From e2f7aee1b14b5fb7bf4dd550f7d0e3b36592d234 Mon Sep 17 00:00:00 2001
From: tangyoupeng <tommy@juicedata.io>
Date: Sun, 18 Feb 2024 16:22:22 +0800
Subject: [PATCH 01/30] Hadoop: make libjfs singleton to avoid possible jnr
 weakhashmap infinity loop (#4407)

---
 .../java/io/juicefs/JuiceFileSystemImpl.java  | 111 +++++++++---------
 .../java/io/juicefs/JuiceFileSystemTest.java  |  23 ++++
 2 files changed, 80 insertions(+), 54 deletions(-)

diff --git a/sdk/java/src/main/java/io/juicefs/JuiceFileSystemImpl.java b/sdk/java/src/main/java/io/juicefs/JuiceFileSystemImpl.java
index 719946dbd135..1196126a05c2 100644
--- a/sdk/java/src/main/java/io/juicefs/JuiceFileSystemImpl.java
+++ b/sdk/java/src/main/java/io/juicefs/JuiceFileSystemImpl.java
@@ -57,12 +57,11 @@
 import java.net.*;
 import java.nio.ByteBuffer;
 import java.nio.ByteOrder;
-import java.nio.charset.Charset;
 import java.nio.file.Files;
 import java.nio.file.Paths;
 import java.nio.file.StandardCopyOption;
 import java.util.*;
-import java.util.concurrent.*;
+import java.util.concurrent.TimeUnit;
 import java.util.jar.JarFile;
 import java.util.stream.Collectors;
 import java.util.zip.GZIPInputStream;
@@ -84,7 +83,8 @@ public class JuiceFileSystemImpl extends FileSystem {
   private int minBufferSize;
   private int cacheReplica;
   private boolean fileChecksumEnabled;
-  private Libjfs lib;
+  private Libjfs lib = loadLibrary();
+
   private long handle;
   private UserGroupInformation ugi;
   private String homeDirPrefix = "/user";
@@ -347,7 +347,6 @@ public void initialize(URI uri, Configuration conf) throws IOException {
     String supergroup = getConf(conf, "supergroup", conf.get("dfs.permissions.superusergroup", "supergroup"));
     String mountpoint = getConf(conf, "mountpoint", "");
 
-    lib = loadLibrary();
     synchronized (JuiceFileSystemImpl.class) {
       if (callBack == null) {
         callBack = new LogCallBackImpl(lib);
@@ -568,7 +567,7 @@ private static void initStubLoader() {
     }
   }
 
-  public static Libjfs loadLibrary() throws IOException {
+  public static Libjfs loadLibrary() {
     initStubLoader();
 
     LibraryLoader<Libjfs> libjfsLibraryLoader = LibraryLoader.create(Libjfs.class);
@@ -607,69 +606,73 @@ public static Libjfs loadLibrary() throws IOException {
     }
     URLConnection con;
     try {
-      con = location.openConnection();
-    } catch (FileNotFoundException e) {
-      // jar may changed
-      return loadExistLib(libjfsLibraryLoader, dir, name, libFile);
-    }
-    if (location.getProtocol().equals("jar") && (con instanceof JarURLConnection)) {
-      LOG.debug("juicefs-hadoop.jar is a nested jar");
-      JarURLConnection connection = (JarURLConnection) con;
-      JarFile jfsJar = connection.getJarFile();
-      ZipEntry entry = jfsJar.getJarEntry(resource);
-      soTime = entry.getLastModifiedTime().toMillis();
-      ins = jfsJar.getInputStream(entry);
-    } else {
-      URI locationUri;
       try {
-        locationUri = location.toURI();
-      } catch (URISyntaxException e) {
+        con = location.openConnection();
+      } catch (FileNotFoundException e) {
+        // jar may changed
         return loadExistLib(libjfsLibraryLoader, dir, name, libFile);
       }
-      if (Files.isDirectory(Paths.get(locationUri))) { // for debug: sdk/java/target/classes
-        soTime = con.getLastModified();
-        ins = JuiceFileSystemImpl.class.getClassLoader().getResourceAsStream(resource);
+      if (location.getProtocol().equals("jar") && (con instanceof JarURLConnection)) {
+        LOG.debug("juicefs-hadoop.jar is a nested jar");
+        JarURLConnection connection = (JarURLConnection) con;
+        JarFile jfsJar = connection.getJarFile();
+        ZipEntry entry = jfsJar.getJarEntry(resource);
+        soTime = entry.getLastModifiedTime().toMillis();
+        ins = jfsJar.getInputStream(entry);
       } else {
-        JarFile jfsJar;
+        URI locationUri;
         try {
-          jfsJar = new JarFile(locationUri.getPath());
-        } catch (FileNotFoundException fne) {
+          locationUri = location.toURI();
+        } catch (URISyntaxException e) {
           return loadExistLib(libjfsLibraryLoader, dir, name, libFile);
         }
-        ZipEntry entry = jfsJar.getJarEntry(resource);
-        soTime = entry.getLastModifiedTime().toMillis();
-        ins = jfsJar.getInputStream(entry);
+        if (Files.isDirectory(Paths.get(locationUri))) { // for debug: sdk/java/target/classes
+          soTime = con.getLastModified();
+          ins = JuiceFileSystemImpl.class.getClassLoader().getResourceAsStream(resource);
+        } else {
+          JarFile jfsJar;
+          try {
+            jfsJar = new JarFile(locationUri.getPath());
+          } catch (FileNotFoundException fne) {
+            return loadExistLib(libjfsLibraryLoader, dir, name, libFile);
+          }
+          ZipEntry entry = jfsJar.getJarEntry(resource);
+          soTime = entry.getLastModifiedTime().toMillis();
+          ins = jfsJar.getInputStream(entry);
+        }
       }
-    }
 
-    synchronized (JuiceFileSystemImpl.class) {
-      if (!libFile.exists() || libFile.lastModified() < soTime) {
-        // try the name for current user
-        libFile = new File(dir, System.getProperty("user.name") + "-" + name);
+      synchronized (JuiceFileSystemImpl.class) {
         if (!libFile.exists() || libFile.lastModified() < soTime) {
-          InputStream reader = new GZIPInputStream(ins);
-          File tmp = File.createTempFile(name, null, dir);
-          FileOutputStream writer = new FileOutputStream(tmp);
-          byte[] buffer = new byte[128 << 10];
-          int bytesRead = 0;
-          while ((bytesRead = reader.read(buffer)) != -1) {
-            writer.write(buffer, 0, bytesRead);
-          }
-          writer.close();
-          reader.close();
-          tmp.setLastModified(soTime);
-          tmp.setReadable(true, false);
-          try {
-            File org = new File(dir, name);
-            Files.move(tmp.toPath(), org.toPath(), StandardCopyOption.ATOMIC_MOVE);
-            libFile = org;
-          } catch (Exception ade) {
-            Files.move(tmp.toPath(), libFile.toPath(), StandardCopyOption.ATOMIC_MOVE);
+          // try the name for current user
+          libFile = new File(dir, System.getProperty("user.name") + "-" + name);
+          if (!libFile.exists() || libFile.lastModified() < soTime) {
+            InputStream reader = new GZIPInputStream(ins);
+            File tmp = File.createTempFile(name, null, dir);
+            FileOutputStream writer = new FileOutputStream(tmp);
+            byte[] buffer = new byte[128 << 10];
+            int bytesRead = 0;
+            while ((bytesRead = reader.read(buffer)) != -1) {
+              writer.write(buffer, 0, bytesRead);
+            }
+            writer.close();
+            reader.close();
+            tmp.setLastModified(soTime);
+            tmp.setReadable(true, false);
+            try {
+              File org = new File(dir, name);
+              Files.move(tmp.toPath(), org.toPath(), StandardCopyOption.ATOMIC_MOVE);
+              libFile = org;
+            } catch (Exception ade) {
+              Files.move(tmp.toPath(), libFile.toPath(), StandardCopyOption.ATOMIC_MOVE);
+            }
           }
         }
       }
+      ins.close();
+    } catch (Exception e) {
+      throw new RuntimeException("Init libjfs failed", e);
     }
-    ins.close();
     return libjfsLibraryLoader.load(libFile.getAbsolutePath());
   }
 
diff --git a/sdk/java/src/test/java/io/juicefs/JuiceFileSystemTest.java b/sdk/java/src/test/java/io/juicefs/JuiceFileSystemTest.java
index 8a059661695d..102a817805a5 100644
--- a/sdk/java/src/test/java/io/juicefs/JuiceFileSystemTest.java
+++ b/sdk/java/src/test/java/io/juicefs/JuiceFileSystemTest.java
@@ -35,6 +35,9 @@
 import java.nio.ByteBuffer;
 import java.security.PrivilegedExceptionAction;
 import java.util.*;
+import java.util.concurrent.ExecutorService;
+import java.util.concurrent.Executors;
+import java.util.concurrent.TimeUnit;
 import java.util.concurrent.atomic.AtomicBoolean;
 import java.util.concurrent.atomic.AtomicLong;
 
@@ -921,4 +924,24 @@ public void testUserWithMultiGroups() throws Exception {
     tomFs.close();
     superFs.close();
   }
+
+  public void testConcurrentCreate() throws Exception {
+    int threads = 100;
+    ExecutorService pool = Executors.newFixedThreadPool(threads);
+    for (int i = 0; i < threads; i++) {
+      pool.submit(() -> {
+        JuiceFileSystem jfs = new JuiceFileSystem();
+        try {
+          jfs.initialize(URI.create("jfs://dev/"), cfg);
+          jfs.listStatus(new Path("/"));
+          jfs.close();
+        } catch (IOException e) {
+          fail("concurrent create failed");
+          System.exit(1);
+        }
+      });
+    }
+    pool.shutdown();
+    pool.awaitTermination(1, TimeUnit.MINUTES);
+  }
 }

From d46cb5b98bb931b8c9afb33b441c0696984a53b2 Mon Sep 17 00:00:00 2001
From: tangyoupeng <tommy@juicedata.io>
Date: Wed, 21 Feb 2024 14:52:14 +0800
Subject: [PATCH 02/30] Hadoop: use auto increment fs handle to avoid possible
 conflict (#4411)

---
 sdk/java/libjfs/main.go | 77 +++++++++++++++++++++--------------------
 1 file changed, 39 insertions(+), 38 deletions(-)

diff --git a/sdk/java/libjfs/main.go b/sdk/java/libjfs/main.go
index 24fe5e43cca4..a06a08ff5f32 100644
--- a/sdk/java/libjfs/main.go
+++ b/sdk/java/libjfs/main.go
@@ -65,14 +65,15 @@ var (
 	openFiles  = make(map[int]*fwrapper)
 	nextHandle = 1
 
-	fslock   sync.Mutex
-	handlers = make(map[uintptr]*wrapper)
-	activefs = make(map[string][]*wrapper)
-	logger   = utils.GetLogger("juicefs")
-	bOnce    sync.Once
-	bridges  []*Bridge
-	pOnce    sync.Once
-	pushers  []*push.Pusher
+	fslock       sync.Mutex
+	handlers           = make(map[int64]*wrapper)
+	nextFsHandle int64 = 0
+	activefs           = make(map[string][]*wrapper)
+	logger             = utils.GetLogger("juicefs")
+	bOnce        sync.Once
+	bridges      []*Bridge
+	pOnce        sync.Once
+	pushers      []*push.Pusher
 )
 
 const (
@@ -312,7 +313,7 @@ type javaConf struct {
 	PushGraphite      string  `json:"pushGraphite"`
 }
 
-func getOrCreate(name, user, group, superuser, supergroup string, f func() *fs.FileSystem) uintptr {
+func getOrCreate(name, user, group, superuser, supergroup string, f func() *fs.FileSystem) int64 {
 	fslock.Lock()
 	defer fslock.Unlock()
 	ws := activefs[name]
@@ -340,9 +341,9 @@ func getOrCreate(name, user, group, superuser, supergroup string, f func() *fs.F
 		w.ctx = meta.NewContext(uint32(os.Getpid()), w.lookupUid(user), w.lookupGids(group))
 	}
 	activefs[name] = append(ws, w)
-	h := uintptr(unsafe.Pointer(w)) & 0x7fffffff // low 32bits
-	handlers[h] = w
-	return h
+	nextFsHandle = nextFsHandle + 1
+	handlers[nextFsHandle] = w
+	return nextFsHandle
 }
 
 func push2Gateway(pushGatewayAddr, pushAuth string, pushInterVal time.Duration, registry *prometheus.Registry, commonLabels map[string]string) {
@@ -401,7 +402,7 @@ func push2Graphite(graphite string, pushInterVal time.Duration, registry *promet
 }
 
 //export jfs_init
-func jfs_init(cname, jsonConf, user, group, superuser, supergroup *C.char) uintptr {
+func jfs_init(cname, jsonConf, user, group, superuser, supergroup *C.char) int64 {
 	name := C.GoString(cname)
 	debug.SetGCPercent(50)
 	object.UserAgent = "JuiceFS-SDK " + version.Version()
@@ -576,14 +577,14 @@ func jfs_init(cname, jsonConf, user, group, superuser, supergroup *C.char) uintp
 	})
 }
 
-func F(p uintptr) *wrapper {
+func F(p int64) *wrapper {
 	fslock.Lock()
 	defer fslock.Unlock()
 	return handlers[p]
 }
 
 //export jfs_update_uid_grouping
-func jfs_update_uid_grouping(h uintptr, uidstr *C.char, grouping *C.char) {
+func jfs_update_uid_grouping(h int64, uidstr *C.char, grouping *C.char) {
 	w := F(h)
 	if w == nil {
 		return
@@ -647,7 +648,7 @@ func jfs_update_uid_grouping(h uintptr, uidstr *C.char, grouping *C.char) {
 }
 
 //export jfs_term
-func jfs_term(pid int, h uintptr) int {
+func jfs_term(pid int, h int64) int {
 	w := F(h)
 	if w == nil {
 		return 0
@@ -705,7 +706,7 @@ func jfs_term(pid int, h uintptr) int {
 }
 
 //export jfs_open
-func jfs_open(pid int, h uintptr, cpath *C.char, lenPtr uintptr, flags int) int {
+func jfs_open(pid int, h int64, cpath *C.char, lenPtr uintptr, flags int) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -728,7 +729,7 @@ func jfs_open(pid int, h uintptr, cpath *C.char, lenPtr uintptr, flags int) int
 }
 
 //export jfs_access
-func jfs_access(pid int, h uintptr, cpath *C.char, flags int) int {
+func jfs_access(pid int, h int64, cpath *C.char, flags int) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -737,7 +738,7 @@ func jfs_access(pid int, h uintptr, cpath *C.char, flags int) int {
 }
 
 //export jfs_create
-func jfs_create(pid int, h uintptr, cpath *C.char, mode uint16, umask uint16) int {
+func jfs_create(pid int, h int64, cpath *C.char, mode uint16, umask uint16) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -755,7 +756,7 @@ func jfs_create(pid int, h uintptr, cpath *C.char, mode uint16, umask uint16) in
 }
 
 //export jfs_mkdir
-func jfs_mkdir(pid int, h uintptr, cpath *C.char, mode uint16, umask uint16) int {
+func jfs_mkdir(pid int, h int64, cpath *C.char, mode uint16, umask uint16) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -769,7 +770,7 @@ func jfs_mkdir(pid int, h uintptr, cpath *C.char, mode uint16, umask uint16) int
 }
 
 //export jfs_delete
-func jfs_delete(pid int, h uintptr, cpath *C.char) int {
+func jfs_delete(pid int, h int64, cpath *C.char) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -778,7 +779,7 @@ func jfs_delete(pid int, h uintptr, cpath *C.char) int {
 }
 
