From e7bbe70f79aa9308c2eb54b057ee5779b22f478e Mon Sep 17 00:00:00 2001 From: Ayush Ranjan Date: Mon, 12 Oct 2020 16:27:33 -0700 Subject: [vfs] kernfs: Fix inode memory leak issue. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This change aims to fix the memory leak issue reported in #3933. Background: VFS2 kernfs kept accumulating invalid dentries if those dentries were not walked on. After substantial consideration of the problem by our team, we decided to have an LRU cache solution. This change is the first part to that solution, where we don't cache anything. The LRU cache can be added on top of this. What has changed: - Introduced the concept of an inode tree in kernfs.OrderedChildren. This is helpful is cases where the lifecycle of an inode is different from that of a dentry. - OrderedChildren now deals with initialized inodes instead of initialized dentries. It now implements Lookup() where it constructs a new dentry using the inode. - OrderedChildren holds a ref on all its children inodes. With this change, now an inode can "outlive" a dentry pointing to it. See comments in kernfs.OrderedChildren. - The kernfs dentry tree is solely maintained by kernfs only. Inode implementations can not modify the dentry tree. - Dentries that reach ref count 0 are removed from the dentry tree. - revalidateChildLocked now defer-DecRefs the newly created dentry from Inode.Lookup(), limiting its life to the current filesystem operation. If refs are picked on the dentry during the FS op (via an FD or something), then it will stick around and will be removed when the FD is closed. So there is essentially _no caching_ for Look()ed up dentries. - kernfs.DecRef does not have the precondition that fs.mu must be locked. Fixes #3933 PiperOrigin-RevId: 336768576 --- pkg/sentry/fsimpl/proc/tasks.go | 56 ++++++++++++++++++++--------------------- 1 file changed, 28 insertions(+), 28 deletions(-) (limited to 'pkg/sentry/fsimpl/proc/tasks.go') diff --git a/pkg/sentry/fsimpl/proc/tasks.go b/pkg/sentry/fsimpl/proc/tasks.go index d8f5dd509..3259c3732 100644 --- a/pkg/sentry/fsimpl/proc/tasks.go +++ b/pkg/sentry/fsimpl/proc/tasks.go @@ -38,10 +38,11 @@ const ( // +stateify savable type tasksInode struct { implStatFS - kernfs.AlwaysValid + kernfs.InodeAlwaysValid kernfs.InodeAttrs kernfs.InodeDirectoryNoNewChildren kernfs.InodeNotSymlink + kernfs.InodeTemporary // This holds no meaning as this inode can't be Looked up and is always valid. kernfs.OrderedChildren tasksInodeRefs @@ -52,8 +53,6 @@ type tasksInode struct { // '/proc/self' and '/proc/thread-self' have custom directory offsets in // Linux. So handle them outside of OrderedChildren. - selfSymlink *kernfs.Dentry - threadSelfSymlink *kernfs.Dentry // cgroupControllers is a map of controller name to directory in the // cgroup hierarchy. These controllers are immutable and will be listed @@ -63,52 +62,53 @@ type tasksInode struct { var _ kernfs.Inode = (*tasksInode)(nil) -func (fs *filesystem) newTasksInode(k *kernel.Kernel, pidns *kernel.PIDNamespace, cgroupControllers map[string]string) (*tasksInode, *kernfs.Dentry) { +func (fs *filesystem) newTasksInode(k *kernel.Kernel, pidns *kernel.PIDNamespace, cgroupControllers map[string]string) *tasksInode { root := auth.NewRootCredentials(pidns.UserNamespace()) - contents := map[string]*kernfs.Dentry{ - "cpuinfo": fs.newDentry(root, fs.NextIno(), 0444, newStaticFileSetStat(cpuInfoData(k))), - "filesystems": fs.newDentry(root, fs.NextIno(), 0444, &filesystemsData{}), - "loadavg": fs.newDentry(root, fs.NextIno(), 