File System Project Seminar File System Interfaces Prof. Andreas Polze Andreas Grapentin, Sven Köhler Max Plauth, Jossekin Beilharz, Felix Eberhardt Hasso Plattner Institute
1 File System Seminar Background Interfaces Polze, Grapentin, Köhler Plauth, Beilharz, Eberhardt 18.10.2017 Chart 2
Interfaces Multiple File Systems May be Present ext4 / proc home usr btrfs procfs nfs share bin File System Seminar Interfaces Polze, Grapentin, Köhler Plauth, Beilharz, Eberhardt The operating system needs to provide a common interface 18.10.2017 Chart 3 to different file system drivers, as well as the user space applications.
Interfaces A Common Abstraction program open readdir kernel Virtual File System nfs socket proc ext4 btrfs fs File System Seminar Interfaces Polze, Grapentin, Köhler Plauth, Beilharz, Eberhardt 18.10.2017 Block Buffer disk Chart 4
Interfaces Virtual File Systems ■ First introduced with SunOS 2.0 in 1985 (for NFS) ■ Nowadays most major operating systems provide a VFS or comparable means (Linux, *BSD, macOS, Windows, …) ■ File System In User Space (FUSE) is a cross-platform VFS, fed from User Space processes ■ Desktop Enviroments often provide their own VFS (KDE – KIO, Gnome – GIO) File System Seminar program FUSE driver GIO Interfaces Polze, Grapentin, Köhler Plauth, Beilharz, Eberhardt 18.10.2017 disk VFS FUSE block buffer Chart 5
Virtual File System Tasks ■ Provides an abstraction layer between applications and file systems ■ Presents a common file model ■ Caches inodes, directory entries, block buffers (e.g. LRU) ■ May include syncronization/locking mechanisms ■ Ease implementation for new file system drivers by auto-completing interfaces (e.g. use mmap to implement read/write) File System Seminar Interfaces Polze, Grapentin, Köhler Plauth, Beilharz, Eberhardt 18.10.2017 Chart 6
2 version 4.0, 2015-06-01 outlines the Linux storage stack as of Kernel version 4.0 Fibre Channel over Ethernet Fibre Channel Virtual Host ISCSI FireWire USB mmap (anonymous pages) Applications (processes) iscsi_target_mod tcm_usb_gadget LIO malloc read(2) write(2) open(2) stat(2) chmod(2) tcm_qla2xxx vfs_writev, vfs_readv, ... sbp_target ... tcm_vhost tcm_fc VFS Block-based FS Pseudo FS Special Network FS ext2 ext3 ext4 xfs NFS coda purpose FS proc sysfs Direct I/O Page target_core_mod ... (O_DIRECT) btrfs ifs iso9660 smbfs tmpfs ramfs pipefs futexfs cache gfs ... ceph target_core_file ocfs ... devtmpfs usbfs Stackable FS target_core_iblock FUSE ecryptfs overlayfs unionfs userspace (e.g. sshfs) target_core_pscsi target_core_user network stackable (optional) struct bio - sector on disk BIOs (block I/Os) Devices on top of “normal” BIOs (block I/Os) - sector cnt block devices - bio_vec cnt drbd LVM - bio_vec index device mapper - bio_vec list mdraid dm-crypt dm-mirror ... bcache dm-cache dm-thin dm-raid dm-delay userspace BIOs BIOs Block Layer BIOs I/O scheduler blkmq Maps BIOs to requests multi queue hooked in device drivers noop Software (they hook in like stacked ... queues cfq devices do) deadline Hardware Hardware dispatch ... dispatch queue queues Request Request BIO based drivers based drivers based drivers File System Seminar Request-based device mapper targets dm-multipath Linux VFS SCSI mid layer Interfaces /dev/rbd* /dev/vd* /dev/rssd* /dev/skd* sysfs scsi-mq /dev/zram* /dev/mmcblk*p* /dev/nullb* (transport attributes) SCSI upper level drivers ... /dev/nvme*n* /dev/rsxx* /dev/sda /dev/sd* Transport classes scsi_transport_fc /dev/sr* /dev/st* Polze, Grapentin, Köhler zram rbd mmc null_blk virtio_blk mtip32xx nvme skd rsxx scsi_transport_sas network scsi_transport_... memory Plauth, Beilharz, Eberhardt SCSI low level drivers libata megaraid_sas qla2xxx pm8001 iscsi_tcp virtio_scsi ufs ... 18.10.2017 ... ahci ata_piix aacraid lpfc mpt3sas vmw_pvscsi Chart 7 network HDD SSD DVD LSI Qlogic PMC-Sierra Micron nvme stec para-virtualized mobile device virtio_pci drive RAID HBA HBA SCSI PCIe card device device flash memory Adaptec Emulex LSI 12Gbs VMware's SD-/MMC-Card SAS HBA para-virtualized IBM flash RAID HBA Physical devices SCSI adapter