 //export jfs_rmr
-func jfs_rmr(pid int, h uintptr, cpath *C.char) int {
+func jfs_rmr(pid int, h int64, cpath *C.char) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -787,7 +788,7 @@ func jfs_rmr(pid int, h uintptr, cpath *C.char) int {
 }
 
 //export jfs_rename
-func jfs_rename(pid int, h uintptr, oldpath *C.char, newpath *C.char) int {
+func jfs_rename(pid int, h int64, oldpath *C.char, newpath *C.char) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -796,7 +797,7 @@ func jfs_rename(pid int, h uintptr, oldpath *C.char, newpath *C.char) int {
 }
 
 //export jfs_truncate
-func jfs_truncate(pid int, h uintptr, path *C.char, length uint64) int {
+func jfs_truncate(pid int, h int64, path *C.char, length uint64) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -805,7 +806,7 @@ func jfs_truncate(pid int, h uintptr, path *C.char, length uint64) int {
 }
 
 //export jfs_setXattr
-func jfs_setXattr(pid int, h uintptr, path *C.char, name *C.char, value uintptr, vlen int, mode int) int {
+func jfs_setXattr(pid int, h int64, path *C.char, name *C.char, value uintptr, vlen int, mode int) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -821,7 +822,7 @@ func jfs_setXattr(pid int, h uintptr, path *C.char, name *C.char, value uintptr,
 }
 
 //export jfs_getXattr
-func jfs_getXattr(pid int, h uintptr, path *C.char, name *C.char, buf uintptr, bufsize int) int {
+func jfs_getXattr(pid int, h int64, path *C.char, name *C.char, buf uintptr, bufsize int) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -838,7 +839,7 @@ func jfs_getXattr(pid int, h uintptr, path *C.char, name *C.char, buf uintptr, b
 }
 
 //export jfs_listXattr
-func jfs_listXattr(pid int, h uintptr, path *C.char, buf uintptr, bufsize int) int {
+func jfs_listXattr(pid int, h int64, path *C.char, buf uintptr, bufsize int) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -855,7 +856,7 @@ func jfs_listXattr(pid int, h uintptr, path *C.char, buf uintptr, bufsize int) i
 }
 
 //export jfs_removeXattr
-func jfs_removeXattr(pid int, h uintptr, path *C.char, name *C.char) int {
+func jfs_removeXattr(pid int, h int64, path *C.char, name *C.char) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -864,7 +865,7 @@ func jfs_removeXattr(pid int, h uintptr, path *C.char, name *C.char) int {
 }
 
 //export jfs_readlink
-func jfs_readlink(pid int, h uintptr, link *C.char, buf uintptr, bufsize int) int {
+func jfs_readlink(pid int, h int64, link *C.char, buf uintptr, bufsize int) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -898,7 +899,7 @@ func fill_stat(w *wrapper, wb *utils.Buffer, st *fs.FileStat) int {
 }
 
 //export jfs_stat1
-func jfs_stat1(pid int, h uintptr, cpath *C.char, buf uintptr) int {
+func jfs_stat1(pid int, h int64, cpath *C.char, buf uintptr) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -911,7 +912,7 @@ func jfs_stat1(pid int, h uintptr, cpath *C.char, buf uintptr) int {
 }
 
 //export jfs_lstat1
-func jfs_lstat1(pid int, h uintptr, cpath *C.char, buf uintptr) int {
+func jfs_lstat1(pid int, h int64, cpath *C.char, buf uintptr) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -924,7 +925,7 @@ func jfs_lstat1(pid int, h uintptr, cpath *C.char, buf uintptr) int {
 }
 
 //export jfs_summary
-func jfs_summary(pid int, h uintptr, cpath *C.char, buf uintptr) int {
+func jfs_summary(pid int, h int64, cpath *C.char, buf uintptr) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -947,7 +948,7 @@ func jfs_summary(pid int, h uintptr, cpath *C.char, buf uintptr) int {
 }
 
 //export jfs_statvfs
-func jfs_statvfs(pid int, h uintptr, buf uintptr) int {
+func jfs_statvfs(pid int, h int64, buf uintptr) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -960,7 +961,7 @@ func jfs_statvfs(pid int, h uintptr, buf uintptr) int {
 }
 
 //export jfs_chmod
-func jfs_chmod(pid int, h uintptr, cpath *C.char, mode C.mode_t) int {
+func jfs_chmod(pid int, h int64, cpath *C.char, mode C.mode_t) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -974,7 +975,7 @@ func jfs_chmod(pid int, h uintptr, cpath *C.char, mode C.mode_t) int {
 }
 
 //export jfs_utime
-func jfs_utime(pid int, h uintptr, cpath *C.char, mtime, atime int64) int {
+func jfs_utime(pid int, h int64, cpath *C.char, mtime, atime int64) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -988,7 +989,7 @@ func jfs_utime(pid int, h uintptr, cpath *C.char, mtime, atime int64) int {
 }
 
 //export jfs_setOwner
-func jfs_setOwner(pid int, h uintptr, cpath *C.char, owner *C.char, group *C.char) int {
+func jfs_setOwner(pid int, h int64, cpath *C.char, owner *C.char, group *C.char) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL
@@ -1015,7 +1016,7 @@ func setOwner(w *wrapper, ctx meta.Context, path string, owner, group string) in
 }
 
 //export jfs_listdir
-func jfs_listdir(pid int, h uintptr, cpath *C.char, offset int, buf uintptr, bufsize int) int {
+func jfs_listdir(pid int, h int64, cpath *C.char, offset int, buf uintptr, bufsize int) int {
 	var ctx meta.Context
 	var f *fs.File
 	var w *wrapper
@@ -1073,7 +1074,7 @@ func toBuf(s uintptr, sz int) []byte {
 }
 
 //export jfs_concat
-func jfs_concat(pid int, h uintptr, _dst *C.char, buf uintptr, bufsize int) int {
+func jfs_concat(pid int, h int64, _dst *C.char, buf uintptr, bufsize int) int {
 	w := F(h)
 	if w == nil {
 		return EINVAL

From b2b13205bad09feccb2d9609e46bb5681f51a583 Mon Sep 17 00:00:00 2001
From: zhijian <djx@juicedata.io>
Date: Fri, 23 Feb 2024 15:39:19 +0800
Subject: [PATCH 03/30] cmd/sync: execute chmod after chown to avoid setuid
 setgid failure (#4418)

---
 pkg/sync/sync.go | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/pkg/sync/sync.go b/pkg/sync/sync.go
index 6b3c9721a3fa..de83b2458a45 100644
--- a/pkg/sync/sync.go
+++ b/pkg/sync/sync.go
@@ -214,14 +214,15 @@ func copyPerms(dst object.ObjectStorage, obj object.Object, config *Config) {
 	start := time.Now()
 	key := obj.Key()
 	fi := obj.(object.File)
+	// chmod needs to be executed after chown, because chown will change setuid setgid to be invalid.
+	if err := dst.(object.FileSystem).Chown(key, fi.Owner(), fi.Group()); err != nil {
+		logger.Warnf("Chown %s to (%s,%s): %s", key, fi.Owner(), fi.Group(), err)
+	}
 	if !fi.IsSymlink() || !config.Links {
 		if err := dst.(object.FileSystem).Chmod(key, fi.Mode()); err != nil {
 			logger.Warnf("Chmod %s to %o: %s", key, fi.Mode(), err)
 		}
 	}
-	if err := dst.(object.FileSystem).Chown(key, fi.Owner(), fi.Group()); err != nil {
-		logger.Warnf("Chown %s to (%s,%s): %s", key, fi.Owner(), fi.Group(), err)
-	}
 	logger.Debugf("Copied permissions (%s:%s:%s) for %s in %s", fi.Owner(), fi.Group(), fi.Mode(), key, time.Since(start))
 }
 

From f642c9181efd83350c8a53f7d80dd813f969cf26 Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Wed, 28 Feb 2024 13:20:56 +0800
Subject: [PATCH 04/30] vfs: fix file length after fallocate & copyfilerange
 (#4427)

Co-authored-by: Davies Liu <davies@juicedata.io>
---
 pkg/fs/fs.go          |  2 +-
 pkg/meta/base_test.go | 32 ++++++++++++++++----------------
 pkg/meta/interface.go |  4 ++--
 pkg/meta/redis.go     | 18 ++++++++++++++----
 pkg/meta/sql.go       | 18 ++++++++++++++----
 pkg/meta/tkv.go       | 18 ++++++++++++++----
 pkg/vfs/vfs.go        | 41 ++++++++++++++++++++++++++++++-----------
 7 files changed, 91 insertions(+), 42 deletions(-)

diff --git a/pkg/fs/fs.go b/pkg/fs/fs.go
index 35dbf1895f8a..3dfb1a508bf0 100644
--- a/pkg/fs/fs.go
+++ b/pkg/fs/fs.go
@@ -575,7 +575,7 @@ func (fs *FileSystem) CopyFileRange(ctx meta.Context, src string, soff uint64, d
 	if err != 0 {
 		return
 	}
-	err = fs.m.CopyFileRange(ctx, sfi.inode, soff, dfi.inode, doff, size, 0, &written)
+	err = fs.m.CopyFileRange(ctx, sfi.inode, soff, dfi.inode, doff, size, 0, &written, nil)
 	return
 }
 
diff --git a/pkg/meta/base_test.go b/pkg/meta/base_test.go
index 521985ff0a8a..b6a1ec7338f8 100644
--- a/pkg/meta/base_test.go
+++ b/pkg/meta/base_test.go
@@ -476,25 +476,25 @@ func testMetaClient(t *testing.T, m Meta) {
 	if len(slices) != 2 || slices[0].Id != 0 || slices[0].Size != 100 || slices[1].Id != sliceId || slices[1].Size != 100 {
 		t.Fatalf("slices: %v", slices)
 	}
-	if st := m.Fallocate(ctx, inode, fallocPunchHole|fallocKeepSize, 100, 50); st != 0 {
+	if st := m.Fallocate(ctx, inode, fallocPunchHole|fallocKeepSize, 100, 50, nil); st != 0 {
 		t.Fatalf("fallocate: %s", st)
 	}
-	if st := m.Fallocate(ctx, inode, fallocPunchHole|fallocCollapesRange, 100, 50); st != syscall.EINVAL {
+	if st := m.Fallocate(ctx, inode, fallocPunchHole|fallocCollapesRange, 100, 50, nil); st != syscall.EINVAL {
 		t.Fatalf("fallocate: %s", st)
 	}
-	if st := m.Fallocate(ctx, inode, fallocPunchHole|fallocInsertRange, 100, 50); st != syscall.EINVAL {
+	if st := m.Fallocate(ctx, inode, fallocPunchHole|fallocInsertRange, 100, 50, nil); st != syscall.EINVAL {
 		t.Fatalf("fallocate: %s", st)
 	}
-	if st := m.Fallocate(ctx, inode, fallocCollapesRange, 100, 50); st != syscall.ENOTSUP {
+	if st := m.Fallocate(ctx, inode, fallocCollapesRange, 100, 50, nil); st != syscall.ENOTSUP {
 		t.Fatalf("fallocate: %s", st)
 	}
-	if st := m.Fallocate(ctx, inode, fallocPunchHole, 100, 50); st != syscall.EINVAL {
+	if st := m.Fallocate(ctx, inode, fallocPunchHole, 100, 50, nil); st != syscall.EINVAL {
 		t.Fatalf("fallocate: %s", st)
 	}
-	if st := m.Fallocate(ctx, inode, fallocPunchHole|fallocKeepSize, 0, 0); st != syscall.EINVAL {
+	if st := m.Fallocate(ctx, inode, fallocPunchHole|fallocKeepSize, 0, 0, nil); st != syscall.EINVAL {
 		t.Fatalf("fallocate: %s", st)
 	}
-	if st := m.Fallocate(ctx, parent, fallocPunchHole|fallocKeepSize, 100, 50); st != syscall.EPERM {
+	if st := m.Fallocate(ctx, parent, fallocPunchHole|fallocKeepSize, 100, 50, nil); st != syscall.EPERM {
 		t.Fatalf("fallocate dir: %s", st)
 	}
 	if st := m.Read(ctx, inode, 0, &slices); st != 0 {
@@ -1413,7 +1413,7 @@ func testCopyFileRange(t *testing.T, m Meta) {
 	m.Write(ctx, iin, 3, 0, Slice{12, 63 << 20, 10 << 20, 30 << 20}, time.Now())
 	m.Write(ctx, iout, 2, 10<<20, Slice{13, 50 << 20, 10 << 20, 30 << 20}, time.Now())
 	var copied uint64
-	if st := m.CopyFileRange(ctx, iin, 150, iout, 30<<20, 200<<20, 0, &copied); st != 0 {
+	if st := m.CopyFileRange(ctx, iin, 150, iout, 30<<20, 200<<20, 0, &copied, nil); st != 0 {
 		t.Fatalf("copy file range: %s", st)
 	}
 	var expected uint64 = 200 << 20
@@ -1964,17 +1964,17 @@ func testAttrFlags(t *testing.T, m Meta) {
 	if st := m.SetAttr(ctx, fallocFile, SetAttrFlag, 0, attr); st != 0 {
 		t.Fatalf("setattr f: %s", st)
 	}
-	if st := m.Fallocate(ctx, fallocFile, fallocKeepSize, 0, 1024); st != 0 {
+	if st := m.Fallocate(ctx, fallocFile, fallocKeepSize, 0, 1024, nil); st != 0 {
 		t.Fatalf("fallocate f: %s", st)
 	}
-	if st := m.Fallocate(ctx, fallocFile, fallocKeepSize|fallocZeroRange, 0, 1024); st != syscall.EPERM {
+	if st := m.Fallocate(ctx, fallocFile, fallocKeepSize|fallocZeroRange, 0, 1024, nil); st != syscall.EPERM {
 		t.Fatalf("fallocate f: %s", st)
 	}
 	attr.Flags = FlagImmutable
 	if st := m.SetAttr(ctx, fallocFile, SetAttrFlag, 0, attr); st != 0 {
 		t.Fatalf("setattr f: %s", st)
 	}
-	if st := m.Fallocate(ctx, fallocFile, fallocKeepSize, 0, 1024); st != syscall.EPERM {
+	if st := m.Fallocate(ctx, fallocFile, fallocKeepSize, 0, 1024, nil); st != syscall.EPERM {
 		t.Fatalf("fallocate f: %s", st)
 	}
 
@@ -1985,21 +1985,21 @@ func testAttrFlags(t *testing.T, m Meta) {
 	if st := m.Create(ctx, 1, "copydstfile", 0644, 022, 0, &copydstFile, nil); st != 0 {
 		t.Fatalf("create f: %s", st)
 	}
-	if st := m.Fallocate(ctx, copysrcFile, 0, 0, 1024); st != 0 {
+	if st := m.Fallocate(ctx, copysrcFile, 0, 0, 1024, nil); st != 0 {
 		t.Fatalf("fallocate f: %s", st)
 	}
 	attr.Flags = FlagAppend
 	if st := m.SetAttr(ctx, copydstFile, SetAttrFlag, 0, attr); st != 0 {
 		t.Fatalf("setattr f: %s", st)
 	}
-	if st := m.CopyFileRange(ctx, copysrcFile, 0, copydstFile, 0, 1024, 0, nil); st != syscall.EPERM {
+	if st := m.CopyFileRange(ctx, copysrcFile, 0, copydstFile, 0, 1024, 0, nil, nil); st != syscall.EPERM {
 		t.Fatalf("copy_file_range f: %s", st)
 	}
 	attr.Flags = FlagImmutable
 	if st := m.SetAttr(ctx, copydstFile, SetAttrFlag, 0, attr); st != 0 {
 		t.Fatalf("setattr f: %s", st)
 	}
-	if st := m.CopyFileRange(ctx, copysrcFile, 0, copydstFile, 0, 1024, 0, nil); st != syscall.EPERM {
+	if st := m.CopyFileRange(ctx, copysrcFile, 0, copydstFile, 0, 1024, 0, nil, nil); st != syscall.EPERM {
 		t.Fatalf("copy_file_range f: %s", st)
 	}
 }
@@ -2182,7 +2182,7 @@ func testDirStat(t *testing.T, m Meta) {
 	checkResult(0, align4K(0), 1)
 
