -387,78 +270,33 @@ ZSTD_cwksp_reserve_internal(ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase
 /**
  * Reserves and returns unaligned memory.
  */
-MEM_STATIC BYTE* ZSTD_cwksp_reserve_buffer(ZSTD_cwksp* ws, size_t bytes)
-{
+MEM_STATIC BYTE* ZSTD_cwksp_reserve_buffer(ZSTD_cwksp* ws, size_t bytes) {
     return (BYTE*)ZSTD_cwksp_reserve_internal(ws, bytes, ZSTD_cwksp_alloc_buffers);
 }
 
 /**
- * Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes).
- * This memory has been initialized at least once in the past.
- * This doesn't mean it has been initialized this time, and it might contain data from previous
- * operations.
- * The main usage is for algorithms that might need read access into uninitialized memory.
- * The algorithm must maintain safety under these conditions and must make sure it doesn't
- * leak any of the past data (directly or in side channels).
+ * Reserves and returns memory sized on and aligned on sizeof(unsigned).
  */
-MEM_STATIC void* ZSTD_cwksp_reserve_aligned_init_once(ZSTD_cwksp* ws, size_t bytes)
-{
-    size_t const alignedBytes = ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES);
-    void* ptr = ZSTD_cwksp_reserve_internal(ws, alignedBytes, ZSTD_cwksp_alloc_aligned_init_once);
-    assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
-    if(ptr && ptr < ws->initOnceStart) {
-        /* We assume the memory following the current allocation is either:
-         * 1. Not usable as initOnce memory (end of workspace)
-         * 2. Another initOnce buffer that has been allocated before (and so was previously memset)
-         * 3. An ASAN redzone, in which case we don't want to write on it
-         * For these reasons it should be fine to not explicitly zero every byte up to ws->initOnceStart.
-         * Note that we assume here that MSAN and ASAN cannot run in the same time. */
-        ZSTD_memset(ptr, 0, MIN((size_t)((U8*)ws->initOnceStart - (U8*)ptr), alignedBytes));
-        ws->initOnceStart = ptr;
-    }
-#if ZSTD_MEMORY_SANITIZER
-    assert(__msan_test_shadow(ptr, bytes) == -1);
-#endif
-    return ptr;
+MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes) {
+    assert((bytes & (sizeof(U32)-1)) == 0);
+    return ZSTD_cwksp_reserve_internal(ws, ZSTD_cwksp_align(bytes, sizeof(U32)), ZSTD_cwksp_alloc_aligned);
 }
 
 /**
- * Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes).
- */
-MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes)
-{
-    void* ptr = ZSTD_cwksp_reserve_internal(ws, ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES),
-                                            ZSTD_cwksp_alloc_aligned);
-    assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
-    return ptr;
-}
-
-/**
- * Aligned on 64 bytes. These buffers have the special property that
- * their values remain constrained, allowing us to reuse them without
+ * Aligned on sizeof(unsigned). These buffers have the special property that
+ * their values remain constrained, allowing us to re-use them without
  * memset()-ing them.
  */
-MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes)
-{
-    const ZSTD_cwksp_alloc_phase_e phase = ZSTD_cwksp_alloc_aligned_init_once;
-    void* alloc;
-    void* end;
-    void* top;
-
-    /* We can only start allocating tables after we are done reserving space for objects at the
-     * start of the workspace */
-    if(ws->phase < phase) {
-        if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase))) {
-            return NULL;
-        }
-    }
-    alloc = ws->tableEnd;
-    end = (BYTE *)alloc + bytes;
-    top = ws->allocStart;
+MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes) {
+    const ZSTD_cwksp_alloc_phase_e phase = ZSTD_cwksp_alloc_aligned;
+    void* alloc = ws->tableEnd;
+    void* end = (BYTE *)alloc + bytes;
+    void* top = ws->allocStart;
 
     DEBUGLOG(5, "cwksp: reserving %p table %zd bytes, %zd bytes remaining",
         alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes);
     assert((bytes & (sizeof(U32)-1)) == 0);
+    ZSTD_cwksp_internal_advance_phase(ws, phase);
     ZSTD_cwksp_assert_internal_consistency(ws);
     assert(end <= top);
     if (end > top) {
@@ -468,41 +306,35 @@ MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes)
     }
     ws->tableEnd = end;
 
-#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
-    if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
-        __asan_unpoison_memory_region(alloc, bytes);
-    }
+#if defined (ADDRESS_SANITIZER) && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
+    __asan_unpoison_memory_region(alloc, bytes);
 #endif
 
-    assert((bytes & (ZSTD_CWKSP_ALIGNMENT_BYTES-1)) == 0);
-    assert(((size_t)alloc & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
     return alloc;
 }
 
 /**
  * Aligned on sizeof(void*).
- * Note : should happen only once, at workspace first initialization
  */
-MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes)
-{
-    size_t const roundedBytes = ZSTD_cwksp_align(bytes, sizeof(void*));
+MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) {
+    size_t roundedBytes = ZSTD_cwksp_align(bytes, sizeof(void*));
     void* alloc = ws->objectEnd;
     void* end = (BYTE*)alloc + roundedBytes;
 
-#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
+#if defined (ADDRESS_SANITIZER) && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
     /* over-reserve space */
     end = (BYTE *)end + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
 #endif
 
-    DEBUGLOG(4,
+    DEBUGLOG(5,
         "cwksp: reserving %p object %zd bytes (rounded to %zd), %zd bytes remaining",
         alloc, bytes, roundedBytes, ZSTD_cwksp_available_space(ws) - roundedBytes);
-    assert((size_t)alloc % ZSTD_ALIGNOF(void*) == 0);
-    assert(bytes % ZSTD_ALIGNOF(void*) == 0);
+    assert(((size_t)alloc & (sizeof(void*)-1)) == 0);
+    assert((bytes & (sizeof(void*)-1)) == 0);
     ZSTD_cwksp_assert_internal_consistency(ws);
     /* we must be in the first phase, no advance is possible */
     if (ws->phase != ZSTD_cwksp_alloc_objects || end > ws->workspaceEnd) {
-        DEBUGLOG(3, "cwksp: object alloc failed!");
+        DEBUGLOG(4, "cwksp: object alloc failed!");
         ws->allocFailed = 1;
         return NULL;
     }
@@ -510,38 +342,27 @@ MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes)
     ws->tableEnd = end;
     ws->tableValidEnd = end;
 
-#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
+#if defined (ADDRESS_SANITIZER) && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
     /* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on
      * either size. */
-    alloc = (BYTE*)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
-    if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
-        __asan_unpoison_memory_region(alloc, bytes);
-    }
+    alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
+    __asan_unpoison_memory_region(alloc, bytes);
 #endif
 
     return alloc;
 }
 
-MEM_STATIC void ZSTD_cwksp_mark_tables_dirty(ZSTD_cwksp* ws)
-{
+MEM_STATIC void ZSTD_cwksp_mark_tables_dirty(ZSTD_cwksp* ws) {
     DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_dirty");
 
-#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
-    /* To validate that the table reuse logic is sound, and that we don't
+#if defined (MEMORY_SANITIZER) && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
+    /* To validate that the table re-use logic is sound, and that we don't
      * access table space that we haven't cleaned, we re-"poison" the table
-     * space every time we mark it dirty.
-     * Since tableValidEnd space and initOnce space may overlap we don't poison
-     * the initOnce portion as it break its promise. This means that this poisoning
-     * check isn't always applied fully. */
+     * space every time we mark it dirty. */
     {
         size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd;
         assert(__msan_test_shadow(ws->objectEnd, size) == -1);
-        if((BYTE*)ws->tableValidEnd < (BYTE*)ws->initOnceStart) {
-            __msan_poison(ws->objectEnd, size);
-        } else {
-            assert(ws->initOnceStart >= ws->objectEnd);
-            __msan_poison(ws->objectEnd, (BYTE*)ws->initOnceStart - (BYTE*)ws->objectEnd);
-        }
+        __msan_poison(ws->objectEnd, size);
     }
 #endif
 
@@ -569,7 +390,7 @@ MEM_STATIC void ZSTD_cwksp_clean_tables(ZSTD_cwksp* ws) {
     assert(ws->tableValidEnd >= ws->objectEnd);
     assert(ws->tableValidEnd <= ws->allocStart);
     if (ws->tableValidEnd < ws->tableEnd) {
-        ZSTD_memset(ws->tableValidEnd, 0, (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->tableValidEnd));
+        memset(ws->tableValidEnd, 0, (BYTE*)ws->tableEnd - (BYTE*)ws->tableValidEnd);
     }
     ZSTD_cwksp_mark_tables_clean(ws);
 }
@@ -581,12 +402,8 @@ MEM_STATIC void ZSTD_cwksp_clean_tables(ZSTD_cwksp* ws) {
 MEM_STATIC void ZSTD_cwksp_clear_tables(ZSTD_cwksp* ws) {
     DEBUGLOG(4, "cwksp: clearing tables!");
 
-#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
-    /* We don't do this when the workspace is statically allocated, because
-     * when that is the case, we have no capability to hook into the end of the
-     * workspace's lifecycle to unpoison the memory.
-     */
-    if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
+#if defined (ADDRESS_SANITIZER) && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
+    {
         size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd;
         __asan_poison_memory_region(ws->objectEnd, size);
     }
@@ -603,96 +420,77 @@ MEM_STATIC void ZSTD_cwksp_clear_tables(ZSTD_cwksp* ws) {
 MEM_STATIC void ZSTD_cwksp_clear(ZSTD_cwksp* ws) {
     DEBUGLOG(4, "cwksp: clearing!");
 
-#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
-    /* To validate that the context reuse logic is sound, and that we don't
+#if defined (MEMORY_SANITIZER) && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
+    /* To validate that the context re-use logic is sound, and that we don't
      * access stuff that this compression hasn't initialized, we re-"poison"
-     * the workspace except for the areas in which we expect memory reuse
-     * without initialization (objects, valid tables area and init once
-     * memory). */
+     * the workspace (or at least the non-static, non-table parts of it)
+     * every time we start a new compression. */
     {
-        if((BYTE*)ws->tableValidEnd < (BYTE*)ws->initOnceStart) {
-            size_t size = (BYTE*)ws->initOnceStart - (BYTE*)ws->tableValidEnd;
-            __msan_poison(ws->tableValidEnd, size);
-        }
+        size_t size = (BYTE*)ws->workspaceEnd - (BYTE*)ws->tableValidEnd;
+        __msan_poison(ws->tableValidEnd, size);
     }
 #endif
 
-#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
-    /* We don't do this when the workspace is statically allocated, because
-     * when that is the case, we have no capability to hook into the end of the
-     * workspace's lifecycle to unpoison the memory.
-     */
-    if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
+#if defined (ADDRESS_SANITIZER) && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
+    {
         size_t size = (BYTE*)ws->workspaceEnd - (BYTE*)ws->objectEnd;
         __asan_poison_memory_region(ws->objectEnd, size);
     }
 #endif
 
     ws->tableEnd = ws->objectEnd;
-    ws->allocStart = ZSTD_cwksp_initialAllocStart(ws);
+    ws->allocStart = ws->workspaceEnd;
     ws->allocFailed = 0;
-    if (ws->phase > ZSTD_cwksp_alloc_aligned_init_once) {
-        ws->phase = ZSTD_cwksp_alloc_aligned_init_once;
+    if (ws->phase > ZSTD_cwksp_alloc_buffers) {
+        ws->phase = ZSTD_cwksp_alloc_buffers;
     }
     ZSTD_cwksp_assert_internal_consistency(ws);
 }
 
-MEM_STATIC size_t ZSTD_cwksp_sizeof(const ZSTD_cwksp* ws) {
-    return (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->workspace);
-}
-
-MEM_STATIC size_t ZSTD_cwksp_used(const ZSTD_cwksp* ws) {
-    return (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->workspace)
-         + (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->allocStart);
-}
-
 /**
  * The provided workspace takes ownership of the buffer [start, start+size).
  * Any existing values in the workspace are ignored (the previously managed
  * buffer, if present, must be separately freed).
  */
-MEM_STATIC void ZSTD_cwksp_init(ZSTD_cwksp* ws, void* start, size_t size, ZSTD_cwksp_static_alloc_e isStatic) {
+MEM_STATIC void ZSTD_cwksp_init(ZSTD_cwksp* ws, void* start, size_t size) {
     DEBUGLOG(4, "cwksp: init'ing workspace with %zd bytes", size);
     assert(((size_t)start & (sizeof(void*)-1)) == 0); /* ensure correct alignment */
     ws->workspace = start;
     ws->workspaceEnd = (BYTE*)start + size;
     ws->objectEnd = ws->workspace;
     ws->tableValidEnd = ws->objectEnd;
-    ws->initOnceStart = ZSTD_cwksp_initialAllocStart(ws);
     ws->phase = ZSTD_cwksp_alloc_objects;
-    ws->isStatic = isStatic;
     ZSTD_cwksp_clear(ws);
     ws->workspaceOversizedDuration = 0;
     ZSTD_cwksp_assert_internal_consistency(ws);
 }
 
 MEM_STATIC size_t ZSTD_cwksp_create(ZSTD_cwksp* ws, size_t size, ZSTD_customMem customMem) {
-    void* workspace = ZSTD_customMalloc(size, customMem);
+    void* workspace = ZSTD_malloc(size, customMem);
     DEBUGLOG(4, "cwksp: creating new workspace with %zd bytes", size);
-    RETURN_ERROR_IF(workspace == NULL, memory_allocation, "NULL pointer!");
-    ZSTD_cwksp_init(ws, workspace, size, ZSTD_cwksp_dynamic_alloc);
+    RETURN_ERROR_IF(workspace == NULL, memory_allocation);
+    ZSTD_cwksp_init(ws, workspace, size);
     return 0;
 }
 
 MEM_STATIC void ZSTD_cwksp_free(ZSTD_cwksp* ws, ZSTD_customMem customMem) {
     void *ptr = ws->workspace;
     DEBUGLOG(4, "cwksp: freeing workspace");
-#if ZSTD_MEMORY_SANITIZER && !defined(ZSTD_MSAN_DONT_POISON_WORKSPACE)
-    if (ptr != NULL && customMem.customFree != NULL) {
-        __msan_unpoison(ptr, ZSTD_cwksp_sizeof(ws));
-    }
-#endif
-    ZSTD_memset(ws, 0, sizeof(ZSTD_cwksp));
-    ZSTD_customFree(ptr, customMem);
+    memset(ws, 0, sizeof(ZSTD_cwksp));
+    ZSTD_free(ptr, customMem);
 }
 
 /**
  * Moves the management of a workspace from one cwksp to another. The src cwksp
- * is left in an invalid state (src must be re-init()'ed before it's used again).
+ * is left in an invalid state (src must be re-init()'ed before its used again).
  */
 MEM_STATIC void ZSTD_cwksp_move(ZSTD_cwksp* dst, ZSTD_cwksp* src) {
     *dst = *src;
-    ZSTD_memset(src, 0, sizeof(ZSTD_cwksp));
+    memset(src, 0, sizeof(ZSTD_cwksp));
+}
+
+MEM_STATIC size_t ZSTD_cwksp_sizeof(const ZSTD_cwksp* ws) {
+    return (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->workspace);
 }
 
 MEM_STATIC int ZSTD_cwksp_reserve_failed(const ZSTD_cwksp* ws) {
@@ -703,18 +501,6 @@ MEM_STATIC int ZSTD_cwksp_reserve_failed(const ZSTD_cwksp* ws) {
 *  Functions Checking Free Space
 ***************************************/
 
-/* ZSTD_alignmentSpaceWithinBounds() :
- * Returns if the estimated space needed for a wksp is within an acceptable limit of the
- * actual amount of space used.
- */
-MEM_STATIC int ZSTD_cwksp_estimated_space_within_bounds(const ZSTD_cwksp *const ws, size_t const estimatedSpace) {
-    /* We have an alignment space between objects and tables between tables and buffers, so we can have up to twice
-     * the alignment bytes difference between estimation and actual usage */
-    return (estimatedSpace - ZSTD_cwksp_slack_space_required()) <= ZSTD_cwksp_used(ws) &&
-           ZSTD_cwksp_used(ws) <= estimatedSpace;
-}
-
-
 MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws) {
     return (size_t)((BYTE*)ws->allocStart - (BYTE*)ws->tableEnd);
 }
@@ -747,5 +533,3 @@ MEM_STATIC void ZSTD_cwksp_bump_oversized_duration(
 #endif
 
 #endif /* ZSTD_CWKSP_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_ddict.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -15,18 +14,18 @@
 /*-*******************************************************
 *  Dependencies
 *********************************************************/
-#include "allocations.h"  /* ZSTD_customMalloc, ZSTD_customFree */
-#include "zstd_deps.h"   /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */
+#include <string.h>      /* memcpy, memmove, memset */
 #include "cpu.h"         /* bmi2 */
 #include "mem.h"         /* low level memory routines */
 #define FSE_STATIC_LINKING_ONLY
 #include "fse.h"
+#define HUF_STATIC_LINKING_ONLY
 #include "huf.h"
 #include "zstd_decompress_internal.h"
 #include "zstd_ddict.h"
 
 #if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1)
-#include "zstd_legacy.h"
+#  include "zstd_legacy.h"
 #endif
 
 
@@ -66,10 +65,6 @@ void ZSTD_copyDDictParameters(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict)
     dctx->virtualStart = ddict->dictContent;
     dctx->dictEnd = (const BYTE*)ddict->dictContent + ddict->dictSize;
     dctx->previousDstEnd = dctx->dictEnd;
-#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
-    dctx->dictContentBeginForFuzzing = dctx->prefixStart;
-    dctx->dictContentEndForFuzzing = dctx->previousDstEnd;
-#endif
     if (ddict->entropyPresent) {
         dctx->litEntropy = 1;
         dctx->fseEntropy = 1;
@@ -112,7 +107,7 @@ ZSTD_loadEntropy_intoDDict(ZSTD_DDict* ddict,
     /* load entropy tables */
     RETURN_ERROR_IF(ZSTD_isError(ZSTD_loadDEntropy(
             &ddict->entropy, ddict->dictContent, ddict->dictSize)),
-        dictionary_corrupted, "");
+        dictionary_corrupted);
     ddict->entropyPresent = 1;
     return 0;
 }
@@ -128,17 +123,17 @@ static size_t ZSTD_initDDict_internal(ZSTD_DDict* ddict,
         ddict->dictContent = dict;
         if (!dict) dictSize = 0;
     } else {
-        void* const internalBuffer = ZSTD_customMalloc(dictSize, ddict->cMem);
+        void* const internalBuffer = ZSTD_malloc(dictSize, ddict->cMem);
         ddict->dictBuffer = internalBuffer;
         ddict->dictContent = internalBuffer;
         if (!internalBuffer) return ERROR(memory_allocation);
-        ZSTD_memcpy(internalBuffer, dict, dictSize);
+        memcpy(internalBuffer, dict, dictSize);
     }
     ddict->dictSize = dictSize;
-    ddict->entropy.hufTable[0] = (HUF_DTable)((ZSTD_HUFFDTABLE_CAPACITY_LOG)*0x1000001);  /* cover both little and big endian */
+    ddict->entropy.hufTable[0] = (HUF_DTable)((HufLog)*0x1000001);  /* cover both little and big endian */
 
     /* parse dictionary content */
-    FORWARD_IF_ERROR( ZSTD_loadEntropy_intoDDict(ddict, dictContentType) , "");
+    FORWARD_IF_ERROR( ZSTD_loadEntropy_intoDDict(ddict, dictContentType) );
 
     return 0;
 }
@@ -148,9 +143,9 @@ ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize,
                                       ZSTD_dictContentType_e dictContentType,
                                       ZSTD_customMem customMem)
 {
-    if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL;
+    if (!customMem.customAlloc ^ !customMem.customFree) return NULL;
 