0444, &loadavgData{}), + contents := map[string]kernfs.Inode{ + "cpuinfo": fs.newInode(root, 0444, newStaticFileSetStat(cpuInfoData(k))), + "filesystems": fs.newInode(root, 0444, &filesystemsData{}), + "loadavg": fs.newInode(root, 0444, &loadavgData{}), "sys": fs.newSysDir(root, k), - "meminfo": fs.newDentry(root, fs.NextIno(), 0444, &meminfoData{}), + "meminfo": fs.newInode(root, 0444, &meminfoData{}), "mounts": kernfs.NewStaticSymlink(root, linux.UNNAMED_MAJOR, fs.devMinor, fs.NextIno(), "self/mounts"), "net": kernfs.NewStaticSymlink(root, linux.UNNAMED_MAJOR, fs.devMinor, fs.NextIno(), "self/net"), - "stat": fs.newDentry(root, fs.NextIno(), 0444, &statData{}), - "uptime": fs.newDentry(root, fs.NextIno(), 0444, &uptimeData{}), - "version": fs.newDentry(root, fs.NextIno(), 0444, &versionData{}), + "stat": fs.newInode(root, 0444, &statData{}), + "uptime": fs.newInode(root, 0444, &uptimeData{}), + "version": fs.newInode(root, 0444, &versionData{}), } inode := &tasksInode{ pidns: pidns, fs: fs, - selfSymlink: fs.newSelfSymlink(root, fs.NextIno(), pidns), - threadSelfSymlink: fs.newThreadSelfSymlink(root, fs.NextIno(), pidns), cgroupControllers: cgroupControllers, } inode.InodeAttrs.Init(root, linux.UNNAMED_MAJOR, fs.devMinor, fs.NextIno(), linux.ModeDirectory|0555) inode.EnableLeakCheck() - dentry := &kernfs.Dentry{} - dentry.Init(inode) - inode.OrderedChildren.Init(kernfs.OrderedChildrenOptions{}) - links := inode.OrderedChildren.Populate(dentry, contents) + links := inode.OrderedChildren.Populate(contents) inode.IncLinks(links) - return inode, dentry + return inode } -// Lookup implements kernfs.inodeDynamicLookup.Lookup. -func (i *tasksInode) Lookup(ctx context.Context, name string) (*kernfs.Dentry, error) { - // Try to lookup a corresponding task. +// Lookup implements kernfs.inodeDirectory.Lookup. +func (i *tasksInode) Lookup(ctx context.Context, name string) (kernfs.Inode, error) { + // Check if a static entry was looked up. + if d, err := i.OrderedChildren.Lookup(ctx, name); err == nil { + return d, nil + } + + // Not a static entry. Try to lookup a corresponding task. tid, err := strconv.ParseUint(name, 10, 64) if err != nil { + root := auth.NewRootCredentials(i.pidns.UserNamespace()) // If it failed to parse, check if it's one of the special handled files. switch name { case selfName: - return i.selfSymlink, nil + return i.newSelfSymlink(root), nil case threadSelfName: - return i.threadSelfSymlink, nil + return i.newThreadSelfSymlink(root), nil } return nil, syserror.ENOENT } @@ -118,10 +118,10 @@ func (i *tasksInode) Lookup(ctx context.Context, name string) (*kernfs.Dentry, e return nil, syserror.ENOENT } - return i.fs.newTaskInode(task, i.pidns, true, i.cgroupControllers), nil + return i.fs.newTaskInode(task, i.pidns, true, i.cgroupControllers) } -// IterDirents implements kernfs.inodeDynamicLookup.IterDirents. +// IterDirents implements kernfs.inodeDirectory.IterDirents. func (i *tasksInode) IterDirents(ctx context.Context, cb vfs.IterDirentsCallback, offset, _ int64) (int64, error) { // fs/proc/internal.h: #define FIRST_PROCESS_ENTRY 256 const FIRST_PROCESS_ENTRY = 256 @@ -229,8 +229,8 @@ func (i *tasksInode) Stat(ctx context.Context, vsfs *vfs.Filesystem, opts vfs.St } // DecRef implements kernfs.Inode.DecRef. -func (i *tasksInode) DecRef(context.Context) { - i.tasksInodeRefs.DecRef(i.Destroy) +func (i *tasksInode) DecRef(ctx context.Context) { + i.tasksInodeRefs.DecRef(func() { i.Destroy(ctx) }) } // staticFileSetStat implements a special static file that allows inode -- cgit v1.2.3