 	// test dir with file and fallocate
-	if st := m.Fallocate(Background, fileInode, 0, 0, 4097); st != 0 {
+	if st := m.Fallocate(Background, fileInode, 0, 0, 4097, nil); st != 0 {
 		t.Fatalf("fallocate: %s", st)
 	}
 	time.Sleep(500 * time.Millisecond)
@@ -2313,7 +2313,7 @@ func testClone(t *testing.T, m Meta) {
 	if eno := m.Mknod(Background, dir3, "file3", TypeFile, 0777, 022, 0, "", &file3, nil); eno != 0 {
 		t.Fatalf("mknod: %s", eno)
 	}
-	if eno := m.Fallocate(Background, file3, 0, 0, 67108864); eno != 0 {
+	if eno := m.Fallocate(Background, file3, 0, 0, 67108864, nil); eno != 0 {
 		t.Fatalf("fallocate: %s", eno)
 	}
 
diff --git a/pkg/meta/interface.go b/pkg/meta/interface.go
index 61dac6fca516..f31b9fb56c29 100644
--- a/pkg/meta/interface.go
+++ b/pkg/meta/interface.go
@@ -354,7 +354,7 @@ type Meta interface {
 	// Truncate changes the length for given file.
 	Truncate(ctx Context, inode Ino, flags uint8, attrlength uint64, attr *Attr, skipPermCheck bool) syscall.Errno
 	// Fallocate preallocate given space for given file.
-	Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size uint64) syscall.Errno
+	Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size uint64, length *uint64) syscall.Errno
 	// ReadLink returns the target of a symlink.
 	ReadLink(ctx Context, inode Ino, path *[]byte) syscall.Errno
 	// Symlink creates a symlink in a directory with given name.
@@ -391,7 +391,7 @@ type Meta interface {
 	// InvalidateChunkCache invalidate chunk cache
 	InvalidateChunkCache(ctx Context, inode Ino, indx uint32) syscall.Errno
 	// CopyFileRange copies part of a file to another one.
-	CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, offOut uint64, size uint64, flags uint32, copied *uint64) syscall.Errno
+	CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, offOut uint64, size uint64, flags uint32, copied, outLength *uint64) syscall.Errno
 	// GetParents returns a map of node parents (> 1 parents if hardlinked)
 	GetParents(ctx Context, inode Ino) map[Ino]int
 	// GetDirStat returns the space and inodes usage of a directory.
diff --git a/pkg/meta/redis.go b/pkg/meta/redis.go
index 34295c775362..b01ed125ce40 100644
--- a/pkg/meta/redis.go
+++ b/pkg/meta/redis.go
@@ -1032,7 +1032,7 @@ func (m *redisMeta) Truncate(ctx Context, inode Ino, flags uint8, length uint64,
 	return errno(err)
 }
 
-func (m *redisMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size uint64) syscall.Errno {
+func (m *redisMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size uint64, flength *uint64) syscall.Errno {
 	if mode&fallocCollapesRange != 0 && mode != fallocCollapesRange {
 		return syscall.EINVAL
 	}
@@ -1090,6 +1090,9 @@ func (m *redisMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, si
 		if err := m.checkQuota(ctx, newSpace, 0, m.getParents(ctx, tx, inode, t.Parent)...); err != 0 {
 			return err
 		}
+		if flength != nil {
+			*flength = length
+		}
 		t.Length = length
 		now := time.Now()
 		t.Mtime = now.Unix()
@@ -2281,7 +2284,7 @@ func (m *redisMeta) Write(ctx Context, inode Ino, indx uint32, off uint32, slice
 	return errno(err)
 }
 
-func (m *redisMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, offOut uint64, size uint64, flags uint32, copied *uint64) syscall.Errno {
+func (m *redisMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, offOut uint64, size uint64, flags uint32, copied, outLength *uint64) syscall.Errno {
 	defer m.timeit("CopyFileRange", time.Now())
 	f := m.of.find(fout)
 	if f != nil {
@@ -2306,7 +2309,9 @@ func (m *redisMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino,
 			return syscall.EINVAL
 		}
 		if offIn >= sattr.Length {
-			*copied = 0
+			if copied != nil {
+				*copied = 0
+			}
 			return nil
 		}
 		size := size
@@ -2336,6 +2341,9 @@ func (m *redisMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino,
 		attr.Mtimensec = uint32(now.Nanosecond())
 		attr.Ctime = now.Unix()
 		attr.Ctimensec = uint32(now.Nanosecond())
+		if outLength != nil {
+			*outLength = attr.Length
+		}
 
 		var vals [][]string
 		for i := offIn / ChunkSize; i <= (offIn+size)/ChunkSize; i++ {
@@ -2402,7 +2410,9 @@ func (m *redisMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino,
 			return nil
 		})
 		if err == nil {
-			*copied = size
+			if copied != nil {
+				*copied = size
+			}
 		}
 		return err
 	}, m.inodeKey(fout), m.inodeKey(fin))
diff --git a/pkg/meta/sql.go b/pkg/meta/sql.go
index 98b5e01b1977..5ce756340b96 100644
--- a/pkg/meta/sql.go
+++ b/pkg/meta/sql.go
@@ -1046,7 +1046,7 @@ func (m *dbMeta) Truncate(ctx Context, inode Ino, flags uint8, length uint64, at
 	return errno(err)
 }
 
-func (m *dbMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size uint64) syscall.Errno {
+func (m *dbMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size uint64, flength *uint64) syscall.Errno {
 	if mode&fallocCollapesRange != 0 && mode != fallocCollapesRange {
 		return syscall.EINVAL
 	}
@@ -1112,6 +1112,9 @@ func (m *dbMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size
 		if _, err := s.Cols("length", "mtime", "ctime", "mtimensec", "ctimensec").Update(&nodeAttr, &node{Inode: inode}); err != nil {
 			return err
 		}
+		if flength != nil {
+			*flength = length
+		}
 		if mode&(fallocZeroRange|fallocPunchHole) != 0 && off < old {
 			off, size := off, size
 			if off+size > old {
@@ -2289,7 +2292,7 @@ func (m *dbMeta) Write(ctx Context, inode Ino, indx uint32, off uint32, slice Sl
 	return errno(err)
 }
 
-func (m *dbMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, offOut uint64, size uint64, flags uint32, copied *uint64) syscall.Errno {
+func (m *dbMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, offOut uint64, size uint64, flags uint32, copied, outLength *uint64) syscall.Errno {
 	defer m.timeit("CopyFileRange", time.Now())
 	f := m.of.find(fout)
 	if f != nil {
@@ -2311,7 +2314,9 @@ func (m *dbMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, off
 			return syscall.EINVAL
 		}
 		if offIn >= nin.Length {
-			*copied = 0
+			if copied != nil {
+				*copied = 0
+			}
 			return nil
 		}
 		size := size
@@ -2339,6 +2344,9 @@ func (m *dbMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, off
 		nout.Ctime = now / 1e3
 		nout.Mtimensec = int16(now % 1e3)
 		nout.Ctimensec = int16(now % 1e3)
+		if outLength != nil {
+			*outLength = nout.Length
+		}
 
 		var cs []chunk
 		err = s.Where("inode = ? AND indx >= ? AND indx <= ?", fin, offIn/ChunkSize, (offIn+size)/ChunkSize).ForUpdate().Find(&cs)
@@ -2409,7 +2417,9 @@ func (m *dbMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, off
 		if _, err := s.Cols("length", "mtime", "ctime", "mtimensec", "ctimensec").Update(&nout, &node{Inode: fout}); err != nil {
 			return err
 		}
-		*copied = size
+		if copied != nil {
+			*copied = size
+		}
 		return nil
 	}, fout)
 	if err == nil {
diff --git a/pkg/meta/tkv.go b/pkg/meta/tkv.go
index c3652573f8ea..b09d9f087c83 100644
--- a/pkg/meta/tkv.go
+++ b/pkg/meta/tkv.go
@@ -965,7 +965,7 @@ func (m *kvMeta) Truncate(ctx Context, inode Ino, flags uint8, length uint64, at
 	return errno(err)
 }
 
-func (m *kvMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size uint64) syscall.Errno {
+func (m *kvMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size uint64, flength *uint64) syscall.Errno {
 	if mode&fallocCollapesRange != 0 && mode != fallocCollapesRange {
 		return syscall.EINVAL
 	}
@@ -1020,6 +1020,9 @@ func (m *kvMeta) Fallocate(ctx Context, inode Ino, mode uint8, off uint64, size
 		if err := m.checkQuota(ctx, newSpace, 0, m.getParents(tx, inode, t.Parent)...); err != 0 {
 			return err
 		}
+		if flength != nil {
+			*flength = length
+		}
 		t.Length = length
 		now := time.Now()
 		t.Mtime = now.Unix()
@@ -1970,7 +1973,7 @@ func (m *kvMeta) Write(ctx Context, inode Ino, indx uint32, off uint32, slice Sl
 	return errno(err)
 }
 
-func (m *kvMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, offOut uint64, size uint64, flags uint32, copied *uint64) syscall.Errno {
+func (m *kvMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, offOut uint64, size uint64, flags uint32, copied, outLength *uint64) syscall.Errno {
 	defer m.timeit("CopyFileRange", time.Now())
 	var newLength, newSpace int64
 	f := m.of.find(fout)
@@ -1992,7 +1995,9 @@ func (m *kvMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, off
 			return syscall.EINVAL
 		}
 		if offIn >= sattr.Length {
-			*copied = 0
+			if copied != nil {
+				*copied = 0
+			}
 			return nil
 		}
 		size := size
@@ -2022,6 +2027,9 @@ func (m *kvMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, off
 		attr.Mtimensec = uint32(now.Nanosecond())
 		attr.Ctime = now.Unix()
 		attr.Ctimensec = uint32(now.Nanosecond())
+		if outLength != nil {
+			*outLength = attr.Length
+		}
 
 		vals := make(map[string][]byte)
 		tx.scan(m.chunkKey(fin, uint32(offIn/ChunkSize)), m.chunkKey(fin, uint32((offIn+size)/ChunkSize)+1),
@@ -2084,7 +2092,9 @@ func (m *kvMeta) CopyFileRange(ctx Context, fin Ino, offIn uint64, fout Ino, off
 			}
 		}
 		tx.set(m.inodeKey(fout), m.marshal(&attr))
-		*copied = size
+		if copied != nil {
+			*copied = size
+		}
 		return nil
 	}, fout)
 	if err == nil {
diff --git a/pkg/vfs/vfs.go b/pkg/vfs/vfs.go
index 8f270816b62f..f335e89cb3a5 100644
--- a/pkg/vfs/vfs.go
+++ b/pkg/vfs/vfs.go
@@ -490,7 +490,7 @@ func (v *VFS) Truncate(ctx Context, ino Ino, size int64, fh uint64, attr *Attr)
 	if err == 0 {
 		v.writer.Truncate(ino, uint64(size))
 		v.reader.Truncate(ino, uint64(size))
-		v.invalidateLength(ino)
+		v.invalidateAttr(ino)
 	}
 	return err
 }
@@ -526,7 +526,7 @@ func (v *VFS) Release(ctx Context, ino Ino, fh uint64) {
 			f.Unlock()
 			if f.writer != nil {
 				_ = f.writer.Flush(ctx)
-				v.invalidateLength(ino)
+				v.invalidateAttr(ino)
 			}
 			if locks&1 != 0 {
 				_ = v.Meta.Flock(ctx, ino, owner, F_UNLCK, false)
@@ -664,14 +664,15 @@ func (v *VFS) Write(ctx Context, ino Ino, buf []byte, off, fh uint64) (err sysca
 
 	if err == 0 {
 		writtenSizeHistogram.Observe(float64(len(buf)))
-		v.reader.Truncate(ino, v.writer.GetLength(ino))
+		v.reader.Invalidate(ino, off, size)
+		v.invalidateAttr(ino)
 	}
 	return
 }
 
-func (v *VFS) Fallocate(ctx Context, ino Ino, mode uint8, off, length int64, fh uint64) (err syscall.Errno) {
-	defer func() { logit(ctx, "fallocate (%d,%d,%d,%d): %s", ino, mode, off, length, strerr(err)) }()
-	if off < 0 || length <= 0 {
+func (v *VFS) Fallocate(ctx Context, ino Ino, mode uint8, off, size int64, fh uint64) (err syscall.Errno) {
+	defer func() { logit(ctx, "fallocate (%d,%d,%d,%d): %s", ino, mode, off, size, strerr(err)) }()
+	if off < 0 || size <= 0 {
 		err = syscall.EINVAL
 		return
 	}
@@ -684,7 +685,7 @@ func (v *VFS) Fallocate(ctx Context, ino Ino, mode uint8, off, length int64, fh
 		err = syscall.EBADF
 		return
 	}
-	if off >= maxFileSize || off+length >= maxFileSize {
+	if off >= maxFileSize || off+size >= maxFileSize {
 		err = syscall.EFBIG
 		return
 	}
@@ -699,7 +700,23 @@ func (v *VFS) Fallocate(ctx Context, ino Ino, mode uint8, off, length int64, fh
 	defer h.Wunlock()
 	defer h.removeOp(ctx)
 
-	err = v.Meta.Fallocate(ctx, ino, mode, uint64(off), uint64(length))
+	err = v.writer.Flush(ctx, ino)
+	if err != 0 {
+		return
+	}
+	var length uint64
+	err = v.Meta.Fallocate(ctx, ino, mode, uint64(off), uint64(size), &length)
+	if err == 0 {
+		v.writer.Truncate(ino, length)
+		s := size
+		if off+size > int64(length) {
+			s = int64(length) - off
+		}
+		if s > 0 {
+			v.reader.Invalidate(ino, uint64(off), uint64(s))
+		}
+		v.invalidateAttr(ino)
+	}
 	return
 }
 
@@ -765,10 +782,12 @@ func (v *VFS) CopyFileRange(ctx Context, nodeIn Ino, fhIn, offIn uint64, nodeOut
 	if err != 0 {
 		return
 	}
-	err = v.Meta.CopyFileRange(ctx, nodeIn, offIn, nodeOut, offOut, size, flags, &copied)
+	var length uint64
+	err = v.Meta.CopyFileRange(ctx, nodeIn, offIn, nodeOut, offOut, size, flags, &copied, &length)
 	if err == 0 {
+		v.writer.Truncate(nodeOut, length)
 		v.reader.Invalidate(nodeOut, offOut, size)
-		v.invalidateLength(nodeOut)
+		v.invalidateAttr(nodeOut)
 	}
 	return
 }
@@ -1006,7 +1025,7 @@ func NewVFS(conf *Config, m meta.Meta, store chunk.ChunkStore, registerer promet
 	return v
 }
 
-func (v *VFS) invalidateLength(ino Ino) {
+func (v *VFS) invalidateAttr(ino Ino) {
 	v.modM.Lock()
 	v.modifiedAt[ino] = time.Now()
 	v.modM.Unlock()

From 6faccc9d0b0cf91a7765e56f5fa2e55b8b1a1554 Mon Sep 17 00:00:00 2001
From: Davies Liu <davies@juicedata.io>
Date: Thu, 29 Feb 2024 10:29:48 +0800
Subject: [PATCH 05/30] flush the source file of copy_file_range (#4436)