-    {   ZSTD_DDict* const ddict = (ZSTD_DDict*) ZSTD_customMalloc(sizeof(ZSTD_DDict), customMem);
+    {   ZSTD_DDict* const ddict = (ZSTD_DDict*) ZSTD_malloc(sizeof(ZSTD_DDict), customMem);
         if (ddict == NULL) return NULL;
         ddict->cMem = customMem;
         {   size_t const initResult = ZSTD_initDDict_internal(ddict,
@@ -199,7 +194,7 @@ const ZSTD_DDict* ZSTD_initStaticDDict(
     if ((size_t)sBuffer & 7) return NULL;   /* 8-aligned */
     if (sBufferSize < neededSpace) return NULL;
     if (dictLoadMethod == ZSTD_dlm_byCopy) {
-        ZSTD_memcpy(ddict+1, dict, dictSize);  /* local copy */
+        memcpy(ddict+1, dict, dictSize);  /* local copy */
         dict = ddict+1;
     }
     if (ZSTD_isError( ZSTD_initDDict_internal(ddict,
@@ -214,8 +209,8 @@ size_t ZSTD_freeDDict(ZSTD_DDict* ddict)
 {
     if (ddict==NULL) return 0;   /* support free on NULL */
     {   ZSTD_customMem const cMem = ddict->cMem;
-        ZSTD_customFree(ddict->dictBuffer, cMem);
-        ZSTD_customFree(ddict, cMem);
+        ZSTD_free(ddict->dictBuffer, cMem);
+        ZSTD_free(ddict, cMem);
         return 0;
     }
 }
@@ -241,7 +236,5 @@ size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict)
 unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict)
 {
     if (ddict==NULL) return 0;
-    return ddict->dictID;
+    return ZSTD_getDictID_fromDict(ddict->dictContent, ddict->dictSize);
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_ddict.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -16,7 +15,7 @@
 /*-*******************************************************
  *  Dependencies
  *********************************************************/
-#include "zstd_deps.h"   /* size_t */
+#include <stddef.h>   /* size_t */
 #include "zstd.h"     /* ZSTD_DDict, and several public functions */
 
 
@@ -43,5 +42,3 @@ void ZSTD_copyDDictParameters(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict);
 
 
 #endif /* ZSTD_DDICT_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_decompress_block.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -16,7 +15,7 @@
 /*-*******************************************************
  *  Dependencies
  *********************************************************/
-#include "zstd_deps.h"   /* size_t */
+#include <stddef.h>   /* size_t */
 #include "zstd.h"    /* DCtx, and some public functions */
 #include "zstd_internal.h"  /* blockProperties_t, and some public functions */
 #include "zstd_decompress_internal.h"  /* ZSTD_seqSymbol */
@@ -34,12 +33,6 @@
  */
 
 
- /* Streaming state is used to inform allocation of the literal buffer */
-typedef enum {
-    not_streaming = 0,
-    is_streaming = 1
-} streaming_operation;
-
 /* ZSTD_decompressBlock_internal() :
  * decompress block, starting at `src`,
  * into destination buffer `dst`.
@@ -48,29 +41,19 @@ typedef enum {
  */
 size_t ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx,
                                void* dst, size_t dstCapacity,
-                         const void* src, size_t srcSize, const streaming_operation streaming);
+                         const void* src, size_t srcSize, const int frame);
 
 /* ZSTD_buildFSETable() :
  * generate FSE decoding table for one symbol (ll, ml or off)
  * this function must be called with valid parameters only
  * (dt is large enough, normalizedCounter distribution total is a power of 2, max is within range, etc.)
  * in which case it cannot fail.
- * The workspace must be 4-byte aligned and at least ZSTD_BUILD_FSE_TABLE_WKSP_SIZE bytes, which is
- * defined in zstd_decompress_internal.h.
  * Internal use only.
  */
 void ZSTD_buildFSETable(ZSTD_seqSymbol* dt,
              const short* normalizedCounter, unsigned maxSymbolValue,
-             const U32* baseValue, const U8* nbAdditionalBits,
-                   unsigned tableLog, void* wksp, size_t wkspSize,
-                   int bmi2);
-
-/* Internal definition of ZSTD_decompressBlock() to avoid deprecation warnings. */
-size_t ZSTD_decompressBlock_deprecated(ZSTD_DCtx* dctx,
-                            void* dst, size_t dstCapacity,
-                      const void* src, size_t srcSize);
+             const U32* baseValue, const U32* nbAdditionalBits,
+                   unsigned tableLog);
 
 
 #endif /* ZSTD_DEC_BLOCK_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_decompress_internal.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -21,33 +20,33 @@
  *  Dependencies
  *********************************************************/
 #include "mem.h"             /* BYTE, U16, U32 */
-#include "zstd_internal.h"   /* constants : MaxLL, MaxML, MaxOff, LLFSELog, etc. */
+#include "zstd_internal.h"   /* ZSTD_seqSymbol */
 
 
 
 /*-*******************************************************
  *  Constants
  *********************************************************/
-static UNUSED_ATTR const U32 LL_base[MaxLL+1] = {
+static const U32 LL_base[MaxLL+1] = {
                  0,    1,    2,     3,     4,     5,     6,      7,
                  8,    9,   10,    11,    12,    13,    14,     15,
                 16,   18,   20,    22,    24,    28,    32,     40,
                 48,   64, 0x80, 0x100, 0x200, 0x400, 0x800, 0x1000,
                 0x2000, 0x4000, 0x8000, 0x10000 };
 
-static UNUSED_ATTR const U32 OF_base[MaxOff+1] = {
+static const U32 OF_base[MaxOff+1] = {
                  0,        1,       1,       5,     0xD,     0x1D,     0x3D,     0x7D,
                  0xFD,   0x1FD,   0x3FD,   0x7FD,   0xFFD,   0x1FFD,   0x3FFD,   0x7FFD,
                  0xFFFD, 0x1FFFD, 0x3FFFD, 0x7FFFD, 0xFFFFD, 0x1FFFFD, 0x3FFFFD, 0x7FFFFD,
                  0xFFFFFD, 0x1FFFFFD, 0x3FFFFFD, 0x7FFFFFD, 0xFFFFFFD, 0x1FFFFFFD, 0x3FFFFFFD, 0x7FFFFFFD };
 
-static UNUSED_ATTR const U8 OF_bits[MaxOff+1] = {
+static const U32 OF_bits[MaxOff+1] = {
                      0,  1,  2,  3,  4,  5,  6,  7,
                      8,  9, 10, 11, 12, 13, 14, 15,
                     16, 17, 18, 19, 20, 21, 22, 23,
                     24, 25, 26, 27, 28, 29, 30, 31 };
 
-static UNUSED_ATTR const U32 ML_base[MaxML+1] = {
+static const U32 ML_base[MaxML+1] = {
                      3,  4,  5,    6,     7,     8,     9,    10,
                     11, 12, 13,   14,    15,    16,    17,    18,
                     19, 20, 21,   22,    23,    24,    25,    26,
@@ -74,17 +73,12 @@ static UNUSED_ATTR const U32 ML_base[MaxML+1] = {
 
  #define SEQSYMBOL_TABLE_SIZE(log)   (1 + (1 << (log)))
 
-#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE (sizeof(S16) * (MaxSeq + 1) + (1u << MaxFSELog) + sizeof(U64))
-#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32 ((ZSTD_BUILD_FSE_TABLE_WKSP_SIZE + sizeof(U32) - 1) / sizeof(U32))
-#define ZSTD_HUFFDTABLE_CAPACITY_LOG 12
-
 typedef struct {
     ZSTD_seqSymbol LLTable[SEQSYMBOL_TABLE_SIZE(LLFSELog)];    /* Note : Space reserved for FSE Tables */
     ZSTD_seqSymbol OFTable[SEQSYMBOL_TABLE_SIZE(OffFSELog)];   /* is also used as temporary workspace while building hufTable during DDict creation */
     ZSTD_seqSymbol MLTable[SEQSYMBOL_TABLE_SIZE(MLFSELog)];    /* and therefore must be at least HUF_DECOMPRESS_WORKSPACE_SIZE large */
-    HUF_DTable hufTable[HUF_DTABLE_SIZE(ZSTD_HUFFDTABLE_CAPACITY_LOG)];  /* can accommodate HUF_decompress4X */
+    HUF_DTable hufTable[HUF_DTABLE_SIZE(HufLog)];  /* can accommodate HUF_decompress4X */
     U32 rep[ZSTD_REP_NUM];
-    U32 workspace[ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32];
 } ZSTD_entropyDTables_t;
 
 typedef enum { ZSTDds_getFrameHeaderSize, ZSTDds_decodeFrameHeader,
@@ -101,29 +95,6 @@ typedef enum {
     ZSTD_use_once = 1            /* Use the dictionary once and set to ZSTD_dont_use */
 } ZSTD_dictUses_e;
 
-/* Hashset for storing references to multiple ZSTD_DDict within ZSTD_DCtx */
-typedef struct {
-    const ZSTD_DDict** ddictPtrTable;
-    size_t ddictPtrTableSize;
-    size_t ddictPtrCount;
-} ZSTD_DDictHashSet;
-
-#ifndef ZSTD_DECODER_INTERNAL_BUFFER
-#  define ZSTD_DECODER_INTERNAL_BUFFER  (1 << 16)
-#endif
-
-#define ZSTD_LBMIN 64
-#define ZSTD_LBMAX (128 << 10)
-
-/* extra buffer, compensates when dst is not large enough to store litBuffer */
-#define ZSTD_LITBUFFEREXTRASIZE  BOUNDED(ZSTD_LBMIN, ZSTD_DECODER_INTERNAL_BUFFER, ZSTD_LBMAX)
-
-typedef enum {
-    ZSTD_not_in_dst = 0,  /* Stored entirely within litExtraBuffer */
-    ZSTD_in_dst = 1,           /* Stored entirely within dst (in memory after current output write) */
-    ZSTD_split = 2            /* Split between litExtraBuffer and dst */
-} ZSTD_litLocation_e;
-
 struct ZSTD_DCtx_s
 {
     const ZSTD_seqSymbol* LLTptr;
@@ -138,7 +109,6 @@ struct ZSTD_DCtx_s
     const void* dictEnd;          /* end of previous segment */
     size_t expected;
     ZSTD_frameHeader fParams;
-    U64 processedCSize;
     U64 decodedSize;
     blockType_e bType;            /* used in ZSTD_decompressContinue(), store blockType between block header decoding and block decompression stages */
     ZSTD_dStage stage;
@@ -147,17 +117,12 @@ struct ZSTD_DCtx_s
     XXH64_state_t xxhState;
     size_t headerSize;
     ZSTD_format_e format;
-    ZSTD_forceIgnoreChecksum_e forceIgnoreChecksum;   /* User specified: if == 1, will ignore checksums in compressed frame. Default == 0 */
-    U32 validateChecksum;         /* if == 1, will validate checksum. Is == 1 if (fParams.checksumFlag == 1) and (forceIgnoreChecksum == 0). */
     const BYTE* litPtr;
     ZSTD_customMem customMem;
     size_t litSize;
     size_t rleSize;
     size_t staticSize;
-    int isFrameDecompression;
-#if DYNAMIC_BMI2 != 0
     int bmi2;                     /* == 1 if the CPU supports BMI2 and 0 otherwise. CPU support is determined dynamically once per context lifetime. */
-#endif
 
     /* dictionary */
     ZSTD_DDict* ddictLocal;
@@ -165,10 +130,6 @@ struct ZSTD_DCtx_s
     U32 dictID;
     int ddictIsCold;             /* if == 1 : dictionary is "new" for working context, and presumed "cold" (not in cpu cache) */
     ZSTD_dictUses_e dictUses;
-    ZSTD_DDictHashSet* ddictSet;                    /* Hash set for multiple ddicts */
-    ZSTD_refMultipleDDicts_e refMultipleDDicts;     /* User specified: if == 1, will allow references to multiple DDicts. Default == 0 (disabled) */
-    int disableHufAsm;
-    int maxBlockSizeParam;
 
     /* streaming */
     ZSTD_dStreamStage streamStage;
@@ -181,44 +142,17 @@ struct ZSTD_DCtx_s
     size_t outStart;
     size_t outEnd;
     size_t lhSize;
-#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1)
     void* legacyContext;
     U32 previousLegacyVersion;
     U32 legacyVersion;
-#endif
     U32 hostageByte;
     int noForwardProgress;
-    ZSTD_bufferMode_e outBufferMode;
-    ZSTD_outBuffer expectedOutBuffer;
 
     /* workspace */
-    BYTE* litBuffer;
-    const BYTE* litBufferEnd;
-    ZSTD_litLocation_e litBufferLocation;
-    BYTE litExtraBuffer[ZSTD_LITBUFFEREXTRASIZE + WILDCOPY_OVERLENGTH]; /* literal buffer can be split between storage within dst and within this scratch buffer */
+    BYTE litBuffer[ZSTD_BLOCKSIZE_MAX + WILDCOPY_OVERLENGTH];
     BYTE headerBuffer[ZSTD_FRAMEHEADERSIZE_MAX];
-
-    size_t oversizedDuration;
-
-#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
-    void const* dictContentBeginForFuzzing;
-    void const* dictContentEndForFuzzing;
-#endif
-
-    /* Tracing */
-#if ZSTD_TRACE
-    ZSTD_TraceCtx traceCtx;
-#endif
 };  /* typedef'd to ZSTD_DCtx within "zstd.h" */
 
-MEM_STATIC int ZSTD_DCtx_get_bmi2(const struct ZSTD_DCtx_s *dctx) {
-#if DYNAMIC_BMI2 != 0
-	return dctx->bmi2;
-#else
-    (void)dctx;
-	return 0;
-#endif
-}
 
 /*-*******************************************************
  *  Shared internal functions
@@ -226,7 +160,7 @@ MEM_STATIC int ZSTD_DCtx_get_bmi2(const struct ZSTD_DCtx_s *dctx) {
 
 /*! ZSTD_loadDEntropy() :
  *  dict : must point at beginning of a valid zstd dictionary.
- * @return : size of dictionary header (size of magic number + dict ID + entropy tables) */
+ * @return : size of entropy tables read */
 size_t ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy,
                    const void* const dict, size_t const dictSize);
 
@@ -235,9 +169,7 @@ size_t ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy,
  *  If yes, do nothing (continue on current segment).
  *  If not, classify previous segment as "external dictionary", and start a new segment.
  *  This function cannot fail. */
-void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst, size_t dstSize);
+void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst);
 
 
 #endif /* ZSTD_DECOMPRESS_INTERNAL_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_deps.h

@@ -1,114 +0,0 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
-/* This file provides common libc dependencies that zstd requires.
- * The purpose is to allow replacing this file with a custom implementation
- * to compile zstd without libc support.
- */
-
-/* Need:
- * NULL
- * INT_MAX
- * UINT_MAX
- * ZSTD_memcpy()
- * ZSTD_memset()
- * ZSTD_memmove()
- */
-#ifndef ZSTD_DEPS_COMMON
-#define ZSTD_DEPS_COMMON
-
-#include <limits.h>
-#include <stddef.h>
-#include <string.h>
-
-#if defined(__GNUC__) && __GNUC__ >= 4
-# define ZSTD_memcpy(d,s,l) __builtin_memcpy((d),(s),(l))
-# define ZSTD_memmove(d,s,l) __builtin_memmove((d),(s),(l))
-# define ZSTD_memset(p,v,l) __builtin_memset((p),(v),(l))
-#else
-# define ZSTD_memcpy(d,s,l) memcpy((d),(s),(l))
-# define ZSTD_memmove(d,s,l) memmove((d),(s),(l))
-# define ZSTD_memset(p,v,l) memset((p),(v),(l))
-#endif
-
-#endif /* ZSTD_DEPS_COMMON */
-
-/* Need:
- * ZSTD_malloc()
- * ZSTD_free()
- * ZSTD_calloc()
- */
-#ifdef ZSTD_DEPS_NEED_MALLOC
-#ifndef ZSTD_DEPS_MALLOC
-#define ZSTD_DEPS_MALLOC
-
-#include <stdlib.h>
-
-#define ZSTD_malloc(s) malloc(s)
-#define ZSTD_calloc(n,s) calloc((n), (s))
-#define ZSTD_free(p) free((p))
-
-#endif /* ZSTD_DEPS_MALLOC */
-#endif /* ZSTD_DEPS_NEED_MALLOC */
-
-/*
- * Provides 64-bit math support.
- * Need:
- * U64 ZSTD_div64(U64 dividend, U32 divisor)
- */
-#ifdef ZSTD_DEPS_NEED_MATH64
-#ifndef ZSTD_DEPS_MATH64
-#define ZSTD_DEPS_MATH64
-
-#define ZSTD_div64(dividend, divisor) ((dividend) / (divisor))
-
-#endif /* ZSTD_DEPS_MATH64 */
-#endif /* ZSTD_DEPS_NEED_MATH64 */
-
-/* Need:
- * assert()
- */
-#ifdef ZSTD_DEPS_NEED_ASSERT
-#ifndef ZSTD_DEPS_ASSERT
-#define ZSTD_DEPS_ASSERT
-
-#include <assert.h>
-
-#endif /* ZSTD_DEPS_ASSERT */
-#endif /* ZSTD_DEPS_NEED_ASSERT */
-
-/* Need:
- * ZSTD_DEBUG_PRINT()
- */
-#ifdef ZSTD_DEPS_NEED_IO
-#ifndef ZSTD_DEPS_IO
-#define ZSTD_DEPS_IO
-
-#include <stdio.h>
-#define ZSTD_DEBUG_PRINT(...) fprintf(stderr, __VA_ARGS__)
-
-#endif /* ZSTD_DEPS_IO */
-#endif /* ZSTD_DEPS_NEED_IO */
-
-/* Only requested when <stdint.h> is known to be present.
- * Need:
- * intptr_t
- */
-#ifdef ZSTD_DEPS_NEED_STDINT
-#ifndef ZSTD_DEPS_STDINT
-#define ZSTD_DEPS_STDINT
-
-#include <stdint.h>
-
-#endif /* ZSTD_DEPS_STDINT */
-#endif /* ZSTD_DEPS_NEED_STDINT */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_double_fast.c

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -12,49 +11,8 @@
 #include "zstd_compress_internal.h"
 #include "zstd_double_fast.h"
 
-#ifndef ZSTD_EXCLUDE_DFAST_BLOCK_COMPRESSOR
 
-static
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-void ZSTD_fillDoubleHashTableForCDict(ZSTD_matchState_t* ms,
-                              void const* end, ZSTD_dictTableLoadMethod_e dtlm)
-{
-    const ZSTD_compressionParameters* const cParams = &ms->cParams;
-    U32* const hashLarge = ms->hashTable;
-    U32  const hBitsL = cParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
-    U32  const mls = cParams->minMatch;
-    U32* const hashSmall = ms->chainTable;
-    U32  const hBitsS = cParams->chainLog + ZSTD_SHORT_CACHE_TAG_BITS;
-    const BYTE* const base = ms->window.base;
-    const BYTE* ip = base + ms->nextToUpdate;
-    const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
-    const U32 fastHashFillStep = 3;
-
-    /* Always insert every fastHashFillStep position into the hash tables.
-     * Insert the other positions into the large hash table if their entry
-     * is empty.
-     */
-    for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) {
-        U32 const curr = (U32)(ip - base);
-        U32 i;
-        for (i = 0; i < fastHashFillStep; ++i) {
-            size_t const smHashAndTag = ZSTD_hashPtr(ip + i, hBitsS, mls);
-            size_t const lgHashAndTag = ZSTD_hashPtr(ip + i, hBitsL, 8);
-            if (i == 0) {
-                ZSTD_writeTaggedIndex(hashSmall, smHashAndTag, curr + i);
-            }
-            if (i == 0 || hashLarge[lgHashAndTag >> ZSTD_SHORT_CACHE_TAG_BITS] == 0) {
-                ZSTD_writeTaggedIndex(hashLarge, lgHashAndTag, curr + i);
-            }
-            /* Only load extra positions for ZSTD_dtlm_full */
-            if (dtlm == ZSTD_dtlm_fast)
-                break;
-    }   }
-}
-
-static
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-void ZSTD_fillDoubleHashTableForCCtx(ZSTD_matchState_t* ms,
+void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
                               void const* end, ZSTD_dictTableLoadMethod_e dtlm)
 {
     const ZSTD_compressionParameters* const cParams = &ms->cParams;
@@ -73,251 +31,27 @@ void ZSTD_fillDoubleHashTableForCCtx(ZSTD_matchState_t* ms,
      * is empty.
      */
     for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) {
-        U32 const curr = (U32)(ip - base);
+        U32 const current = (U32)(ip - base);
         U32 i;
         for (i = 0; i < fastHashFillStep; ++i) {
             size_t const smHash = ZSTD_hashPtr(ip + i, hBitsS, mls);
             size_t const lgHash = ZSTD_hashPtr(ip + i, hBitsL, 8);
             if (i == 0)
-                hashSmall[smHash] = curr + i;
+                hashSmall[smHash] = current + i;
             if (i == 0 || hashLarge[lgHash] == 0)
-                hashLarge[lgHash] = curr + i;
+                hashLarge[lgHash] = current + i;
             /* Only load extra positions for ZSTD_dtlm_full */
             if (dtlm == ZSTD_dtlm_fast)
                 break;
-        }   }
-}
-
-void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
-                        const void* const end,
-                        ZSTD_dictTableLoadMethod_e dtlm,
-                        ZSTD_tableFillPurpose_e tfp)
-{
-    if (tfp == ZSTD_tfp_forCDict) {
-        ZSTD_fillDoubleHashTableForCDict(ms, end, dtlm);
-    } else {
-        ZSTD_fillDoubleHashTableForCCtx(ms, end, dtlm);
-    }
+    }   }
 }
 
