diff options
73 files changed, 264 insertions, 9721 deletions
diff --git a/Documentation/filesystems/00-INDEX b/Documentation/filesystems/00-INDEX index f66e748fc5e4..b7bd6c9009cc 100644 --- a/Documentation/filesystems/00-INDEX +++ b/Documentation/filesystems/00-INDEX @@ -87,8 +87,6 @@ jfs.txt - info and mount options for the JFS filesystem. locks.txt - info on file locking implementations, flock() vs. fcntl(), etc. -logfs.txt - - info on the LogFS flash filesystem. mandatory-locking.txt - info on the Linux implementation of Sys V mandatory file locking. ncpfs.txt diff --git a/Documentation/filesystems/logfs.txt b/Documentation/filesystems/logfs.txt deleted file mode 100644 index bca42c22a143..000000000000 --- a/Documentation/filesystems/logfs.txt +++ /dev/null @@ -1,241 +0,0 @@ - -The LogFS Flash Filesystem -========================== - -Specification -============= - -Superblocks ------------ - -Two superblocks exist at the beginning and end of the filesystem. -Each superblock is 256 Bytes large, with another 3840 Bytes reserved -for future purposes, making a total of 4096 Bytes. - -Superblock locations may differ for MTD and block devices. On MTD the -first non-bad block contains a superblock in the first 4096 Bytes and -the last non-bad block contains a superblock in the last 4096 Bytes. -On block devices, the first 4096 Bytes of the device contain the first -superblock and the last aligned 4096 Byte-block contains the second -superblock. - -For the most part, the superblocks can be considered read-only. They -are written only to correct errors detected within the superblocks, -move the journal and change the filesystem parameters through tunefs. -As a result, the superblock does not contain any fields that require -constant updates, like the amount of free space, etc. - -Segments --------- - -The space in the device is split up into equal-sized segments. -Segments are the primary write unit of LogFS. Within each segments, -writes happen from front (low addresses) to back (high addresses. If -only a partial segment has been written, the segment number, the -current position within and optionally a write buffer are stored in -the journal. - -Segments are erased as a whole. Therefore Garbage Collection may be -required to completely free a segment before doing so. - -Journal --------- - -The journal contains all global information about the filesystem that -is subject to frequent change. At mount time, it has to be scanned -for the most recent commit entry, which contains a list of pointers to -all currently valid entries. - -Object Store ------------- - -All space except for the superblocks and journal is part of the object -store. Each segment contains a segment header and a number of -objects, each consisting of the object header and the payload. -Objects are either inodes, directory entries (dentries), file data -blocks or indirect blocks. - -Levels ------- - -Garbage collection (GC) may fail if all data is written -indiscriminately. One requirement of GC is that data is separated -roughly according to the distance between the tree root and the data. -Effectively that means all file data is on level 0, indirect blocks -are on levels 1, 2, 3 4 or 5 for 1x, 2x, 3x, 4x or 5x indirect blocks, -respectively. Inode file data is on level 6 for the inodes and 7-11 -for indirect blocks. - -Each segment contains objects of a single level only. As a result, -each level requires its own separate segment to be open for writing. - -Inode File ----------- - -All inodes are stored in a special file, the inode file. Single -exception is the inode file's inode (master inode) which for obvious -reasons is stored in the journal instead. Instead of data blocks, the -leaf nodes of the inode files are inodes. - -Aliases -------- - -Writes in LogFS are done by means of a wandering tree. A naïve -implementation would require that for each write or a block, all -parent blocks are written as well, since the block pointers have -changed. Such an implementation would not be very efficient. - -In LogFS, the block pointer changes are cached in the journal by means -of alias entries. Each alias consists of its logical address - inode -number, block index, level and child number (index into block) - and -the changed data. Any 8-byte word can be changes in this manner. - -Currently aliases are used for block pointers, file size, file used -bytes and the height of an inodes indirect tree. - -Segment Aliases ---------------- - -Related to regular aliases, these are used to handle bad blocks. -Initially, bad blocks are handled by moving the affected segment -content to a spare segment and noting this move in the journal with a -segment alias, a simple (to, from) tupel. GC will later empty this -segment and the alias can be removed again. This is used on MTD only. - -Vim ---- - -By cleverly predicting the life time of data, it is possible to -separate long-living data from short-living data and thereby reduce -the GC overhead later. Each type of distinc life expectency (vim) can -have a separate segment open for writing. Each (level, vim) tupel can -be open just once. If an open segment with unknown vim is encountered -at mount time, it is closed and ignored henceforth. - -Indirect Tree -------------- - -Inodes in LogFS are similar to FFS-style filesystems with direct and -indirect block pointers. One difference is that LogFS uses a single -indirect pointer that can be either a 1x, 2x, etc. indirect pointer. -A height field in the inode defines the height of the indirect tree -and thereby the indirection of the pointer. - -Another difference is the addressing of indirect blocks. In LogFS, -the first 16 pointers in the first indirect block are left empty, -corresponding to the 16 direct pointers in the inode. In ext2 (maybe -others as well) the first pointer in the first indirect block -corresponds to logical block 12, skipping the 12 direct pointers. -So where ext2 is using arithmetic to better utilize space, LogFS keeps -arithmetic simple and uses compression to save space. - -Compression ------------ - -Both file data and metadata can be compressed. Compression for file -data can be enabled with chattr +c and disabled with chattr -c. Doing -so has no effect on existing data, but new data will be stored -accordingly. New inodes will inherit the compression flag of the -parent directory. - -Metadata is always compressed. However, the space accounting ignores -this and charges for the uncompressed size. Failing to do so could -result in GC failures when, after moving some data, indirect blocks -compress worse than previously. Even on a 100% full medium, GC may -not consume any extra space, so the compression gains are lost space -to the user. - -However, they are not lost space to the filesystem internals. By -cheating the user for those bytes, the filesystem gained some slack -space and GC will run less often and faster. - -Garbage Collection and Wear Leveling ------------------------------------- - -Garbage collection is invoked whenever the number of free segments -falls below a threshold. The best (known) candidate is picked based -on the least amount of valid data contained in the segment. All -remaining valid data is copied elsewhere, thereby invalidating it. - -The GC code also checks for aliases and writes then back if their -number gets too large. - -Wear leveling is done by occasionally picking a suboptimal segment for -garbage collection. If a stale segments erase count is significantly -lower than the active segments' erase counts, it will be picked. Wear -leveling is rate limited, so it will never monopolize the device for -more than one segment worth at a time. - -Values for "occasionally", "significantly lower" are compile time -constants. - -Hashed directories ------------------- - -To satisfy efficient lookup(), directory entries are hashed and -located based on the hash. In order to both support large directories -and not be overly inefficient for small directories, several hash -tables of increasing size are used. For each table, the hash value -modulo the table size gives the table index. - -Tables sizes are chosen to limit the number of indirect blocks with a -fully populated table to 0, 1, 2 or 3 respectively. So the first -table contains 16 entries, the second 512-16, etc. - -The last table is special in several ways. First its size depends on -the effective 32bit limit on telldir/seekdir cookies. Since logfs -uses the upper half of the address space for indirect blocks, the size -is limited to 2^31. Secondly the table contains hash buckets with 16 -entries each. - -Using single-entry buckets would result in birthday "attacks". At -just 2^16 used entries, hash collisions would be likely (P >= 0.5). -My math skills are insufficient to do the combinatorics for the 17x -collisions necessary to overflow a bucket, but testing showed that in -10,000 runs the lowest directory fill before a bucket overflow was -188,057,130 entries with an average of 315,149,915 entries. So for -directory sizes of up to a million, bucket overflows should be -virtually impossible under normal circumstances. - -With carefully chosen filenames, it is obviously possible to cause an -overflow with just 21 entries (4 higher tables + 16 entries + 1). So -there may be a security concern if a malicious user has write access -to a directory. - -Open For Discussion -=================== - -Device Address Space --------------------- - -A device address space is used for caching. Both block devices