---
 pkg/vfs/vfs.go | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/pkg/vfs/vfs.go b/pkg/vfs/vfs.go
index f335e89cb3a5..ceaea18fca85 100644
--- a/pkg/vfs/vfs.go
+++ b/pkg/vfs/vfs.go
@@ -778,6 +778,10 @@ func (v *VFS) CopyFileRange(ctx Context, nodeIn Ino, fhIn, offIn uint64, nodeOut
 		defer hi.removeOp(ctx)
 	}
 
+	err = v.writer.Flush(ctx, nodeIn)
+	if err != 0 {
+		return
+	}
 	err = v.writer.Flush(ctx, nodeOut)
 	if err != 0 {
 		return

From e7f99fcacebcfdf7cf2dd49c2fed1c371c57eb29 Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Wed, 6 Mar 2024 21:11:19 +0800
Subject: [PATCH 06/30] vfs: invalidate entry cache in readdir (#4453)

---
 pkg/fuse/fuse.go    |  6 ++--
 pkg/vfs/handle.go   | 34 ++++++++++++++++++++++
 pkg/vfs/vfs.go      | 37 +++++++++++++++++++++++-
 pkg/vfs/vfs_test.go | 70 +++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 144 insertions(+), 3 deletions(-)

diff --git a/pkg/fuse/fuse.go b/pkg/fuse/fuse.go
index 10d11e9b609a..a4e25c3be2e1 100644
--- a/pkg/fuse/fuse.go
+++ b/pkg/fuse/fuse.go
@@ -354,11 +354,12 @@ func (fs *fileSystem) ReadDir(cancel <-chan struct{}, in *fuse.ReadIn, out *fuse
 	defer releaseContext(ctx)
 	entries, _, err := fs.v.Readdir(ctx, Ino(in.NodeId), in.Size, int(in.Offset), in.Fh, false)
 	var de fuse.DirEntry
-	for _, e := range entries {
+	for i, e := range entries {
 		de.Ino = uint64(e.Inode)
 		de.Name = string(e.Name)
 		de.Mode = e.Attr.SMode()
 		if !out.AddDirEntry(de) {
+			fs.v.UpdateReaddirOffset(ctx, Ino(in.NodeId), in.Fh, int(in.Offset)+i)
 			break
 		}
 	}
@@ -371,12 +372,13 @@ func (fs *fileSystem) ReadDirPlus(cancel <-chan struct{}, in *fuse.ReadIn, out *
 	entries, readAt, err := fs.v.Readdir(ctx, Ino(in.NodeId), in.Size, int(in.Offset), in.Fh, true)
 	ctx.start = readAt
 	var de fuse.DirEntry
-	for _, e := range entries {
+	for i, e := range entries {
 		de.Ino = uint64(e.Inode)
 		de.Name = string(e.Name)
 		de.Mode = e.Attr.SMode()
 		eo := out.AddDirLookupEntry(de)
 		if eo == nil {
+			fs.v.UpdateReaddirOffset(ctx, Ino(in.NodeId), in.Fh, int(in.Offset)+i)
 			break
 		}
 		if e.Attr.Full {
diff --git a/pkg/vfs/handle.go b/pkg/vfs/handle.go
index 98fe7e73461a..79aff9940c3d 100644
--- a/pkg/vfs/handle.go
+++ b/pkg/vfs/handle.go
@@ -33,6 +33,8 @@ type handle struct {
 	// for dir
 	children []*meta.Entry
 	readAt   time.Time
+	readOff  int
+	index    map[string]int
 
 	// for file
 	locks      uint8
@@ -234,3 +236,35 @@ func (v *VFS) releaseFileHandle(ino Ino, fh uint64) {
 		h.Close()
 	}
 }
+
+func (v *VFS) invalidateDirHandle(parent Ino, name string, inode Ino, attr *Attr) {
+	v.hanleM.Lock()
+	hs := v.handles[parent]
+	v.hanleM.Unlock()
+	for _, h := range hs {
+		h.Lock()
+		if h.children != nil && h.index != nil {
+			if inode > 0 {
+				h.children = append(h.children, &meta.Entry{
+					Inode: inode,
+					Name:  []byte(name),
+					Attr:  attr,
+				})
+				h.index[name] = len(h.children) - 1
+			} else {
+				i, ok := h.index[name]
+				if ok {
+					delete(h.index, name)
+					h.children[i].Inode = 0 // invalid
+					if i >= h.readOff {
+						// not read yet, remove it
+						h.children[i] = h.children[len(h.children)-1]
+						h.index[string(h.children[i].Name)] = i
+						h.children = h.children[:len(h.children)-1]
+					}
+				}
+			}
+		}
+		h.Unlock()
+	}
+}
diff --git a/pkg/vfs/vfs.go b/pkg/vfs/vfs.go
index ceaea18fca85..fabbe869ad63 100644
--- a/pkg/vfs/vfs.go
+++ b/pkg/vfs/vfs.go
@@ -172,6 +172,7 @@ func (v *VFS) Mknod(ctx Context, parent Ino, name string, mode uint16, cumask ui
 	err = v.Meta.Mknod(ctx, parent, name, _type, mode&07777, cumask, rdev, "", &inode, attr)
 	if err == 0 {
 		entry = &meta.Entry{Inode: inode, Attr: attr}
+		v.invalidateDirHandle(parent, name, inode, attr)
 	}
 	return
 }
@@ -187,6 +188,9 @@ func (v *VFS) Unlink(ctx Context, parent Ino, name string) (err syscall.Errno) {
 		return
 	}
 	err = v.Meta.Unlink(ctx, parent, name)
+	if err == 0 {
+		v.invalidateDirHandle(parent, name, 0, nil)
+	}
 	return
 }
 
@@ -208,6 +212,7 @@ func (v *VFS) Mkdir(ctx Context, parent Ino, name string, mode uint16, cumask ui
 	err = v.Meta.Mkdir(ctx, parent, name, mode, cumask, 0, &inode, attr)
 	if err == 0 {
 		entry = &meta.Entry{Inode: inode, Attr: attr}
+		v.invalidateDirHandle(parent, name, inode, attr)
 	}
 	return
 }
@@ -219,6 +224,9 @@ func (v *VFS) Rmdir(ctx Context, parent Ino, name string) (err syscall.Errno) {
 		return
 	}
 	err = v.Meta.Rmdir(ctx, parent, name)
+	if err == 0 {
+		v.invalidateDirHandle(parent, name, 0, nil)
+	}
 	return
 }
 
@@ -240,6 +248,7 @@ func (v *VFS) Symlink(ctx Context, path string, parent Ino, name string) (entry
 	err = v.Meta.Symlink(ctx, parent, name, path, &inode, attr)
 	if err == 0 {
 		entry = &meta.Entry{Inode: inode, Attr: attr}
+		v.invalidateDirHandle(parent, name, inode, attr)
 	}
 	return
 }
@@ -267,7 +276,14 @@ func (v *VFS) Rename(ctx Context, parent Ino, name string, newparent Ino, newnam
 		return
 	}
 
-	err = v.Meta.Rename(ctx, parent, name, newparent, newname, flags, nil, nil)
+	var inode Ino
+	var attr = &Attr{}
+	err = v.Meta.Rename(ctx, parent, name, newparent, newname, flags, &inode, attr)
+	if err == 0 {
+		v.invalidateDirHandle(parent, name, 0, nil)
+		v.invalidateDirHandle(newparent, newname, 0, nil)
+		v.invalidateDirHandle(newparent, newname, inode, attr)
+	}
 	return
 }
 
@@ -292,6 +308,7 @@ func (v *VFS) Link(ctx Context, ino Ino, newparent Ino, newname string) (entry *
 	err = v.Meta.Link(ctx, ino, newparent, newname, attr)
 	if err == 0 {
 		entry = &meta.Entry{Inode: ino, Attr: attr}
+		v.invalidateDirHandle(newparent, newname, ino, attr)
 	}
 	return
 }
@@ -357,14 +374,31 @@ func (v *VFS) Readdir(ctx Context, ino Ino, size uint32, off int, fh uint64, plu
 				})
 			}
 		}
+		index := make(map[string]int)
+		for i, e := range inodes {
+			index[string(e.Name)] = i
+		}
+		h.index = index
 	}
 	if off < len(h.children) {
 		entries = h.children[off:]
+		// we don't know how much of them will be sent, assume all of them
+		h.readOff = len(h.children) - 1
 	}
 	readAt = h.readAt
 	return
 }
 
+func (v *VFS) UpdateReaddirOffset(ctx Context, ino Ino, fh uint64, off int) {
+	h := v.findHandle(ino, fh)
+	if h == nil {
+		return
+	}
+	h.Lock()
+	defer h.Unlock()
+	h.readOff = off
+}
+
 func (v *VFS) Releasedir(ctx Context, ino Ino, fh uint64) int {
 	h := v.findHandle(ino, fh)
 	if h == nil {
@@ -398,6 +432,7 @@ func (v *VFS) Create(ctx Context, parent Ino, name string, mode uint16, cumask u
 		v.UpdateLength(inode, attr)
 		fh = v.newFileHandle(inode, attr.Length, flags)
 		entry = &meta.Entry{Inode: inode, Attr: attr}
+		v.invalidateDirHandle(parent, name, inode, attr)
 	}
 	return
 }
diff --git a/pkg/vfs/vfs_test.go b/pkg/vfs/vfs_test.go
index a35ae77381dd..407cacda4543 100644
--- a/pkg/vfs/vfs_test.go
+++ b/pkg/vfs/vfs_test.go
@@ -860,3 +860,73 @@ func TestInternalFile(t *testing.T) {
 		t.Fatalf("result: %s", string(resp[:n]))
 	}
 }
+
+func TestReaddirCache(t *testing.T) {
+	v, _ := createTestVFS()
+	ctx := NewLogContext(meta.Background)
+	entry, st := v.Mkdir(ctx, 1, "testdir", 0777, 022)
+	if st != 0 {
+		t.Fatalf("mkdir testdir: %s", st)
+	}
+	parent := entry.Inode
+	for i := 0; i < 100; i++ {
+		_, _ = v.Mkdir(ctx, parent, fmt.Sprintf("d%d", i), 0777, 022)
+	}
+	fh, _ := v.Opendir(ctx, parent, 0)
+	_, _ = v.Mkdir(ctx, parent, fmt.Sprintf("d%d", 100), 0777, 022)
+	defer v.Releasedir(ctx, parent, fh)
+	var off = 20
+	var files = make(map[string]bool)
+	// read first 20
+	entries, _, _ := v.Readdir(ctx, parent, 20, 0, fh, true)
+	for _, e := range entries[:off] {
+		files[string(e.Name)] = true
+	}
+	v.UpdateReaddirOffset(ctx, parent, fh, off)
+	for i := 0; i < 100; i += 10 {
+		name := fmt.Sprintf("d%d", i)
+		_ = v.Rmdir(ctx, parent, name)
+		delete(files, name)
+	}
+	for i := 100; i < 110; i++ {
+		_, _ = v.Mkdir(ctx, parent, fmt.Sprintf("d%d", i), 0777, 022)
+		_ = v.Rename(ctx, parent, fmt.Sprintf("d%d", i), parent, fmt.Sprintf("d%d", i+10), 0)
+		delete(files, fmt.Sprintf("d%d", i))
+	}
+	for {
+		entries, _, _ := v.Readdir(ctx, parent, 20, off, fh, true)
+		if len(entries) == 0 {
+			break
+		}
+		if len(entries) > 20 {
+			entries = entries[:20]
+		}
+		for _, e := range entries {
+			if e.Inode > 0 {
+				files[string(e.Name)] = true
+			} else {
+				t.Logf("invalid entry %s", e.Name)
+			}
+		}
+		off += len(entries)
+		v.UpdateReaddirOffset(ctx, parent, fh, off)
+	}
+	for i := 0; i < 100; i += 10 {
+		name := fmt.Sprintf("d%d", i)
+		if _, ok := files[name]; ok {
+			t.Fatalf("dir %s should be deleted", name)
+		}
+	}
+	for i := 100; i < 110; i++ {
+		name := fmt.Sprintf("d%d", i)
+		if _, ok := files[name]; ok {
+			t.Fatalf("dir %s should be deleted", name)
+		}
+	}
+	for i := 110; i < 120; i++ {
+		name := fmt.Sprintf("d%d", i)
+		if _, ok := files[name]; !ok {
+			t.Fatalf("dir %s should be added", name)
+		}
+	}
+}

From 64219aab898fbbe66ade80ad28e896d83dd200d3 Mon Sep 17 00:00:00 2001
From: li1213987842 <30381726+li1213987842@users.noreply.github.com>
Date: Tue, 12 Mar 2024 10:42:03 +0800
Subject: [PATCH 07/30] gateway: fix gateway head dir return 200 but expect 404
 (#4486)

---
 pkg/gateway/gateway.go | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/pkg/gateway/gateway.go b/pkg/gateway/gateway.go
index 735e94225151..7aea7e10744c 100644
--- a/pkg/gateway/gateway.go
+++ b/pkg/gateway/gateway.go
@@ -601,8 +601,15 @@ func (n *jfsObjects) GetObjectInfo(ctx context.Context, bucket, object string, o
 		err = jfsToObjectErr(ctx, eno, bucket, object)
 		return
 	}
-	// put /dir1/key1; head /dir1 return 404; head /dir1/ return 200
-	if strings.HasSuffix(object, sep) && !fi.IsDir() || !strings.HasSuffix(object, sep) && fi.IsDir() {
+	// put /dir1/key1; head /dir1 return 404; head /dir1/ return 404; head /dir1/key1 return 200
+	// put /dir1/key1/; head /dir1/key1 return 404; head /dir1/key1/ return 200
+	var isObject bool
+	if strings.HasSuffix(object, sep) && fi.IsDir() && fi.Atime() == 0 {
+		isObject = true
+	} else if !strings.HasSuffix(object, sep) && !fi.IsDir() {
+		isObject = true
+	}
+	if !isObject {
 		err = jfsToObjectErr(ctx, syscall.ENOENT, bucket, object)
 		return
 	}

From 59394e9178786af2bb5ac1d3ba7c114000a9ba12 Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Thu, 14 Mar 2024 18:06:36 +0800
Subject: [PATCH 08/30] gateway: add open flag when downloading an object
 (#4507)

---
 cmd/object.go          | 2 +-
 pkg/gateway/gateway.go | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/cmd/object.go b/cmd/object.go
index a82ba0fbeac0..9e17518eb8b8 100644
--- a/cmd/object.go
+++ b/cmd/object.go
@@ -96,7 +96,7 @@ func (f *jFile) Close() error {
 }
 
 func (j *juiceFS) Get(key string, off, limit int64) (io.ReadCloser, error) {
-	f, err := j.jfs.Open(ctx, j.path(key), 0)
+	f, err := j.jfs.Open(ctx, j.path(key), vfs.MODE_MASK_R)
 	if err != 0 {
 		return nil, err
 	}
diff --git a/pkg/gateway/gateway.go b/pkg/gateway/gateway.go
index 7aea7e10744c..caff44dc78ee 100644
--- a/pkg/gateway/gateway.go
+++ b/pkg/gateway/gateway.go
@@ -468,7 +468,7 @@ func (n *jfsObjects) GetObjectNInfo(ctx context.Context, bucket, object string,
 	if err != nil {
 		return
 	}
-	f, eno := n.fs.Open(mctx, n.path(bucket, object), 0)
+	f, eno := n.fs.Open(mctx, n.path(bucket, object), vfs.MODE_MASK_R)
 	if eno != 0 {
 		return nil, jfsToObjectErr(ctx, eno, bucket, object)
 	}