 
 FORCE_INLINE_TEMPLATE
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-size_t ZSTD_compressBlock_doubleFast_noDict_generic(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize, U32 const mls /* template */)
-{
-    ZSTD_compressionParameters const* cParams = &ms->cParams;
-    U32* const hashLong = ms->hashTable;
-    const U32 hBitsL = cParams->hashLog;
-    U32* const hashSmall = ms->chainTable;
-    const U32 hBitsS = cParams->chainLog;
-    const BYTE* const base = ms->window.base;
-    const BYTE* const istart = (const BYTE*)src;
-    const BYTE* anchor = istart;
-    const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
-    /* presumes that, if there is a dictionary, it must be using Attach mode */
-    const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
-    const BYTE* const prefixLowest = base + prefixLowestIndex;
-    const BYTE* const iend = istart + srcSize;
-    const BYTE* const ilimit = iend - HASH_READ_SIZE;
-    U32 offset_1=rep[0], offset_2=rep[1];
-    U32 offsetSaved1 = 0, offsetSaved2 = 0;
-
-    size_t mLength;
-    U32 offset;
-    U32 curr;
-
-    /* how many positions to search before increasing step size */
-    const size_t kStepIncr = 1 << kSearchStrength;
-    /* the position at which to increment the step size if no match is found */
-    const BYTE* nextStep;
-    size_t step; /* the current step size */
-
-    size_t hl0; /* the long hash at ip */
-    size_t hl1; /* the long hash at ip1 */
-
-    U32 idxl0; /* the long match index for ip */
-    U32 idxl1; /* the long match index for ip1 */
-
-    const BYTE* matchl0; /* the long match for ip */
-    const BYTE* matchs0; /* the short match for ip */
-    const BYTE* matchl1; /* the long match for ip1 */
-
-    const BYTE* ip = istart; /* the current position */
-    const BYTE* ip1; /* the next position */
-
-    DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_noDict_generic");
-
-    /* init */
-    ip += ((ip - prefixLowest) == 0);
-    {
-        U32 const current = (U32)(ip - base);
-        U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, current, cParams->windowLog);
-        U32 const maxRep = current - windowLow;
-        if (offset_2 > maxRep) offsetSaved2 = offset_2, offset_2 = 0;
-        if (offset_1 > maxRep) offsetSaved1 = offset_1, offset_1 = 0;
-    }
-
-    /* Outer Loop: one iteration per match found and stored */
-    while (1) {
-        step = 1;
-        nextStep = ip + kStepIncr;
-        ip1 = ip + step;
-
-        if (ip1 > ilimit) {
-            goto _cleanup;
-        }
-
-        hl0 = ZSTD_hashPtr(ip, hBitsL, 8);
-        idxl0 = hashLong[hl0];
-        matchl0 = base + idxl0;
-
-        /* Inner Loop: one iteration per search / position */
-        do {
-            const size_t hs0 = ZSTD_hashPtr(ip, hBitsS, mls);
-            const U32 idxs0 = hashSmall[hs0];
-            curr = (U32)(ip-base);
-            matchs0 = base + idxs0;
-
-            hashLong[hl0] = hashSmall[hs0] = curr;   /* update hash tables */
-
-            /* check noDict repcode */
-            if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
-                mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
-                ip++;
-                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
-                goto _match_stored;
-            }
-
-            hl1 = ZSTD_hashPtr(ip1, hBitsL, 8);
-
-            if (idxl0 > prefixLowestIndex) {
-                /* check prefix long match */
-                if (MEM_read64(matchl0) == MEM_read64(ip)) {
-                    mLength = ZSTD_count(ip+8, matchl0+8, iend) + 8;
-                    offset = (U32)(ip-matchl0);
-                    while (((ip>anchor) & (matchl0>prefixLowest)) && (ip[-1] == matchl0[-1])) { ip--; matchl0--; mLength++; } /* catch up */
-                    goto _match_found;
-                }
-            }
-
-            idxl1 = hashLong[hl1];
-            matchl1 = base + idxl1;
-
-            if (idxs0 > prefixLowestIndex) {
-                /* check prefix short match */
-                if (MEM_read32(matchs0) == MEM_read32(ip)) {
-                    goto _search_next_long;
-                }
-            }
-
-            if (ip1 >= nextStep) {
-                PREFETCH_L1(ip1 + 64);
-                PREFETCH_L1(ip1 + 128);
-                step++;
-                nextStep += kStepIncr;
-            }
-            ip = ip1;
-            ip1 += step;
-
-            hl0 = hl1;
-            idxl0 = idxl1;
-            matchl0 = matchl1;
-    #if defined(__aarch64__)
-            PREFETCH_L1(ip+256);
-    #endif
-        } while (ip1 <= ilimit);
-
-_cleanup:
-        /* If offset_1 started invalid (offsetSaved1 != 0) and became valid (offset_1 != 0),
-         * rotate saved offsets. See comment in ZSTD_compressBlock_fast_noDict for more context. */
-        offsetSaved2 = ((offsetSaved1 != 0) && (offset_1 != 0)) ? offsetSaved1 : offsetSaved2;
-
-        /* save reps for next block */
-        rep[0] = offset_1 ? offset_1 : offsetSaved1;
-        rep[1] = offset_2 ? offset_2 : offsetSaved2;
-
-        /* Return the last literals size */
-        return (size_t)(iend - anchor);
-
-_search_next_long:
-
-        /* check prefix long +1 match */
-        if (idxl1 > prefixLowestIndex) {
-            if (MEM_read64(matchl1) == MEM_read64(ip1)) {
-                ip = ip1;
-                mLength = ZSTD_count(ip+8, matchl1+8, iend) + 8;
-                offset = (U32)(ip-matchl1);
-                while (((ip>anchor) & (matchl1>prefixLowest)) && (ip[-1] == matchl1[-1])) { ip--; matchl1--; mLength++; } /* catch up */
-                goto _match_found;
-            }
-        }
-
-        /* if no long +1 match, explore the short match we found */
-        mLength = ZSTD_count(ip+4, matchs0+4, iend) + 4;
-        offset = (U32)(ip - matchs0);
-        while (((ip>anchor) & (matchs0>prefixLowest)) && (ip[-1] == matchs0[-1])) { ip--; matchs0--; mLength++; } /* catch up */
-
-        /* fall-through */
-
-_match_found: /* requires ip, offset, mLength */
-        offset_2 = offset_1;
-        offset_1 = offset;
-
-        if (step < 4) {
-            /* It is unsafe to write this value back to the hashtable when ip1 is
-             * greater than or equal to the new ip we will have after we're done
-             * processing this match. Rather than perform that test directly
-             * (ip1 >= ip + mLength), which costs speed in practice, we do a simpler
-             * more predictable test. The minmatch even if we take a short match is
-             * 4 bytes, so as long as step, the distance between ip and ip1
-             * (initially) is less than 4, we know ip1 < new ip. */
-            hashLong[hl1] = (U32)(ip1 - base);
-        }
-
-        ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
-
-_match_stored:
-        /* match found */
-        ip += mLength;
-        anchor = ip;
-
-        if (ip <= ilimit) {
-            /* Complementary insertion */
-            /* done after iLimit test, as candidates could be > iend-8 */
-            {   U32 const indexToInsert = curr+2;
-                hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
-                hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
-                hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
-                hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
-            }
-
-            /* check immediate repcode */
-            while ( (ip <= ilimit)
-                 && ( (offset_2>0)
-                    & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
-                /* store sequence */
-                size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
-                U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff;  /* swap offset_2 <=> offset_1 */
-                hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
-                hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
-                ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, rLength);
-                ip += rLength;
-                anchor = ip;
-                continue;   /* faster when present ... (?) */
-            }
-        }
-    }
-}
-
-
-FORCE_INLINE_TEMPLATE
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
+size_t ZSTD_compressBlock_doubleFast_generic(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize,
-        U32 const mls /* template */)
+        U32 const mls /* template */, ZSTD_dictMode_e const dictMode)
 {
     ZSTD_compressionParameters const* cParams = &ms->cParams;
     U32* const hashLong = ms->hashTable;
@@ -329,45 +63,63 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
     const BYTE* ip = istart;
     const BYTE* anchor = istart;
     const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
+    const U32 lowestValid = ms->window.dictLimit;
+    const U32 maxDistance = 1U << cParams->windowLog;
     /* presumes that, if there is a dictionary, it must be using Attach mode */
-    const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
+    const U32 prefixLowestIndex = (endIndex - lowestValid > maxDistance) ? endIndex - maxDistance : lowestValid;
     const BYTE* const prefixLowest = base + prefixLowestIndex;
     const BYTE* const iend = istart + srcSize;
     const BYTE* const ilimit = iend - HASH_READ_SIZE;
     U32 offset_1=rep[0], offset_2=rep[1];
+    U32 offsetSaved = 0;
 
     const ZSTD_matchState_t* const dms = ms->dictMatchState;
-    const ZSTD_compressionParameters* const dictCParams = &dms->cParams;
-    const U32* const dictHashLong  = dms->hashTable;
-    const U32* const dictHashSmall = dms->chainTable;
-    const U32 dictStartIndex       = dms->window.dictLimit;
-    const BYTE* const dictBase     = dms->window.base;
-    const BYTE* const dictStart    = dictBase + dictStartIndex;
-    const BYTE* const dictEnd      = dms->window.nextSrc;
-    const U32 dictIndexDelta       = prefixLowestIndex - (U32)(dictEnd - dictBase);
-    const U32 dictHBitsL           = dictCParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
-    const U32 dictHBitsS           = dictCParams->chainLog + ZSTD_SHORT_CACHE_TAG_BITS;
-    const U32 dictAndPrefixLength  = (U32)((ip - prefixLowest) + (dictEnd - dictStart));
+    const ZSTD_compressionParameters* const dictCParams =
+                                     dictMode == ZSTD_dictMatchState ?
+                                     &dms->cParams : NULL;
+    const U32* const dictHashLong  = dictMode == ZSTD_dictMatchState ?
+                                     dms->hashTable : NULL;
+    const U32* const dictHashSmall = dictMode == ZSTD_dictMatchState ?
+                                     dms->chainTable : NULL;
+    const U32 dictStartIndex       = dictMode == ZSTD_dictMatchState ?
+                                     dms->window.dictLimit : 0;
+    const BYTE* const dictBase     = dictMode == ZSTD_dictMatchState ?
+                                     dms->window.base : NULL;
+    const BYTE* const dictStart    = dictMode == ZSTD_dictMatchState ?
+                                     dictBase + dictStartIndex : NULL;
+    const BYTE* const dictEnd      = dictMode == ZSTD_dictMatchState ?
+                                     dms->window.nextSrc : NULL;
+    const U32 dictIndexDelta       = dictMode == ZSTD_dictMatchState ?
+                                     prefixLowestIndex - (U32)(dictEnd - dictBase) :
+                                     0;
+    const U32 dictHBitsL           = dictMode == ZSTD_dictMatchState ?
+                                     dictCParams->hashLog : hBitsL;
+    const U32 dictHBitsS           = dictMode == ZSTD_dictMatchState ?
+                                     dictCParams->chainLog : hBitsS;
+    const U32 dictAndPrefixLength  = (U32)(ip - prefixLowest + dictEnd - dictStart);
 
-    DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_dictMatchState_generic");
+    DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_generic");
+
+    assert(dictMode == ZSTD_noDict || dictMode == ZSTD_dictMatchState);
 
     /* if a dictionary is attached, it must be within window range */
-    assert(ms->window.dictLimit + (1U << cParams->windowLog) >= endIndex);
-
-    if (ms->prefetchCDictTables) {
-        size_t const hashTableBytes = (((size_t)1) << dictCParams->hashLog) * sizeof(U32);
-        size_t const chainTableBytes = (((size_t)1) << dictCParams->chainLog) * sizeof(U32);
-        PREFETCH_AREA(dictHashLong, hashTableBytes);
-        PREFETCH_AREA(dictHashSmall, chainTableBytes);
+    if (dictMode == ZSTD_dictMatchState) {
+        assert(lowestValid + maxDistance >= endIndex);
     }
 
     /* init */
     ip += (dictAndPrefixLength == 0);
-
-    /* dictMatchState repCode checks don't currently handle repCode == 0
-     * disabling. */
-    assert(offset_1 <= dictAndPrefixLength);
-    assert(offset_2 <= dictAndPrefixLength);
+    if (dictMode == ZSTD_noDict) {
+        U32 const maxRep = (U32)(ip - prefixLowest);
+        if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
+        if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
+    }
+    if (dictMode == ZSTD_dictMatchState) {
+        /* dictMatchState repCode checks don't currently handle repCode == 0
+         * disabling. */
+        assert(offset_1 <= dictAndPrefixLength);
+        assert(offset_2 <= dictAndPrefixLength);
+    }
 
     /* Main Search Loop */
     while (ip < ilimit) {   /* < instead of <=, because repcode check at (ip+1) */
@@ -375,30 +127,37 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
         U32 offset;
         size_t const h2 = ZSTD_hashPtr(ip, hBitsL, 8);
         size_t const h = ZSTD_hashPtr(ip, hBitsS, mls);
-        size_t const dictHashAndTagL = ZSTD_hashPtr(ip, dictHBitsL, 8);
-        size_t const dictHashAndTagS = ZSTD_hashPtr(ip, dictHBitsS, mls);
-        U32 const dictMatchIndexAndTagL = dictHashLong[dictHashAndTagL >> ZSTD_SHORT_CACHE_TAG_BITS];
-        U32 const dictMatchIndexAndTagS = dictHashSmall[dictHashAndTagS >> ZSTD_SHORT_CACHE_TAG_BITS];
-        int const dictTagsMatchL = ZSTD_comparePackedTags(dictMatchIndexAndTagL, dictHashAndTagL);
-        int const dictTagsMatchS = ZSTD_comparePackedTags(dictMatchIndexAndTagS, dictHashAndTagS);
-        U32 const curr = (U32)(ip-base);
+        size_t const dictHL = ZSTD_hashPtr(ip, dictHBitsL, 8);
+        size_t const dictHS = ZSTD_hashPtr(ip, dictHBitsS, mls);
+        U32 const current = (U32)(ip-base);
         U32 const matchIndexL = hashLong[h2];
         U32 matchIndexS = hashSmall[h];
         const BYTE* matchLong = base + matchIndexL;
         const BYTE* match = base + matchIndexS;
-        const U32 repIndex = curr + 1 - offset_1;
-        const BYTE* repMatch = (repIndex < prefixLowestIndex) ?
+        const U32 repIndex = current + 1 - offset_1;
+        const BYTE* repMatch = (dictMode == ZSTD_dictMatchState
+                            && repIndex < prefixLowestIndex) ?
                                dictBase + (repIndex - dictIndexDelta) :
                                base + repIndex;
-        hashLong[h2] = hashSmall[h] = curr;   /* update hash tables */
+        hashLong[h2] = hashSmall[h] = current;   /* update hash tables */
 
-        /* check repcode */
-        if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */)
+        /* check dictMatchState repcode */
+        if (dictMode == ZSTD_dictMatchState
+            && ((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */)
             && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
             const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
             mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
             ip++;
-            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
+            goto _match_stored;
+        }
+
+        /* check noDict repcode */
+        if ( dictMode == ZSTD_noDict
+          && ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1)))) {
+            mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
+            ip++;
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
             goto _match_stored;
         }
 
@@ -410,15 +169,15 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
                 while (((ip>anchor) & (matchLong>prefixLowest)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
                 goto _match_found;
             }
-        } else if (dictTagsMatchL) {
+        } else if (dictMode == ZSTD_dictMatchState) {
             /* check dictMatchState long match */
-            U32 const dictMatchIndexL = dictMatchIndexAndTagL >> ZSTD_SHORT_CACHE_TAG_BITS;
+            U32 const dictMatchIndexL = dictHashLong[dictHL];
             const BYTE* dictMatchL = dictBase + dictMatchIndexL;
             assert(dictMatchL < dictEnd);
 
             if (dictMatchL > dictStart && MEM_read64(dictMatchL) == MEM_read64(ip)) {
                 mLength = ZSTD_count_2segments(ip+8, dictMatchL+8, iend, dictEnd, prefixLowest) + 8;
-                offset = (U32)(curr - dictMatchIndexL - dictIndexDelta);
+                offset = (U32)(current - dictMatchIndexL - dictIndexDelta);
                 while (((ip>anchor) & (dictMatchL>dictStart)) && (ip[-1] == dictMatchL[-1])) { ip--; dictMatchL--; mLength++; } /* catch up */
                 goto _match_found;
         }   }
@@ -428,9 +187,9 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
             if (MEM_read32(match) == MEM_read32(ip)) {
                 goto _search_next_long;
             }
-        } else if (dictTagsMatchS) {
+        } else if (dictMode == ZSTD_dictMatchState) {
             /* check dictMatchState short match */
-            U32 const dictMatchIndexS = dictMatchIndexAndTagS >> ZSTD_SHORT_CACHE_TAG_BITS;
+            U32 const dictMatchIndexS = dictHashSmall[dictHS];
             match = dictBase + dictMatchIndexS;
             matchIndexS = dictMatchIndexS + dictIndexDelta;
 
@@ -439,19 +198,15 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
         }   }
 
         ip += ((ip-anchor) >> kSearchStrength) + 1;
-#if defined(__aarch64__)
-        PREFETCH_L1(ip+256);
-#endif
         continue;
 
 _search_next_long:
+
         {   size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
-            size_t const dictHashAndTagL3 = ZSTD_hashPtr(ip+1, dictHBitsL, 8);
+            size_t const dictHLNext = ZSTD_hashPtr(ip+1, dictHBitsL, 8);
             U32 const matchIndexL3 = hashLong[hl3];
-            U32 const dictMatchIndexAndTagL3 = dictHashLong[dictHashAndTagL3 >> ZSTD_SHORT_CACHE_TAG_BITS];
-            int const dictTagsMatchL3 = ZSTD_comparePackedTags(dictMatchIndexAndTagL3, dictHashAndTagL3);
             const BYTE* matchL3 = base + matchIndexL3;
-            hashLong[hl3] = curr + 1;
+            hashLong[hl3] = current + 1;
 