Virtual File System in Linux The Basic Building Blocks (4+2) ■ struct super_block – represents a mounted system, links the root directory. ■ struct inode – represents an existing file ■ struct file – represents an open file, with pointer ■ struct dentry – represents a directory entry, links a name to an inode ■ struct file_system_type – used to mount a system, File System Seminar Interfaces builds a superblock Polze, Grapentin, Köhler ■ struct vfsmount – a device+mountpoint pair Plauth, Beilharz, Eberhardt 18.10.2017 Chart 8
Virtual File System in Linux Relation Of The Four Basic Objects i_sb superblock Storage Device inode Hardlink f_dentry d_inode fd file proc1 dentry dentry File System Seminar fd Interfaces file proc2 Dentry cache f_dentry Polze, Grapentin, Köhler Plauth, Beilharz, Eberhardt 18.10.2017 Chart 9
Virtual File System in Linux OOP-ish? Object oriented programming looks AbstractInode like the ideal paradigm here. ________________ But Linux, as most kernels, uses C. + size + permission Virtual methods, … ________________ polymorphism and inheritance + create = 0 + link = 0 require additional effort in C. + unlink = 0 … File System Seminar Interfaces Polze, Grapentin, Köhler InodeExt4 InodeNFS InodeProcFS Plauth, Beilharz, Eberhardt 18.10.2017 Chart 10
) Virtual File System in Linux / How To Do Virtual Methods and Objects in C ( GObject File System Seminar Interfaces Polze, Grapentin, Köhler PT2 <= 2013 Plauth, Beilharz, Eberhardt 18.10.2017 Chart 11
Virtual File System in Linux How To Do Virtual Methods The Linux Way (I) struct inode { unsigned long i_ino; umode_t i_mode; uid_t i_uid; gid_t i_gid; kdev_t i_rdev; loff_t i_size; struct timespec i_atime; struct timespec i_ctime; struct timespec i_mtime; struct super_block *i_sb; struct inode_operations *i_op; File System Seminar const struct file_operations *i_fop; Interfaces void *i_private; Polze, Grapentin, Köhler } Plauth, Beilharz, Eberhardt 18.10.2017 Chart 12 include/linux/fs.h#L566
Virtual File System in Linux How To Do Virtual Methods The Linux Way (II) struct inode_operations { int (*create) (struct inode *, struct dentry *, int); struct dentry * (*lookup) (struct inode *, struct dentry *); int (*link) (struct dentry *, struct inode *, struct dentry *); int (*unlink) (struct inode *, struct dentry *); int (*symlink) (struct inode *, struct dentry *, const char *); int (*mkdir) (struct inode *, struct dentry *, int); int (*rmdir) (struct inode *, struct dentry *); int (*mknod) (struct inode *, struct dentry *, int, dev_t); int (*rename) (struct inode *, struct dentry *, struct inode *, struct dentry *); int (*readlink) (struct dentry *, char *,int); int (*follow_link) (struct dentry *, struct nameidata *); void (*truncate) (struct inode *); int (*permission) (struct inode *, int); File System Seminar int (*setattr) (struct dentry *, struct iattr *); Interfaces int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *); Polze, Grapentin, Köhler ... Plauth, Beilharz, Eberhardt }; 18.10.2017 Chart 13 inode->i_op->create(...)
Virtual File System in Linux How To Do Virtual Methods The Linux Way (III) “this”-pointer struct file_operations { struct module *owner; loff_t (*llseek) (struct file *, loff_t, int); ssize_t (*read) (struct file *, char *, size_t, loff_t *); ssize_t (*write) (struct file *, const char *, size_t, loff_t *); int (*readdir) (struct file *, void *, filldir_t); int (*ioctl) (struct inode *, struct file *, unsigned int, unsigned long); int (*mmap) (struct file *, struct vm_area_struct *); int (*open) (struct inode *, struct file *); int (*flush) (struct file *); unsigned long (*get_unmapped_area)(struct file *, ...); ... }; File System Seminar Your new FS driver needs to fill in those function pointers and provide Interfaces Polze, Grapentin, Köhler {super,inode,file,dentry}_operations to the corresponding data struct. Plauth, Beilharz, Eberhardt 18.10.2017 Chart 14
Virtual File System in Linux Adding A New File System (Overview) 1. Create a new loadable kernel module 2. Fill a struct file_system_type 3. Register your file system 4. Implement get_sb to load and create a superblock 5. Implement file, inode, superblock and dentry operations 6. Implement iget to load an arbitrary inode and iput to save them 7. Load super block and create inode for your root directory File System Seminar Interfaces Polze, Grapentin, Köhler Plauth, Beilharz, Eberhardt 18.10.2017 Chart 15
More recommend