From 11fd1a28d7601f0b3bba4f30431a3f84b7c3d339 Mon Sep 17 00:00:00 2001
From: tangyoupeng <tommy@juicedata.io>
Date: Mon, 18 Mar 2024 15:43:13 +0800
Subject: [PATCH 09/30] fix guid lookupGroup (#4532)

---
 sdk/java/libjfs/guid.go | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sdk/java/libjfs/guid.go b/sdk/java/libjfs/guid.go
index db14d349879e..95f743a7756b 100644
--- a/sdk/java/libjfs/guid.go
+++ b/sdk/java/libjfs/guid.go
@@ -115,7 +115,7 @@ func (m *mapping) lookupGroup(name string) uint32 {
 	}
 	logger.Debugf("update group to %s:%d by lookup group", name, id)
 	m.updateGroup(name, id)
-	return 0
+	return id
 }
 
 func (m *mapping) lookupUserID(id uint32) string {

From ecb0765ef40803f45f6f510a4597d8c040c4aaa2 Mon Sep 17 00:00:00 2001
From: tangyoupeng <tommy@juicedata.io>
Date: Wed, 20 Mar 2024 10:12:23 +0800
Subject: [PATCH 10/30] Hadoop: rmr check perm only when is not empty (#4539)

---
 pkg/meta/utils.go | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/pkg/meta/utils.go b/pkg/meta/utils.go
index 30a0e9c20253..a69d38f6111d 100644
--- a/pkg/meta/utils.go
+++ b/pkg/meta/utils.go
@@ -256,9 +256,6 @@ func updateLocks(ls []plockRecord, nl plockRecord) []plockRecord {
 }
 
 func (m *baseMeta) emptyDir(ctx Context, inode Ino, skipCheckTrash bool, count *uint64, concurrent chan int) syscall.Errno {
-	if st := m.Access(ctx, inode, MODE_MASK_W|MODE_MASK_X, nil); st != 0 {
-		return st
-	}
 	for {
 		var entries []*Entry
 		if st := m.en.doReaddir(ctx, inode, 0, &entries, 10000); st != 0 && st != syscall.ENOENT {
@@ -267,6 +264,9 @@ func (m *baseMeta) emptyDir(ctx Context, inode Ino, skipCheckTrash bool, count *
 		if len(entries) == 0 {
 			return 0
 		}
+		if st := m.Access(ctx, inode, MODE_MASK_W|MODE_MASK_X, nil); st != 0 {
+			return st
+		}
 		var wg sync.WaitGroup
 		var status syscall.Errno
 		// try directories first to increase parallel

From a4a58f12a2d2ce8e267bb117b931e139c0a97dc9 Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Wed, 27 Mar 2024 17:10:37 +0800
Subject: [PATCH 11/30] meta: fix negative values in quota before used in
 statfs (#4591)

---
 pkg/meta/base.go      | 19 +++++++------------
 pkg/meta/base_test.go | 21 +++++++++++++++++++++
 pkg/meta/interface.go | 16 ++++++++++++++++
 3 files changed, 44 insertions(+), 12 deletions(-)

diff --git a/pkg/meta/base.go b/pkg/meta/base.go
index a8e582079f44..686d0fe96128 100644
--- a/pkg/meta/base.go
+++ b/pkg/meta/base.go
@@ -986,25 +986,20 @@ func (m *baseMeta) StatFS(ctx Context, ino Ino, totalspace, availspace, iused, i
 		if q == nil {
 			continue
 		}
+		q.sanitize()
 		if usage == nil {
 			usage = q
 		}
 		if q.MaxSpace > 0 {
-			ls := q.MaxSpace - q.UsedSpace
-			if ls < 0 {
-				ls = 0
-			}
-			if uint64(ls) < *availspace {
-				*availspace = uint64(ls)
+			ls := uint64(q.MaxSpace - q.UsedSpace)
+			if ls < *availspace {
+				*availspace = ls
 			}
 		}
 		if q.MaxInodes > 0 {
-			li := q.MaxInodes - q.UsedInodes
-			if li < 0 {
-				li = 0
-			}
-			if uint64(li) < *iavail {
-				*iavail = uint64(li)
+			li := uint64(q.MaxInodes - q.UsedInodes)
+			if li < *iavail {
+				*iavail = li
 			}
 		}
 	}
diff --git a/pkg/meta/base_test.go b/pkg/meta/base_test.go
index b6a1ec7338f8..6df45774ef01 100644
--- a/pkg/meta/base_test.go
+++ b/pkg/meta/base_test.go
@@ -659,6 +659,27 @@ func testMetaClient(t *testing.T, m Meta) {
 		t.Fatalf("total space %d, iavail %d", totalspace, iavail)
 	}
 
+	base.loadQuotas()
+	base.quotaMu.RLock()
+	q := base.dirQuotas[subIno]
+	base.quotaMu.RUnlock()
+	q.update(4<<10, 15) // used > max
+	base.doFlushQuotas()
+	if st := m.StatFS(ctx, subIno, &totalspace, &availspace, &iused, &iavail); st != 0 {
+		t.Fatalf("statfs: %s", st)
+	}
+	if totalspace != 4<<10 || availspace != 0 || iused != 15 || iavail != 0 {
+		t.Fatalf("total space %d, availspace %d, iused %d, iavail %d", totalspace, availspace, iused, iavail)
+	}
+	q.update(-8<<10, -20) // used < 0
+	base.doFlushQuotas()
+	if st := m.StatFS(ctx, subIno, &totalspace, &availspace, &iused, &iavail); st != 0 {
+		t.Fatalf("statfs: %s", st)
+	}
+	if totalspace != 1<<10 || availspace != 1<<10 || iused != 0 || iavail != 10 {
+		t.Fatalf("total space %d, availspace %d, iused %d, iavail %d", totalspace, availspace, iused, iavail)
+	}
+
 	if st := m.Rmdir(ctx, 1, "subdir"); st != 0 {
 		t.Fatalf("rmdir subdir: %s", st)
 	}
diff --git a/pkg/meta/interface.go b/pkg/meta/interface.go
index f31b9fb56c29..50febddccaa4 100644
--- a/pkg/meta/interface.go
+++ b/pkg/meta/interface.go
@@ -304,6 +304,22 @@ func (q *Quota) update(space, inodes int64) {
 	atomic.AddInt64(&q.newInodes, inodes)
 }
 
+// not thread safe
+func (q *Quota) sanitize() {
+	if q.UsedSpace < 0 {
+		q.UsedSpace = 0
+	}
+	if q.MaxSpace > 0 && q.MaxSpace < q.UsedSpace {
+		q.MaxSpace = q.UsedSpace
+	}
+	if q.UsedInodes < 0 {
+		q.UsedInodes = 0
+	}
+	if q.MaxInodes > 0 && q.MaxInodes < q.UsedInodes {
+		q.MaxInodes = q.UsedInodes
+	}
+}
+
 // Meta is a interface for a meta service for file system.
 type Meta interface {
 	// Name of database

From c42509bc936045ddbd5d3f163ee4ca7a098fc194 Mon Sep 17 00:00:00 2001
From: jiefenghuang <jiefeng@juicedata.io>
Date: Mon, 1 Apr 2024 21:42:14 +0800
Subject: [PATCH 12/30] enhance: check for link (#4625)

Signed-off-by: jiefenghuang <jiefeng@juicedata.io>
---
 pkg/meta/base.go  | 3 +++
 pkg/meta/redis.go | 2 +-
 pkg/meta/sql.go   | 2 +-
 pkg/meta/tkv.go   | 2 +-
 4 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/pkg/meta/base.go b/pkg/meta/base.go
index 686d0fe96128..c43396e1d26e 100644
--- a/pkg/meta/base.go
+++ b/pkg/meta/base.go
@@ -1397,6 +1397,9 @@ func (m *baseMeta) Link(ctx Context, inode, parent Ino, name string, attr *Attr)
 	if name == "" {
 		return syscall.ENOENT
 	}
+	if name == "." || name == ".." {
+		return syscall.EEXIST
+	}
 
 	defer m.timeit("Link", time.Now())
 	if attr == nil {
diff --git a/pkg/meta/redis.go b/pkg/meta/redis.go
index b01ed125ce40..d5cac93817cb 100644
--- a/pkg/meta/redis.go
+++ b/pkg/meta/redis.go
@@ -1896,7 +1896,7 @@ func (m *redisMeta) doLink(ctx Context, inode, parent Ino, name string, attr *At
 		if pattr.Parent > TrashInode {
 			return syscall.ENOENT
 		}
-		if st := m.Access(ctx, parent, MODE_MASK_W, &pattr); st != 0 {
+		if st := m.Access(ctx, parent, MODE_MASK_W|MODE_MASK_X, &pattr); st != 0 {
 			return st
 		}
 		if pattr.Flags&FlagImmutable != 0 {
diff --git a/pkg/meta/sql.go b/pkg/meta/sql.go
index 5ce756340b96..0a760112c2f2 100644
--- a/pkg/meta/sql.go
+++ b/pkg/meta/sql.go
@@ -1954,7 +1954,7 @@ func (m *dbMeta) doLink(ctx Context, inode, parent Ino, name string, attr *Attr)
 		}
 		var pattr Attr
 		m.parseAttr(&pn, &pattr)
-		if st := m.Access(ctx, parent, MODE_MASK_W, &pattr); st != 0 {
+		if st := m.Access(ctx, parent, MODE_MASK_W|MODE_MASK_X, &pattr); st != 0 {
 			return st
 		}
 		if pn.Flags&FlagImmutable != 0 {
diff --git a/pkg/meta/tkv.go b/pkg/meta/tkv.go
index b09d9f087c83..7e917730de09 100644
--- a/pkg/meta/tkv.go
+++ b/pkg/meta/tkv.go
@@ -1729,7 +1729,7 @@ func (m *kvMeta) doLink(ctx Context, inode, parent Ino, name string, attr *Attr)
 		if pattr.Parent > TrashInode {
 			return syscall.ENOENT
 		}
-		if st := m.Access(ctx, parent, MODE_MASK_W, &pattr); st != 0 {
+		if st := m.Access(ctx, parent, MODE_MASK_W|MODE_MASK_X, &pattr); st != 0 {
 			return st
 		}
 		if pattr.Flags&FlagImmutable != 0 {

From ef24d00640d545cb97dbdfbff5f806f84d407e2b Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Mon, 1 Apr 2024 22:21:46 +0800
Subject: [PATCH 13/30] meta/sql: fix nanosecond part in Write (#4618)

---
 pkg/meta/sql.go | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pkg/meta/sql.go b/pkg/meta/sql.go
index 0a760112c2f2..cac46df19cc2 100644
--- a/pkg/meta/sql.go
+++ b/pkg/meta/sql.go
@@ -2253,10 +2253,10 @@ func (m *dbMeta) Write(ctx Context, inode Ino, indx uint32, off uint32, slice Sl
 			return err
 		}
 		nodeAttr.Mtime = mtime.UnixNano() / 1e3
-		nodeAttr.Mtimensec = int16(mtime.Nanosecond())
+		nodeAttr.Mtimensec = int16(mtime.Nanosecond() % 1e3)
 		ctime := time.Now()
 		nodeAttr.Ctime = ctime.UnixNano() / 1e3
-		nodeAttr.Ctimensec = int16(ctime.Nanosecond())
+		nodeAttr.Ctimensec = int16(ctime.Nanosecond() % 1e3)
 
 		var ck = chunk{Inode: inode, Indx: indx}
 		ok, err = s.ForUpdate().MustCols("indx").Get(&ck)

From 87ba05de7582b0cf6af183466a1bc9f479e344b8 Mon Sep 17 00:00:00 2001
From: Davies Liu <davies@juicedata.io>
Date: Wed, 3 Apr 2024 15:48:08 +0800
Subject: [PATCH 14/30] pkg/object: fix storage class of BOS (#4636)

Co-authored-by: zhijian <djx@juicedata.io>
---
 pkg/object/bos.go                 |  4 +++
 pkg/object/object_storage_test.go | 50 ++++++++++---------------------
 2 files changed, 20 insertions(+), 34 deletions(-)

diff --git a/pkg/object/bos.go b/pkg/object/bos.go
index 2c1a0b4bce87..6070d7fbff88 100644
--- a/pkg/object/bos.go
+++ b/pkg/object/bos.go
@@ -57,6 +57,10 @@ func (q *bosclient) Limits() Limits {
 	}
 }
 
+func (q *bosclient) SetStorageClass(sc string) {
+	q.sc = sc
+}
+
 func (q *bosclient) Create() error {
 	_, err := q.c.PutBucket(q.bucket)
 	if err == nil && q.sc != "" {
diff --git a/pkg/object/object_storage_test.go b/pkg/object/object_storage_test.go
index b4522ddad59e..e16d4ec71d63 100644
--- a/pkg/object/object_storage_test.go
+++ b/pkg/object/object_storage_test.go
@@ -73,40 +73,22 @@ func listAll(s ObjectStorage, prefix, marker string, limit int64, followLink boo
 }
 
 func setStorageClass(o ObjectStorage) string {
-	switch s := o.(type) {
-	case *wasb:
-		s.sc = string(blob2.AccessTierCool)
-		return s.sc
-	case *bosclient:
-		s.sc = "STANDARD_IA"
-		return s.sc
-	case *COS:
-		s.sc = "STANDARD_IA"
-		return s.sc
-	case *ks3:
-		s.sc = "STANDARD_IA"
-		return s.sc
-	case *gs:
-		s.sc = "NEARLINE"
-		return s.sc
-	case *obsClient:
-		s.sc = "STANDARD_IA"
-		return s.sc
-	case *ossClient:
-		s.sc = string(oss.StorageIA)
-		return s.sc
-	case *qingstor:
-		s.sc = "STANDARD_IA"
-		return s.sc
-	case *s3client:
-		s.sc = "STANDARD_IA"
-		return s.sc
-	case *tosClient:
-		s.sc = string(enum.StorageClassIa)
-		return s.sc
-	default:
-		return ""
-	}
+	if osc, ok := o.(SupportStorageClass); ok {
+		var sc = "STANDARD_IA"
+		switch o.(type) {
+		case *wasb:
+			sc = string(blob2.AccessTierCool)
+		case *gs:
+			sc = "NEARLINE"
+		case *ossClient:
+			sc = string(oss.StorageIA)
+		case *tosClient:
+			sc = string(enum.StorageClassIa)
+		}
+		osc.SetStorageClass(sc)
+		return sc
+	}
+	return ""
 }
 
 // nolint:errcheck

From f1e6577ab3ddc3b9c71f0e4925161a9ccce59e3b Mon Sep 17 00:00:00 2001
From: zhijian <djx@juicedata.io>
Date: Mon, 15 Apr 2024 11:32:29 +0800
Subject: [PATCH 15/30] =?UTF-8?q?gateway=EF=BC=9Afix=20the=20way=20bucket?=
 =?UTF-8?q?=20paths=20are=20calculated=20in=20multi-bucket=20mode=20(#4687?=
 =?UTF-8?q?)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 pkg/gateway/gateway.go | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pkg/gateway/gateway.go b/pkg/gateway/gateway.go
index caff44dc78ee..ce97e89c06f8 100644
--- a/pkg/gateway/gateway.go
+++ b/pkg/gateway/gateway.go
@@ -157,7 +157,7 @@ func (n *jfsObjects) isValidBucketName(bucket string) error {
 }
 
 func (n *jfsObjects) path(p ...string) string {
-	if len(p) > 0 && p[0] == n.conf.Format.Name {
+	if !n.gConf.MultiBucket && len(p) > 0 && p[0] == n.conf.Format.Name {
 		p = p[1:]
 	}
 	return sep + minio.PathJoin(p...)