             /* check prefix long +1 match */
             if (matchIndexL3 > prefixLowestIndex) {
@@ -462,23 +217,23 @@ _search_next_long:
                     while (((ip>anchor) & (matchL3>prefixLowest)) && (ip[-1] == matchL3[-1])) { ip--; matchL3--; mLength++; } /* catch up */
                     goto _match_found;
                 }
-            } else if (dictTagsMatchL3) {
+            } else if (dictMode == ZSTD_dictMatchState) {
                 /* check dict long +1 match */
-                U32 const dictMatchIndexL3 = dictMatchIndexAndTagL3 >> ZSTD_SHORT_CACHE_TAG_BITS;
+                U32 const dictMatchIndexL3 = dictHashLong[dictHLNext];
                 const BYTE* dictMatchL3 = dictBase + dictMatchIndexL3;
                 assert(dictMatchL3 < dictEnd);
                 if (dictMatchL3 > dictStart && MEM_read64(dictMatchL3) == MEM_read64(ip+1)) {
                     mLength = ZSTD_count_2segments(ip+1+8, dictMatchL3+8, iend, dictEnd, prefixLowest) + 8;
                     ip++;
-                    offset = (U32)(curr + 1 - dictMatchIndexL3 - dictIndexDelta);
+                    offset = (U32)(current + 1 - dictMatchIndexL3 - dictIndexDelta);
                     while (((ip>anchor) & (dictMatchL3>dictStart)) && (ip[-1] == dictMatchL3[-1])) { ip--; dictMatchL3--; mLength++; } /* catch up */
                     goto _match_found;
         }   }   }
 
         /* if no long +1 match, explore the short match we found */
-        if (matchIndexS < prefixLowestIndex) {
+        if (dictMode == ZSTD_dictMatchState && matchIndexS < prefixLowestIndex) {
             mLength = ZSTD_count_2segments(ip+4, match+4, iend, dictEnd, prefixLowest) + 4;
-            offset = (U32)(curr - matchIndexS);
+            offset = (U32)(current - matchIndexS);
             while (((ip>anchor) & (match>dictStart)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
         } else {
             mLength = ZSTD_count(ip+4, match+4, iend) + 4;
@@ -486,11 +241,13 @@ _search_next_long:
             while (((ip>anchor) & (match>prefixLowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
         }
 
+        /* fall-through */
+
 _match_found:
         offset_2 = offset_1;
         offset_1 = offset;
 
-        ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
+        ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
 _match_stored:
         /* match found */
@@ -500,7 +257,7 @@ _match_stored:
         if (ip <= ilimit) {
             /* Complementary insertion */
             /* done after iLimit test, as candidates could be > iend-8 */
-            {   U32 const indexToInsert = curr+2;
+            {   U32 const indexToInsert = current+2;
                 hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
                 hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
                 hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
@@ -508,55 +265,53 @@ _match_stored:
             }
 
             /* check immediate repcode */
-            while (ip <= ilimit) {
-                U32 const current2 = (U32)(ip-base);
-                U32 const repIndex2 = current2 - offset_2;
-                const BYTE* repMatch2 = repIndex2 < prefixLowestIndex ?
-                        dictBase + repIndex2 - dictIndexDelta :
-                        base + repIndex2;
-                if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
-                   && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
-                    const BYTE* const repEnd2 = repIndex2 < prefixLowestIndex ? dictEnd : iend;
-                    size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixLowest) + 4;
-                    U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset;   /* swap offset_2 <=> offset_1 */
-                    ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
-                    hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
-                    hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
-                    ip += repLength2;
+            if (dictMode == ZSTD_dictMatchState) {
+                while (ip <= ilimit) {
+                    U32 const current2 = (U32)(ip-base);
+                    U32 const repIndex2 = current2 - offset_2;
+                    const BYTE* repMatch2 = dictMode == ZSTD_dictMatchState
+                        && repIndex2 < prefixLowestIndex ?
+                            dictBase - dictIndexDelta + repIndex2 :
+                            base + repIndex2;
+                    if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
+                       && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
+                        const BYTE* const repEnd2 = repIndex2 < prefixLowestIndex ? dictEnd : iend;
+                        size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixLowest) + 4;
+                        U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset;   /* swap offset_2 <=> offset_1 */
+                        ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
+                        hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
+                        hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
+                        ip += repLength2;
+                        anchor = ip;
+                        continue;
+                    }
+                    break;
+            }   }
+
+            if (dictMode == ZSTD_noDict) {
+                while ( (ip <= ilimit)
+                     && ( (offset_2>0)
+                        & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
+                    /* store sequence */
+                    size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
+                    U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff;  /* swap offset_2 <=> offset_1 */
+                    hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
+                    hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
+                    ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, rLength-MINMATCH);
+                    ip += rLength;
                     anchor = ip;
-                    continue;
-                }
-                break;
-            }
-        }
+                    continue;   /* faster when present ... (?) */
+        }   }   }
     }   /* while (ip < ilimit) */
 
     /* save reps for next block */
-    rep[0] = offset_1;
-    rep[1] = offset_2;
+    rep[0] = offset_1 ? offset_1 : offsetSaved;
+    rep[1] = offset_2 ? offset_2 : offsetSaved;
 
     /* Return the last literals size */
     return (size_t)(iend - anchor);
 }
 
-#define ZSTD_GEN_DFAST_FN(dictMode, mls)                                                                 \
-    static size_t ZSTD_compressBlock_doubleFast_##dictMode##_##mls(                                      \
-            ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],                          \
-            void const* src, size_t srcSize)                                                             \
-    {                                                                                                    \
-        return ZSTD_compressBlock_doubleFast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mls); \
-    }
-
-ZSTD_GEN_DFAST_FN(noDict, 4)
-ZSTD_GEN_DFAST_FN(noDict, 5)
-ZSTD_GEN_DFAST_FN(noDict, 6)
-ZSTD_GEN_DFAST_FN(noDict, 7)
-
-ZSTD_GEN_DFAST_FN(dictMatchState, 4)
-ZSTD_GEN_DFAST_FN(dictMatchState, 5)
-ZSTD_GEN_DFAST_FN(dictMatchState, 6)
-ZSTD_GEN_DFAST_FN(dictMatchState, 7)
-
 
 size_t ZSTD_compressBlock_doubleFast(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
@@ -567,13 +322,13 @@ size_t ZSTD_compressBlock_doubleFast(
     {
     default: /* includes case 3 */
     case 4 :
-        return ZSTD_compressBlock_doubleFast_noDict_4(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 4, ZSTD_noDict);
     case 5 :
-        return ZSTD_compressBlock_doubleFast_noDict_5(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 5, ZSTD_noDict);
     case 6 :
-        return ZSTD_compressBlock_doubleFast_noDict_6(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 6, ZSTD_noDict);
     case 7 :
-        return ZSTD_compressBlock_doubleFast_noDict_7(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 7, ZSTD_noDict);
     }
 }
 
@@ -587,20 +342,18 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState(
     {
     default: /* includes case 3 */
     case 4 :
-        return ZSTD_compressBlock_doubleFast_dictMatchState_4(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 4, ZSTD_dictMatchState);
     case 5 :
-        return ZSTD_compressBlock_doubleFast_dictMatchState_5(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 5, ZSTD_dictMatchState);
     case 6 :
-        return ZSTD_compressBlock_doubleFast_dictMatchState_6(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 6, ZSTD_dictMatchState);
     case 7 :
-        return ZSTD_compressBlock_doubleFast_dictMatchState_7(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 7, ZSTD_dictMatchState);
     }
 }
 
 
-static
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-size_t ZSTD_compressBlock_doubleFast_extDict_generic(
+static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize,
         U32 const mls /* template */)
@@ -631,7 +384,7 @@ size_t ZSTD_compressBlock_doubleFast_extDict_generic(
 
     /* if extDict is invalidated due to maxDistance, switch to "regular" variant */
     if (prefixStartIndex == dictStartIndex)
-        return ZSTD_compressBlock_doubleFast(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, mls, ZSTD_noDict);
 
     /* Search Loop */
     while (ip < ilimit) {  /* < instead of <=, because (ip+1) */
@@ -645,31 +398,31 @@ size_t ZSTD_compressBlock_doubleFast_extDict_generic(
         const BYTE* const matchLongBase = matchLongIndex < prefixStartIndex ? dictBase : base;
         const BYTE* matchLong = matchLongBase + matchLongIndex;
 
-        const U32 curr = (U32)(ip-base);
-        const U32 repIndex = curr + 1 - offset_1;   /* offset_1 expected <= curr +1 */
+        const U32 current = (U32)(ip-base);
+        const U32 repIndex = current + 1 - offset_1;   /* offset_1 expected <= current +1 */
         const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
         const BYTE* const repMatch = repBase + repIndex;
         size_t mLength;
-        hashSmall[hSmall] = hashLong[hLong] = curr;   /* update hash table */
+        hashSmall[hSmall] = hashLong[hLong] = current;   /* update hash table */
 
         if ((((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex doesn't overlap dict + prefix */
-            & (offset_1 <= curr+1 - dictStartIndex)) /* note: we are searching at curr+1 */
+            & (repIndex > dictStartIndex))
           && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
             const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
             mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
             ip++;
-            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
+            ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
         } else {
             if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
                 const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend;
                 const BYTE* const lowMatchPtr = matchLongIndex < prefixStartIndex ? dictStart : prefixStart;
                 U32 offset;
                 mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, prefixStart) + 8;
-                offset = curr - matchLongIndex;
+                offset = current - matchLongIndex;
                 while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; }   /* catch up */
                 offset_2 = offset_1;
                 offset_1 = offset;
-                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
+                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
             } else if ((matchIndex > dictStartIndex) && (MEM_read32(match) == MEM_read32(ip))) {
                 size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
@@ -677,24 +430,24 @@ size_t ZSTD_compressBlock_doubleFast_extDict_generic(
                 const BYTE* const match3Base = matchIndex3 < prefixStartIndex ? dictBase : base;
                 const BYTE* match3 = match3Base + matchIndex3;
                 U32 offset;
-                hashLong[h3] = curr + 1;
+                hashLong[h3] = current + 1;
                 if ( (matchIndex3 > dictStartIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) {
                     const BYTE* const matchEnd = matchIndex3 < prefixStartIndex ? dictEnd : iend;
                     const BYTE* const lowMatchPtr = matchIndex3 < prefixStartIndex ? dictStart : prefixStart;
                     mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, prefixStart) + 8;
                     ip++;
-                    offset = curr+1 - matchIndex3;
+                    offset = current+1 - matchIndex3;
                     while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */
                 } else {
                     const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend;
                     const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart;
                     mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
-                    offset = curr - matchIndex;
+                    offset = current - matchIndex;
                     while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; }   /* catch up */
                 }
                 offset_2 = offset_1;
                 offset_1 = offset;
-                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
+                ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
 
             } else {
                 ip += ((ip-anchor) >> kSearchStrength) + 1;
@@ -708,7 +461,7 @@ size_t ZSTD_compressBlock_doubleFast_extDict_generic(
         if (ip <= ilimit) {
             /* Complementary insertion */
             /* done after iLimit test, as candidates could be > iend-8 */
-            {   U32 const indexToInsert = curr+2;
+            {   U32 const indexToInsert = current+2;
                 hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
                 hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
                 hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
@@ -721,12 +474,12 @@ size_t ZSTD_compressBlock_doubleFast_extDict_generic(
                 U32 const repIndex2 = current2 - offset_2;
                 const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
                 if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3)   /* intentional overflow : ensure repIndex2 doesn't overlap dict + prefix */
-                    & (offset_2 <= current2 - dictStartIndex))
+                    & (repIndex2 > dictStartIndex))
                   && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
                     const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
                     size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
                     U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset;   /* swap offset_2 <=> offset_1 */
-                    ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
+                    ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
                     hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
                     hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
                     ip += repLength2;
@@ -744,10 +497,6 @@ size_t ZSTD_compressBlock_doubleFast_extDict_generic(
     return (size_t)(iend - anchor);
 }
 
-ZSTD_GEN_DFAST_FN(extDict, 4)
-ZSTD_GEN_DFAST_FN(extDict, 5)
-ZSTD_GEN_DFAST_FN(extDict, 6)
-ZSTD_GEN_DFAST_FN(extDict, 7)
 
 size_t ZSTD_compressBlock_doubleFast_extDict(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
@@ -758,16 +507,12 @@ size_t ZSTD_compressBlock_doubleFast_extDict(
     {
     default: /* includes case 3 */
     case 4 :
-        return ZSTD_compressBlock_doubleFast_extDict_4(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 4);
     case 5 :
-        return ZSTD_compressBlock_doubleFast_extDict_5(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 5);
     case 6 :
-        return ZSTD_compressBlock_doubleFast_extDict_6(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 6);
     case 7 :
-        return ZSTD_compressBlock_doubleFast_extDict_7(ms, seqStore, rep, src, srcSize);
+        return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 7);
     }
 }
-
-#endif /* ZSTD_EXCLUDE_DFAST_BLOCK_COMPRESSOR */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_double_fast.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -19,12 +18,8 @@ extern "C" {
 #include "mem.h"      /* U32 */
 #include "zstd_compress_internal.h"     /* ZSTD_CCtx, size_t */
 
-#ifndef ZSTD_EXCLUDE_DFAST_BLOCK_COMPRESSOR
-
 void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
-                              void const* end, ZSTD_dictTableLoadMethod_e dtlm,
-                              ZSTD_tableFillPurpose_e tfp);
-
+                              void const* end, ZSTD_dictTableLoadMethod_e dtlm);
 size_t ZSTD_compressBlock_doubleFast(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
@@ -35,19 +30,9 @@ size_t ZSTD_compressBlock_doubleFast_extDict(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
 
-#define ZSTD_COMPRESSBLOCK_DOUBLEFAST ZSTD_compressBlock_doubleFast
-#define ZSTD_COMPRESSBLOCK_DOUBLEFAST_DICTMATCHSTATE ZSTD_compressBlock_doubleFast_dictMatchState
-#define ZSTD_COMPRESSBLOCK_DOUBLEFAST_EXTDICT ZSTD_compressBlock_doubleFast_extDict
-#else
-#define ZSTD_COMPRESSBLOCK_DOUBLEFAST NULL
-#define ZSTD_COMPRESSBLOCK_DOUBLEFAST_DICTMATCHSTATE NULL
-#define ZSTD_COMPRESSBLOCK_DOUBLEFAST_EXTDICT NULL
-#endif /* ZSTD_EXCLUDE_DFAST_BLOCK_COMPRESSOR */
 
 #if defined (__cplusplus)
 }
 #endif
 
 #endif /* ZSTD_DOUBLE_FAST_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_errors.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -21,31 +20,19 @@ extern "C" {
 
 
 /* =====   ZSTDERRORLIB_API : control library symbols visibility   ===== */
-#ifndef ZSTDERRORLIB_VISIBLE
-   /* Backwards compatibility with old macro name */
-#  ifdef ZSTDERRORLIB_VISIBILITY
-#    define ZSTDERRORLIB_VISIBLE ZSTDERRORLIB_VISIBILITY
-#  elif defined(__GNUC__) && (__GNUC__ >= 4) && !defined(__MINGW32__)
-#    define ZSTDERRORLIB_VISIBLE __attribute__ ((visibility ("default")))
+#ifndef ZSTDERRORLIB_VISIBILITY
+#  if defined(__GNUC__) && (__GNUC__ >= 4)
+#    define ZSTDERRORLIB_VISIBILITY __attribute__ ((visibility ("default")))
 #  else
-#    define ZSTDERRORLIB_VISIBLE
+#    define ZSTDERRORLIB_VISIBILITY
 #  endif
 #endif
-
-#ifndef ZSTDERRORLIB_HIDDEN
-#  if defined(__GNUC__) && (__GNUC__ >= 4) && !defined(__MINGW32__)
-#    define ZSTDERRORLIB_HIDDEN __attribute__ ((visibility ("hidden")))
-#  else
-#    define ZSTDERRORLIB_HIDDEN
-#  endif
-#endif
-
 #if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1)
-#  define ZSTDERRORLIB_API __declspec(dllexport) ZSTDERRORLIB_VISIBLE
+#  define ZSTDERRORLIB_API __declspec(dllexport) ZSTDERRORLIB_VISIBILITY
 #elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1)
-#  define ZSTDERRORLIB_API __declspec(dllimport) ZSTDERRORLIB_VISIBLE /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
+#  define ZSTDERRORLIB_API __declspec(dllimport) ZSTDERRORLIB_VISIBILITY /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
 #else
-#  define ZSTDERRORLIB_API ZSTDERRORLIB_VISIBLE
+#  define ZSTDERRORLIB_API ZSTDERRORLIB_VISIBILITY
 #endif
 
 /*-*********************************************
@@ -71,17 +58,14 @@ typedef enum {
   ZSTD_error_frameParameter_windowTooLarge = 16,
   ZSTD_error_corruption_detected = 20,
   ZSTD_error_checksum_wrong      = 22,
-  ZSTD_error_literals_headerWrong = 24,
   ZSTD_error_dictionary_corrupted      = 30,
   ZSTD_error_dictionary_wrong          = 32,
   ZSTD_error_dictionaryCreation_failed = 34,
   ZSTD_error_parameter_unsupported   = 40,
-  ZSTD_error_parameter_combination_unsupported = 41,
   ZSTD_error_parameter_outOfBound    = 42,
   ZSTD_error_tableLog_tooLarge       = 44,
   ZSTD_error_maxSymbolValue_tooLarge = 46,
   ZSTD_error_maxSymbolValue_tooSmall = 48,
-  ZSTD_error_stabilityCondition_notRespected = 50,
   ZSTD_error_stage_wrong       = 60,
   ZSTD_error_init_missing      = 62,
   ZSTD_error_memory_allocation = 64,
@@ -89,15 +73,9 @@ typedef enum {
   ZSTD_error_dstSize_tooSmall = 70,
   ZSTD_error_srcSize_wrong    = 72,
   ZSTD_error_dstBuffer_null   = 74,
-  ZSTD_error_noForwardProgress_destFull = 80,
-  ZSTD_error_noForwardProgress_inputEmpty = 82,
   /* following error codes are __NOT STABLE__, they can be removed or changed in future versions */
   ZSTD_error_frameIndex_tooLarge = 100,
   ZSTD_error_seekableIO          = 102,
-  ZSTD_error_dstBuffer_wrong     = 104,
-  ZSTD_error_srcBuffer_wrong     = 105,
-  ZSTD_error_sequenceProducer_failed = 106,
-  ZSTD_error_externalSequences_invalid = 107,
   ZSTD_error_maxCode = 120  /* never EVER use this value directly, it can change in future versions! Use ZSTD_isError() instead */
 } ZSTD_ErrorCode;
 
@@ -113,5 +91,3 @@ ZSTDERRORLIB_API const char* ZSTD_getErrorString(ZSTD_ErrorCode code);   /**< Sa
 #endif
 
 #endif /* ZSTD_ERRORS_H_398273423 */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_fast.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -20,8 +19,7 @@ extern "C" {
 #include "zstd_compress_internal.h"
 
 void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
-                        void const* end, ZSTD_dictTableLoadMethod_e dtlm,
-                        ZSTD_tableFillPurpose_e tfp);
+                        void const* end, ZSTD_dictTableLoadMethod_e dtlm);
 size_t ZSTD_compressBlock_fast(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
@@ -37,5 +35,3 @@ size_t ZSTD_compressBlock_fast_extDict(
 #endif
 
 #endif /* ZSTD_FAST_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_internal.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -21,7 +20,6 @@
 *  Dependencies
 ***************************************/
 #include "compiler.h"
-#include "cpu.h"
 #include "mem.h"
 #include "debug.h"                 /* assert, DEBUGLOG, RAWLOG, g_debuglevel */
 #include "error_private.h"
@@ -29,16 +27,12 @@
 #include "zstd.h"
 #define FSE_STATIC_LINKING_ONLY
 #include "fse.h"
+#define HUF_STATIC_LINKING_ONLY
 #include "huf.h"
 #ifndef XXH_STATIC_LINKING_ONLY
 #  define XXH_STATIC_LINKING_ONLY  /* XXH64_state_t */
 #endif
 #include "xxhash.h"                /* XXH_reset, update, digest */
-#ifndef ZSTD_NO_TRACE
-#  include "zstd_trace.h"
-#else
-#  define ZSTD_TRACE 0
-#endif
 