and -MTD provide functions to either read a single page or write a segment. -Partial segments may be written for data integrity, but where possible -complete segments are written for performance on simple block device -flash media. - -Meta Inodes ------------ - -Inodes are stored in the inode file, which is just a regular file for -most purposes. At umount time, however, the inode file needs to -remain open until all dirty inodes are written. So -generic_shutdown_super() may not close this inode, but shouldn't -complain about remaining inodes due to the inode file either. Same -goes for mapping inode of the device address space. - -Currently logfs uses a hack that essentially copies part of fs/inode.c -code over. A general solution would be preferred. - -Indirect block mapping ----------------------- - -With compression, the block device (or mapping inode) cannot be used -to cache indirect blocks. Some other place is required. Currently -logfs uses the top half of each inode's address space. The low 8TB -(on 32bit) are filled with file data, the high 8TB are used for -indirect blocks. - -One problem is that 16TB files created on 64bit systems actually have -data in the top 8TB. But files >16TB would cause problems anyway, so -only the limit has changed. diff --git a/MAINTAINERS b/MAINTAINERS index 6a671ce87762..bf8690d0a1e1 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7564,14 +7564,6 @@ S: Maintained F: Documentation/ldm.txt F: block/partitions/ldm.* -LogFS -M: Joern Engel <joern@logfs.org> -M: Prasad Joshi <prasadjoshi.linux@gmail.com> -L: logfs@logfs.org -W: logfs.org -S: Maintained -F: fs/logfs/ - LSILOGIC MPT FUSION DRIVERS (FC/SAS/SPI) M: Sathya Prakash <sathya.prakash@broadcom.com> M: Chaitra P B <chaitra.basappa@broadcom.com> diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c index c4a75a18dcae..233e850fdac7 100644 --- a/drivers/bluetooth/hci_vhci.c +++ b/drivers/bluetooth/hci_vhci.c @@ -181,7 +181,7 @@ static inline ssize_t vhci_get_user(struct vhci_data *data, if (!skb) return -ENOMEM; - if (copy_from_iter(skb_put(skb, len), len, from) != len) { + if (!copy_from_iter_full(skb_put(skb, len), len, from)) { kfree_skb(skb); return -EFAULT; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index cc8aafd9cb0d..60bd4afe45c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2523,7 +2523,7 @@ static void amdgpu_debugfs_remove_files(struct amdgpu_device *adev) static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; bool pm_pg_lock, use_bank; @@ -2599,7 +2599,7 @@ end: static ssize_t amdgpu_debugfs_regs_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; bool pm_pg_lock, use_bank; @@ -2673,7 +2673,7 @@ static ssize_t amdgpu_debugfs_regs_write(struct file *f, const char __user *buf, static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2700,7 +2700,7 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf, static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2728,7 +2728,7 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2755,7 +2755,7 @@ static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf, static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2783,7 +2783,7 @@ static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2810,7 +2810,7 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf, static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -2838,7 +2838,7 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user * static ssize_t amdgpu_debugfs_gca_config_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; uint32_t *config, no_regs = 0; @@ -2908,7 +2908,7 @@ static ssize_t amdgpu_debugfs_gca_config_read(struct file *f, char __user *buf, static ssize_t amdgpu_debugfs_sensor_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; int idx, r; int32_t value; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 4c992826d2d6..a47628395914 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -280,7 +280,7 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring) static ssize_t amdgpu_debugfs_ring_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_ring *ring = (struct amdgpu_ring*)f->f_inode->i_private; + struct amdgpu_ring *ring = file_inode(f)->i_private; int r, i; uint32_t value, result, early[3]; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index bc70f80260d8..8e35c1ff59e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -1511,7 +1511,7 @@ static const struct drm_info_list amdgpu_ttm_debugfs_list[] = { static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; @@ -1555,7 +1555,7 @@ static const struct file_operations amdgpu_ttm_vram_fops = { static ssize_t amdgpu_ttm_gtt_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { - struct amdgpu_device *adev = f->f_inode->i_private; + struct amdgpu_device *adev = file_inode(f)->i_private; ssize_t result = 0; int r; diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 52a9d811be06..5c26653eceb5 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -679,7 +679,6 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m, int depth; bool zerocopy = false; size_t linear; - ssize_t n; if (q->flags & IFF_VNET_HDR) { vnet_hdr_len = q->vnet_hdr_sz; @@ -690,8 +689,7 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m, len -= vnet_hdr_len; err = -EFAULT; - n = copy_from_iter(&vnet_hdr, sizeof(vnet_hdr), from); - if (n != sizeof(vnet_hdr)) + if (!copy_from_iter_full(&vnet_hdr, sizeof(vnet_hdr), from)) goto err; iov_iter_advance(from, vnet_hdr_len - sizeof(vnet_hdr)); if ((vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) && diff --git a/drivers/net/tun.c b/drivers/net/tun.c index a569e61bc1d9..57e88b814700 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1156,7 +1156,6 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, bool zerocopy = false; int err; u32 rxhash; - ssize_t n; if (!(tun->dev->flags & IFF_UP)) return -EIO; @@ -1166,8 +1165,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, return -EINVAL; len -= sizeof(pi); - n = copy_from_iter(&pi, sizeof(pi), from); - if (n != sizeof(pi)) + if (!copy_from_iter_full(&pi, sizeof(pi), from)) return -EFAULT; } @@ -1176,8 +1174,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, return -EINVAL; len -= tun->vnet_hdr_sz; - n = copy_from_iter(&gso, sizeof(gso), from); - if (n != sizeof(gso)) + if (!copy_from_iter_full(&gso, sizeof(gso), from)) return -EFAULT; if ((gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) && diff --git a/drivers/staging/greybus/camera.c b/drivers/staging/greybus/camera.c index 1c5b41ae6774..0ee291ca2c72 100644 --- a/drivers/staging/greybus/camera.c +++ b/drivers/staging/greybus/camera.c @@ -1092,7 +1092,7 @@ static ssize_t gb_camera_debugfs_read(struct file *file, char __user *buf, size_t len, loff_t *offset) { const struct gb_camera_debugfs_entry *op = file->private_data; - struct gb_camera *gcam = file->f_inode->i_private; + struct gb_camera *gcam = file_inode(file)->i_private; struct gb_camera_debugfs_buffer *buffer; ssize_t ret; @@ -1114,7 +1114,7 @@ static ssize_t gb_camera_debugfs_write(struct file *file, loff_t *offset) { const struct gb_camera_debugfs_entry *op = file->private_data; - struct gb_camera *gcam = file->f_inode->i_private; + struct gb_camera *gcam = file_inode(file)->i_private; ssize_t ret; char *kbuf; diff --git a/drivers/staging/greybus/es2.c b/drivers/staging/greybus/es2.c index f1d256df06d5..c1929dfa9b31 100644 --- a/drivers/staging/greybus/es2.c +++ b/drivers/staging/greybus/es2.c @@ -1249,7 +1249,7 @@ static int apb_log_poll(void *data) static ssize_t apb_log_read(struct file *f, char __user *buf, size_t count, loff_t *ppos) { - struct es2_ap_dev *es2 = f->f_inode->i_private; + struct es2_ap_dev *es2 = file_inode(f)->i_private; ssize_t ret; size_t copied; char *tmp_buf; @@ -1303,7 +1303,7 @@ static void usb_log_disable(struct es2_ap_dev *es2) static ssize_t apb_log_enable_read(struct file *f, char __user *buf, size_t count, loff_t *ppos) { - struct es2_ap_dev *es2 = f->f_inode->i_private; + struct es2_ap_dev *es2 = file_inode(f)->i_private; int enable = !IS_ERR_OR_NULL(es2->apb_log_task); char tmp_buf[3]; @@ -1316,7 +1316,7 @@ static ssize_t apb_log_enable_write(struct file *f, const char __user *buf, { int enable; ssize_t retval; - struct es2_ap_dev *es2 = f->f_inode->i_private; + struct es2_ap_dev *es2 = file_inode(f)->i_private; retval = kstrtoint_from_user(buf, count, 10, &enable); if (retval) diff --git a/drivers/staging/greybus/svc.c b/drivers/staging/greybus/svc.c index 550055ec27a5..8779270cadc1 100644 --- a/drivers/staging/greybus/svc.c +++ b/drivers/staging/greybus/svc.c @@ -757,7 +757,7 @@ static int gb_svc_version_request(struct gb_operation *op) static ssize_t pwr_debugfs_voltage_read(struct file *file, char __user *buf, size_t len, loff_t *offset) { - struct svc_debugfs_pwrmon_rail *pwrmon_rails = file->f_inode->i_private; + struct svc_debugfs_pwrmon_rail *pwrmon_rails = file_inode(file)->i_private; struct gb_svc *svc = pwrmon_rails->svc; int ret, desc; u32 value; @@ -780,7 +780,7 @@ static ssize_t pwr_debugfs_voltage_read(struct file *file, char __user *buf, static ssize_t pwr_debugfs_current_read(struct file *file, char __user *buf, size_t len, loff_t *offset) { - struct svc_debugfs_pwrmon_rail *pwrmon_rails = file->f_inode->i_private; + struct svc_debugfs_pwrmon_rail *pwrmon_rails = file_inode(file)->i_private; struct gb_svc *svc = pwrmon_rails->svc; int ret, desc; u32 value; @@ -803,7 +803,7 @@ static ssize_t pwr_debugfs_current_read(struct file *file, char __user *buf, static ssize_t pwr_debugfs_power_read(struct