From 89986f8a90704227fbc6010eadd93c3e31414972 Mon Sep 17 00:00:00 2001
From: Changxin Miao <miaochangxin@step.ai>
Date: Wed, 17 Apr 2024 16:53:32 +0800
Subject: [PATCH 16/30] Fix panic when reading and writing s.writer
 concurrently (#4710)

Signed-off-by: Changxin Miao <miaochangxin@step.ai>
Co-authored-by: Davies Liu <davies@juicedata.io>
---
 pkg/vfs/writer.go | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pkg/vfs/writer.go b/pkg/vfs/writer.go
index aaedc00ffbb3..2ab7331cad0c 100644
--- a/pkg/vfs/writer.go
+++ b/pkg/vfs/writer.go
@@ -119,7 +119,6 @@ func (s *sliceWriter) flushData() {
 		s.writer.Abort()
 		s.err = syscall.EIO
 	}
-	s.writer = nil
 }
 
 // protected by s.chunk.file

From 4e3639b18fcd84b764925f8b21ab6b0f57ea6ef8 Mon Sep 17 00:00:00 2001
From: zhijian <djx@juicedata.io>
Date: Thu, 18 Apr 2024 18:45:36 +0800
Subject: [PATCH 17/30] postgres: fix postgres address parse (#4731)

---
 pkg/object/object_storage_test.go | 2 +-
 pkg/object/sql.go                 | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/pkg/object/object_storage_test.go b/pkg/object/object_storage_test.go
index e16d4ec71d63..047b72247266 100644
--- a/pkg/object/object_storage_test.go
+++ b/pkg/object/object_storage_test.go
@@ -911,7 +911,7 @@ func TestPG(t *testing.T) { //skip mutate
 
 }
 func TestPGWithSearchPath(t *testing.T) { //skip mutate
-	_, err := newSQLStore("postgres", "localhost:5432/test?sslmode=disable&search_path=juicefs,public", "", "")
+	_, err := newSQLStore("postgres", "127.0.0.1:5432/test?sslmode=disable&search_path=juicefs,public", "", "")
 	if !strings.Contains(err.Error(), "currently, only one schema is supported in search_path") {
 		t.Fatalf("TestPGWithSearchPath error: %s", err)
 	}
diff --git a/pkg/object/sql.go b/pkg/object/sql.go
index 21b7b1df1a3d..622f358eeeab 100644
--- a/pkg/object/sql.go
+++ b/pkg/object/sql.go
@@ -168,9 +168,9 @@ func newSQLStore(driver, addr, user, password string) (ObjectStorage, error) {
 		uri = "postgres://" + uri
 		driver = "pgx"
 
-		parse, err := url.Parse(addr)
+		parse, err := url.Parse(uri)
 		if err != nil {
-			return nil, fmt.Errorf("parse url %s failed: %s", addr, err)
+			return nil, fmt.Errorf("parse url %s failed: %s", uri, err)
 		}
 		searchPath = parse.Query().Get("search_path")
 		if searchPath != "" {

From f902cc6ec6df1d0824b51a6366aa3dac8a2abeef Mon Sep 17 00:00:00 2001
From: Changxin Miao <miaochangxin@step.ai>
Date: Mon, 22 Apr 2024 16:09:59 +0800
Subject: [PATCH 18/30] meta: Fix error message in quota always shows inode 0
 (#4741)

Signed-off-by: Changxin Miao <miaochangxin@step.ai>
---
 pkg/meta/base.go | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/pkg/meta/base.go b/pkg/meta/base.go
index c43396e1d26e..1a8d9702566b 100644
--- a/pkg/meta/base.go
+++ b/pkg/meta/base.go
@@ -708,8 +708,9 @@ func (m *baseMeta) getQuotaParent(ctx Context, inode Ino) Ino {
 		if inode <= RootInode {
 			break
 		}
+		lastInode := inode
 		if inode, st = m.getDirParent(ctx, inode); st != 0 {
-			logger.Warnf("Get directory parent of inode %d: %s", inode, st)
+			logger.Warnf("Get directory parent of inode %d: %s", lastInode, st)
 			break
 		}
 	}
@@ -732,8 +733,9 @@ func (m *baseMeta) checkDirQuota(ctx Context, inode Ino, space, inodes int64) bo
 		if inode <= RootInode {
 			break
 		}
+		lastInode := inode
 		if inode, st = m.getDirParent(ctx, inode); st != 0 {
-			logger.Warnf("Get directory parent of inode %d: %s", inode, st)
+			logger.Warnf("Get directory parent of inode %d: %s", lastInode, st)
 			break
 		}
 	}
@@ -756,8 +758,9 @@ func (m *baseMeta) updateDirQuota(ctx Context, inode Ino, space, inodes int64) {
 		if inode <= RootInode {
 			break
 		}
+		lastInode := inode
 		if inode, st = m.getDirParent(ctx, inode); st != 0 {
-			logger.Warnf("Get directory parent of inode %d: %s", inode, st)
+			logger.Warnf("Get directory parent of inode %d: %s", lastInode, st)
 			break
 		}
 	}

From 790db4184802d93393abb9fd0ffee9ec8c1d9269 Mon Sep 17 00:00:00 2001
From: zhijian <djx@juicedata.io>
Date: Mon, 22 Apr 2024 16:12:47 +0800
Subject: [PATCH 19/30] cmd/clone: set the attr should be done after checking
 permissions (#4745)

---
 pkg/meta/sql.go | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/pkg/meta/sql.go b/pkg/meta/sql.go
index cac46df19cc2..e9594ba89e3d 100644
--- a/pkg/meta/sql.go
+++ b/pkg/meta/sql.go
@@ -3907,6 +3907,12 @@ func (m *dbMeta) doCloneEntry(ctx Context, srcIno Ino, parent Ino, name string,
 		n.Inode = ino
 		n.Parent = parent
 		now := time.Now()
+
+		m.parseAttr(&n, attr)
+		if eno := m.Access(ctx, srcIno, MODE_MASK_R, attr); eno != 0 {
+			return eno
+		}
+
 		if cmode&CLONE_MODE_PRESERVE_ATTR == 0 {
 			n.Uid = ctx.Uid()
 			n.Gid = ctx.Gid()
@@ -3922,10 +3928,6 @@ func (m *dbMeta) doCloneEntry(ctx Context, srcIno Ino, parent Ino, name string,
 		if n.Type == TypeFile && n.Nlink > 1 {
 			n.Nlink = 1
 		}
-		m.parseAttr(&n, attr)
-		if eno := m.Access(ctx, srcIno, MODE_MASK_R, attr); eno != 0 {
-			return eno
-		}
 
 		if top {
 			var pattr Attr

From ab138028e36038171da6129b378fe3240c5a104a Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Wed, 1 May 2024 11:48:56 +0800
Subject: [PATCH 20/30] meta/redis: fix some transactions that are not wrapped
 by multi/exec (#4801)

---
 pkg/meta/redis.go | 35 +++++++++++++++++++++++++++--------
 1 file changed, 27 insertions(+), 8 deletions(-)

diff --git a/pkg/meta/redis.go b/pkg/meta/redis.go
index d5cac93817cb..ac66f373fe97 100644
--- a/pkg/meta/redis.go
+++ b/pkg/meta/redis.go
@@ -389,10 +389,11 @@ func (m *redisMeta) incrCounter(name string, value int64) (int64, error) {
 
 func (m *redisMeta) setIfSmall(name string, value, diff int64) (bool, error) {
 	var changed bool
+	ctx := Background
 	name = m.prefix + name
-	err := m.txn(Background, func(tx *redis.Tx) error {
+	err := m.txn(ctx, func(tx *redis.Tx) error {
 		changed = false
-		old, err := tx.Get(Background, name).Int64()
+		old, err := tx.Get(ctx, name).Int64()
 		if err != nil && err != redis.Nil {
 			return err
 		}
@@ -400,7 +401,11 @@ func (m *redisMeta) setIfSmall(name string, value, diff int64) (bool, error) {
 			return nil
 		} else {
 			changed = true
-			return tx.Set(Background, name, value, 0).Err()
+			_, err = tx.TxPipelined(ctx, func(pipe redis.Pipeliner) error {
+				pipe.Set(ctx, name, value, 0)
+				return nil
+			})
+			return err
 		}
 	}, name)
 
@@ -1184,7 +1189,11 @@ func (m *redisMeta) doReadlink(ctx Context, inode Ino, noatime bool) (atime int6
 		}
 		attr.Atime = now.Unix()
 		attr.Atimensec = uint32(now.Nanosecond())
-		return tx.Set(ctx, m.inodeKey(inode), m.marshal(attr), 0).Err()
+		_, e = tx.TxPipelined(ctx, func(pipe redis.Pipeliner) error {
+			pipe.Set(ctx, m.inodeKey(inode), m.marshal(attr), 0)
+			return nil
+		})
+		return e
 	}, m.inodeKey(inode))
 	atime = attr.Atime
 	return
@@ -3386,7 +3395,11 @@ func (m *redisMeta) doRepair(ctx Context, inode Ino, attr *Attr) syscall.Errno {
 				attr.Nlink++
 			}
 		}
-		return tx.Set(ctx, m.inodeKey(inode), m.marshal(attr), 0).Err()
+		_, err = tx.TxPipelined(ctx, func(pipe redis.Pipeliner) error {
+			pipe.Set(ctx, m.inodeKey(inode), m.marshal(attr), 0)
+			return nil
+		})
+		return err
 	}, m.inodeKey(inode), m.entryKey(inode)))
 }
 
@@ -3435,10 +3448,13 @@ func (m *redisMeta) doSetXattr(ctx Context, inode Ino, name string, value []byte
 			} else if !ok {
 				return ENOATTR
 			}
-			_, err := m.rdb.HSet(ctx, key, name, value).Result()
+			_, err := tx.TxPipelined(ctx, func(pipe redis.Pipeliner) error {
+				pipe.HSet(ctx, key, name, value)
+				return nil
+			})
 			return err
 		default: // XattrCreateOrReplace
-			_, err := m.rdb.HSet(ctx, key, name, value).Result()
+			_, err := tx.HSet(ctx, key, name, value).Result()
 			return err
 		}
 	}, key))
@@ -4413,7 +4429,10 @@ func (m *redisMeta) doTouchAtime(ctx Context, inode Ino, attr *Attr, now time.Ti
 		}
 		attr.Atime = now.Unix()
 		attr.Atimensec = uint32(now.Nanosecond())
-		if err = tx.Set(ctx, m.inodeKey(inode), m.marshal(attr), 0).Err(); err == nil {
+		if _, err = tx.TxPipelined(ctx, func(pipe redis.Pipeliner) error {
+			pipe.Set(ctx, m.inodeKey(inode), m.marshal(attr), 0)
+			return nil
+		}); err == nil {
 			updated = true
 		}
 		return err

From c37e3d4cb885bf9820301c9b3f44d8981a3d1162 Mon Sep 17 00:00:00 2001
From: Davies Liu <davies@juicedata.io>
Date: Mon, 13 May 2024 15:38:19 +0800
Subject: [PATCH 21/30] sync: fix concurrent stats update in cluster mode
 (#4843)

---
 pkg/sync/cluster.go | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/pkg/sync/cluster.go b/pkg/sync/cluster.go
index 656e820f4d41..e0e60d509d57 100644
--- a/pkg/sync/cluster.go
+++ b/pkg/sync/cluster.go
@@ -83,7 +83,11 @@ func httpRequest(url string, body []byte) (ans []byte, err error) {
 	return io.ReadAll(resp.Body)
 }
 
+var sendStatMu sync.Mutex
+
 func sendStats(addr string) {
+	sendStatMu.Lock()
+	defer sendStatMu.Unlock()
 	var r Stat
 	r.Skipped = skipped.Current()
 	r.Copied = copied.Current()

From 159a402992d08d274f15b49924d8768d219709da Mon Sep 17 00:00:00 2001
From: zhijian <djx@juicedata.io>
Date: Wed, 15 May 2024 19:50:08 +0800
Subject: [PATCH 22/30] cmd/clone: fix src mode check (#4856)

---
 pkg/meta/redis.go | 9 +++++----
 pkg/meta/tkv.go   | 9 +++++----
 2 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/pkg/meta/redis.go b/pkg/meta/redis.go
index ac66f373fe97..2088e0eba40f 100644
--- a/pkg/meta/redis.go
+++ b/pkg/meta/redis.go
@@ -4212,14 +4212,15 @@ func (m *redisMeta) LoadMeta(r io.Reader) (err error) {
 	return err
 }
 
-func (m *redisMeta) doCloneEntry(ctx Context, srcIno Ino, parent Ino, name string, ino Ino, attr *Attr, cmode uint8, cumask uint16, top bool) syscall.Errno {
+func (m *redisMeta) doCloneEntry(ctx Context, srcIno Ino, parent Ino, name string, ino Ino, originAttr *Attr, cmode uint8, cumask uint16, top bool) syscall.Errno {
 	return errno(m.txn(ctx, func(tx *redis.Tx) error {
 		a, err := tx.Get(ctx, m.inodeKey(srcIno)).Bytes()
 		if err != nil {
 			return err
 		}
-		m.parseAttr(a, attr)
-		if eno := m.Access(ctx, srcIno, MODE_MASK_R, attr); eno != 0 {
+		m.parseAttr(a, originAttr)
+		attr := *originAttr
+		if eno := m.Access(ctx, srcIno, MODE_MASK_R, &attr); eno != 0 {
 			return eno
 		}
 		attr.Parent = parent
@@ -4268,7 +4269,7 @@ func (m *redisMeta) doCloneEntry(ctx Context, srcIno Ino, parent Ino, name strin
 		}
 
 		_, err = tx.TxPipelined(ctx, func(p redis.Pipeliner) error {
-			p.Set(ctx, m.inodeKey(ino), m.marshal(attr), 0)
+			p.Set(ctx, m.inodeKey(ino), m.marshal(&attr), 0)
 			p.IncrBy(ctx, m.usedSpaceKey(), align4K(attr.Length))
 			p.Incr(ctx, m.totalInodesKey())
 			if len(srcXattr) > 0 {
diff --git a/pkg/meta/tkv.go b/pkg/meta/tkv.go
index 7e917730de09..38b6d1f358e1 100644
--- a/pkg/meta/tkv.go
+++ b/pkg/meta/tkv.go
@@ -3359,14 +3359,15 @@ func (m *kvMeta) LoadMeta(r io.Reader) error {
 	})
 }
 
-func (m *kvMeta) doCloneEntry(ctx Context, srcIno Ino, parent Ino, name string, ino Ino, attr *Attr, cmode uint8, cumask uint16, top bool) syscall.Errno {
+func (m *kvMeta) doCloneEntry(ctx Context, srcIno Ino, parent Ino, name string, ino Ino, originAttr *Attr, cmode uint8, cumask uint16, top bool) syscall.Errno {
 	return errno(m.txn(func(tx *kvTxn) error {
 		a := tx.get(m.inodeKey(srcIno))
 		if a == nil {
 			return syscall.ENOENT
 		}
-		m.parseAttr(a, attr)
-		if eno := m.Access(ctx, srcIno, MODE_MASK_R, attr); eno != 0 {
+		m.parseAttr(a, originAttr)
+		attr := *originAttr
+		if eno := m.Access(ctx, srcIno, MODE_MASK_R, &attr); eno != 0 {
 			return eno
 		}
 		attr.Parent = parent
@@ -3416,7 +3417,7 @@ func (m *kvMeta) doCloneEntry(ctx Context, srcIno Ino, parent Ino, name string,
 			}
 		}
 
-		tx.set(m.inodeKey(ino), m.marshal(attr))
+		tx.set(m.inodeKey(ino), m.marshal(&attr))
 		prefix := m.xattrKey(srcIno, "")
 		tx.scan(prefix, nextKey(prefix), false, func(k, v []byte) bool {
 			tx.set(m.xattrKey(ino, string(k[len(prefix):])), v)