 #if defined (__cplusplus)
 extern "C" {
@@ -58,7 +52,50 @@ extern "C" {
 #undef MAX
 #define MIN(a,b) ((a)<(b) ? (a) : (b))
 #define MAX(a,b) ((a)>(b) ? (a) : (b))
-#define BOUNDED(min,val,max) (MAX(min,MIN(val,max)))
+
+/**
+ * Return the specified error if the condition evaluates to true.
+ *
+ * In debug modes, prints additional information.
+ * In order to do that (particularly, printing the conditional that failed),
+ * this can't just wrap RETURN_ERROR().
+ */
+#define RETURN_ERROR_IF(cond, err, ...) \
+  if (cond) { \
+    RAWLOG(3, "%s:%d: ERROR!: check %s failed, returning %s", __FILE__, __LINE__, ZSTD_QUOTE(cond), ZSTD_QUOTE(ERROR(err))); \
+    RAWLOG(3, ": " __VA_ARGS__); \
+    RAWLOG(3, "\n"); \
+    return ERROR(err); \
+  }
+
+/**
+ * Unconditionally return the specified error.
+ *
+ * In debug modes, prints additional information.
+ */
+#define RETURN_ERROR(err, ...) \
+  do { \
+    RAWLOG(3, "%s:%d: ERROR!: unconditional check failed, returning %s", __FILE__, __LINE__, ZSTD_QUOTE(ERROR(err))); \
+    RAWLOG(3, ": " __VA_ARGS__); \
+    RAWLOG(3, "\n"); \
+    return ERROR(err); \
+  } while(0);
+
+/**
+ * If the provided expression evaluates to an error code, returns that error code.
+ *
+ * In debug modes, prints additional information.
+ */
+#define FORWARD_IF_ERROR(err, ...) \
+  do { \
+    size_t const err_code = (err); \
+    if (ERR_isError(err_code)) { \
+      RAWLOG(3, "%s:%d: ERROR!: forwarding error in %s: %s", __FILE__, __LINE__, ZSTD_QUOTE(err), ERR_getErrorName(err_code)); \
+      RAWLOG(3, ": " __VA_ARGS__); \
+      RAWLOG(3, "\n"); \
+      return err_code; \
+    } \
+  } while(0);
 
 
 /*-*************************************
@@ -67,7 +104,8 @@ extern "C" {
 #define ZSTD_OPT_NUM    (1<<12)
 
 #define ZSTD_REP_NUM      3                 /* number of repcodes */
-static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 };
+#define ZSTD_REP_MOVE     (ZSTD_REP_NUM-1)
+static const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 };
 
 #define KB *(1 <<10)
 #define MB *(1 <<20)
@@ -81,21 +119,19 @@ static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 };
 #define BIT0   1
 
 #define ZSTD_WINDOWLOG_ABSOLUTEMIN 10
-static UNUSED_ATTR const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 };
-static UNUSED_ATTR const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 };
+static const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 };
+static const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 };
 
 #define ZSTD_FRAMEIDSIZE 4   /* magic number size */
 
 #define ZSTD_BLOCKHEADERSIZE 3   /* C standard doesn't allow `static const` variable to be init using another `static const` variable */
-static UNUSED_ATTR const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE;
+static const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE;
 typedef enum { bt_raw, bt_rle, bt_compressed, bt_reserved } blockType_e;
 
-#define ZSTD_FRAMECHECKSUMSIZE 4
-
 #define MIN_SEQUENCES_SIZE 1 /* nbSeq==0 */
-#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */)   /* for a non-null block */
-#define MIN_LITERALS_FOR_4_STREAMS 6
+#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */ + MIN_SEQUENCES_SIZE /* nbSeq==0 */)   /* for a non-null block */
 
+#define HufLog 12
 typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingType_e;
 
 #define LONGNBSEQ 0x7F00
@@ -103,7 +139,6 @@ typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingTy
 #define MINMATCH 3
 
 #define Litbits  8
-#define LitHufLog 11
 #define MaxLit ((1<<Litbits) - 1)
 #define MaxML   52
 #define MaxLL   35
@@ -114,92 +149,53 @@ typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingTy
 #define LLFSELog    9
 #define OffFSELog   8
 #define MaxFSELog  MAX(MAX(MLFSELog, LLFSELog), OffFSELog)
-#define MaxMLBits 16
-#define MaxLLBits 16
 
-#define ZSTD_MAX_HUF_HEADER_SIZE 128 /* header + <= 127 byte tree description */
-/* Each table cannot take more than #symbols * FSELog bits */
-#define ZSTD_MAX_FSE_HEADERS_SIZE (((MaxML + 1) * MLFSELog + (MaxLL + 1) * LLFSELog + (MaxOff + 1) * OffFSELog + 7) / 8)
-
-static UNUSED_ATTR const U8 LL_bits[MaxLL+1] = {
-     0, 0, 0, 0, 0, 0, 0, 0,
-     0, 0, 0, 0, 0, 0, 0, 0,
-     1, 1, 1, 1, 2, 2, 3, 3,
-     4, 6, 7, 8, 9,10,11,12,
-    13,14,15,16
-};
-static UNUSED_ATTR const S16 LL_defaultNorm[MaxLL+1] = {
-     4, 3, 2, 2, 2, 2, 2, 2,
-     2, 2, 2, 2, 2, 1, 1, 1,
-     2, 2, 2, 2, 2, 2, 2, 2,
-     2, 3, 2, 1, 1, 1, 1, 1,
-    -1,-1,-1,-1
-};
+static const U32 LL_bits[MaxLL+1] = { 0, 0, 0, 0, 0, 0, 0, 0,
+                                      0, 0, 0, 0, 0, 0, 0, 0,
+                                      1, 1, 1, 1, 2, 2, 3, 3,
+                                      4, 6, 7, 8, 9,10,11,12,
+                                     13,14,15,16 };
+static const S16 LL_defaultNorm[MaxLL+1] = { 4, 3, 2, 2, 2, 2, 2, 2,
+                                             2, 2, 2, 2, 2, 1, 1, 1,
+                                             2, 2, 2, 2, 2, 2, 2, 2,
+                                             2, 3, 2, 1, 1, 1, 1, 1,
+                                            -1,-1,-1,-1 };
 #define LL_DEFAULTNORMLOG 6  /* for static allocation */
-static UNUSED_ATTR const U32 LL_defaultNormLog = LL_DEFAULTNORMLOG;
+static const U32 LL_defaultNormLog = LL_DEFAULTNORMLOG;
 
-static UNUSED_ATTR const U8 ML_bits[MaxML+1] = {
-     0, 0, 0, 0, 0, 0, 0, 0,
-     0, 0, 0, 0, 0, 0, 0, 0,
-     0, 0, 0, 0, 0, 0, 0, 0,
-     0, 0, 0, 0, 0, 0, 0, 0,
-     1, 1, 1, 1, 2, 2, 3, 3,
-     4, 4, 5, 7, 8, 9,10,11,
-    12,13,14,15,16
-};
-static UNUSED_ATTR const S16 ML_defaultNorm[MaxML+1] = {
-     1, 4, 3, 2, 2, 2, 2, 2,
-     2, 1, 1, 1, 1, 1, 1, 1,
-     1, 1, 1, 1, 1, 1, 1, 1,
-     1, 1, 1, 1, 1, 1, 1, 1,
-     1, 1, 1, 1, 1, 1, 1, 1,
-     1, 1, 1, 1, 1, 1,-1,-1,
-    -1,-1,-1,-1,-1
-};
+static const U32 ML_bits[MaxML+1] = { 0, 0, 0, 0, 0, 0, 0, 0,
+                                      0, 0, 0, 0, 0, 0, 0, 0,
+                                      0, 0, 0, 0, 0, 0, 0, 0,
+                                      0, 0, 0, 0, 0, 0, 0, 0,
+                                      1, 1, 1, 1, 2, 2, 3, 3,
+                                      4, 4, 5, 7, 8, 9,10,11,
+                                     12,13,14,15,16 };
+static const S16 ML_defaultNorm[MaxML+1] = { 1, 4, 3, 2, 2, 2, 2, 2,
+                                             2, 1, 1, 1, 1, 1, 1, 1,
+                                             1, 1, 1, 1, 1, 1, 1, 1,
+                                             1, 1, 1, 1, 1, 1, 1, 1,
+                                             1, 1, 1, 1, 1, 1, 1, 1,
+                                             1, 1, 1, 1, 1, 1,-1,-1,
+                                            -1,-1,-1,-1,-1 };
 #define ML_DEFAULTNORMLOG 6  /* for static allocation */
-static UNUSED_ATTR const U32 ML_defaultNormLog = ML_DEFAULTNORMLOG;
+static const U32 ML_defaultNormLog = ML_DEFAULTNORMLOG;
 
-static UNUSED_ATTR const S16 OF_defaultNorm[DefaultMaxOff+1] = {
-     1, 1, 1, 1, 1, 1, 2, 2,
-     2, 1, 1, 1, 1, 1, 1, 1,
-     1, 1, 1, 1, 1, 1, 1, 1,
-    -1,-1,-1,-1,-1
-};
+static const S16 OF_defaultNorm[DefaultMaxOff+1] = { 1, 1, 1, 1, 1, 1, 2, 2,
+                                                     2, 1, 1, 1, 1, 1, 1, 1,
+                                                     1, 1, 1, 1, 1, 1, 1, 1,
+                                                    -1,-1,-1,-1,-1 };
 #define OF_DEFAULTNORMLOG 5  /* for static allocation */
-static UNUSED_ATTR const U32 OF_defaultNormLog = OF_DEFAULTNORMLOG;
+static const U32 OF_defaultNormLog = OF_DEFAULTNORMLOG;
 
 
 /*-*******************************************
 *  Shared functions to include for inlining
 *********************************************/
-static void ZSTD_copy8(void* dst, const void* src) {
-#if defined(ZSTD_ARCH_ARM_NEON)
-    vst1_u8((uint8_t*)dst, vld1_u8((const uint8_t*)src));
-#else
-    ZSTD_memcpy(dst, src, 8);
-#endif
-}
-#define COPY8(d,s) do { ZSTD_copy8(d,s); d+=8; s+=8; } while (0)
+static void ZSTD_copy8(void* dst, const void* src) { memcpy(dst, src, 8); }
 
-/* Need to use memmove here since the literal buffer can now be located within
-   the dst buffer. In circumstances where the op "catches up" to where the
-   literal buffer is, there can be partial overlaps in this call on the final
-   copy if the literal is being shifted by less than 16 bytes. */
-static void ZSTD_copy16(void* dst, const void* src) {
-#if defined(ZSTD_ARCH_ARM_NEON)
-    vst1q_u8((uint8_t*)dst, vld1q_u8((const uint8_t*)src));
-#elif defined(ZSTD_ARCH_X86_SSE2)
-    _mm_storeu_si128((__m128i*)dst, _mm_loadu_si128((const __m128i*)src));
-#elif defined(__clang__)
-    ZSTD_memmove(dst, src, 16);
-#else
-    /* ZSTD_memmove is not inlined properly by gcc */
-    BYTE copy16_buf[16];
-    ZSTD_memcpy(copy16_buf, src, 16);
-    ZSTD_memcpy(dst, copy16_buf, 16);
-#endif
-}
-#define COPY16(d,s) do { ZSTD_copy16(d,s); d+=16; s+=16; } while (0)
+#define COPY8(d,s) { ZSTD_copy8(d,s); d+=8; s+=8; }
+static void ZSTD_copy16(void* dst, const void* src) { memcpy(dst, src, 16); }
+#define COPY16(d,s) { ZSTD_copy16(d,s); d+=16; s+=16; }
 
 #define WILDCOPY_OVERLENGTH 32
 #define WILDCOPY_VECLEN 16
@@ -211,13 +207,13 @@ typedef enum {
 } ZSTD_overlap_e;
 
 /*! ZSTD_wildcopy() :
- *  Custom version of ZSTD_memcpy(), can over read/write up to WILDCOPY_OVERLENGTH bytes (if length==0)
+ *  Custom version of memcpy(), can over read/write up to WILDCOPY_OVERLENGTH bytes (if length==0)
  *  @param ovtype controls the overlap detection
  *         - ZSTD_no_overlap: The source and destination are guaranteed to be at least WILDCOPY_VECLEN bytes apart.
  *         - ZSTD_overlap_src_before_dst: The src and dst may overlap, but they MUST be at least 8 bytes apart.
  *           The src buffer must be before the dst buffer.
  */
-MEM_STATIC FORCE_INLINE_ATTR
+MEM_STATIC FORCE_INLINE_ATTR DONT_VECTORIZE
 void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e const ovtype)
 {
     ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src;
@@ -225,23 +221,24 @@ void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e
     BYTE* op = (BYTE*)dst;
     BYTE* const oend = op + length;
 
+    assert(diff >= 8 || (ovtype == ZSTD_no_overlap && diff <= -WILDCOPY_VECLEN));
+
     if (ovtype == ZSTD_overlap_src_before_dst && diff < WILDCOPY_VECLEN) {
         /* Handle short offset copies. */
         do {
-            COPY8(op, ip);
+            COPY8(op, ip)
         } while (op < oend);
     } else {
         assert(diff >= WILDCOPY_VECLEN || diff <= -WILDCOPY_VECLEN);
-        /* Separate out the first COPY16() call because the copy length is
+        /* Separate out the first two COPY16() calls because the copy length is
          * almost certain to be short, so the branches have different
-         * probabilities. Since it is almost certain to be short, only do
-         * one COPY16() in the first call. Then, do two calls per loop since
-         * at that point it is more likely to have a high trip count.
+         * probabilities.
+         * On gcc-9 unrolling once is +1.6%, twice is +2%, thrice is +1.8%.
+         * On clang-8 unrolling once is +1.4%, twice is +3.3%, thrice is +3%.
          */
-        ZSTD_copy16(op, ip);
-        if (16 >= length) return;
-        op += 16;
-        ip += 16;
+        COPY16(op, ip);
+        COPY16(op, ip);
+        if (op >= oend) return;
         do {
             COPY16(op, ip);
             COPY16(op, ip);
@@ -250,92 +247,30 @@ void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e
     }
 }
 
-MEM_STATIC size_t ZSTD_limitCopy(void* dst, size_t dstCapacity, const void* src, size_t srcSize)
-{
-    size_t const length = MIN(dstCapacity, srcSize);
-    if (length > 0) {
-        ZSTD_memcpy(dst, src, length);
-    }
-    return length;
-}
-
-/* define "workspace is too large" as this number of times larger than needed */
-#define ZSTD_WORKSPACETOOLARGE_FACTOR 3
-
-/* when workspace is continuously too large
- * during at least this number of times,
- * context's memory usage is considered wasteful,
- * because it's sized to handle a worst case scenario which rarely happens.
- * In which case, resize it down to free some memory */
-#define ZSTD_WORKSPACETOOLARGE_MAXDURATION 128
-
-/* Controls whether the input/output buffer is buffered or stable. */
-typedef enum {
-    ZSTD_bm_buffered = 0,  /* Buffer the input/output */
-    ZSTD_bm_stable = 1     /* ZSTD_inBuffer/ZSTD_outBuffer is stable */
-} ZSTD_bufferMode_e;
-
 
 /*-*******************************************
 *  Private declarations
 *********************************************/
 typedef struct seqDef_s {
-    U32 offBase;   /* offBase == Offset + ZSTD_REP_NUM, or repcode 1,2,3 */
+    U32 offset;
     U16 litLength;
-    U16 mlBase;    /* mlBase == matchLength - MINMATCH */
+    U16 matchLength;
 } seqDef;
 
-/* Controls whether seqStore has a single "long" litLength or matchLength. See seqStore_t. */
-typedef enum {
-    ZSTD_llt_none = 0,             /* no longLengthType */
-    ZSTD_llt_literalLength = 1,    /* represents a long literal */
-    ZSTD_llt_matchLength = 2       /* represents a long match */
-} ZSTD_longLengthType_e;
-
 typedef struct {
     seqDef* sequencesStart;
-    seqDef* sequences;      /* ptr to end of sequences */
-    BYTE*  litStart;
-    BYTE*  lit;             /* ptr to end of literals */
-    BYTE*  llCode;
-    BYTE*  mlCode;
-    BYTE*  ofCode;
+    seqDef* sequences;
+    BYTE* litStart;
+    BYTE* lit;
+    BYTE* llCode;
+    BYTE* mlCode;
+    BYTE* ofCode;
     size_t maxNbSeq;
     size_t maxNbLit;
-
-    /* longLengthPos and longLengthType to allow us to represent either a single litLength or matchLength
-     * in the seqStore that has a value larger than U16 (if it exists). To do so, we increment
-     * the existing value of the litLength or matchLength by 0x10000.
-     */
-    ZSTD_longLengthType_e longLengthType;
-    U32                   longLengthPos;  /* Index of the sequence to apply long length modification to */
+    U32   longLengthID;   /* 0 == no longLength; 1 == Lit.longLength; 2 == Match.longLength; */
+    U32   longLengthPos;
 } seqStore_t;
 
-typedef struct {
-    U32 litLength;
-    U32 matchLength;
-} ZSTD_sequenceLength;
-
-/**
- * Returns the ZSTD_sequenceLength for the given sequences. It handles the decoding of long sequences
- * indicated by longLengthPos and longLengthType, and adds MINMATCH back to matchLength.
- */
-MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore, seqDef const* seq)
-{
-    ZSTD_sequenceLength seqLen;
-    seqLen.litLength = seq->litLength;
-    seqLen.matchLength = seq->mlBase + MINMATCH;
-    if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) {
-        if (seqStore->longLengthType == ZSTD_llt_literalLength) {
-            seqLen.litLength += 0x10000;
-        }
-        if (seqStore->longLengthType == ZSTD_llt_matchLength) {
-            seqLen.matchLength += 0x10000;
-        }
-    }
-    return seqLen;
-}
-
 /**
  * Contains the compressed frame size and an upper-bound for the decompressed frame size.
  * Note: before using `compressedSize`, check for errors using ZSTD_isError().
@@ -343,13 +278,43 @@ MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore
  *          `decompressedBound != ZSTD_CONTENTSIZE_ERROR`
  */
 typedef struct {
-    size_t nbBlocks;
     size_t compressedSize;
     unsigned long long decompressedBound;
 } ZSTD_frameSizeInfo;   /* decompress & legacy */
 
 const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx);   /* compress & dictBuilder */
-int ZSTD_seqToCodes(const seqStore_t* seqStorePtr);   /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */
+void ZSTD_seqToCodes(const seqStore_t* seqStorePtr);   /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */
+
+/* custom memory allocation functions */
+void* ZSTD_malloc(size_t size, ZSTD_customMem customMem);
+void* ZSTD_calloc(size_t size, ZSTD_customMem customMem);
+void ZSTD_free(void* ptr, ZSTD_customMem customMem);
+
+
+MEM_STATIC U32 ZSTD_highbit32(U32 val)   /* compress, dictBuilder, decodeCorpus */
+{
+    assert(val != 0);
+    {
+#   if defined(_MSC_VER)   /* Visual */
+        unsigned long r=0;
+        _BitScanReverse(&r, val);
+        return (unsigned)r;
+#   elif defined(__GNUC__) && (__GNUC__ >= 3)   /* GCC Intrinsic */
+        return __builtin_clz (val) ^ 31;
+#   elif defined(__ICCARM__)    /* IAR Intrinsic */
+        return 31 - __CLZ(val);
+#   else   /* Software version */
+        static const U32 DeBruijnClz[32] = { 0, 9, 1, 10, 13, 21, 2, 29, 11, 14, 16, 18, 22, 25, 3, 30, 8, 12, 20, 28, 15, 17, 24, 7, 19, 27, 23, 6, 26, 5, 4, 31 };
+        U32 v = val;
+        v |= v >> 1;
+        v |= v >> 2;
+        v |= v >> 4;
+        v |= v >> 8;
+        v |= v >> 16;
+        return DeBruijnClz[(v * 0x07C4ACDDU) >> 27];
+#   endif
+    }
+}
 
 
 /* ZSTD_invalidateRepCodes() :
@@ -367,29 +332,19 @@ typedef struct {
 
 /*! ZSTD_getcBlockSize() :
  *  Provides the size of compressed block from block header `src` */
-/*  Used by: decompress, fullbench */
+/* Used by: decompress, fullbench (does not get its definition from here) */
 size_t ZSTD_getcBlockSize(const void* src, size_t srcSize,
                           blockProperties_t* bpPtr);
 