file *file, char __user *buf, size_t len, loff_t *offset) { - struct svc_debugfs_pwrmon_rail *pwrmon_rails = file->f_inode->i_private; + struct svc_debugfs_pwrmon_rail *pwrmon_rails = file_inode(file)->i_private; struct gb_svc *svc = pwrmon_rails->svc; int ret, desc; u32 value; diff --git a/drivers/staging/greybus/timesync.c b/drivers/staging/greybus/timesync.c index e586627f4bbc..29e6c1c12807 100644 --- a/drivers/staging/greybus/timesync.c +++ b/drivers/staging/greybus/timesync.c @@ -921,7 +921,7 @@ EXPORT_SYMBOL_GPL(gb_timesync_schedule_asynchronous); static ssize_t gb_timesync_ping_read(struct file *file, char __user *ubuf, size_t len, loff_t *offset, bool ktime) { - struct gb_timesync_svc *timesync_svc = file->f_inode->i_private; + struct gb_timesync_svc *timesync_svc = file_inode(file)->i_private; char *buf; ssize_t ret = 0; diff --git a/drivers/staging/lustre/lustre/llite/dcache.c b/drivers/staging/lustre/lustre/llite/dcache.c index 0e45d8fc4d7c..65bf0c401b44 100644 --- a/drivers/staging/lustre/lustre/llite/dcache.c +++ b/drivers/staging/lustre/lustre/llite/dcache.c @@ -57,9 +57,6 @@ static void ll_release(struct dentry *de) LASSERT(de); lld = ll_d2d(de); - if (!lld) /* NFS copies the de->d_op methods (bug 4655) */ - return; - if (lld->lld_it) { ll_intent_release(lld->lld_it); kfree(lld->lld_it); @@ -126,30 +123,13 @@ static int ll_ddelete(const struct dentry *de) return 0; } -int ll_d_init(struct dentry *de) +static int ll_d_init(struct dentry *de) { - CDEBUG(D_DENTRY, "ldd on dentry %pd (%p) parent %p inode %p refc %d\n", - de, de, de->d_parent, d_inode(de), d_count(de)); - - if (!de->d_fsdata) { - struct ll_dentry_data *lld; - - lld = kzalloc(sizeof(*lld), GFP_NOFS); - if (likely(lld)) { - spin_lock(&de->d_lock); - if (likely(!de->d_fsdata)) { - de->d_fsdata = lld; - __d_lustre_invalidate(de); - } else { - kfree(lld); - } - spin_unlock(&de->d_lock); - } else { - return -ENOMEM; - } - } - LASSERT(de->d_op == &ll_d_ops); - + struct ll_dentry_data *lld = kzalloc(sizeof(*lld), GFP_KERNEL); + if (unlikely(!lld)) + return -ENOMEM; + lld->lld_invalid = 1; + de->d_fsdata = lld; return 0; } @@ -300,6 +280,7 @@ static int ll_revalidate_nd(struct dentry *dentry, unsigned int flags) } const struct dentry_operations ll_d_ops = { + .d_init = ll_d_init, .d_revalidate = ll_revalidate_nd, .d_release = ll_release, .d_delete = ll_ddelete, diff --git a/drivers/staging/lustre/lustre/llite/llite_internal.h b/drivers/staging/lustre/lustre/llite/llite_internal.h index 2f46d475cd7d..065a9a7e120a 100644 --- a/drivers/staging/lustre/lustre/llite/llite_internal.h +++ b/drivers/staging/lustre/lustre/llite/llite_internal.h @@ -769,7 +769,6 @@ int ll_hsm_release(struct inode *inode); /* llite/dcache.c */ -int ll_d_init(struct dentry *de); extern const struct dentry_operations ll_d_ops; void ll_intent_drop_lock(struct lookup_intent *); void ll_intent_release(struct lookup_intent *); @@ -1148,7 +1147,7 @@ dentry_may_statahead(struct inode *dir, struct dentry *dentry) * 'lld_sa_generation == lli->lli_sa_generation'. */ ldd = ll_d2d(dentry); - if (ldd && ldd->lld_sa_generation == lli->lli_sa_generation) + if (ldd->lld_sa_generation == lli->lli_sa_generation) return false; return true; @@ -1267,17 +1266,7 @@ static inline void ll_set_lock_data(struct obd_export *exp, struct inode *inode, static inline int d_lustre_invalid(const struct dentry *dentry) { - struct ll_dentry_data *lld = ll_d2d(dentry); - - return !lld || lld->lld_invalid; -} - -static inline void __d_lustre_invalidate(struct dentry *dentry) -{ - struct ll_dentry_data *lld = ll_d2d(dentry); - - if (lld) - lld->lld_invalid = 1; + return ll_d2d(dentry)->lld_invalid; } /* @@ -1293,7 +1282,7 @@ static inline void d_lustre_invalidate(struct dentry *dentry, int nested) spin_lock_nested(&dentry->d_lock, nested ? DENTRY_D_LOCK_NESTED : DENTRY_D_LOCK_NORMAL); - __d_lustre_invalidate(dentry); + ll_d2d(dentry)->lld_invalid = 1; /* * We should be careful about dentries created by d_obtain_alias(). * These dentries are not put in the dentry tree, instead they are diff --git a/drivers/staging/lustre/lustre/llite/llite_nfs.c b/drivers/staging/lustre/lustre/llite/llite_nfs.c index c63236580b0f..49a930f0fc5d 100644 --- a/drivers/staging/lustre/lustre/llite/llite_nfs.c +++ b/drivers/staging/lustre/lustre/llite/llite_nfs.c @@ -169,22 +169,12 @@ ll_iget_for_nfs(struct super_block *sb, struct lu_fid *fid, struct lu_fid *paren /* N.B. d_obtain_alias() drops inode ref on error */ result = d_obtain_alias(inode); if (!IS_ERR(result)) { - int rc; - - rc = ll_d_init(result); - if (rc < 0) { - dput(result); - result = ERR_PTR(rc); - } else { - struct ll_dentry_data *ldd = ll_d2d(result); - - /* - * Need to signal to the ll_intent_file_open that - * we came from NFS and so opencache needs to be - * enabled for this one - */ - ldd->lld_nfs_dentry = 1; - } + /* + * Need to signal to the ll_intent_file_open that + * we came from NFS and so opencache needs to be + * enabled for this one + */ + ll_d2d(result)->lld_nfs_dentry = 1; } return result; diff --git a/drivers/staging/lustre/lustre/llite/namei.c b/drivers/staging/lustre/lustre/llite/namei.c index 9426759aedc9..a8f4e7fb0a46 100644 --- a/drivers/staging/lustre/lustre/llite/namei.c +++ b/drivers/staging/lustre/lustre/llite/namei.c @@ -432,17 +432,9 @@ static struct dentry *ll_find_alias(struct inode *inode, struct dentry *dentry) */ struct dentry *ll_splice_alias(struct inode *inode, struct dentry *de) { - struct dentry *new; - int rc; - if (inode) { - new = ll_find_alias(inode, de); + struct dentry *new = ll_find_alias(inode, de); if (new) { - rc = ll_d_init(new); - if (rc < 0) { - dput(new); - return ERR_PTR(rc); - } d_move(new, de); iput(inode); CDEBUG(D_DENTRY, @@ -451,9 +443,6 @@ struct dentry *ll_splice_alias(struct inode *inode, struct dentry *de) return new; } } - rc = ll_d_init(de); - if (rc < 0) - return ERR_PTR(rc); d_add(de, inode); CDEBUG(D_DENTRY, "Add dentry %p inode %p refc %d flags %#x\n", de, d_inode(de), d_count(de), de->d_flags); diff --git a/drivers/staging/lustre/lustre/llite/statahead.c b/drivers/staging/lustre/lustre/llite/statahead.c index 4769a2230ae1..f1ee17f9ec0d 100644 --- a/drivers/staging/lustre/lustre/llite/statahead.c +++ b/drivers/staging/lustre/lustre/llite/statahead.c @@ -1519,9 +1519,7 @@ out_unplug: * dentry_may_statahead(). */ ldd = ll_d2d(*dentryp); - /* ldd can be NULL if llite lookup failed. */ - if (ldd) - ldd->lld_sa_generation = lli->lli_sa_generation; + ldd->lld_sa_generation = lli->lli_sa_generation; sa_put(sai, entry); return rc; } diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c index 2001005bef45..a35a347ec357 100644 --- a/drivers/target/target_core_configfs.c +++ b/drivers/target/target_core_configfs.c @@ -143,7 +143,7 @@ static ssize_t target_core_item_dbroot_store(struct config_item *item, pr_err("db_root: cannot open: %s\n", db_root_stage); return -EINVAL; } - if (!S_ISDIR(fp->f_inode->i_mode)) { + if (!S_ISDIR(file_inode(fp)->i_mode)) { filp_close(fp, 0); mutex_unlock(&g_tf_lock); pr_err("db_root: not a directory: %s\n", db_root_stage); diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index 0780d8311ec6..aab3fc1dbb94 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -949,7 +949,7 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data) goto error_mutex; } if (!io_data->read && - copy_from_iter(data, data_len, &io_data->data) != data_len) { + !copy_from_iter_full(data, data_len, &io_data->data)) { ret = -EFAULT; goto error_mutex; } diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c index bd82dd12deff..10b2576f8b6a 100644 --- a/drivers/usb/gadget/legacy/inode.c +++ b/drivers/usb/gadget/legacy/inode.c @@ -667,7 +667,7 @@ ep_write_iter(struct kiocb *iocb, struct iov_iter *from) return -ENOMEM; } - if (unlikely(copy_from_iter(buf, len, from) != len)) { + if (unlikely(!copy_from_iter_full(buf, len, from))) { value = -EFAULT; goto out; } diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c index e2be447752c2..253310cdaaca 100644 --- a/drivers/vhost/scsi.c +++ b/drivers/vhost/scsi.c @@ -922,8 +922,7 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq) */ iov_iter_init(&out_iter, WRITE, vq->iov, out, out_size); - ret = copy_from_iter(req, req_size, &out_iter); - if (unlikely(ret != req_size)) { + if (unlikely(!copy_from_iter_full(req, req_size, &out_iter))) { vq_err(vq, "Faulted on copy_from_iter\n"); vhost_scsi_send_bad_target(vs, vq, head, out); continue; diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index c0f81e8cb50c..d6432603880c 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -1863,8 +1863,7 @@ static int get_indirect(struct vhost_virtqueue *vq, i, count); return -EINVAL; } - if (unlikely(copy_from_iter(&desc, sizeof(desc), &from) != - sizeof(desc))) { + if (unlikely(!copy_from_iter_full(&desc, sizeof(desc), &from))) { vq_err(vq, "Failed indirect descriptor: idx %d, %zx\n", i, (size_t)vhost64_to_cpu(vq, indirect->addr) + i * sizeof desc); return -EINVAL; diff --git a/fs/Kconfig b/fs/Kconfig index 8e9e5f4104f4..c2a377cdda2b 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -234,7 +234,6 @@ source "fs/efs/Kconfig" source "fs/jffs2/Kconfig" # UBIFS File system configuration source "fs/ubifs/Kconfig" -source "fs/logfs/Kconfig" source "fs/cramfs/Kconfig" source "fs/squashfs/Kconfig" source "fs/freevxfs/Kconfig" diff --git a/fs/Makefile b/fs/Makefile index ed2b63257ba9..7bbaca9c67b1 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -97,7 +97,6 @@ obj-$(CONFIG_NTFS_FS) += ntfs/ obj-$(CONFIG_UFS_FS) += ufs/ obj-$(CONFIG_EFS_FS) += efs/ obj-$(CONFIG_JFFS2_FS) += jffs2/ -obj-$(CONFIG_LOGFS) += logfs/ obj-$(CONFIG_UBIFS_FS) += ubifs/ obj-$(CONFIG_AFFS_FS) += affs/ obj-$(CONFIG_ROMFS_FS) += romfs/ @@ -277,10 +277,10 @@ static void put_aio_ring_file(struct kioctx *ctx) struct address_space *i_mapping; if (aio_ring_file) { - truncate_setsize(aio_ring_file->f_inode, 0); + truncate_setsize(file_inode(aio_ring_file), 0); /* Prevent further access to the kioctx from migratepages */ - i_mapping = aio_ring_file->f_inode->i_mapping; + i_mapping = aio_ring_file->f_mapping; spin_lock(&i_mapping->private_lock); i_mapping->private_data = NULL; ctx->aio_ring_file = NULL; @@ -483,7 +483,7 @@ static int aio_setup_ring(struct kioctx *ctx) for (i = 0; i < nr_pages; i++) { struct page *page; - page = find_or_create_page(file->f_inode->i_mapping, + page = find_or_create_page(file->f_mapping, i, GFP_HIGHUSER | __GFP_ZERO); if (!page) break; diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 438b5bf675b6..09e7d68dff02 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -94,7 +94,7 @@ static int autofs4_show_options(struct seq_file *m, struct dentry *root) seq_printf(m, ",indirect"); #ifdef CONFIG_CHECKPOINT_RESTORE if (sbi->pipe) - seq_printf(m, ",pipe_ino=%ld", sbi->pipe->f_inode->i_ino); + seq_printf(m, ",pipe_ino=%ld", file_inode(sbi->pipe)->i_ino); else seq_printf(m, ",pipe_ino=-1"); #endif diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index a594c7879cc2..d7a93696663b 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -32,40 +32,19 @@ const struct dentry_operations ceph_dentry_ops; /* * Initialize ceph dentry state. */ -int ceph_init_dentry(struct dentry *dentry) +static int ceph_d_init(struct dentry *dentry) { struct ceph_dentry_info *di; - if (dentry->d_fsdata) - return 0; - di = kmem_cache_zalloc(ceph_dentry_cachep, GFP_KERNEL); if (!di) return -ENOMEM; /* oh well */ - spin_lock(&dentry->d_lock); - if (dentry->d_fsdata) { - /* lost a race */ - kmem_cache_free(ceph_dentry_cachep, di); - goto out_unlock; - } - - if (ceph_snap(d_inode(dentry->d_parent)) == CEPH_NOSNAP) - d_set_d_op(dentry, &ceph_dentry_ops); - else if (ceph_snap(d_inode(dentry->d_parent)) == CEPH_SNAPDIR) - d_set_d_op(dentry, &ceph_snapdir_dentry_ops); - else - d_set_d_op(dentry, &ceph_snap_dentry_ops); - di->dentry = dentry; di->lease_session = NULL; di->time = jiffies; - /* avoid reordering d_fsdata setup so that the check above is safe */ - smp_mb(); dentry->d_fsdata = di; ceph_dentry_lru_add(dentry); -out_unlock: - spin_unlock(&dentry->d_lock); return 0; } @@ -737,10 +716,6 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry, if (dentry->d_name.len > NAME_MAX) return ERR_PTR(-ENAMETOOLONG); - err = ceph_init_dentry(dentry); - if (err < 0) - return ERR_PTR(err); - /* can we conclude ENOENT locally? */ if (d_really_is_negative(dentry)) { struct ceph_inode_info *ci = ceph_inode(dir); @@ -1323,16 +1298,6 @@ static void ceph_d_release(struct dentry *dentry) kmem_cache_free(ceph_dentry_cachep, di); } -static int ceph_snapdir_d_revalidate(struct dentry *dentry, - unsigned int flags) -{ - /* - * Eventually, we'll want to revalidate snapped metadata - * too... probably... - */ - return 1; -} - /* * When the VFS prunes a dentry from the cache, we need to clear the * complete flag on the parent directory. @@ -1351,6 +1316,9 @@ static void ceph_d_prune(struct dentry *dentry) if (d_unhashed(dentry)) return; + if (ceph_snap(d_inode(dentry->d_parent)) == CEPH_SNAPDIR) + return; + /* * we hold d_lock, so d_parent is stable, and d_fsdata is never * cleared until d_release @@ -1521,14 +1489,5 @@ const struct dentry_operations ceph_dentry_ops = { .d_revalidate = ceph_d_revalidate, .d_release = ceph_d_release, .d_prune = ceph_d_prune, -}; - -const struct dentry_operations ceph_snapdir_dentry_ops = { - .d_revalidate = ceph_snapdir_d_revalidate, - .d_release = ceph_d_release, -}; - -const struct dentry_operations ceph_snap_dentry_ops = { - .d_release = ceph_d_release, - .d_prune = ceph_d_prune, + .d_init = ceph_d_init, }; diff --git a/fs/ceph/export.c b/fs/ceph/export.c index 1780218a48f0..180bbef760f2 100644 --- a/fs/ceph/export.c +++ b/fs/ceph/export.c @@ -62,7 +62,6 @@ static struct dentry *__fh_to_dentry(struct super_block *sb, u64 ino) { struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc; struct inode *inode; - struct dentry *dentry; struct ceph_vino vino; int err; @@ -94,16 +93,7 @@ static struct dentry *__fh_to_dentry(struct super_block *sb, u64 ino) return ERR_PTR(-ESTALE); } - dentry = d_obtain_alias(inode); - if (IS_ERR(dentry)) - return dentry; - err = ceph_init_dentry(dentry); - if (err < 0) { - dput(dentry); - return ERR_PTR(err); - } - dout("__fh_to_dentry %llx %p dentry %p\n", ino, inode, dentry); - return dentry; + return d_obtain_alias(inode); } /* @@ -131,7 +121,6 @@ static struct dentry *__get_parent(struct super_block *sb, struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc; struct ceph_mds_request *req; struct inode *inode; - struct dentry *dentry; int mask; int err; @@ -164,18 +153,7 @@ static struct dentry *__get_parent(struct super_block *sb, if (!inode) return ERR_PTR(-ENOENT); - dentry = d_obtain_alias(inode); - if (IS_ERR(dentry)) - return dentry; - err = ceph_init_dentry(dentry); - if (err < 0) { - dput(dentry); - return ERR_PTR(err); - } - dout("__get_parent ino %llx parent %p ino %llx.%llx\n", - child ? ceph_ino(d_inode(child)) : ino, - dentry, ceph_vinop(inode)); - return dentry; + return d_obtain_alias(inode); } static struct dentry *ceph_get_parent(struct dentry *child) diff --git a/fs/ceph/file.c b/fs/ceph/file.c index f995e3528a33..159fc8f1a6a0 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -351,10 +351,6 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, if (dentry->d_name.len > NAME_MAX) return -ENAMETOOLONG; - err = ceph_init_dentry(dentry); - if (err < 0) - return err; - if (flags & O_CREAT) { err = ceph_pre_init_acls(dir, &mode, &acls); if (err < 0) diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index ef4d04647325..284f0d807151 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -1023,16 +1023,17 @@ static void update_dentry_lease(struct dentry *dentry, long unsigned half_ttl = from_time + (duration * HZ / 2) / 1000; struct inode *dir; - /* only track leases on regular dentries */ - if (dentry->d_op != &ceph_dentry_ops) - return; - spin_lock(&dentry->d_lock); dout("update_dentry_lease %p duration %lu ms ttl %lu\n", dentry, duration, ttl); /* make lease_rdcache_gen match directory */ dir = d_inode(dentry->d_parent); + + /* only track leases on regular dentries */ + if (ceph_snap(dir) != CEPH_NOSNAP) + goto out_unlock; + di->lease_shared_gen = ceph_inode(dir)->i_shared_gen; if (duration == 0) @@ -1202,12 +1203,7 @@ retry_lookup: err = -ENOMEM; goto done; } - err = ceph_init_dentry(dn); - if (err < 0) { - dput(dn); - dput(parent); - goto done; - } + err = 0; } else if (d_really_is_positive(dn) && (ceph_ino(d_inode(dn)) != vino.ino || ceph_snap(d_inode(dn)) != vino.snap)) { @@ -1561,12 +1557,6 @@ retry_lookup: err = -ENOMEM; goto out; } - ret = ceph_init_dentry(dn); - if (ret < 0) { - dput(dn); - err = ret; - goto out; - } } else if (d_really_is_positive(dn) && (ceph_ino(d_inode(dn)) != vino.ino || ceph_snap(d_inode(dn)) != vino.snap)) { diff --git a/fs/ceph/super.c b/fs/ceph/super.c index b382e5910eea..f2f76696ddca 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -795,7 +795,6 @@ static struct dentry *open_root_dentry(struct ceph_fs_client *fsc, root = ERR_PTR(-ENOMEM); goto out; } - ceph_init_dentry(root); dout("open_root_inode success, root dentry is %p\n", root); } else { root = ERR_PTR(err); @@ -879,6 +878,7 @@ static int ceph_set_super(struct super_block *s, void *data) fsc->sb = s; s->s_op = &ceph_super_ops; + s->s_d_op = &ceph_dentry_ops; s->s_export_op = &ceph_export_ops; s->s_time_gran = 1000; /* 1000 ns == 1 us */ diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 3e3fa9163059..931687f71a7c 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -934,8 +934,7 @@ extern const struct file_operations ceph_dir_fops; extern const struct file_operations ceph_snapdir_fops; extern const struct inode_operations ceph_dir_iops; extern const struct inode_operations ceph_snapdir_iops; -extern const struct dentry_operations ceph_dentry_ops, ceph_snap_dentry_ops, - ceph_snapdir_dentry_ops; +extern const struct dentry_operations ceph_dentry_ops; extern loff_t ceph_make_fpos(unsigned high, unsigned off, bool hash_order); extern int ceph_handle_notrace_create(struct inode *dir, struct dentry *dentry); @@ -951,13 +950,6 @@ extern void ceph_invalidate_dentry_lease(struct dentry *dentry); extern unsigned ceph_dentry_hash(struct inode *dir, struct dentry *dn); extern void ceph_readdir_cache_release(struct ceph_readdir_cache_control *ctl); -/* - * our d_ops vary depending on whether the inode is live, - * snapshotted (read-only), or a virtual ".snap" directory. - */ -int ceph_init_dentry(struct dentry *dentry); - - /* ioctl.c */ extern long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg); diff --git a/fs/compat.c b/fs/compat.c index bd064a2c3550..543b48c29ac3 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -253,9 +253,9 @@ COMPAT_SYSCALL_DEFINE2(fstatfs, unsigned int, fd, struct compat_statfs __user *, static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstatfs *kbuf) { - if (sizeof ubuf->f_blocks == 4) { - if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail | - kbuf->f_bsize | kbuf->f_frsize) & 