From 2cd2e016546b3e3cdd829981c82353e6ab619db5 Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Tue, 21 May 2024 10:28:31 +0800
Subject: [PATCH 23/30] vfs: fix the race of getting file length (#4876)

---
 pkg/vfs/reader.go | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/pkg/vfs/reader.go b/pkg/vfs/reader.go
index 72c67240c5b5..35a14ec51b65 100644
--- a/pkg/vfs/reader.go
+++ b/pkg/vfs/reader.go
@@ -168,12 +168,10 @@ func (s *sliceReader) run() {
 	inode := f.inode
 	f.Unlock()
 
-	f.Lock()
-	length := f.length
-	f.Unlock()
 	var slices []meta.Slice
 	err := f.r.m.Read(meta.Background, inode, indx, &slices)
 	f.Lock()
+	length := f.length
 	if s.state != BUSY || f.err != 0 || f.closing {
 		s.done(0, 0)
 	}

From 5d39bcdb2c2004b4246b08815c529f560db534ce Mon Sep 17 00:00:00 2001
From: prOOrc <prOOrc@users.noreply.github.com>
Date: Tue, 28 May 2024 04:55:17 +0300
Subject: [PATCH 24/30] fix: JfsObjects.CompleteMultipartUpload (#4893)

Co-authored-by: zhijian <djx@juicedata.io>
---
 pkg/gateway/gateway.go | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pkg/gateway/gateway.go b/pkg/gateway/gateway.go
index ce97e89c06f8..cad4f1ce5ac0 100644
--- a/pkg/gateway/gateway.go
+++ b/pkg/gateway/gateway.go
@@ -932,7 +932,7 @@ func (n *jfsObjects) CompleteMultipartUpload(ctx context.Context, bucket, object
 	var total uint64
 	for _, part := range parts {
 		p := n.ppath(bucket, uploadID, strconv.Itoa(part.PartNumber))
-		copied, eno := n.fs.CopyFileRange(mctx, p, 0, tmp, total, 1<<30)
+		copied, eno := n.fs.CopyFileRange(mctx, p, 0, tmp, total, 5<<30)
 		if eno != 0 {
 			err = jfsToObjectErr(ctx, eno, bucket, object, uploadID)
 			logger.Errorf("merge parts: %s", err)

From 6d5126665e13d9def3e32ca510b368d51c232339 Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Thu, 30 May 2024 16:14:03 +0800
Subject: [PATCH 25/30] meta: fix protection check for compaction (#4901)

---
 pkg/meta/base_test.go | 20 ++++++++++++++++++++
 pkg/meta/redis.go     | 32 +++++++++++++++++---------------
 pkg/meta/sql.go       | 30 ++++++++++++++++--------------
 pkg/meta/tkv.go       | 30 ++++++++++++++++--------------
 4 files changed, 69 insertions(+), 43 deletions(-)

diff --git a/pkg/meta/base_test.go b/pkg/meta/base_test.go
index 6df45774ef01..193e1b825f0e 100644
--- a/pkg/meta/base_test.go
+++ b/pkg/meta/base_test.go
@@ -1304,6 +1304,26 @@ func testCompaction(t *testing.T, m Meta, trash bool) {
 	if len(slices) != 1 || slices[0].Len != 3<<20 {
 		t.Fatalf("inode %d should be compacted, but have %d slices, size %d", inode, len(slices), slices[0].Len)
 	}
+
+	m.NewSlice(ctx, &sliceId)
+	_ = m.Write(ctx, inode, 2, 0, Slice{Id: sliceId, Size: 2338508, Len: 2338508}, time.Now())
+	m.NewSlice(ctx, &sliceId)
+	_ = m.Write(ctx, inode, 2, 8829056, Slice{Id: sliceId, Size: 1074933, Len: 1074933}, time.Now())
+	m.NewSlice(ctx, &sliceId)
+	_ = m.Write(ctx, inode, 2, 7663608, Slice{Id: sliceId, Size: 41480, Len: 4148}, time.Now())
+	_ = m.Fallocate(ctx, inode, fallocZeroRange, 2*ChunkSize+4515328, 3152428, nil)
+	_ = m.Fallocate(ctx, inode, fallocZeroRange, 2*ChunkSize+4515328, 2607724, nil)
+	if c, ok := m.(compactor); ok {
+		c.compactChunk(inode, 2, true)
+	}
+	if st := m.Read(ctx, inode, 2, &slices); st != 0 {
+		t.Fatalf("read 1: %s", st)
+	}
+	// compact twice: 4515328+2607724-2338508 = 4784544; 8829056+1074933-2338508-4784544=2780937
+	if len(slices) != 3 || slices[0].Len != 2338508 || slices[1].Len != 4784544 || slices[2].Len != 2780937 {
+		t.Fatalf("inode %d should be compacted, but have %d slices, size %d,%d,%d",
+			inode, len(slices), slices[0].Len, slices[1].Len, slices[2].Len)
+	}
 }
 
 func testConcurrentWrite(t *testing.T, m Meta) {
diff --git a/pkg/meta/redis.go b/pkg/meta/redis.go
index 2088e0eba40f..ecd790f3b653 100644
--- a/pkg/meta/redis.go
+++ b/pkg/meta/redis.go
@@ -2918,17 +2918,19 @@ func (m *redisMeta) compactChunk(inode Ino, indx uint32, force bool) {
 		return
 	}
 	skipped := skipSome(ss)
-	var first, last *slice
-	if skipped > 0 {
-		first, last = ss[0], ss[skipped-1]
-	}
-	ss = ss[skipped:]
-	pos, size, slices := compactChunk(ss)
-	if len(ss) < 2 || size == 0 {
+	compacted := ss[skipped:]
+	pos, size, slices := compactChunk(compacted)
+	if len(compacted) < 2 || size == 0 {
 		return
 	}
-	if first != nil && last != nil && pos+size > first.pos && last.pos+last.len > pos {
-		panic(fmt.Sprintf("invalid compaction: skipped slices [%+v, %+v], pos %d, size %d", *first, *last, pos, size))
+	for _, s := range ss[:skipped] {
+		if pos+size > s.pos && s.pos+s.len > pos {
+			var sstring string
+			for _, s := range ss {
+				sstring += fmt.Sprintf("\n%+v", *s)
+			}
+			panic(fmt.Sprintf("invalid compaction skipped %d, pos %d, size %d; slices: %s", skipped, pos, size, sstring))
+		}
 	}
 
 	var id uint64
@@ -2936,11 +2938,11 @@ func (m *redisMeta) compactChunk(inode Ino, indx uint32, force bool) {
 	if st != 0 {
 		return
 	}
-	logger.Debugf("compact %d:%d: skipped %d slices (%d bytes) %d slices (%d bytes)", inode, indx, skipped, pos, len(ss), size)
+	logger.Debugf("compact %d:%d: skipped %d slices (%d bytes) %d slices (%d bytes)", inode, indx, skipped, pos, len(compacted), size)
 	err = m.newMsg(CompactChunk, slices, id)
 	if err != nil {
 		if !strings.Contains(err.Error(), "not exist") && !strings.Contains(err.Error(), "not found") {
-			logger.Warnf("compact %d %d with %d slices: %s", inode, indx, len(ss), err)
+			logger.Warnf("compact %d %d with %d slices: %s", inode, indx, len(compacted), err)
 		}
 		return
 	}
@@ -2948,13 +2950,13 @@ func (m *redisMeta) compactChunk(inode Ino, indx uint32, force bool) {
 	var rs []*redis.IntCmd // trash disabled: check reference of slices
 	trash := m.toTrash(0)
 	if trash {
-		for _, s := range ss {
+		for _, s := range compacted {
 			if s.id > 0 {
 				buf = append(buf, m.encodeDelayedSlice(s.id, s.size)...)
 			}
 		}
 	} else {
-		rs = make([]*redis.IntCmd, len(ss))
+		rs = make([]*redis.IntCmd, len(compacted))
 	}
 	key := m.chunkKey(inode, indx)
 	errno := errno(m.txn(ctx, func(tx *redis.Tx) error {
@@ -2983,7 +2985,7 @@ func (m *redisMeta) compactChunk(inode Ino, indx uint32, force bool) {
 					pipe.HSet(ctx, m.delSlices(), fmt.Sprintf("%d_%d", id, time.Now().Unix()), buf)
 				}
 			} else {
-				for i, s := range ss {
+				for i, s := range compacted {
 					if s.id > 0 {
 						rs[i] = pipe.HIncrBy(ctx, m.sliceRefs(), m.sliceKey(s.id, s.size), -1)
 					}
@@ -3010,7 +3012,7 @@ func (m *redisMeta) compactChunk(inode Ino, indx uint32, force bool) {
 		m.of.InvalidateChunk(inode, indx)
 		m.cleanupZeroRef(m.sliceKey(id, size))
 		if !trash {
-			for i, s := range ss {
+			for i, s := range compacted {
 				if s.id > 0 && rs[i].Err() == nil && rs[i].Val() < 0 {
 					m.deleteSlice(s.id, s.size)
 				}
diff --git a/pkg/meta/sql.go b/pkg/meta/sql.go
index e9594ba89e3d..57d9bc929f5a 100644
--- a/pkg/meta/sql.go
+++ b/pkg/meta/sql.go
@@ -2774,17 +2774,19 @@ func (m *dbMeta) compactChunk(inode Ino, indx uint32, force bool) {
 		return
 	}
 	skipped := skipSome(ss)
-	var first, last *slice
-	if skipped > 0 {
-		first, last = ss[0], ss[skipped-1]
-	}
-	ss = ss[skipped:]
-	pos, size, slices := compactChunk(ss)
-	if len(ss) < 2 || size == 0 {
+	compacted := ss[skipped:]
+	pos, size, slices := compactChunk(compacted)
+	if len(compacted) < 2 || size == 0 {
 		return
 	}
-	if first != nil && last != nil && pos+size > first.pos && last.pos+last.len > pos {
-		panic(fmt.Sprintf("invalid compaction: skipped slices [%+v, %+v], pos %d, size %d", *first, *last, pos, size))
+	for _, s := range ss[:skipped] {
+		if pos+size > s.pos && s.pos+s.len > pos {
+			var sstring string
+			for _, s := range ss {
+				sstring += fmt.Sprintf("\n%+v", *s)
+			}
+			panic(fmt.Sprintf("invalid compaction skipped %d, pos %d, size %d; slices: %s", skipped, pos, size, sstring))
+		}
 	}
 
 	var id uint64
@@ -2792,18 +2794,18 @@ func (m *dbMeta) compactChunk(inode Ino, indx uint32, force bool) {
 	if st != 0 {
 		return
 	}
-	logger.Debugf("compact %d:%d: skipped %d slices (%d bytes) %d slices (%d bytes)", inode, indx, skipped, pos, len(ss), size)
+	logger.Debugf("compact %d:%d: skipped %d slices (%d bytes) %d slices (%d bytes)", inode, indx, skipped, pos, len(compacted), size)
 	err = m.newMsg(CompactChunk, slices, id)
 	if err != nil {
 		if !strings.Contains(err.Error(), "not exist") && !strings.Contains(err.Error(), "not found") {
-			logger.Warnf("compact %d %d with %d slices: %s", inode, indx, len(ss), err)
+			logger.Warnf("compact %d %d with %d slices: %s", inode, indx, len(compacted), err)
 		}
 		return
 	}
 	var buf []byte
 	trash := m.toTrash(0)
 	if trash {
-		for _, s := range ss {
+		for _, s := range compacted {
 			if s.id > 0 {
 				buf = append(buf, m.encodeDelayedSlice(s.id, s.size)...)
 			}
@@ -2835,7 +2837,7 @@ func (m *dbMeta) compactChunk(inode Ino, indx uint32, force bool) {
 				}
 			}
 		} else {
-			for _, s_ := range ss {
+			for _, s_ := range compacted {
 				if s_.id == 0 {
 					continue
 				}
@@ -2871,7 +2873,7 @@ func (m *dbMeta) compactChunk(inode Ino, indx uint32, force bool) {
 	} else if err == nil {
 		m.of.InvalidateChunk(inode, indx)
 		if !trash {
-			for _, s := range ss {
+			for _, s := range compacted {
 				if s.id == 0 {
 					continue
 				}
diff --git a/pkg/meta/tkv.go b/pkg/meta/tkv.go
index 38b6d1f358e1..98ccf9260fd7 100644
--- a/pkg/meta/tkv.go
+++ b/pkg/meta/tkv.go
@@ -2399,17 +2399,19 @@ func (m *kvMeta) compactChunk(inode Ino, indx uint32, force bool) {
 		return
 	}
 	skipped := skipSome(ss)
-	var first, last *slice
-	if skipped > 0 {
-		first, last = ss[0], ss[skipped-1]
-	}
-	ss = ss[skipped:]
-	pos, size, slices := compactChunk(ss)
-	if len(ss) < 2 || size == 0 {
+	compacted := ss[skipped:]
+	pos, size, slices := compactChunk(compacted)
+	if len(compacted) < 2 || size == 0 {
 		return
 	}
-	if first != nil && last != nil && pos+size > first.pos && last.pos+last.len > pos {
-		panic(fmt.Sprintf("invalid compaction: skipped slices [%+v, %+v], pos %d, size %d", *first, *last, pos, size))
+	for _, s := range ss[:skipped] {
+		if pos+size > s.pos && s.pos+s.len > pos {
+			var sstring string
+			for _, s := range ss {
+				sstring += fmt.Sprintf("\n%+v", *s)
+			}
+			panic(fmt.Sprintf("invalid compaction skipped %d, pos %d, size %d; slices: %s", skipped, pos, size, sstring))
+		}
 	}
 
 	var id uint64
@@ -2417,18 +2419,18 @@ func (m *kvMeta) compactChunk(inode Ino, indx uint32, force bool) {
 	if st != 0 {
 		return
 	}
-	logger.Debugf("compact %d:%d: skipped %d slices (%d bytes) %d slices (%d bytes)", inode, indx, skipped, pos, len(ss), size)
+	logger.Debugf("compact %d:%d: skipped %d slices (%d bytes) %d slices (%d bytes)", inode, indx, skipped, pos, len(compacted), size)
 	err = m.newMsg(CompactChunk, slices, id)
 	if err != nil {
 		if !strings.Contains(err.Error(), "not exist") && !strings.Contains(err.Error(), "not found") {
-			logger.Warnf("compact %d %d with %d slices: %s", inode, indx, len(ss), err)
+			logger.Warnf("compact %d %d with %d slices: %s", inode, indx, len(compacted), err)
 		}
 		return
 	}
 	var dsbuf []byte
 	trash := m.toTrash(0)
 	if trash {
-		for _, s := range ss {
+		for _, s := range compacted {
 			if s.id > 0 {
 				dsbuf = append(dsbuf, m.encodeDelayedSlice(s.id, s.size)...)
 			}
@@ -2450,7 +2452,7 @@ func (m *kvMeta) compactChunk(inode Ino, indx uint32, force bool) {
 				tx.set(m.delSliceKey(time.Now().Unix(), id), dsbuf)
 			}
 		} else {
-			for _, s := range ss {
+			for _, s := range compacted {
 				if s.id > 0 {
 					tx.incrBy(m.sliceKey(s.id, s.size), -1)
 				}
@@ -2480,7 +2482,7 @@ func (m *kvMeta) compactChunk(inode Ino, indx uint32, force bool) {
 		m.cleanupZeroRef(id, size)
 		if !trash {
 			var refs int64
-			for _, s := range ss {
+			for _, s := range compacted {
 				if s.id > 0 && m.client.txn(func(tx *kvTxn) error {
 					refs = tx.incrBy(m.sliceKey(s.id, s.size), 0)
 					return nil