 /*! ZSTD_decodeSeqHeaders() :
  *  decode sequence header from src */
-/*  Used by: zstd_decompress_block, fullbench */
+/* Used by: decompress, fullbench (does not get its definition from here) */
 size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr,
                        const void* src, size_t srcSize);
 
-/**
- * @returns true iff the CPU supports dynamic BMI2 dispatch.
- */
-MEM_STATIC int ZSTD_cpuSupportsBmi2(void)
-{
-    ZSTD_cpuid_t cpuid = ZSTD_cpuid();
-    return ZSTD_cpuid_bmi1(cpuid) && ZSTD_cpuid_bmi2(cpuid);
-}
 
 #if defined (__cplusplus)
 }
 #endif
 
 #endif   /* ZSTD_CCOMMON_H_MODULE */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_lazy.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -18,188 +17,51 @@ extern "C" {
 
 #include "zstd_compress_internal.h"
 
-/**
- * Dedicated Dictionary Search Structure bucket log. In the
- * ZSTD_dedicatedDictSearch mode, the hashTable has
- * 2 ** ZSTD_LAZY_DDSS_BUCKET_LOG entries in each bucket, rather than just
- * one.
- */
-#define ZSTD_LAZY_DDSS_BUCKET_LOG 2
-
-#define ZSTD_ROW_HASH_TAG_BITS 8        /* nb bits to use for the tag */
-
-#if !defined(ZSTD_EXCLUDE_GREEDY_BLOCK_COMPRESSOR) \
- || !defined(ZSTD_EXCLUDE_LAZY_BLOCK_COMPRESSOR) \
- || !defined(ZSTD_EXCLUDE_LAZY2_BLOCK_COMPRESSOR) \
- || !defined(ZSTD_EXCLUDE_BTLAZY2_BLOCK_COMPRESSOR)
 U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip);
-void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip);
-
-void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip);
 
 void ZSTD_preserveUnsortedMark (U32* const table, U32 const size, U32 const reducerValue);  /*! used in ZSTD_reduceIndex(). preemptively increase value of ZSTD_DUBT_UNSORTED_MARK */
-#endif
 
-#ifndef ZSTD_EXCLUDE_GREEDY_BLOCK_COMPRESSOR
-size_t ZSTD_compressBlock_greedy(
+size_t ZSTD_compressBlock_btlazy2(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_greedy_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_greedy_dictMatchState(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_greedy_dictMatchState_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_greedy_dedicatedDictSearch(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_greedy_extDict(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_greedy_extDict_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-
-#define ZSTD_COMPRESSBLOCK_GREEDY ZSTD_compressBlock_greedy
-#define ZSTD_COMPRESSBLOCK_GREEDY_ROW ZSTD_compressBlock_greedy_row
-#define ZSTD_COMPRESSBLOCK_GREEDY_DICTMATCHSTATE ZSTD_compressBlock_greedy_dictMatchState
-#define ZSTD_COMPRESSBLOCK_GREEDY_DICTMATCHSTATE_ROW ZSTD_compressBlock_greedy_dictMatchState_row
-#define ZSTD_COMPRESSBLOCK_GREEDY_DEDICATEDDICTSEARCH ZSTD_compressBlock_greedy_dedicatedDictSearch
-#define ZSTD_COMPRESSBLOCK_GREEDY_DEDICATEDDICTSEARCH_ROW ZSTD_compressBlock_greedy_dedicatedDictSearch_row
-#define ZSTD_COMPRESSBLOCK_GREEDY_EXTDICT ZSTD_compressBlock_greedy_extDict
-#define ZSTD_COMPRESSBLOCK_GREEDY_EXTDICT_ROW ZSTD_compressBlock_greedy_extDict_row
-#else
-#define ZSTD_COMPRESSBLOCK_GREEDY NULL
-#define ZSTD_COMPRESSBLOCK_GREEDY_ROW NULL
-#define ZSTD_COMPRESSBLOCK_GREEDY_DICTMATCHSTATE NULL
-#define ZSTD_COMPRESSBLOCK_GREEDY_DICTMATCHSTATE_ROW NULL
-#define ZSTD_COMPRESSBLOCK_GREEDY_DEDICATEDDICTSEARCH NULL
-#define ZSTD_COMPRESSBLOCK_GREEDY_DEDICATEDDICTSEARCH_ROW NULL
-#define ZSTD_COMPRESSBLOCK_GREEDY_EXTDICT NULL
-#define ZSTD_COMPRESSBLOCK_GREEDY_EXTDICT_ROW NULL
-#endif
-
-#ifndef ZSTD_EXCLUDE_LAZY_BLOCK_COMPRESSOR
-size_t ZSTD_compressBlock_lazy(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy_dictMatchState(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy_dictMatchState_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy_dedicatedDictSearch(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy_extDict(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy_extDict_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-
-#define ZSTD_COMPRESSBLOCK_LAZY ZSTD_compressBlock_lazy
-#define ZSTD_COMPRESSBLOCK_LAZY_ROW ZSTD_compressBlock_lazy_row
-#define ZSTD_COMPRESSBLOCK_LAZY_DICTMATCHSTATE ZSTD_compressBlock_lazy_dictMatchState
-#define ZSTD_COMPRESSBLOCK_LAZY_DICTMATCHSTATE_ROW ZSTD_compressBlock_lazy_dictMatchState_row
-#define ZSTD_COMPRESSBLOCK_LAZY_DEDICATEDDICTSEARCH ZSTD_compressBlock_lazy_dedicatedDictSearch
-#define ZSTD_COMPRESSBLOCK_LAZY_DEDICATEDDICTSEARCH_ROW ZSTD_compressBlock_lazy_dedicatedDictSearch_row
-#define ZSTD_COMPRESSBLOCK_LAZY_EXTDICT ZSTD_compressBlock_lazy_extDict
-#define ZSTD_COMPRESSBLOCK_LAZY_EXTDICT_ROW ZSTD_compressBlock_lazy_extDict_row
-#else
-#define ZSTD_COMPRESSBLOCK_LAZY NULL
-#define ZSTD_COMPRESSBLOCK_LAZY_ROW NULL
-#define ZSTD_COMPRESSBLOCK_LAZY_DICTMATCHSTATE NULL
-#define ZSTD_COMPRESSBLOCK_LAZY_DICTMATCHSTATE_ROW NULL
-#define ZSTD_COMPRESSBLOCK_LAZY_DEDICATEDDICTSEARCH NULL
-#define ZSTD_COMPRESSBLOCK_LAZY_DEDICATEDDICTSEARCH_ROW NULL
-#define ZSTD_COMPRESSBLOCK_LAZY_EXTDICT NULL
-#define ZSTD_COMPRESSBLOCK_LAZY_EXTDICT_ROW NULL
-#endif
-
-#ifndef ZSTD_EXCLUDE_LAZY2_BLOCK_COMPRESSOR
 size_t ZSTD_compressBlock_lazy2(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy2_row(
+size_t ZSTD_compressBlock_lazy(
+        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+        void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_greedy(
+        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+        void const* src, size_t srcSize);
+
+size_t ZSTD_compressBlock_btlazy2_dictMatchState(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
 size_t ZSTD_compressBlock_lazy2_dictMatchState(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy2_dictMatchState_row(
+size_t ZSTD_compressBlock_lazy_dictMatchState(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch(
+size_t ZSTD_compressBlock_greedy_dictMatchState(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row(
+
+size_t ZSTD_compressBlock_greedy_extDict(
+        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
+        void const* src, size_t srcSize);
+size_t ZSTD_compressBlock_lazy_extDict(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
 size_t ZSTD_compressBlock_lazy2_extDict(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_lazy2_extDict_row(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-
-#define ZSTD_COMPRESSBLOCK_LAZY2 ZSTD_compressBlock_lazy2
-#define ZSTD_COMPRESSBLOCK_LAZY2_ROW ZSTD_compressBlock_lazy2_row
-#define ZSTD_COMPRESSBLOCK_LAZY2_DICTMATCHSTATE ZSTD_compressBlock_lazy2_dictMatchState
-#define ZSTD_COMPRESSBLOCK_LAZY2_DICTMATCHSTATE_ROW ZSTD_compressBlock_lazy2_dictMatchState_row
-#define ZSTD_COMPRESSBLOCK_LAZY2_DEDICATEDDICTSEARCH ZSTD_compressBlock_lazy2_dedicatedDictSearch
-#define ZSTD_COMPRESSBLOCK_LAZY2_DEDICATEDDICTSEARCH_ROW ZSTD_compressBlock_lazy2_dedicatedDictSearch_row
-#define ZSTD_COMPRESSBLOCK_LAZY2_EXTDICT ZSTD_compressBlock_lazy2_extDict
-#define ZSTD_COMPRESSBLOCK_LAZY2_EXTDICT_ROW ZSTD_compressBlock_lazy2_extDict_row
-#else
-#define ZSTD_COMPRESSBLOCK_LAZY2 NULL
-#define ZSTD_COMPRESSBLOCK_LAZY2_ROW NULL
-#define ZSTD_COMPRESSBLOCK_LAZY2_DICTMATCHSTATE NULL
-#define ZSTD_COMPRESSBLOCK_LAZY2_DICTMATCHSTATE_ROW NULL
-#define ZSTD_COMPRESSBLOCK_LAZY2_DEDICATEDDICTSEARCH NULL
-#define ZSTD_COMPRESSBLOCK_LAZY2_DEDICATEDDICTSEARCH_ROW NULL
-#define ZSTD_COMPRESSBLOCK_LAZY2_EXTDICT NULL
-#define ZSTD_COMPRESSBLOCK_LAZY2_EXTDICT_ROW NULL
-#endif
-
-#ifndef ZSTD_EXCLUDE_BTLAZY2_BLOCK_COMPRESSOR
-size_t ZSTD_compressBlock_btlazy2(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
-size_t ZSTD_compressBlock_btlazy2_dictMatchState(
-        ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-        void const* src, size_t srcSize);
 size_t ZSTD_compressBlock_btlazy2_extDict(
         ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
         void const* src, size_t srcSize);
 
-#define ZSTD_COMPRESSBLOCK_BTLAZY2 ZSTD_compressBlock_btlazy2
-#define ZSTD_COMPRESSBLOCK_BTLAZY2_DICTMATCHSTATE ZSTD_compressBlock_btlazy2_dictMatchState
-#define ZSTD_COMPRESSBLOCK_BTLAZY2_EXTDICT ZSTD_compressBlock_btlazy2_extDict
-#else
-#define ZSTD_COMPRESSBLOCK_BTLAZY2 NULL
-#define ZSTD_COMPRESSBLOCK_BTLAZY2_DICTMATCHSTATE NULL
-#define ZSTD_COMPRESSBLOCK_BTLAZY2_EXTDICT NULL
-#endif
-
-
 #if defined (__cplusplus)
 }
 #endif
 
 #endif /* ZSTD_LAZY_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_ldm.c

@@ -1,137 +1,22 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
  * LICENSE file in the root directory of this source tree) and the GPLv2 (found
  * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
  */
 
 #include "zstd_ldm.h"
 
 #include "debug.h"
-#include "xxhash.h"
 #include "zstd_fast.h"          /* ZSTD_fillHashTable() */
 #include "zstd_double_fast.h"   /* ZSTD_fillDoubleHashTable() */
-#include "zstd_ldm_geartab.h"
 
 #define LDM_BUCKET_SIZE_LOG 3
 #define LDM_MIN_MATCH_LENGTH 64
 #define LDM_HASH_RLOG 7
-
-typedef struct {
-    U64 rolling;
-    U64 stopMask;
-} ldmRollingHashState_t;
-
-/** ZSTD_ldm_gear_init():
- *
- * Initializes the rolling hash state such that it will honor the
- * settings in params. */
-static void ZSTD_ldm_gear_init(ldmRollingHashState_t* state, ldmParams_t const* params)
-{
-    unsigned maxBitsInMask = MIN(params->minMatchLength, 64);
-    unsigned hashRateLog = params->hashRateLog;
-
-    state->rolling = ~(U32)0;
-
-    /* The choice of the splitting criterion is subject to two conditions:
-     *   1. it has to trigger on average every 2^(hashRateLog) bytes;
-     *   2. ideally, it has to depend on a window of minMatchLength bytes.
-     *
-     * In the gear hash algorithm, bit n depends on the last n bytes;
-     * so in order to obtain a good quality splitting criterion it is
-     * preferable to use bits with high weight.
-     *
-     * To match condition 1 we use a mask with hashRateLog bits set
-     * and, because of the previous remark, we make sure these bits
-     * have the highest possible weight while still respecting
-     * condition 2.
-     */
-    if (hashRateLog > 0 && hashRateLog <= maxBitsInMask) {
-        state->stopMask = (((U64)1 << hashRateLog) - 1) << (maxBitsInMask - hashRateLog);
-    } else {
-        /* In this degenerate case we simply honor the hash rate. */
-        state->stopMask = ((U64)1 << hashRateLog) - 1;
-    }
-}
-
-/** ZSTD_ldm_gear_reset()
- * Feeds [data, data + minMatchLength) into the hash without registering any
- * splits. This effectively resets the hash state. This is used when skipping
- * over data, either at the beginning of a block, or skipping sections.
- */
-static void ZSTD_ldm_gear_reset(ldmRollingHashState_t* state,
-                                BYTE const* data, size_t minMatchLength)
-{
-    U64 hash = state->rolling;
-    size_t n = 0;
-
-#define GEAR_ITER_ONCE() do {                                  \
-        hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \
-        n += 1;                                                \
-    } while (0)
-    while (n + 3 < minMatchLength) {
-        GEAR_ITER_ONCE();
-        GEAR_ITER_ONCE();
-        GEAR_ITER_ONCE();
-        GEAR_ITER_ONCE();
-    }
-    while (n < minMatchLength) {
-        GEAR_ITER_ONCE();
-    }
-#undef GEAR_ITER_ONCE
-}
-
-/** ZSTD_ldm_gear_feed():
- *
- * Registers in the splits array all the split points found in the first
- * size bytes following the data pointer. This function terminates when
- * either all the data has been processed or LDM_BATCH_SIZE splits are
- * present in the splits array.
- *
- * Precondition: The splits array must not be full.
- * Returns: The number of bytes processed. */
-static size_t ZSTD_ldm_gear_feed(ldmRollingHashState_t* state,
-                                 BYTE const* data, size_t size,
-                                 size_t* splits, unsigned* numSplits)
-{
-    size_t n;
-    U64 hash, mask;
-
-    hash = state->rolling;
-    mask = state->stopMask;
-    n = 0;
-
-#define GEAR_ITER_ONCE() do { \
-        hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \
-        n += 1; \
-        if (UNLIKELY((hash & mask) == 0)) { \
-            splits[*numSplits] = n; \
-            *numSplits += 1; \
-            if (*numSplits == LDM_BATCH_SIZE) \
-                goto done; \
-        } \
-    } while (0)
-
-    while (n + 3 < size) {
-        GEAR_ITER_ONCE();
-        GEAR_ITER_ONCE();
-        GEAR_ITER_ONCE();
-        GEAR_ITER_ONCE();
-    }
-    while (n < size) {
-        GEAR_ITER_ONCE();
-    }
-
-#undef GEAR_ITER_ONCE
-
-done:
-    state->rolling = hash;
-    return n;
-}
+#define LDM_HASH_CHAR_OFFSET 10
 
 void ZSTD_ldm_adjustParameters(ldmParams_t* params,
                                ZSTD_compressionParameters const* cParams)
@@ -141,6 +26,13 @@ void ZSTD_ldm_adjustParameters(ldmParams_t* params,
     DEBUGLOG(4, "ZSTD_ldm_adjustParameters");
     if (!params->bucketSizeLog) params->bucketSizeLog = LDM_BUCKET_SIZE_LOG;
     if (!params->minMatchLength) params->minMatchLength = LDM_MIN_MATCH_LENGTH;
+    if (cParams->strategy >= ZSTD_btopt) {
+      /* Get out of the way of the optimal parser */
+      U32 const minMatch = MAX(cParams->targetLength, params->minMatchLength);
+      assert(minMatch >= ZSTD_LDM_MINMATCH_MIN);
+      assert(minMatch <= ZSTD_LDM_MINMATCH_MAX);
+      params->minMatchLength = minMatch;
+    }
     if (params->hashLog == 0) {
         params->hashLog = MAX(ZSTD_HASHLOG_MIN, params->windowLog - LDM_HASH_RLOG);
         assert(params->hashLog <= ZSTD_HASHLOG_MAX);
@@ -160,12 +52,47 @@ size_t ZSTD_ldm_getTableSize(ldmParams_t params)
     size_t const ldmBucketSize = ((size_t)1) << (params.hashLog - ldmBucketSizeLog);
     size_t const totalSize = ZSTD_cwksp_alloc_size(ldmBucketSize)
                            + ZSTD_cwksp_alloc_size(ldmHSize * sizeof(ldmEntry_t));
-    return params.enableLdm == ZSTD_ps_enable ? totalSize : 0;
+    return params.enableLdm ? totalSize : 0;
 }
 
 size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize)
 {
-    return params.enableLdm == ZSTD_ps_enable ? (maxChunkSize / params.minMatchLength) : 0;
+    return params.enableLdm ? (maxChunkSize / params.minMatchLength) : 0;
+}
+
+/** ZSTD_ldm_getSmallHash() :
+ *  numBits should be <= 32
+ *  If numBits==0, returns 0.
+ *  @return : the most significant numBits of value. */
+static U32 ZSTD_ldm_getSmallHash(U64 value, U32 numBits)
+{
+    assert(numBits <= 32);
+    return numBits == 0 ? 0 : (U32)(value >> (64 - numBits));
+}
+
+/** ZSTD_ldm_getChecksum() :
+ *  numBitsToDiscard should be <= 32
+ *  @return : the next most significant 32 bits after numBitsToDiscard */
+static U32 ZSTD_ldm_getChecksum(U64 hash, U32 numBitsToDiscard)
+{
+    assert(numBitsToDiscard <= 32);
+    return (hash >> (64 - 32 - numBitsToDiscard)) & 0xFFFFFFFF;
+}
+
+/** ZSTD_ldm_getTag() ;
+ *  Given the hash, returns the most significant numTagBits bits
+ *  after (32 + hbits) bits.
+ *
+ *  If there are not enough bits remaining, return the last
+ *  numTagBits bits. */
+static U32 ZSTD_ldm_getTag(U64 hash, U32 hbits, U32 numTagBits)
+{
+    assert(numTagBits < 32 && hbits <= 32);
+    if (32 - hbits < numTagBits) {
+        return hash & (((U32)1 << numTagBits) - 1);
+    } else {
+        return (hash >> (32 - hbits - numTagBits)) & (((U32)1 << numTagBits) - 1);
+    }
 }
 
 /** ZSTD_ldm_getBucket() :
@@ -182,12 +109,38 @@ static void ZSTD_ldm_insertEntry(ldmState_t* ldmState,
                                  size_t const hash, const ldmEntry_t entry,
                                  ldmParams_t const ldmParams)
 {
-    BYTE* const pOffset = ldmState->bucketOffsets + hash;
-    unsigned const offset = *pOffset;
-
-    *(ZSTD_ldm_getBucket(ldmState, hash, ldmParams) + offset) = entry;
-    *pOffset = (BYTE)((offset + 1) & ((1u << ldmParams.bucketSizeLog) - 1));
+    BYTE* const bucketOffsets = ldmState->bucketOffsets;
+    *(ZSTD_ldm_getBucket(ldmState, hash, ldmParams) + bucketOffsets[hash]) = entry;
+    bucketOffsets[hash]++;
+    bucketOffsets[hash] &= ((U32)1 << ldmParams.bucketSizeLog) - 1;
+}
 