0xffffffff00000000ULL) + if (sizeof(ubuf->f_bsize) == 4) { + if ((kbuf->f_type | kbuf->f_bsize | kbuf->f_namelen | + kbuf->f_frsize | kbuf->f_flags) & 0xffffffff00000000ULL) return -EOVERFLOW; /* f_files and f_ffree may be -1; it's okay * to stuff that into 32 bits */ diff --git a/fs/fcntl.c b/fs/fcntl.c index 350a2c8cfd28..6e2771c210f6 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -52,7 +52,7 @@ static int setfl(int fd, struct file * filp, unsigned long arg) arg |= O_NONBLOCK; /* Pipe packetized mode is controlled by O_DIRECT flag */ - if (!S_ISFIFO(filp->f_inode->i_mode) && (arg & O_DIRECT)) { + if (!S_ISFIFO(inode->i_mode) && (arg & O_DIRECT)) { if (!filp->f_mapping || !filp->f_mapping->a_ops || !filp->f_mapping->a_ops->direct_IO) return -EINVAL; diff --git a/fs/logfs/Kconfig b/fs/logfs/Kconfig deleted file mode 100644 index 2b4503163930..000000000000 --- a/fs/logfs/Kconfig +++ /dev/null @@ -1,17 +0,0 @@ -config LOGFS - tristate "LogFS file system" - depends on MTD || (!MTD && BLOCK) - select ZLIB_INFLATE - select ZLIB_DEFLATE - select CRC32 - select BTREE - help - Flash filesystem aimed to scale efficiently to large devices. - In comparison to JFFS2 it offers significantly faster mount - times and potentially less RAM usage, although the latter has - not been measured yet. - - In its current state it is still very experimental and should - not be used for other than testing purposes. - - If unsure, say N. diff --git a/fs/logfs/Makefile b/fs/logfs/Makefile deleted file mode 100644 index 4820027787ee..000000000000 --- a/fs/logfs/Makefile +++ /dev/null @@ -1,13 +0,0 @@ -obj-$(CONFIG_LOGFS) += logfs.o - -logfs-y += compr.o -logfs-y += dir.o -logfs-y += file.o -logfs-y += gc.o -logfs-y += inode.o -logfs-y += journal.o -logfs-y += readwrite.o -logfs-y += segment.o -logfs-y += super.o -logfs-$(CONFIG_BLOCK) += dev_bdev.o -logfs-$(CONFIG_MTD) += dev_mtd.o diff --git a/fs/logfs/compr.c b/fs/logfs/compr.c deleted file mode 100644 index 961f02b86d97..000000000000 --- a/fs/logfs/compr.c +++ /dev/null @@ -1,95 +0,0 @@ -/* - * fs/logfs/compr.c - compression routines - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - */ -#include "logfs.h" -#include <linux/vmalloc.h> -#include <linux/zlib.h> - -#define COMPR_LEVEL 3 - -static DEFINE_MUTEX(compr_mutex); -static struct z_stream_s stream; - -int logfs_compress(void *in, void *out, size_t inlen, size_t outlen) -{ - int err, ret; - - ret = -EIO; - mutex_lock(&compr_mutex); - err = zlib_deflateInit(&stream, COMPR_LEVEL); - if (err != Z_OK) - goto error; - - stream.next_in = in; - stream.avail_in = inlen; - stream.total_in = 0; - stream.next_out = out; - stream.avail_out = outlen; - stream.total_out = 0; - - err = zlib_deflate(&stream, Z_FINISH); - if (err != Z_STREAM_END) - goto error; - - err = zlib_deflateEnd(&stream); - if (err != Z_OK) - goto error; - - if (stream.total_out >= stream.total_in) - goto error; - - ret = stream.total_out; -error: - mutex_unlock(&compr_mutex); - return ret; -} - -int logfs_uncompress(void *in, void *out, size_t inlen, size_t outlen) -{ - int err, ret; - - ret = -EIO; - mutex_lock(&compr_mutex); - err = zlib_inflateInit(&stream); - if (err != Z_OK) - goto error; - - stream.next_in = in; - stream.avail_in = inlen; - stream.total_in = 0; - stream.next_out = out; - stream.avail_out = outlen; - stream.total_out = 0; - - err = zlib_inflate(&stream, Z_FINISH); - if (err != Z_STREAM_END) - goto error; - - err = zlib_inflateEnd(&stream); - if (err != Z_OK) - goto error; - - ret = 0; -error: - mutex_unlock(&compr_mutex); - return ret; -} - -int __init logfs_compr_init(void) -{ - size_t size = max(zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL), - zlib_inflate_workspacesize()); - stream.workspace = vmalloc(size); - if (!stream.workspace) - return -ENOMEM; - return 0; -} - -void logfs_compr_exit(void) -{ - vfree(stream.workspace); -} diff --git a/fs/logfs/dev_bdev.c b/fs/logfs/dev_bdev.c deleted file mode 100644 index 9bfa0151d7c9..000000000000 --- a/fs/logfs/dev_bdev.c +++ /dev/null @@ -1,290 +0,0 @@ -/* - * fs/logfs/dev_bdev.c - Device access methods for block devices - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - */ -#include "logfs.h" -#include <linux/bio.h> -#include <linux/blkdev.h> -#include <linux/buffer_head.h> -#include <linux/gfp.h> -#include <linux/prefetch.h> - -#define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1)) - -static int sync_request(struct page *page, struct block_device *bdev, int op) -{ - struct bio bio; - struct bio_vec bio_vec; - - bio_init(&bio, &bio_vec, 1); - bio.bi_bdev = bdev; - bio_add_page(&bio, page, PAGE_SIZE, 0); - bio.bi_iter.bi_sector = page->index * (PAGE_SIZE >> 9); - bio_set_op_attrs(&bio, op, 0); - - return submit_bio_wait(&bio); -} - -static int bdev_readpage(void *_sb, struct page *page) -{ - struct super_block *sb = _sb; - struct block_device *bdev = logfs_super(sb)->s_bdev; - int err; - - err = sync_request(page, bdev, READ); - if (err) { - ClearPageUptodate(page); - SetPageError(page); - } else { - SetPageUptodate(page); - ClearPageError(page); - } - unlock_page(page); - return err; -} - -static DECLARE_WAIT_QUEUE_HEAD(wq); - -static void writeseg_end_io(struct bio *bio) -{ - struct bio_vec *bvec; - int i; - struct super_block *sb = bio->bi_private; - struct logfs_super *super = logfs_super(sb); - - BUG_ON(bio->bi_error); /* FIXME: Retry io or write elsewhere */ - - bio_for_each_segment_all(bvec, bio, i) { - end_page_writeback(bvec->bv_page); - put_page(bvec->bv_page); - } - bio_put(bio); - if (atomic_dec_and_test(&super->s_pending_writes)) - wake_up(&wq); -} - -static int __bdev_writeseg(struct super_block *sb, u64 ofs, pgoff_t index, - size_t nr_pages) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - struct bio *bio = NULL; - struct page *page; - unsigned int max_pages; - int i, ret; - - max_pages = min_t(size_t, nr_pages, BIO_MAX_PAGES); - - for (i = 0; i < nr_pages; i++) { - if (!bio) { - bio = bio_alloc(GFP_NOFS, max_pages); - BUG_ON(!bio); - - bio->bi_bdev = super->s_bdev; - bio->bi_iter.bi_sector = ofs >> 9; - bio->bi_private = sb; - bio->bi_end_io = writeseg_end_io; - bio_set_op_attrs(bio, REQ_OP_WRITE, 0); - } - page = find_lock_page(mapping, index + i); - BUG_ON(!page); - ret = bio_add_page(bio, page, PAGE_SIZE, 0); - - BUG_ON(PageWriteback(page)); - set_page_writeback(page); - unlock_page(page); - - if (!ret) { - /* Block layer cannot split bios :( */ - ofs += bio->bi_iter.bi_size; - atomic_inc(&super->s_pending_writes); - submit_bio(bio); - bio = NULL; - } - } - - if (bio) { - atomic_inc(&super->s_pending_writes); - submit_bio(bio); - } - return 0; -} - -static void bdev_writeseg(struct super_block *sb, u64 ofs, size_t len) -{ - struct logfs_super *super = logfs_super(sb); - int head; - - BUG_ON(super->s_flags & LOGFS_SB_FLAG_RO); - - if (len == 0) { - /* This can happen when the object fit perfectly into a - * segment, the segment gets written per sync and subsequently - * closed. - */ - return; - } - head = ofs & (PAGE_SIZE - 1); - if (head) { - ofs -= head; - len += head; - } - len = PAGE_ALIGN(len); - __bdev_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT); -} - - -static void erase_end_io(struct bio *bio) -{ - struct super_block *sb = bio->bi_private; - struct logfs_super *super = logfs_super(sb); - - BUG_ON(bio->bi_error); /* FIXME: Retry io or write elsewhere */ - bio_put(bio); - if (atomic_dec_and_test(&super->s_pending_writes)) - wake_up(&wq); -} - -static int do_erase(struct super_block *sb, u64 ofs, pgoff_t index, - size_t nr_pages) -{ - struct logfs_super *super = logfs_super(sb); - struct bio *bio = NULL; - unsigned int max_pages; - int i, ret; - - max_pages = min_t(size_t, nr_pages, BIO_MAX_PAGES); - - for (i = 0; i < nr_pages; i++) { - if (!bio) { - bio = bio_alloc(GFP_NOFS, max_pages); - BUG_ON(!bio); - - bio->bi_bdev = super->s_bdev; - bio->bi_iter.bi_sector = ofs >> 9; - bio->bi_private = sb; - bio->bi_end_io = erase_end_io; - bio_set_op_attrs(bio, REQ_OP_WRITE, 0); - } - ret = bio_add_page(bio, super->s_erase_page, PAGE_SIZE, 0); - if (!ret) { - /* Block layer cannot split bios :( */ - ofs += bio->bi_iter.bi_size; - atomic_inc(&super->s_pending_writes); - submit_bio(bio); - } - } - if (bio) { - atomic_inc(&super->s_pending_writes); - submit_bio(bio); - } - return 0; -} - -static int bdev_erase(struct super_block *sb, loff_t to, size_t len, - int ensure_write) -{ - struct logfs_super *super = logfs_super(sb); - - BUG_ON(to & (PAGE_SIZE - 1)); - BUG_ON(len & (PAGE_SIZE - 1)); - - if (super->s_flags & LOGFS_SB_FLAG_RO) - return -EROFS; - - if (ensure_write) { - /* - * Object store doesn't care whether erases happen or not. - * But for the journal they are required. Otherwise a scan - * can find an old commit entry and assume it is the current - * one, travelling back in time. - */ - do_erase(sb, to, to >> PAGE_SHIFT, len >> PAGE_SHIFT); - } - - return 0; -} - -static void bdev_sync(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - - wait_event(wq, atomic_read(&super->s_pending_writes) == 0); -} - -static struct page *bdev_find_first_sb(struct super_block *sb, u64 *ofs) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - filler_t *filler = bdev_readpage; - - *ofs = 0; - return read_cache_page(mapping, 0, filler, sb); -} - -static struct page *bdev_find_last_sb(struct super_block *sb, u64 *ofs) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - filler_t *filler = bdev_readpage; - u64 pos = (super->s_bdev->bd_inode->i_size & ~0xfffULL) - 0x1000; - pgoff_t index = pos >> PAGE_SHIFT; - - *ofs = pos; - return read_cache_page(mapping, index, filler, sb); -} - -static int bdev_write_sb(struct super_block *sb, struct page *page) -{ - struct block_device *bdev = logfs_super(sb)->s_bdev; - - /* Nothing special to do for block devices. */ - return sync_request(page, bdev, WRITE); -} - -static void bdev_put_device(struct logfs_super *s) -{ - blkdev_put(s->s_bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL); -} - -static int bdev_can_write_buf(struct super_block *sb, u64 ofs) -{ - return 0; -} - -static const struct logfs_device_ops bd_devops = { - .find_first_sb = bdev_find_first_sb, - .find_last_sb = bdev_find_last_sb, - .write_sb = bdev_write_sb, - .readpage = bdev_readpage, - .writeseg = bdev_writeseg, - .erase = bdev_erase, - .can_write_buf = bdev_can_write_buf, - .sync = bdev_sync, - .put_device = bdev_put_device, -}; - -int logfs_get_sb_bdev(struct logfs_super *p, struct file_system_type *type, - const char *devname) -{ - struct block_device *bdev; - - bdev = blkdev_get_by_path(devname, FMODE_READ|FMODE_WRITE|FMODE_EXCL, - type); - if (IS_ERR(bdev)) - return PTR_ERR(bdev); - - if (MAJOR(bdev->bd_dev) == MTD_BLOCK_MAJOR) { - int mtdnr = MINOR(bdev->bd_dev); - blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL); - return logfs_get_sb_mtd(p, mtdnr); - } - - p->s_bdev = bdev; - p->s_mtd = NULL; - p->s_devops = &bd_devops; - return 0; -} diff --git a/fs/logfs/dev_mtd.c b/fs/logfs/dev_mtd.c deleted file mode 100644 index b76a62b1978f..000000000000 --- a/fs/logfs/dev_mtd.c +++ /dev/null @@ -1,274 +0,0 @@ -/* - * fs/logfs/dev_mtd.c - Device access methods for MTD - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - */ -#include "logfs.h" -#include <linux/completion.h> -#include <linux/mount.h> -#include <linux/sched.h> -#include <linux/slab.h> - -#define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1)) - -static int logfs_mtd_read(struct super_block *sb, loff_t ofs, size_t len, - void *buf) -{ - struct mtd_info *mtd = logfs_super(sb)->s_mtd; - size_t retlen; - int ret; - - ret = mtd_read(mtd, ofs, len, &retlen, buf); - BUG_ON(ret == -EINVAL); - if (ret) - return ret; - - /* Not sure if we should loop instead. */ - if (retlen != len) - return -EIO; - - return 0; -} - -static int loffs_mtd_write(struct super_block *sb, loff_t ofs, size_t len, - void *buf) -{ - struct logfs_super *super = logfs_super(sb); - struct mtd_info *mtd = super->s_mtd; - size_t retlen; - loff_t page_start, page_end; - int ret; - - if (super->s_flags & LOGFS_SB_FLAG_RO) - return -EROFS; - - BUG_ON((ofs >= mtd->size) || (len > mtd->size - ofs)); - BUG_ON(ofs != (ofs >> super->s_writeshift) << super->s_writeshift); - BUG_ON(len > PAGE_SIZE); - page_start = ofs & PAGE_MASK; - page_end = PAGE_ALIGN(ofs + len) - 1; - ret = mtd_write(mtd, ofs, len, &retlen, buf); - if (ret || (retlen != len)) - return -EIO; - - return 0; -} - -/* - * For as long as I can remember (since about 2001) mtd->erase has been an - * asynchronous interface lacking the first driver to actually use the - * asynchronous properties. So just to prevent the first implementor of such - * a thing from breaking logfs in 2350, we do the usual pointless dance to - * declare a completion variable and wait for completion before returning - * from logfs_mtd_erase(). What an exercise in futility! - */ -static void logfs_erase_callback(struct erase_info *ei) -{ - complete((struct completion *)ei->priv); -} - -static int logfs_mtd_erase_mapping(struct super_block *sb, loff_t ofs, - size_t len) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - struct page *page; - pgoff_t index = ofs >> PAGE_SHIFT; - - for (index = ofs >> PAGE_SHIFT; index < (ofs + len) >> PAGE_SHIFT; index++) { - page = find_get_page(mapping, index); - if (!page) - continue; - memset(page_address(page), 0xFF, PAGE_SIZE); - put_page(page); - } - return 0; -} - -static int logfs_mtd_erase(struct super_block *sb, loff_t ofs, size_t len, - int ensure_write) -{ - struct mtd_info *mtd = logfs_super(sb)->s_mtd; - struct erase_info ei; - DECLARE_COMPLETION_ONSTACK(complete); - int ret; - - BUG_ON(len % mtd->erasesize); - if (logfs_super(sb)->s_flags & LOGFS_SB_FLAG_RO) - return -EROFS; - - memset(&ei, 0, sizeof(ei)); - ei.mtd = mtd; - ei.addr = ofs; - ei.len = len; - ei.callback = logfs_erase_callback; - ei.priv = (long)&complete; - ret = mtd_erase(mtd, &ei); - if (ret) - return -EIO; - - wait_for_completion(&complete); - if (ei.state != MTD_ERASE_DONE) - return -EIO; - return logfs_mtd_erase_mapping(sb, ofs, len); -} - -static void logfs_mtd_sync(struct super_block *sb) -{ - struct mtd_info *mtd = logfs_super(sb)->s_mtd; - - mtd_sync(mtd); -} - -static int logfs_mtd_readpage(void *_sb, struct page *page) -{ - struct super_block *sb = _sb; - int err; - - err = logfs_mtd_read(sb, page->index << PAGE_SHIFT, PAGE_SIZE, - page_address(page)); - if (err == -EUCLEAN || err == -EBADMSG) { - /* -EBADMSG happens regularly on power failures */ - err = 0; - /* FIXME: force GC this segment */ - } - if (err) { - ClearPageUptodate(page); - SetPageError(page); - } else { - SetPageUptodate(page); - ClearPageError(page); - } - unlock_page(page); - return err; -} - -static struct page *logfs_mtd_find_first_sb(struct super_block *sb, u64 *ofs) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - filler_t *filler = logfs_mtd_readpage; - struct mtd_info *mtd = super->s_mtd; - - *ofs = 0; - while (mtd_block_isbad(mtd, *ofs)) { - *ofs += mtd->erasesize; - if (*ofs >= mtd->size) - return NULL; - } - BUG_ON(*ofs & ~PAGE_MASK); - return read_cache_page(mapping, *ofs >> PAGE_SHIFT, filler, sb); -} - -static struct page *logfs_mtd_find_last_sb(struct super_block *sb, u64 *ofs) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - filler_t *filler = logfs_mtd_readpage; - struct mtd_info *mtd = super->s_mtd; - - *ofs = mtd->size - mtd->erasesize; - while (mtd_block_isbad(mtd, *ofs)) { - *ofs -= mtd->erasesize; - if (*ofs <= 0) - return NULL; - } - *ofs = *ofs + mtd->erasesize - 0x1000; - BUG_ON(*ofs & ~PAGE_MASK); - return read_cache_page(mapping, *ofs >> PAGE_SHIFT, filler, sb); -} - -static int __logfs_mtd_writeseg(struct super_block *sb, u64 ofs, pgoff_t index, - size_t nr_pages) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - struct page *page; - int i, err; - - for (i = 0; i < nr_pages; i++) { - page = find_lock_page(mapping, index + i); - BUG_ON(!page); - - err = loffs_mtd_write(sb, page->index << PAGE_SHIFT, PAGE_SIZE, - page_address(page)); - unlock_page(page); - put_page(page); - if (err) - return err; - } - return 0; -} - -static void logfs_mtd_writeseg(struct super_block *sb, u64 ofs, size_t len) -{ - struct logfs_super *super = logfs_super(sb); - int head; - - if (super->s_flags & LOGFS_SB_FLAG_RO) - return; - - if (len == 0) { - /* This can happen when the object fit perfectly into a - * segment, the segment gets written per sync and subsequently - * closed. - */ - return; - } - head = ofs & (PAGE_SIZE - 1); - if (head) { - ofs -= head; - len += head; - } - len = PAGE_ALIGN(len); - __logfs_mtd_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT); -} - -static void logfs_mtd_put_device(struct logfs_super *s) -{ - put_mtd_device(s->s_mtd); -} - -static int logfs_mtd_can_write_buf(struct super_block *sb, u64 ofs) -{ - struct logfs_super *super = logfs_super(sb); - void *buf; - int err; - - buf = kmalloc(super->s_writesize, GFP_KERNEL); - if (!buf) - return -ENOMEM; - err = logfs_mtd_read(sb, ofs, super->s_writesize, buf); - if (err) - goto out; - if (memchr_inv(buf, 0xff, super->s_writesize)) - err = -EIO; - kfree(buf); -out: - return err; -} - -static const struct logfs_device_ops mtd_devops = { - .find_first_sb = logfs_mtd_find_first_sb, - .find_last_sb = logfs_mtd_find_last_sb, - .readpage = logfs_mtd_readpage, - .writeseg = logfs_mtd_writeseg, - .erase = logfs_mtd_erase, - .can_write_buf = logfs_mtd_can_write_buf, - .sync = logfs_mtd_sync, - .put_device = logfs_mtd_put_device, -}; - -int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr) -{ - struct mtd_info *mtd = get_mtd_device(NULL, mtdnr); - if (IS_ERR(mtd)) - return PTR_ERR(mtd); - - s->s_bdev = NULL; - s->s_mtd = mtd; - s->s_devops = &mtd_devops; - return 0; -} diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c deleted file mode 100644 index c87ea52de3d9..000000000000 --- a/fs/logfs/dir.c +++ /dev/null @@ -1,801 +0,0 @@ -/* - * fs/logfs/dir.c - directory-related code - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - */ -#include "logfs.h" -#include <linux/slab.h> - -/* - * Atomic dir operations - * - * Directory operations are by default not atomic. Dentries and Inodes are - * created/removed/altered in separate operations. Therefore we need to do - * a small amount of journaling. - * - * Create, link, mkdir, mknod and symlink all share the same function to do - * the work: __logfs_create. This function works in two atomic steps: - * 1. allocate inode (remember in journal) - * 2. allocate dentry (clear journal) - * - * As we can only get interrupted between the two, when the inode we just - * created is simply stored in the anchor. On next mount, if we were - * interrupted, we delete the inode. From a users point of view the - * operation never happened. - * - * Unlink and rmdir also share the same function: unlink. Again, this - * function works in two atomic steps - * 1. remove dentry (remember inode in journal) - * 2. unlink inode (clear journal) - * - * And again, on the next mount, if we were interrupted, we delete the inode. - * From a users