From fe546ffacd0418a5a145be54d7d2ee1e16e65b41 Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Thu, 30 May 2024 20:07:25 +0800
Subject: [PATCH 26/30] vfs: fix the issue that plock may be unlocked by
 mistake (#4910)

---
 pkg/vfs/vfs.go      | 2 +-
 pkg/vfs/vfs_unix.go | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/pkg/vfs/vfs.go b/pkg/vfs/vfs.go
index fabbe869ad63..7adf706d07f0 100644
--- a/pkg/vfs/vfs.go
+++ b/pkg/vfs/vfs.go
@@ -564,7 +564,7 @@ func (v *VFS) Release(ctx Context, ino Ino, fh uint64) {
 				v.invalidateAttr(ino)
 			}
 			if locks&1 != 0 {
-				_ = v.Meta.Flock(ctx, ino, owner, F_UNLCK, false)
+				_ = v.Meta.Flock(ctx, ino, owner^fh, F_UNLCK, false)
 			}
 		}
 		_ = v.Meta.Close(ctx, ino)
diff --git a/pkg/vfs/vfs_unix.go b/pkg/vfs/vfs_unix.go
index 701d3e57099d..f2b58e5e0189 100644
--- a/pkg/vfs/vfs_unix.go
+++ b/pkg/vfs/vfs_unix.go
@@ -292,7 +292,7 @@ func (v *VFS) Flock(ctx Context, ino Ino, fh uint64, owner uint64, typ uint32, b
 	}
 	h.addOp(ctx)
 	defer h.removeOp(ctx)
-	err = v.Meta.Flock(ctx, ino, owner, typ, block)
+	err = v.Meta.Flock(ctx, ino, owner^fh, typ, block)
 	if err == 0 {
 		h.Lock()
 		if typ == syscall.F_UNLCK {

From cf2cc3bd3ab06cd94dccbd3645b48d335527cfdf Mon Sep 17 00:00:00 2001
From: Davies Liu <davies@juicedata.io>
Date: Thu, 30 May 2024 23:11:21 +0800
Subject: [PATCH 27/30] upgrade golang.org/x/net to 0.23 (#4911)

---
 go.mod |  8 ++++----
 go.sum | 16 ++++++++--------
 2 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/go.mod b/go.mod
index 68320171ed53..e237a6b2fa44 100644
--- a/go.mod
+++ b/go.mod
@@ -67,12 +67,12 @@ require (
 	go.etcd.io/etcd/client/v3 v3.5.9
 	go.uber.org/automaxprocs v1.5.2
 	go.uber.org/zap v1.20.0
-	golang.org/x/crypto v0.17.0
-	golang.org/x/net v0.17.0
+	golang.org/x/crypto v0.21.0
+	golang.org/x/net v0.23.0
 	golang.org/x/oauth2 v0.7.0
 	golang.org/x/sync v0.2.0
-	golang.org/x/sys v0.15.0
-	golang.org/x/term v0.15.0
+	golang.org/x/sys v0.18.0
+	golang.org/x/term v0.18.0
 	golang.org/x/text v0.14.0
 	google.golang.org/api v0.94.0
 	google.golang.org/protobuf v1.30.0
diff --git a/go.sum b/go.sum
index 07dda592e88d..8b79467f3c6a 100644
--- a/go.sum
+++ b/go.sum
@@ -1123,8 +1123,8 @@ golang.org/x/crypto v0.0.0-20211215153901-e495a2d5b3d3/go.mod h1:IxCIyHEi3zRg3s0
 golang.org/x/crypto v0.0.0-20220112180741-5e0467b6c7ce/go.mod h1:IxCIyHEi3zRg3s0A5j5BB6A9Jmi73HwBIUl50j+osU4=
 golang.org/x/crypto v0.1.0/go.mod h1:RecgLatLF4+eUMCP1PoPZQb+cVrJcOPbHkTkbkB9sbw=
 golang.org/x/crypto v0.6.0/go.mod h1:OFC/31mSvZgRz0V1QTNCzfAI1aIRzbiufJtkMIlEp58=
-golang.org/x/crypto v0.17.0 h1:r8bRNjWL3GshPW3gkd+RpvzWrZAwPS49OmTGZ/uhM4k=
-golang.org/x/crypto v0.17.0/go.mod h1:gCAAfMLgwOJRpTjQ2zCCt2OcSfYMTeZVSRtQlPC7Nq4=
+golang.org/x/crypto v0.21.0 h1:X31++rzVUdKhX5sWmSOFZxx8UW/ldWx55cbf08iNAMA=
+golang.org/x/crypto v0.21.0/go.mod h1:0BP7YvVV9gBbVKyeTG0Gyn+gZm94bibOW5BjDEYAOMs=
 golang.org/x/exp v0.0.0-20190121172915-509febef88a4/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
 golang.org/x/exp v0.0.0-20190306152737-a1d7652674e8/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
 golang.org/x/exp v0.0.0-20190510132918-efd6b22b2522/go.mod h1:ZjyILWgesfNpC6sMxTJOJm9Kp84zZh5NQWvqDGG3Qr8=
@@ -1225,8 +1225,8 @@ golang.org/x/net v0.1.0/go.mod h1:Cx3nUiGt4eDBEyega/BKRp+/AlGL8hYe7U9odMt2Cco=
 golang.org/x/net v0.6.0/go.mod h1:2Tu9+aMcznHK/AK1HMvgo6xiTLG5rD5rZLDS+rp2Bjs=
 golang.org/x/net v0.7.0/go.mod h1:2Tu9+aMcznHK/AK1HMvgo6xiTLG5rD5rZLDS+rp2Bjs=
 golang.org/x/net v0.8.0/go.mod h1:QVkue5JL9kW//ek3r6jTKnTFis1tRmNAW2P1shuFdJc=
-golang.org/x/net v0.17.0 h1:pVaXccu2ozPjCXewfr1S7xza/zcXTity9cCdXQYSjIM=
-golang.org/x/net v0.17.0/go.mod h1:NxSsAGuq816PNPmqtQdLE42eU2Fs7NoRIZrHJAlaCOE=
+golang.org/x/net v0.23.0 h1:7EYJ93RZ9vYSZAIb2x3lnuvqO5zneoD6IvWjuhfxjTs=
+golang.org/x/net v0.23.0/go.mod h1:JKghWKKOSdJwpW2GEx0Ja7fmaKnMsbu+MWVZTokSYmg=
 golang.org/x/oauth2 v0.0.0-20180821212333-d2e6202438be/go.mod h1:N/0e6XlmueqKjAGxoOufVs8QHGRruUQn6yWY3a++T0U=
 golang.org/x/oauth2 v0.0.0-20190226205417-e64efc72b421/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
 golang.org/x/oauth2 v0.0.0-20190402181905-9f3314589c9a/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
@@ -1369,15 +1369,15 @@ golang.org/x/sys v0.0.0-20221010170243-090e33056c14/go.mod h1:oPkhp1MJrh7nUepCBc
 golang.org/x/sys v0.1.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.5.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.15.0 h1:h48lPFYpsTvQJZF4EKyI4aLHaev3CxivZmv7yZig9pc=
-golang.org/x/sys v0.15.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
+golang.org/x/sys v0.18.0 h1:DBdB3niSjOA/O0blCZBqDefyWNYveAYMNF1Wum0DYQ4=
+golang.org/x/sys v0.18.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
 golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
 golang.org/x/term v0.0.0-20210927222741-03fcf44c2211/go.mod h1:jbD1KX2456YbFQfuXm/mYQcufACuNUgVhRMnK/tPxf8=
 golang.org/x/term v0.1.0/go.mod h1:jbD1KX2456YbFQfuXm/mYQcufACuNUgVhRMnK/tPxf8=
 golang.org/x/term v0.5.0/go.mod h1:jMB1sMXY+tzblOD4FWmEbocvup2/aLOaQEp7JmGp78k=
 golang.org/x/term v0.6.0/go.mod h1:m6U89DPEgQRMq3DNkDClhWw02AUbt2daBVO4cn4Hv9U=
-golang.org/x/term v0.15.0 h1:y/Oo/a/q3IXu26lQgl04j/gjuBDOBlx7X6Om1j2CPW4=
-golang.org/x/term v0.15.0/go.mod h1:BDl952bC7+uMoWR75FIrCDx79TPU9oHkTZ9yRbYOrX0=
+golang.org/x/term v0.18.0 h1:FcHjZXDMxI8mM3nwhX9HlKop4C0YQvCVCdwYl2wOtE8=
+golang.org/x/term v0.18.0/go.mod h1:ILwASektA3OnRv7amZ1xhE/KTR+u50pbXfZ03+6Nx58=
 golang.org/x/text v0.0.0-20170915032832-14c0d48ead0c/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.1-0.20180807135948-17ff2d5776d2/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=

From a18582680e83e18ec532bebc0281872ee0f623d8 Mon Sep 17 00:00:00 2001
From: Sandy Xu <sandy@juicedata.io>
Date: Tue, 4 Jun 2024 18:27:48 +0800
Subject: [PATCH 28/30] meta: fix the error code of baseMeta.Open (#4920)

---
 pkg/meta/base.go | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/pkg/meta/base.go b/pkg/meta/base.go
index 1a8d9702566b..db70363cdb66 100644
--- a/pkg/meta/base.go
+++ b/pkg/meta/base.go
@@ -1648,22 +1648,23 @@ func (m *baseMeta) touchAtime(ctx Context, inode Ino, attr *Attr) {
 	}
 }
 
-func (m *baseMeta) Open(ctx Context, inode Ino, flags uint32, attr *Attr) (rerr syscall.Errno) {
+func (m *baseMeta) Open(ctx Context, inode Ino, flags uint32, attr *Attr) (st syscall.Errno) {
 	if m.conf.ReadOnly && flags&(syscall.O_WRONLY|syscall.O_RDWR|syscall.O_TRUNC|syscall.O_APPEND) != 0 {
 		return syscall.EROFS
 	}
 	defer func() {
-		if rerr == 0 {
+		if st == 0 {
 			m.touchAtime(ctx, inode, attr)
 		}
 	}()
 	if m.conf.OpenCache > 0 && m.of.OpenCheck(inode, attr) {
 		return 0
 	}
-	var err syscall.Errno
 	// attr may be valid, see fs.Open()
 	if attr != nil && !attr.Full {
-		err = m.GetAttr(ctx, inode, attr)
+		if st = m.GetAttr(ctx, inode, attr); st != 0 {
+			return
+		}
 	}
 	var mmask uint8 = 0
 	switch flags & (syscall.O_RDONLY | syscall.O_WRONLY | syscall.O_RDWR) {
@@ -1674,7 +1675,7 @@ func (m *baseMeta) Open(ctx Context, inode Ino, flags uint32, attr *Attr) (rerr
 	case syscall.O_RDWR:
 		mmask = MODE_MASK_R | MODE_MASK_W
 	}
-	if rerr = m.Access(ctx, inode, mmask, attr); rerr != 0 {
+	if st = m.Access(ctx, inode, mmask, attr); st != 0 {
 		return
 	}
 
@@ -1691,10 +1692,8 @@ func (m *baseMeta) Open(ctx Context, inode Ino, flags uint32, attr *Attr) (rerr
 			return syscall.EPERM
 		}
 	}
-	if err == 0 {
-		m.of.Open(inode, attr)
-	}
-	return err
+	m.of.Open(inode, attr)
+	return 0
 }
 
 func (m *baseMeta) InvalidateChunkCache(ctx Context, inode Ino, indx uint32) syscall.Errno {

From 2647b03980ba98ff0566c7b0de56c8df345a7955 Mon Sep 17 00:00:00 2001
From: zhijian <djx@juicedata.io>
Date: Wed, 5 Jun 2024 17:48:26 +0800
Subject: [PATCH 29/30] gateway: list object follow symlink (#4919)

---
 .github/scripts/command/sync_minio.sh | 15 ++++++++++++++-
 pkg/gateway/gateway.go                |  8 ++++++++
 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/.github/scripts/command/sync_minio.sh b/.github/scripts/command/sync_minio.sh
index db556bf2d767..80bd4b917c84 100755
--- a/.github/scripts/command/sync_minio.sh
+++ b/.github/scripts/command/sync_minio.sh
@@ -79,11 +79,12 @@ test_sync_external_link(){
     [ -z $(./mc cat myminio/myjfs/hello) ]
 }
 
+# list object should be skipped when encountering a loop symlink
 test_sync_loop_symlink(){
     prepare_test
     touch hello
     ln -s hello /jfs/hello
-    ./juicefs sync minio://minioadmin:minioadmin@localhost:9005/myjfs/ minio://minioadmin:minioadmin@localhost:9000/myjfs/ && exit 1 || true
+    ./juicefs sync minio://minioadmin:minioadmin@localhost:9005/myjfs/ minio://minioadmin:minioadmin@localhost:9000/myjfs/
     rm -rf /jfs/hello
     ./juicefs sync minio://minioadmin:minioadmin@localhost:9005/myjfs/ minio://minioadmin:minioadmin@localhost:9000/myjfs/
 }
@@ -107,6 +108,18 @@ test_sync_deep_symlink(){
     done
 }
 
+test_sync_list_object_symlink(){
+    prepare_test
+    cd /jfs
+    mkdir dir1
+    mkdir -p dir2/src_dir
+    echo abc > dir2/src_dir/afile
+    ln -s ./../dir2/src_dir dir1/symlink_dir
+    cd -
+    ./juicefs sync minio://minioadmin:minioadmin@localhost:9005/myjfs/dir1/ minio://minioadmin:minioadmin@localhost:9000/myjfs/dir3/
+    ./mc cat myminio/myjfs/dir3/symlink_dir/afile | grep abc || (echo "content should be abc" && exit 1)
+}
+
 prepare_test(){
     umount_jfs /jfs $META_URL
     python3 .github/scripts/flush_meta.py $META_URL
diff --git a/pkg/gateway/gateway.go b/pkg/gateway/gateway.go
index cad4f1ce5ac0..3cb0c7451909 100644
--- a/pkg/gateway/gateway.go
+++ b/pkg/gateway/gateway.go
@@ -303,6 +303,14 @@ func (n *jfsObjects) listDirFactory() minio.ListDirFunc {
 			if root && len(fi.Name()) == len(metaBucket) && fi.Name() == metaBucket {
 				continue
 			}
+			if stat, ok := fi.(*fs.FileStat); ok && stat.IsSymlink() {
+				var err syscall.Errno
+				p := n.path(bucket, prefixDir, fi.Name())
+				if fi, err = n.fs.Stat(mctx, p); err != 0 {
+					logger.Errorf("stat %s: %s", p, err)
+					continue
+				}
+			}
 			if fi.IsDir() {
 				entries = append(entries, fi.Name()+sep)
 			} else {

From 9abfc79a34557515a53980436aafb8129bf4fd75 Mon Sep 17 00:00:00 2001
From: Changxin Miao <miaochangxin@step.ai>
Date: Thu, 13 Jun 2024 18:08:06 +0800
Subject: [PATCH 30/30] Manager's IP address is never set in sync cluster mode
 (#4939)

Signed-off-by: Changxin Miao <miaochangxin@step.ai>
---
 pkg/sync/cluster.go | 1 +
 1 file changed, 1 insertion(+)

diff --git a/pkg/sync/cluster.go b/pkg/sync/cluster.go
index e0e60d509d57..344424f427a8 100644
--- a/pkg/sync/cluster.go
+++ b/pkg/sync/cluster.go
@@ -193,6 +193,7 @@ func startManager(config *Config, tasks <-chan object.Object) (string, error) {
 		if ip == "" {
 			return "", fmt.Errorf("no local ip found")
 		}
+		addr = ip
 	}
 
 	if !strings.Contains(addr, ":") {