+/** ZSTD_ldm_makeEntryAndInsertByTag() :
+ *
+ *  Gets the small hash, checksum, and tag from the rollingHash.
+ *
+ *  If the tag matches (1 << ldmParams.hashRateLog)-1, then
+ *  creates an ldmEntry from the offset, and inserts it into the hash table.
+ *
+ *  hBits is the length of the small hash, which is the most significant hBits
+ *  of rollingHash. The checksum is the next 32 most significant bits, followed
+ *  by ldmParams.hashRateLog bits that make up the tag. */
+static void ZSTD_ldm_makeEntryAndInsertByTag(ldmState_t* ldmState,
+                                             U64 const rollingHash,
+                                             U32 const hBits,
+                                             U32 const offset,
+                                             ldmParams_t const ldmParams)
+{
+    U32 const tag = ZSTD_ldm_getTag(rollingHash, hBits, ldmParams.hashRateLog);
+    U32 const tagMask = ((U32)1 << ldmParams.hashRateLog) - 1;
+    if (tag == tagMask) {
+        U32 const hash = ZSTD_ldm_getSmallHash(rollingHash, hBits);
+        U32 const checksum = ZSTD_ldm_getChecksum(rollingHash, hBits);
+        ldmEntry_t entry;
+        entry.offset = offset;
+        entry.checksum = checksum;
+        ZSTD_ldm_insertEntry(ldmState, hash, entry, ldmParams);
+    }
 }
 
 /** ZSTD_ldm_countBackwardsMatch() :
@@ -196,10 +149,10 @@ static void ZSTD_ldm_insertEntry(ldmState_t* ldmState,
  *  We count only bytes where pMatch >= pBase and pIn >= pAnchor. */
 static size_t ZSTD_ldm_countBackwardsMatch(
             const BYTE* pIn, const BYTE* pAnchor,
-            const BYTE* pMatch, const BYTE* pMatchBase)
+            const BYTE* pMatch, const BYTE* pBase)
 {
     size_t matchLength = 0;
-    while (pIn > pAnchor && pMatch > pMatchBase && pIn[-1] == pMatch[-1]) {
+    while (pIn > pAnchor && pMatch > pBase && pIn[-1] == pMatch[-1]) {
         pIn--;
         pMatch--;
         matchLength++;
@@ -207,27 +160,6 @@ static size_t ZSTD_ldm_countBackwardsMatch(
     return matchLength;
 }
 
-/** ZSTD_ldm_countBackwardsMatch_2segments() :
- *  Returns the number of bytes that match backwards from pMatch,
- *  even with the backwards match spanning 2 different segments.
- *
- *  On reaching `pMatchBase`, start counting from mEnd */
-static size_t ZSTD_ldm_countBackwardsMatch_2segments(
-                    const BYTE* pIn, const BYTE* pAnchor,
-                    const BYTE* pMatch, const BYTE* pMatchBase,
-                    const BYTE* pExtDictStart, const BYTE* pExtDictEnd)
-{
-    size_t matchLength = ZSTD_ldm_countBackwardsMatch(pIn, pAnchor, pMatch, pMatchBase);
-    if (pMatch - matchLength != pMatchBase || pMatchBase == pExtDictStart) {
-        /* If backwards match is entirely in the extDict or prefix, immediately return */
-        return matchLength;
-    }
-    DEBUGLOG(7, "ZSTD_ldm_countBackwardsMatch_2segments: found 2-parts backwards match (length in prefix==%zu)", matchLength);
-    matchLength += ZSTD_ldm_countBackwardsMatch(pIn - matchLength, pAnchor, pExtDictEnd, pExtDictStart);
-    DEBUGLOG(7, "final backwards match length = %zu", matchLength);
-    return matchLength;
-}
-
 /** ZSTD_ldm_fillFastTables() :
  *
  *  Fills the relevant tables for the ZSTD_fast and ZSTD_dfast strategies.
@@ -243,15 +175,11 @@ static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms,
     switch(ms->cParams.strategy)
     {
     case ZSTD_fast:
-        ZSTD_fillHashTable(ms, iend, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
+        ZSTD_fillHashTable(ms, iend, ZSTD_dtlm_fast);
         break;
 
     case ZSTD_dfast:
-#ifndef ZSTD_EXCLUDE_DFAST_BLOCK_COMPRESSOR
-        ZSTD_fillDoubleHashTable(ms, iend, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
-#else
-        assert(0); /* shouldn't be called: cparams should've been adjusted. */
-#endif
+        ZSTD_fillDoubleHashTable(ms, iend, ZSTD_dtlm_fast);
         break;
 
     case ZSTD_greedy:
@@ -269,43 +197,30 @@ static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms,
     return 0;
 }
 
-void ZSTD_ldm_fillHashTable(
-            ldmState_t* ldmState, const BYTE* ip,
-            const BYTE* iend, ldmParams_t const* params)
+/** ZSTD_ldm_fillLdmHashTable() :
+ *
+ *  Fills hashTable from (lastHashed + 1) to iend (non-inclusive).
+ *  lastHash is the rolling hash that corresponds to lastHashed.
+ *
+ *  Returns the rolling hash corresponding to position iend-1. */
+static U64 ZSTD_ldm_fillLdmHashTable(ldmState_t* state,
+                                     U64 lastHash, const BYTE* lastHashed,
+                                     const BYTE* iend, const BYTE* base,
+                                     U32 hBits, ldmParams_t const ldmParams)
 {
-    U32 const minMatchLength = params->minMatchLength;
-    U32 const hBits = params->hashLog - params->bucketSizeLog;
-    BYTE const* const base = ldmState->window.base;
-    BYTE const* const istart = ip;
-    ldmRollingHashState_t hashState;
-    size_t* const splits = ldmState->splitIndices;
-    unsigned numSplits;
+    U64 rollingHash = lastHash;
+    const BYTE* cur = lastHashed + 1;
 
-    DEBUGLOG(5, "ZSTD_ldm_fillHashTable");
-
-    ZSTD_ldm_gear_init(&hashState, params);
-    while (ip < iend) {
-        size_t hashed;
-        unsigned n;
-
-        numSplits = 0;
-        hashed = ZSTD_ldm_gear_feed(&hashState, ip, iend - ip, splits, &numSplits);
-
-        for (n = 0; n < numSplits; n++) {
-            if (ip + splits[n] >= istart + minMatchLength) {
-                BYTE const* const split = ip + splits[n] - minMatchLength;
-                U64 const xxhash = XXH64(split, minMatchLength, 0);
-                U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1));
-                ldmEntry_t entry;
-
-                entry.offset = (U32)(split - base);
-                entry.checksum = (U32)(xxhash >> 32);
-                ZSTD_ldm_insertEntry(ldmState, hash, entry, *params);
-            }
-        }
-
-        ip += hashed;
+    while (cur < iend) {
+        rollingHash = ZSTD_rollingHash_rotate(rollingHash, cur[-1],
+                                              cur[ldmParams.minMatchLength-1],
+                                              state->hashPower);
+        ZSTD_ldm_makeEntryAndInsertByTag(state,
+                                         rollingHash, hBits,
+                                         (U32)(cur - base), ldmParams);
+        ++cur;
     }
+    return rollingHash;
 }
 
 
@@ -316,24 +231,25 @@ void ZSTD_ldm_fillHashTable(
  *  (after a long match, only update tables a limited amount). */
 static void ZSTD_ldm_limitTableUpdate(ZSTD_matchState_t* ms, const BYTE* anchor)
 {
-    U32 const curr = (U32)(anchor - ms->window.base);
-    if (curr > ms->nextToUpdate + 1024) {
+    U32 const current = (U32)(anchor - ms->window.base);
+    if (current > ms->nextToUpdate + 1024) {
         ms->nextToUpdate =
-            curr - MIN(512, curr - ms->nextToUpdate - 1024);
+            current - MIN(512, current - ms->nextToUpdate - 1024);
     }
 }
 
-static
-ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
-size_t ZSTD_ldm_generateSequences_internal(
+static size_t ZSTD_ldm_generateSequences_internal(
         ldmState_t* ldmState, rawSeqStore_t* rawSeqStore,
         ldmParams_t const* params, void const* src, size_t srcSize)
 {
     /* LDM parameters */
     int const extDict = ZSTD_window_hasExtDict(ldmState->window);
     U32 const minMatchLength = params->minMatchLength;
-    U32 const entsPerBucket = 1U << params->bucketSizeLog;
+    U64 const hashPower = ldmState->hashPower;
     U32 const hBits = params->hashLog - params->bucketSizeLog;
+    U32 const ldmBucketSize = 1U << params->bucketSizeLog;
+    U32 const hashRateLog = params->hashRateLog;
+    U32 const ldmTagMask = (1U << params->hashRateLog) - 1;
     /* Prefix and extDict parameters */
     U32 const dictLimit = ldmState->window.dictLimit;
     U32 const lowestIndex = extDict ? ldmState->window.lowLimit : dictLimit;
@@ -345,69 +261,45 @@ size_t ZSTD_ldm_generateSequences_internal(
     /* Input bounds */
     BYTE const* const istart = (BYTE const*)src;
     BYTE const* const iend = istart + srcSize;
-    BYTE const* const ilimit = iend - HASH_READ_SIZE;
+    BYTE const* const ilimit = iend - MAX(minMatchLength, HASH_READ_SIZE);
     /* Input positions */
     BYTE const* anchor = istart;
     BYTE const* ip = istart;
-    /* Rolling hash state */
-    ldmRollingHashState_t hashState;
-    /* Arrays for staged-processing */
-    size_t* const splits = ldmState->splitIndices;
-    ldmMatchCandidate_t* const candidates = ldmState->matchCandidates;
-    unsigned numSplits;
+    /* Rolling hash */
+    BYTE const* lastHashed = NULL;
+    U64 rollingHash = 0;
 
-    if (srcSize < minMatchLength)
-        return iend - anchor;
+    while (ip <= ilimit) {
+        size_t mLength;
+        U32 const current = (U32)(ip - base);
+        size_t forwardMatchLength = 0, backwardMatchLength = 0;
+        ldmEntry_t* bestEntry = NULL;
+        if (ip != istart) {
+            rollingHash = ZSTD_rollingHash_rotate(rollingHash, lastHashed[0],
+                                                  lastHashed[minMatchLength],
+                                                  hashPower);
+        } else {
+            rollingHash = ZSTD_rollingHash_compute(ip, minMatchLength);
+        }
+        lastHashed = ip;
 
-    /* Initialize the rolling hash state with the first minMatchLength bytes */
-    ZSTD_ldm_gear_init(&hashState, params);
-    ZSTD_ldm_gear_reset(&hashState, ip, minMatchLength);
-    ip += minMatchLength;
-
-    while (ip < ilimit) {
-        size_t hashed;
-        unsigned n;
-
-        numSplits = 0;
-        hashed = ZSTD_ldm_gear_feed(&hashState, ip, ilimit - ip,
-                                    splits, &numSplits);
-
-        for (n = 0; n < numSplits; n++) {
-            BYTE const* const split = ip + splits[n] - minMatchLength;
-            U64 const xxhash = XXH64(split, minMatchLength, 0);
-            U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1));
-
-            candidates[n].split = split;
-            candidates[n].hash = hash;
-            candidates[n].checksum = (U32)(xxhash >> 32);
-            candidates[n].bucket = ZSTD_ldm_getBucket(ldmState, hash, *params);
-            PREFETCH_L1(candidates[n].bucket);
+        /* Do not insert and do not look for a match */
+        if (ZSTD_ldm_getTag(rollingHash, hBits, hashRateLog) != ldmTagMask) {
+           ip++;
+           continue;
         }
 
-        for (n = 0; n < numSplits; n++) {
-            size_t forwardMatchLength = 0, backwardMatchLength = 0,
-                   bestMatchLength = 0, mLength;
-            U32 offset;
-            BYTE const* const split = candidates[n].split;
-            U32 const checksum = candidates[n].checksum;
-            U32 const hash = candidates[n].hash;
-            ldmEntry_t* const bucket = candidates[n].bucket;
-            ldmEntry_t const* cur;
-            ldmEntry_t const* bestEntry = NULL;
-            ldmEntry_t newEntry;
+        /* Get the best entry and compute the match lengths */
+        {
+            ldmEntry_t* const bucket =
+                ZSTD_ldm_getBucket(ldmState,
+                                   ZSTD_ldm_getSmallHash(rollingHash, hBits),
+                                   *params);
+            ldmEntry_t* cur;
+            size_t bestMatchLength = 0;
+            U32 const checksum = ZSTD_ldm_getChecksum(rollingHash, hBits);
 
-            newEntry.offset = (U32)(split - base);
-            newEntry.checksum = checksum;
-
-            /* If a split point would generate a sequence overlapping with
-             * the previous one, we merely register it in the hash table and
-             * move on */
-            if (split < anchor) {
-                ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
-                continue;
-            }
-
-            for (cur = bucket; cur < bucket + entsPerBucket; cur++) {
+            for (cur = bucket; cur < bucket + ldmBucketSize; ++cur) {
                 size_t curForwardMatchLength, curBackwardMatchLength,
                        curTotalMatchLength;
                 if (cur->checksum != checksum || cur->offset <= lowestIndex) {
@@ -421,23 +313,30 @@ size_t ZSTD_ldm_generateSequences_internal(
                         cur->offset < dictLimit ? dictEnd : iend;
                     BYTE const* const lowMatchPtr =
                         cur->offset < dictLimit ? dictStart : lowPrefixPtr;
-                    curForwardMatchLength =
-                        ZSTD_count_2segments(split, pMatch, iend, matchEnd, lowPrefixPtr);
-                    if (curForwardMatchLength < minMatchLength) {
-                        continue;
-                    }
-                    curBackwardMatchLength = ZSTD_ldm_countBackwardsMatch_2segments(
-                            split, anchor, pMatch, lowMatchPtr, dictStart, dictEnd);
-                } else { /* !extDict */
-                    BYTE const* const pMatch = base + cur->offset;
-                    curForwardMatchLength = ZSTD_count(split, pMatch, iend);
+
+                    curForwardMatchLength = ZSTD_count_2segments(
+                                                ip, pMatch, iend,
+                                                matchEnd, lowPrefixPtr);
                     if (curForwardMatchLength < minMatchLength) {
                         continue;
                     }
                     curBackwardMatchLength =
-                        ZSTD_ldm_countBackwardsMatch(split, anchor, pMatch, lowPrefixPtr);
+                        ZSTD_ldm_countBackwardsMatch(ip, anchor, pMatch,
+                                                     lowMatchPtr);
+                    curTotalMatchLength = curForwardMatchLength +
+                                          curBackwardMatchLength;
+                } else { /* !extDict */
+                    BYTE const* const pMatch = base + cur->offset;
+                    curForwardMatchLength = ZSTD_count(ip, pMatch, iend);
+                    if (curForwardMatchLength < minMatchLength) {
+                        continue;
+                    }
+                    curBackwardMatchLength =
+                        ZSTD_ldm_countBackwardsMatch(ip, anchor, pMatch,
+                                                     lowPrefixPtr);
+                    curTotalMatchLength = curForwardMatchLength +
+                                          curBackwardMatchLength;
                 }
-                curTotalMatchLength = curForwardMatchLength + curBackwardMatchLength;
 
                 if (curTotalMatchLength > bestMatchLength) {
                     bestMatchLength = curTotalMatchLength;
@@ -446,54 +345,57 @@ size_t ZSTD_ldm_generateSequences_internal(
                     bestEntry = cur;
                 }
             }
-
-            /* No match found -- insert an entry into the hash table
-             * and process the next candidate match */
-            if (bestEntry == NULL) {
-                ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
-                continue;
-            }
-
-            /* Match found */
-            offset = (U32)(split - base) - bestEntry->offset;
-            mLength = forwardMatchLength + backwardMatchLength;
-            {
-                rawSeq* const seq = rawSeqStore->seq + rawSeqStore->size;
-
-                /* Out of sequence storage */
-                if (rawSeqStore->size == rawSeqStore->capacity)
-                    return ERROR(dstSize_tooSmall);
-                seq->litLength = (U32)(split - backwardMatchLength - anchor);
-                seq->matchLength = (U32)mLength;
-                seq->offset = offset;
-                rawSeqStore->size++;
-            }
-
-            /* Insert the current entry into the hash table --- it must be
-             * done after the previous block to avoid clobbering bestEntry */
-            ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
-
-            anchor = split + forwardMatchLength;
-
-            /* If we find a match that ends after the data that we've hashed
-             * then we have a repeating, overlapping, pattern. E.g. all zeros.
-             * If one repetition of the pattern matches our `stopMask` then all
-             * repetitions will. We don't need to insert them all into out table,
-             * only the first one. So skip over overlapping matches.
-             * This is a major speed boost (20x) for compressing a single byte
-             * repeated, when that byte ends up in the table.
-             */
-            if (anchor > ip + hashed) {
-                ZSTD_ldm_gear_reset(&hashState, anchor - minMatchLength, minMatchLength);
-                /* Continue the outer loop at anchor (ip + hashed == anchor). */
-                ip = anchor - hashed;
-                break;
-            }
         }
 
-        ip += hashed;
-    }
+        /* No match found -- continue searching */
+        if (bestEntry == NULL) {
+            ZSTD_ldm_makeEntryAndInsertByTag(ldmState, rollingHash,
+                                             hBits, current,
+                                             *params);
+            ip++;
+            continue;
+        }
 
+        /* Match found */
+        mLength = forwardMatchLength + backwardMatchLength;
+        ip -= backwardMatchLength;
+
+        {
+            /* Store the sequence:
+             * ip = current - backwardMatchLength
+             * The match is at (bestEntry->offset - backwardMatchLength)
+             */
+            U32 const matchIndex = bestEntry->offset;
+            U32 const offset = current - matchIndex;
+            rawSeq* const seq = rawSeqStore->seq + rawSeqStore->size;
+
+            /* Out of sequence storage */
+            if (rawSeqStore->size == rawSeqStore->capacity)
+                return ERROR(dstSize_tooSmall);
+            seq->litLength = (U32)(ip - anchor);
+            seq->matchLength = (U32)mLength;
+            seq->offset = offset;
+            rawSeqStore->size++;
+        }
+
+        /* Insert the current entry into the hash table */
+        ZSTD_ldm_makeEntryAndInsertByTag(ldmState, rollingHash, hBits,
+                                         (U32)(lastHashed - base),
+                                         *params);
+
+        assert(ip + backwardMatchLength == lastHashed);
+
+        /* Fill the hash table from lastHashed+1 to ip+mLength*/
+        /* Heuristic: don't need to fill the entire table at end of block */
+        if (ip + mLength <= ilimit) {
+            rollingHash = ZSTD_ldm_fillLdmHashTable(
+                              ldmState, rollingHash, lastHashed,
+                              ip + mLength, base, hBits, *params);
+            lastHashed = ip + mLength - 1;
+        }
+        ip += mLength;
+        anchor = ip;
+    }
     return iend - anchor;
 }
 
@@ -542,13 +444,11 @@ size_t ZSTD_ldm_generateSequences(
 
         assert(chunkStart < iend);
         /* 1. Perform overflow correction if necessary. */
-        if (ZSTD_window_needOverflowCorrection(ldmState->window, 0, maxDist, ldmState->loadedDictEnd, chunkStart, chunkEnd)) {
+        if (ZSTD_window_needOverflowCorrection(ldmState->window, chunkEnd)) {
             U32 const ldmHSize = 1U << params->hashLog;
             U32 const correction = ZSTD_window_correctOverflow(
                 &ldmState->window, /* cycleLog */ 0, maxDist, chunkStart);
             ZSTD_ldm_reduceTable(ldmState->hashTable, ldmHSize, correction);
-            /* invalidate dictionaries on overflow correction */
-            ldmState->loadedDictEnd = 0;
         }
         /* 2. We enforce the maximum offset allowed.
          *
@@ -556,15 +456,9 @@ size_t ZSTD_ldm_generateSequences(
          * the window through early invalidation.
          * TODO: * Test the chunk size.
          *       * Try invalidation after the sequence generation and test the
-         *         offset against maxDist directly.
-         *
-         * NOTE: Because of dictionaries + sequence splitting we MUST make sure
-         * that any offset used is valid at the END of the sequence, since it may
-         * be split into two sequences. This condition holds when using
-         * ZSTD_window_enforceMaxDist(), but if we move to checking offsets
-         * against maxDist directly, we'll have to carefully handle that case.
+         *         the offset against maxDist directly.
          */
-        ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, &ldmState->loadedDictEnd, NULL);
+        ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, NULL, NULL);
         /* 3. Generate the sequences for the chunk, and get newLeftoverSize. */
         newLeftoverSize = ZSTD_ldm_generateSequences_internal(
             ldmState, sequences, params, chunkStart, chunkSize);
@@ -586,9 +480,7 @@ size_t ZSTD_ldm_generateSequences(
     return 0;
 }
 
-void
-ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch)
-{
+void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch) {
     while (srcSize > 0 && rawSeqStore->pos < rawSeqStore->size) {
         rawSeq* seq = rawSeqStore->seq + rawSeqStore->pos;
         if (srcSize <= seq->litLength) {
@@ -647,32 +539,14 @@ static rawSeq maybeSplitSequence(rawSeqStore_t* rawSeqStore,
     return sequence;
 }
 
-void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) {
-    U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes);
-    while (currPos && rawSeqStore->pos < rawSeqStore->size) {
-        rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos];
-        if (currPos >= currSeq.litLength + currSeq.matchLength) {
-            currPos -= currSeq.litLength + currSeq.matchLength;
-            rawSeqStore->pos++;
-        } else {
-            rawSeqStore->posInSequence = currPos;
-            break;
-        }
-    }
-    if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) {
-        rawSeqStore->posInSequence = 0;
-    }
-}
-
 size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
     ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-    ZSTD_paramSwitch_e useRowMatchFinder,
     void const* src, size_t srcSize)
 {
     const ZSTD_compressionParameters* const cParams = &ms->cParams;
     unsigned const minMatch = cParams->minMatch;
     ZSTD_blockCompressor const blockCompressor =
-        ZSTD_selectBlockCompressor(cParams->strategy, useRowMatchFinder, ZSTD_matchState_dictMode(ms));
+        ZSTD_selectBlockCompressor(cParams->strategy, ZSTD_matchState_dictMode(ms));
     /* Input bounds */
     BYTE const* const istart = (BYTE const*)src;
     BYTE const* const iend = istart + srcSize;
@@ -680,35 +554,27 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
     BYTE const* ip = istart;
 
     DEBUGLOG(5, "ZSTD_ldm_blockCompress: srcSize=%zu", srcSize);
-    /* If using opt parser, use LDMs only as candidates rather than always accepting them */
-    if (cParams->strategy >= ZSTD_btopt) {
-        size_t lastLLSize;
-        ms->ldmSeqStore = rawSeqStore;
-        lastLLSize = blockCompressor(ms, seqStore, rep, src, srcSize);
-        ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore, srcSize);
-        return lastLLSize;
-    }
-
     assert(rawSeqStore->pos <= rawSeqStore->size);
     assert(rawSeqStore->size <= rawSeqStore->capacity);
-    /* Loop through each sequence and apply the block compressor to the literals */
+    /* Loop through each sequence and apply the block compressor to the lits */
     while (rawSeqStore->pos < rawSeqStore->size && ip < iend) {
         /* maybeSplitSequence updates rawSeqStore->pos */
         rawSeq const sequence = maybeSplitSequence(rawSeqStore,
                                                    (U32)(iend - ip), minMatch);
+        int i;
         /* End signal */
         if (sequence.offset == 0)
             break;
 
+        assert(sequence.offset <= (1U << cParams->windowLog));
         assert(ip + sequence.litLength + sequence.matchLength <= iend);
 
         /* Fill tables for block compressor */
         ZSTD_ldm_limitTableUpdate(ms, ip);
         ZSTD_ldm_fillFastTables(ms, ip);
         /* Run the block compressor */
-        DEBUGLOG(5, "pos %u : calling block compressor on segment of size %u", (unsigned)(ip-istart), sequence.litLength);
+        DEBUGLOG(5, "calling block compressor on segment of size %u", sequence.litLength);
         {
-            int i;
             size_t const newLitLength =
                 blockCompressor(ms, seqStore, rep, ip, sequence.litLength);
             ip += sequence.litLength;
@@ -718,8 +584,8 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
             rep[0] = sequence.offset;
             /* Store the sequence */
             ZSTD_storeSeq(seqStore, newLitLength, ip - newLitLength, iend,
-                          OFFSET_TO_OFFBASE(sequence.offset),
-                          sequence.matchLength);
+                          sequence.offset + ZSTD_REP_MOVE,
+                          sequence.matchLength - MINMATCH);
             ip += sequence.matchLength;
         }
     }
@@ -729,5 +595,3 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
     /* Compress the last literals */
     return blockCompressor(ms, seqStore, rep, ip, iend - ip);
 }
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_ldm.h

@@ -1,12 +1,10 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
  * LICENSE file in the root directory of this source tree) and the GPLv2 (found
  * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
  */
 
 #ifndef ZSTD_LDM_H
@@ -25,10 +23,6 @@ extern "C" {
 
 #define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_LIMIT_DEFAULT
 
-void ZSTD_ldm_fillHashTable(
-            ldmState_t* state, const BYTE* ip,
-            const BYTE* iend, ldmParams_t const* params);
-
 /**
  * ZSTD_ldm_generateSequences():
  *
@@ -67,7 +61,6 @@ size_t ZSTD_ldm_generateSequences(
  */
 size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
             ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
-            ZSTD_paramSwitch_e useRowMatchFinder,
             void const* src, size_t srcSize);
 
 /**
@@ -75,17 +68,11 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
  *
  * Skip past `srcSize` bytes worth of sequences in `rawSeqStore`.
  * Avoids emitting matches less than `minMatch` bytes.
- * Must be called for data that is not passed to ZSTD_ldm_blockCompress().
+ * Must be called for data with is not passed to ZSTD_ldm_blockCompress().
  */
 void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize,
     U32 const minMatch);
 
-/* ZSTD_ldm_skipRawSeqStoreBytes():
- * Moves forward in rawSeqStore by nbBytes, updating fields 'pos' and 'posInSequence'.
- * Not to be used in conjunction with ZSTD_ldm_skipSequences().
- * Must be called for data with is not passed to ZSTD_ldm_blockCompress().
- */
-void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes);
 
 /** ZSTD_ldm_getTableSize() :
  *  Estimate the space needed for long distance matching tables or 0 if LDM is
@@ -116,5 +103,3 @@ void ZSTD_ldm_adjustParameters(ldmParams_t* params,
 #endif
 
 #endif /* ZSTD_FAST_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_ldm_geartab.h

@@ -1,109 +0,0 @@
-#ifndef USE_EXTERNAL_ZSTD
-/*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
- * All rights reserved.
- *
- * This source code is licensed under both the BSD-style license (found in the
- * LICENSE file in the root directory of this source tree) and the GPLv2 (found
- * in the COPYING file in the root directory of this source tree).
- * You may select, at your option, one of the above-listed licenses.
- */
-
-#ifndef ZSTD_LDM_GEARTAB_H
-#define ZSTD_LDM_GEARTAB_H
-
-#include "compiler.h" /* UNUSED_ATTR */
-#include "mem.h"      /* U64 */
-
-static UNUSED_ATTR const U64 ZSTD_ldm_gearTab[256] = {
-    0xf5b8f72c5f77775c, 0x84935f266b7ac412, 0xb647ada9ca730ccc,
-    0xb065bb4b114fb1de, 0x34584e7e8c3a9fd0, 0x4e97e17c6ae26b05,
-    0x3a03d743bc99a604, 0xcecd042422c4044f, 0x76de76c58524259e,
-    0x9c8528f65badeaca, 0x86563706e2097529, 0x2902475fa375d889,
-    0xafb32a9739a5ebe6, 0xce2714da3883e639, 0x21eaf821722e69e,
-    0x37b628620b628,    0x49a8d455d88caf5,  0x8556d711e6958140,
-    0x4f7ae74fc605c1f,  0x829f0c3468bd3a20, 0x4ffdc885c625179e,
-    0x8473de048a3daf1b, 0x51008822b05646b2, 0x69d75d12b2d1cc5f,
-    0x8c9d4a19159154bc, 0xc3cc10f4abbd4003, 0xd06ddc1cecb97391,
-    0xbe48e6e7ed80302e, 0x3481db31cee03547, 0xacc3f67cdaa1d210,
-    0x65cb771d8c7f96cc, 0x8eb27177055723dd, 0xc789950d44cd94be,
-    0x934feadc3700b12b, 0x5e485f11edbdf182, 0x1e2e2a46fd64767a,
-    0x2969ca71d82efa7c, 0x9d46e9935ebbba2e, 0xe056b67e05e6822b,
-    0x94d73f55739d03a0, 0xcd7010bdb69b5a03, 0x455ef9fcd79b82f4,
-    0x869cb54a8749c161, 0x38d1a4fa6185d225, 0xb475166f94bbe9bb,
-    0xa4143548720959f1, 0x7aed4780ba6b26ba, 0xd0ce264439e02312,
-    0x84366d746078d508, 0xa8ce973c72ed17be, 0x21c323a29a430b01,
-    0x9962d617e3af80ee, 0xab0ce91d9c8cf75b, 0x530e8ee6d19a4dbc,
-    0x2ef68c0cf53f5d72, 0xc03a681640a85506, 0x496e4e9f9c310967,
-    0x78580472b59b14a0, 0x273824c23b388577, 0x66bf923ad45cb553,
-    0x47ae1a5a2492ba86, 0x35e304569e229659, 0x4765182a46870b6f,
-    0x6cbab625e9099412, 0xddac9a2e598522c1, 0x7172086e666624f2,
-    0xdf5003ca503b7837, 0x88c0c1db78563d09, 0x58d51865acfc289d,
-    0x177671aec65224f1, 0xfb79d8a241e967d7, 0x2be1e101cad9a49a,
-    0x6625682f6e29186b, 0x399553457ac06e50, 0x35dffb4c23abb74,
-    0x429db2591f54aade, 0xc52802a8037d1009, 0x6acb27381f0b25f3,
-    0xf45e2551ee4f823b, 0x8b0ea2d99580c2f7, 0x3bed519cbcb4e1e1,
-    0xff452823dbb010a,  0x9d42ed614f3dd267, 0x5b9313c06257c57b,
-    0xa114b8008b5e1442, 0xc1fe311c11c13d4b, 0x66e8763ea34c5568,
-    0x8b982af1c262f05d, 0xee8876faaa75fbb7, 0x8a62a4d0d172bb2a,
-    0xc13d94a3b7449a97, 0x6dbbba9dc15d037c, 0xc786101f1d92e0f1,
-    0xd78681a907a0b79b, 0xf61aaf2962c9abb9, 0x2cfd16fcd3cb7ad9,
-    0x868c5b6744624d21, 0x25e650899c74ddd7, 0xba042af4a7c37463,
-    0x4eb1a539465a3eca, 0xbe09dbf03b05d5ca, 0x774e5a362b5472ba,
-    0x47a1221229d183cd, 0x504b0ca18ef5a2df, 0xdffbdfbde2456eb9,
-    0x46cd2b2fbee34634, 0xf2aef8fe819d98c3, 0x357f5276d4599d61,
-    0x24a5483879c453e3, 0x88026889192b4b9,  0x28da96671782dbec,
-    0x4ef37c40588e9aaa, 0x8837b90651bc9fb3, 0xc164f741d3f0e5d6,
-    0xbc135a0a704b70ba, 0x69cd868f7622ada,  0xbc37ba89e0b9c0ab,
-    0x47c14a01323552f6, 0x4f00794bacee98bb, 0x7107de7d637a69d5,
-    0x88af793bb6f2255e, 0xf3c6466b8799b598, 0xc288c616aa7f3b59,
-    0x81ca63cf42fca3fd, 0x88d85ace36a2674b, 0xd056bd3792389e7,
-    0xe55c396c4e9dd32d, 0xbefb504571e6c0a6, 0x96ab32115e91e8cc,
-    0xbf8acb18de8f38d1, 0x66dae58801672606, 0x833b6017872317fb,
-    0xb87c16f2d1c92864, 0xdb766a74e58b669c, 0x89659f85c61417be,
-    0xc8daad856011ea0c, 0x76a4b565b6fe7eae, 0xa469d085f6237312,
-    0xaaf0365683a3e96c, 0x4dbb746f8424f7b8, 0x638755af4e4acc1,
-    0x3d7807f5bde64486, 0x17be6d8f5bbb7639, 0x903f0cd44dc35dc,
-    0x67b672eafdf1196c, 0xa676ff93ed4c82f1, 0x521d1004c5053d9d,
-    0x37ba9ad09ccc9202, 0x84e54d297aacfb51, 0xa0b4b776a143445,
-    0x820d471e20b348e,  0x1874383cb83d46dc, 0x97edeec7a1efe11c,
-    0xb330e50b1bdc42aa, 0x1dd91955ce70e032, 0xa514cdb88f2939d5,
-    0x2791233fd90db9d3, 0x7b670a4cc50f7a9b, 0x77c07d2a05c6dfa5,
-    0xe3778b6646d0a6fa, 0xb39c8eda47b56749, 0x933ed448addbef28,
-    0xaf846af6ab7d0bf4, 0xe5af208eb666e49,  0x5e6622f73534cd6a,
-    0x297daeca42ef5b6e, 0x862daef3d35539a6, 0xe68722498f8e1ea9,
-    0x981c53093dc0d572, 0xfa09b0bfbf86fbf5, 0x30b1e96166219f15,
-    0x70e7d466bdc4fb83, 0x5a66736e35f2a8e9, 0xcddb59d2b7c1baef,
-    0xd6c7d247d26d8996, 0xea4e39eac8de1ba3, 0x539c8bb19fa3aff2,
-    0x9f90e4c5fd508d8,  0xa34e5956fbaf3385, 0x2e2f8e151d3ef375,
-    0x173691e9b83faec1, 0xb85a8d56bf016379, 0x8382381267408ae3,
-    0xb90f901bbdc0096d, 0x7c6ad32933bcec65, 0x76bb5e2f2c8ad595,
-    0x390f851a6cf46d28, 0xc3e6064da1c2da72, 0xc52a0c101cfa5389,
-    0xd78eaf84a3fbc530, 0x3781b9e2288b997e, 0x73c2f6dea83d05c4,
-    0x4228e364c5b5ed7,  0x9d7a3edf0da43911, 0x8edcfeda24686756,
-    0x5e7667a7b7a9b3a1, 0x4c4f389fa143791d, 0xb08bc1023da7cddc,
-    0x7ab4be3ae529b1cc, 0x754e6132dbe74ff9, 0x71635442a839df45,
-    0x2f6fb1643fbe52de, 0x961e0a42cf7a8177, 0xf3b45d83d89ef2ea,
-    0xee3de4cf4a6e3e9b, 0xcd6848542c3295e7, 0xe4cee1664c78662f,
-    0x9947548b474c68c4, 0x25d73777a5ed8b0b, 0xc915b1d636b7fc,
-    0x21c2ba75d9b0d2da, 0x5f6b5dcf608a64a1, 0xdcf333255ff9570c,
-    0x633b922418ced4ee, 0xc136dde0b004b34a, 0x58cc83b05d4b2f5a,
-    0x5eb424dda28e42d2, 0x62df47369739cd98, 0xb4e0b42485e4ce17,
-    0x16e1f0c1f9a8d1e7, 0x8ec3916707560ebf, 0x62ba6e2df2cc9db3,
-    0xcbf9f4ff77d83a16, 0x78d9d7d07d2bbcc4, 0xef554ce1e02c41f4,
-    0x8d7581127eccf94d, 0xa9b53336cb3c8a05, 0x38c42c0bf45c4f91,
-    0x640893cdf4488863, 0x80ec34bc575ea568, 0x39f324f5b48eaa40,
-    0xe9d9ed1f8eff527f, 0x9224fc058cc5a214, 0xbaba00b04cfe7741,
-    0x309a9f120fcf52af, 0xa558f3ec65626212, 0x424bec8b7adabe2f,
-    0x41622513a6aea433, 0xb88da2d5324ca798, 0xd287733b245528a4,
-    0x9a44697e6d68aec3, 0x7b1093be2f49bb28, 0x50bbec632e3d8aad,
-    0x6cd90723e1ea8283, 0x897b9e7431b02bf3, 0x219efdcb338a7047,
-    0x3b0311f0a27c0656, 0xdb17bf91c0db96e7, 0x8cd4fd6b4e85a5b2,
-    0xfab071054ba6409d, 0x40d6fe831fa9dfd9, 0xaf358debad7d791e,
-    0xeb8d0e25a65e3e58, 0xbbcbd3df14e08580, 0xcf751f27ecdab2b,
-    0x2b4da14f2613d8f4
-};
-
-#endif /* ZSTD_LDM_GEARTAB_H */
-
-#endif /* USE_EXTERNAL_ZSTD */

vendor/github.com/DataDog/zstd/zstd_legacy.h

@@ -1,6 +1,5 @@
-#ifndef USE_EXTERNAL_ZSTD
 /*
- * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
  * All rights reserved.
  *
  * This source code is licensed under both the BSD-style license (found in the
@@ -125,20 +124,6 @@ MEM_STATIC size_t ZSTD_decompressLegacy(
                const void* dict,size_t dictSize)
 {
     U32 const version = ZSTD_isLegacy(src, compressedSize);
-    char x;
-    /* Avoid passing NULL to legacy decoding. */
-    if (dst == NULL) {
-        assert(dstCapacity == 0);
-        dst = &x;
-    }
-    if (src == NULL) {
-        assert(compressedSize == 0);
-        src = &x;
-    }
-    if (dict == NULL) {
-        assert(dictSize == 0);
-        dict = &x;
-    }
     (void)dst; (void)dstCapacity; (void)dict; (void)dictSize;  /* unused when ZSTD_LEGACY_SUPPORT >= 8 */
     switch(version)
     {
@@ -257,13 +242,6 @@ MEM_STATIC ZSTD_frameSizeInfo ZSTD_findFrameSizeInfoLegacy(const void *src, size
         frameSizeInfo.compressedSize = ERROR(srcSize_wrong);
         frameSizeInfo.decompressedBound = ZSTD_CONTENTSIZE_ERROR;
     }
-    /* In all cases, decompressedBound == nbBlocks * ZSTD_BLOCKSIZE_MAX.
-     * So we can compute nbBlocks without having to change every function.
-     */
-    if (frameSizeInfo.decompressedBound != ZSTD_CONTENTSIZE_ERROR) {
-        assert((frameSizeInfo.decompressedBound & (ZSTD_BLOCKSIZE_MAX - 1)) == 0);
-        frameSizeInfo.nbBlocks = (size_t)(frameSizeInfo.decompressedBound / ZSTD_BLOCKSIZE_MAX);
-    }
     return frameSizeInfo;
 }
 
@@ -302,12 +280,6 @@ MEM_STATIC size_t ZSTD_freeLegacyStreamContext(void* legacyContext, U32 version)
 MEM_STATIC size_t ZSTD_initLegacyStream(void** legacyContext, U32 prevVersion, U32 newVersion,
                                         const void* dict, size_t dictSize)
 {
-    char x;
-    /* Avoid passing NULL to legacy decoding. */
-    if (dict == NULL) {
-        assert(dictSize == 0);
-        dict = &x;
-    }
     DEBUGLOG(5, "ZSTD_initLegacyStream for v0.%u", newVersion);
     if (prevVersion != newVersion) ZSTD_freeLegacyStreamContext(*legacyContext, prevVersion);
     switch(newVersion)
@@ -367,16 +339,6 @@ MEM_STATIC size_t ZSTD_initLegacyStream(void** legacyContext, U32 prevVersion, U
 MEM_STATIC size_t ZSTD_decompressLegacyStream(void* legacyContext, U32 version,
                                               ZSTD_outBuffer* output, ZSTD_inBuffer* input)
 {
-    static char x;
-    /* Avoid passing NULL to legacy decoding. */
-    if (output->dst == NULL) {
-        assert(output->size == 0);
-        output->dst = &x;
-    }
-    if (input->src == NULL) {
-        assert(input->size == 0);
-        input->src = &x;
-    }
     DEBUGLOG(5, "ZSTD_decompressLegacyStream for v0.%u", version);
     switch(version)
     {
@@ -451,5 +413,3 @@ MEM_STATIC size_t ZSTD_decompressLegacyStream(void* legacyContext, U32 version,
 #endif
 
 #endif   /* ZSTD_LEGACY_H */
-
-#endif /* USE_EXTERNAL_ZSTD */