point of view the operation succeeded. - * - * Rename is the real pain to deal with, harder than all the other methods - * combined. Depending on the circumstances we can run into three cases. - * A "target rename" where the target dentry already existed, a "local - * rename" where both parent directories are identical or a "cross-directory - * rename" in the remaining case. - * - * Local rename is atomic, as the old dentry is simply rewritten with a new - * name. - * - * Cross-directory rename works in two steps, similar to __logfs_create and - * logfs_unlink: - * 1. Write new dentry (remember old dentry in journal) - * 2. Remove old dentry (clear journal) - * - * Here we remember a dentry instead of an inode. On next mount, if we were - * interrupted, we delete the dentry. From a users point of view, the - * operation succeeded. - * - * Target rename works in three atomic steps: - * 1. Attach old inode to new dentry (remember old dentry and new inode) - * 2. Remove old dentry (still remember the new inode) - * 3. Remove victim inode - * - * Here we remember both an inode an a dentry. If we get interrupted - * between steps 1 and 2, we delete both the dentry and the inode. If - * we get interrupted between steps 2 and 3, we delete just the inode. - * In either case, the remaining objects are deleted on next mount. From - * a users point of view, the operation succeeded. - */ - -static int write_dir(struct inode *dir, struct logfs_disk_dentry *dd, - loff_t pos) -{ - return logfs_inode_write(dir, dd, sizeof(*dd), pos, WF_LOCK, NULL); -} - -static int write_inode(struct inode *inode) -{ - return __logfs_write_inode(inode, NULL, WF_LOCK); -} - -static s64 dir_seek_data(struct inode *inode, s64 pos) -{ - s64 new_pos = logfs_seek_data(inode, pos); - - return max(pos, new_pos - 1); -} - -static int beyond_eof(struct inode *inode, loff_t bix) -{ - loff_t pos = bix << inode->i_sb->s_blocksize_bits; - return pos >= i_size_read(inode); -} - -/* - * Prime value was chosen to be roughly 256 + 26. r5 hash uses 11, - * so short names (len <= 9) don't even occupy the complete 32bit name - * space. A prime >256 ensures short names quickly spread the 32bit - * name space. Add about 26 for the estimated amount of information - * of each character and pick a prime nearby, preferably a bit-sparse - * one. - */ -static u32 logfs_hash_32(const char *s, int len, u32 seed) -{ - u32 hash = seed; - int i; - - for (i = 0; i < len; i++) - hash = hash * 293 + s[i]; - return hash; -} - -/* - * We have to satisfy several conflicting requirements here. Small - * directories should stay fairly compact and not require too many - * indirect blocks. The number of possible locations for a given hash - * should be small to make lookup() fast. And we should try hard not - * to overflow the 32bit name space or nfs and 32bit host systems will - * be unhappy. - * - * So we use the following scheme. First we reduce the hash to 0..15 - * and try a direct block. If that is occupied we reduce the hash to - * 16..255 and try an indirect block. Same for 2x and 3x indirect - * blocks. Lastly we reduce the hash to 0x800_0000 .. 0xffff_ffff, - * but use buckets containing eight entries instead of a single one. - * - * Using 16 entries should allow for a reasonable amount of hash - * collisions, so the 32bit name space can be packed fairly tight - * before overflowing. Oh and currently we don't overflow but return - * and error. - * - * How likely are collisions? Doing the appropriate math is beyond me - * and the Bronstein textbook. But running a test program to brute - * force collisions for a couple of days showed that on average the - * first collision occurs after 598M entries, with 290M being the - * smallest result. Obviously 21 entries could already cause a - * collision if all entries are carefully chosen. - */ -static pgoff_t hash_index(u32 hash, int round) -{ - u32 i0_blocks = I0_BLOCKS; - u32 i1_blocks = I1_BLOCKS; - u32 i2_blocks = I2_BLOCKS; - u32 i3_blocks = I3_BLOCKS; - - switch (round) { - case 0: - return hash % i0_blocks; - case 1: - return i0_blocks + hash % (i1_blocks - i0_blocks); - case 2: - return i1_blocks + hash % (i2_blocks - i1_blocks); - case 3: - return i2_blocks + hash % (i3_blocks - i2_blocks); - case 4 ... 19: - return i3_blocks + 16 * (hash % (((1<<31) - i3_blocks) / 16)) - + round - 4; - } - BUG(); -} - -static struct page *logfs_get_dd_page(struct inode *dir, struct dentry *dentry) -{ - const struct qstr *name = &dentry->d_name; - struct page *page; - struct logfs_disk_dentry *dd; - u32 hash = logfs_hash_32(name->name, name->len, 0); - pgoff_t index; - int round; - - if (name->len > LOGFS_MAX_NAMELEN) - return ERR_PTR(-ENAMETOOLONG); - - for (round = 0; round < 20; round++) { - index = hash_index(hash, round); - - if (beyond_eof(dir, index)) - return NULL; - if (!logfs_exist_block(dir, index)) - continue; - page = read_cache_page(dir->i_mapping, index, - (filler_t *)logfs_readpage, NULL); - if (IS_ERR(page)) - return page; - dd = kmap_atomic(page); - BUG_ON(dd->namelen == 0); - - if (name->len != be16_to_cpu(dd->namelen) || - memcmp(name->name, dd->name, name->len)) { - kunmap_atomic(dd); - put_page(page); - continue; - } - - kunmap_atomic(dd); - return page; - } - return NULL; -} - -static int logfs_remove_inode(struct inode *inode) -{ - int ret; - - drop_nlink(inode); - ret = write_inode(inode); - LOGFS_BUG_ON(ret, inode->i_sb); - return ret; -} - -static void abort_transaction(struct inode *inode, struct logfs_transaction *ta) -{ - if (logfs_inode(inode)->li_block) - logfs_inode(inode)->li_block->ta = NULL; - kfree(ta); -} - -static int logfs_unlink(struct inode *dir, struct dentry *dentry) -{ - struct logfs_super *super = logfs_super(dir->i_sb); - struct inode *inode = d_inode(dentry); - struct logfs_transaction *ta; - struct page *page; - pgoff_t index; - int ret; - - ta = kzalloc(sizeof(*ta), GFP_KERNEL); - if (!ta) - return -ENOMEM; - - ta->state = UNLINK_1; - ta->ino = inode->i_ino; - - inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode); - - page = logfs_get_dd_page(dir, dentry); - if (!page) { - kfree(ta); - return -ENOENT; - } - if (IS_ERR(page)) { - kfree(ta); - return PTR_ERR(page); - } - index = page->index; - put_page(page); - - mutex_lock(&super->s_dirop_mutex); - logfs_add_transaction(dir, ta); - - ret = logfs_delete(dir, index, NULL); - if (!ret) - ret = write_inode(dir); - - if (ret) { - abort_transaction(dir, ta); - printk(KERN_ERR"LOGFS: unable to delete inode\n"); - goto out; - } - - ta->state = UNLINK_2; - logfs_add_transaction(inode, ta); - ret = logfs_remove_inode(inode); -out: - mutex_unlock(&super->s_dirop_mutex); - return ret; -} - -static inline int logfs_empty_dir(struct inode *dir) -{ - u64 data; - - data = logfs_seek_data(dir, 0) << dir->i_sb->s_blocksize_bits; - return data >= i_size_read(dir); -} - -static int logfs_rmdir(struct inode *dir, struct dentry *dentry) -{ - struct inode *inode = d_inode(dentry); - - if (!logfs_empty_dir(inode)) - return -ENOTEMPTY; - - return logfs_unlink(dir, dentry); -} - -/* FIXME: readdir currently has it's own dir_walk code. I don't see a good - * way to combine the two copies */ -static int logfs_readdir(struct file *file, struct dir_context *ctx) -{ - struct inode *dir = file_inode(file); - loff_t pos; - struct page *page; - struct logfs_disk_dentry *dd; - - if (ctx->pos < 0) - return -EINVAL; - - if (!dir_emit_dots(file, ctx)) - return 0; - - pos = ctx->pos - 2; - BUG_ON(pos < 0); - for (;; pos++, ctx->pos++) { - bool full; - if (beyond_eof(dir, pos)) - break; - if (!logfs_exist_block(dir, pos)) { - /* deleted dentry */ - pos = dir_seek_data(dir, pos); - continue; - } - page = read_cache_page(dir->i_mapping, pos, - (filler_t *)logfs_readpage, NULL); - if (IS_ERR(page)) - return PTR_ERR(page); - dd = kmap(page); - BUG_ON(dd->namelen == 0); - - full = !dir_emit(ctx, (char *)dd->name, - be16_to_cpu(dd->namelen), - be64_to_cpu(dd->ino), dd->type); - kunmap(page); - put_page(page); - if (full) - break; - } - return 0; -} - -static void logfs_set_name(struct logfs_disk_dentry *dd, const struct qstr *name) -{ - dd->namelen = cpu_to_be16(name->len); - memcpy(dd->name, name->name, name->len); -} - -static struct dentry *logfs_lookup(struct inode *dir, struct dentry *dentry, - unsigned int flags) -{ - struct page *page; - struct logfs_disk_dentry *dd; - pgoff_t index; - u64 ino = 0; - struct inode *inode; - - page = logfs_get_dd_page(dir, dentry); - if (IS_ERR(page)) - return ERR_CAST(page); - if (!page) { - d_add(dentry, NULL); - return NULL; - } - index = page->index; - dd = kmap_atomic(page); - ino = be64_to_cpu(dd->ino); - kunmap_atomic(dd); - put_page(page); - - inode = logfs_iget(dir->i_sb, ino); - if (IS_ERR(inode)) - printk(KERN_ERR"LogFS: Cannot read inode #%llx for dentry (%lx, %lx)n", - ino, dir->i_ino, index); - return d_splice_alias(inode, dentry); -} - -static void grow_dir(struct inode *dir, loff_t index) -{ - index = (index + 1) << dir->i_sb->s_blocksize_bits; - if (i_size_read(dir) < index) - i_size_write(dir, index); -} - -static int logfs_write_dir(struct inode *dir, struct dentry *dentry, - struct inode *inode) -{ - struct page *page; - struct logfs_disk_dentry *dd; - u32 hash = logfs_hash_32(dentry->d_name.name, dentry->d_name.len, 0); - pgoff_t index; - int round, err; - - for (round = 0; round < 20; round++) { - index = hash_index(hash, round); - - if (logfs_exist_block(dir, index)) - continue; - page = find_or_create_page(dir->i_mapping, index, GFP_KERNEL); - if (!page) - return -ENOMEM; - - dd = kmap_atomic(page); - memset(dd, 0, sizeof(*dd)); - dd->ino = cpu_to_be64(inode->i_ino); - dd->type = logfs_type(inode); - logfs_set_name(dd, &dentry->d_name); - kunmap_atomic(dd); - - err = logfs_write_buf(dir, page, WF_LOCK); - unlock_page(page); - put_page(page); - if (!err) - grow_dir(dir, index); - return err; - } - /* FIXME: Is there a better return value? In most cases neither - * the filesystem nor the directory are full. But we have had - * too many collisions for this particular hash and no fallback. - */ - return -ENOSPC; -} - -static int __logfs_create(struct inode *dir, struct dentry *dentry, - struct inode *inode, const char *dest, long destlen) -{ - struct logfs_super *super = logfs_super(dir->i_sb); - struct logfs_inode *li = logfs_inode(inode); - struct logfs_transaction *ta; - int ret; - - ta = kzalloc(sizeof(*ta), GFP_KERNEL); - if (!ta) { - drop_nlink(inode); - iput(inode); - return -ENOMEM; - } - - ta->state = CREATE_1; - ta->ino = inode->i_ino; - mutex_lock(&super->s_dirop_mutex); - logfs_add_transaction(inode, ta); - - if (dest) { - /* symlink */ - ret = logfs_inode_write(inode, dest, destlen, 0, WF_LOCK, NULL); - if (!ret) - ret = write_inode(inode); - } else { - /* creat/mkdir/mknod */ - ret = write_inode(inode); - } - if (ret) { - abort_transaction(inode, ta); - li->li_flags |= LOGFS_IF_STILLBORN; - /* FIXME: truncate symlink */ - drop_nlink(inode); - iput(inode); - goto out; - } - - ta->state = CREATE_2; - logfs_add_transaction(dir, ta); - ret = logfs_write_dir(dir, dentry, inode); - /* sync directory */ - if (!ret) - ret = write_inode(dir); - - if (ret) { - logfs_del_transaction(dir, ta); - ta->state = CREATE_2; - logfs_add_transaction(inode, ta); - logfs_remove_inode(inode); - iput(inode); - goto out; - } - d_instantiate(dentry, inode); -out: - mutex_unlock(&super->s_dirop_mutex); - return ret; -} - -static int logfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) -{ - struct inode *inode; - - /* - * FIXME: why do we have to fill in S_IFDIR, while the mode is - * correct for mknod, creat, etc.? Smells like the vfs *should* - * do it for us but for some reason fails to do so. - */ - inode = logfs_new_inode(dir, S_IFDIR | mode); - if (IS_ERR(inode)) - return PTR_ERR(inode); - - inode->i_op = &logfs_dir_iops; - inode->i_fop = &logfs_dir_fops; - - return __logfs_create(dir, dentry, inode, NULL, 0); -} - -static int logfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) -{ - struct inode *inode; - - inode = logfs_new_inode(dir, mode); - if (IS_ERR(inode)) - return PTR_ERR(inode); - - inode->i_op = &logfs_reg_iops; - inode->i_fop = &logfs_reg_fops; - inode->i_mapping->a_ops = &logfs_reg_aops; - - return __logfs_create(dir, dentry, inode, NULL, 0); -} - -static int logfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, - dev_t rdev) -{ - struct inode *inode; - - if (dentry->d_name.len > LOGFS_MAX_NAMELEN) - return -ENAMETOOLONG; - - inode = logfs_new_inode(dir, mode); - if (IS_ERR(inode)) - return PTR_ERR(inode); - - init_special_inode(inode, mode, rdev); - - return __logfs_create(dir, dentry, inode, NULL, 0); -} - -static int logfs_symlink(struct inode *dir, struct dentry *dentry, - const char *target) -{ - struct inode *inode; - size_t destlen = strlen(target) + 1; - - if (destlen > dir->i_sb->s_blocksize) - return -ENAMETOOLONG; - - inode = logfs_new_inode(dir, S_IFLNK | 0777); - if (IS_ERR(inode)) - return PTR_ERR(inode); - - inode->i_op = &page_symlink_inode_operations; - inode_nohighmem(inode); - inode->i_mapping->a_ops = &logfs_reg_aops; - - return __logfs_create(dir, dentry, inode, target, destlen); -} - -static int logfs_link(struct dentry *old_dentry, struct inode *dir, - struct dentry *dentry) -{ - struct inode *inode = d_inode(old_dentry); - - inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode); - ihold(inode); - inc_nlink(inode); - mark_inode_dirty_sync(inode); - - return __logfs_create(dir, dentry, inode, NULL, 0); -} - -static int logfs_get_dd(struct inode *dir, struct dentry *dentry, - struct logfs_disk_dentry *dd, loff_t *pos) -{ - struct page *page; - void *map; - - page = logfs_get_dd_page(dir, dentry); - if (IS_ERR(page)) - return PTR_ERR(page); - *pos = page->index; - map = kmap_atomic(page); - memcpy(dd, map, sizeof(*dd)); - kunmap_atomic(map); - put_page(page); - return 0; -} - -static int logfs_delete_dd(struct inode *dir, loff_t pos) -{ - /* - * Getting called with pos somewhere beyond eof is either a goofup - * within this file or means someone maliciously edited the - * (crc-protected) journal. - */ - BUG_ON(beyond_eof(dir, pos)); - dir->i_ctime = dir->i_mtime = current_time(dir); - log_dir(" Delete dentry (%lx, %llx)\n", dir->i_ino, pos); - return logfs_delete(dir, pos, NULL); -} - -/* - * Cross-directory rename, target does not exist. Just a little nasty. - * Create a new dentry in the target dir, then remove the old dentry, - * all the while taking care to remember our operation in the journal. - */ -static int logfs_rename_cross(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry) -{ - struct logfs_super *super = logfs_super(old_dir->i_sb); - struct logfs_disk_dentry dd; - struct logfs_transaction *ta; - loff_t pos; - int err; - - /* 1. locate source dd */ - err = logfs_get_dd(old_dir, old_dentry, &dd, &pos); - if (err) - return err; - - ta = kzalloc(sizeof(*ta), GFP_KERNEL); - if (!ta) - return -ENOMEM; - - ta->state = CROSS_RENAME_1; - ta->dir = old_dir->i_ino; - ta->pos = pos; - - /* 2. write target dd */ - mutex_lock(&super->s_dirop_mutex); - logfs_add_transaction(new_dir, ta); - err = logfs_write_dir(new_dir, new_dentry, d_inode(old_dentry)); - if (!err) - err = write_inode(new_dir); - - if (err) { - super->s_rename_dir = 0; - super->s_rename_pos = 0; - abort_transaction(new_dir, ta); - goto out; - } - - /* 3. remove source dd */ - ta->state = CROSS_RENAME_2; - logfs_add_transaction(old_dir, ta); - err = logfs_delete_dd(old_dir, pos); - if (!err) - err = write_inode(old_dir); - LOGFS_BUG_ON(err, old_dir->i_sb); -out: - mutex_unlock(&super->s_dirop_mutex); - return err; -} - -static int logfs_replace_inode(struct inode *dir, struct dentry *dentry, - struct logfs_disk_dentry *dd, struct inode *inode) -{ - loff_t pos; - int err; - - err = logfs_get_dd(dir, dentry, dd, &pos); - if (err) - return err; - dd->ino = cpu_to_be64(inode->i_ino); - dd->type = logfs_type(inode); - - err = write_dir(dir, dd, pos); - if (err) - return err; - log_dir("Replace dentry (%lx, %llx) %s -> %llx\n", dir->i_ino, pos, - dd->name, be64_to_cpu(dd->ino)); - return write_inode(dir); -} - -/* Target dentry exists - the worst case. We need to attach the source - * inode to the target dentry, then remove the orphaned target inode and - * source dentry. - */ -static int logfs_rename_target(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry) -{ - struct logfs_super *super = logfs_super(old_dir->i_sb); - struct inode *old_inode = d_inode(old_dentry); - struct inode *new_inode = d_inode(new_dentry); - int isdir = S_ISDIR(old_inode->i_mode); - struct logfs_disk_dentry dd; - struct logfs_transaction *ta; - loff_t pos; - int err; - - BUG_ON(isdir != S_ISDIR(new_inode->i_mode)); - if (isdir) { - if (!logfs_empty_dir(new_inode)) - return -ENOTEMPTY; - } - - /* 1. locate source dd */ - err = logfs_get_dd(old_dir, old_dentry, &dd, &pos); - if (err) - return err; - - ta = kzalloc(sizeof(*ta), GFP_KERNEL); - if (!ta) - return -ENOMEM; - - ta->state = TARGET_RENAME_1; - ta->dir = old_dir->i_ino; - ta->pos = pos; - ta->ino = new_inode->i_ino; - - /* 2. attach source inode to target dd */ - mutex_lock(&super->s_dirop_mutex); - logfs_add_transaction(new_dir, ta); - err = logfs_replace_inode(new_dir, new_dentry, &dd, old_inode); - if (err) { - super->s_rename_dir = 0; - super->s_rename_pos = 0; - super->s_victim_ino = 0; - abort_transaction(new_dir, ta); - goto out; - } - - /* 3. remove source dd */ - ta->state = TARGET_RENAME_2; - logfs_add_transaction(old_dir, ta); - err = logfs_delete_dd(old_dir, pos); - if (!err) - err = write_inode(old_dir); - LOGFS_BUG_ON(err, old_dir->i_sb); - - /* 4. remove target inode */ - ta->state = TARGET_RENAME_3; - logfs_add_transaction(new_inode, ta); - err = logfs_remove_inode(new_inode); - -out: - mutex_unlock(&super->s_dirop_mutex); - return err; -} - -static int logfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) -{ - if (flags & ~RENAME_NOREPLACE) - return -EINVAL; - - if (d_really_is_positive(new_dentry)) - return logfs_rename_target(old_dir, old_dentry, - new_dir, new_dentry); - return logfs_rename_cross(old_dir, old_dentry, new_dir, new_dentry); -} - -/* No locking done here, as this is called before .get_sb() returns. */ -int logfs_replay_journal(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct inode *inode; - u64 ino, pos; - int err; - - if (super->s_victim_ino) { - /* delete victim inode */ - ino = super->s_victim_ino; - printk(KERN_INFO"LogFS: delete unmapped inode #%llx\n", ino); - inode = logfs_iget(sb, ino); - if (IS_ERR(inode)) - goto fail; - - LOGFS_BUG_ON(i_size_read(inode) > 0, sb); - super->s_victim_ino = 0; - err = logfs_remove_inode(inode); - iput(inode); - if (err) { - super->s_victim_ino = ino; - goto fail; - } - } - if (super->s_rename_dir) { - /* delete old dd from rename */ - ino = super->s_rename_dir; - pos = super->s_rename_pos; - printk(KERN_INFO"LogFS: delete unbacked dentry (%llx, %llx)\n", - ino, pos); - inode = logfs_iget(sb, ino); - if (IS_ERR(inode)) - goto fail; - - super->s_rename_dir = 0; - super->s_rename_pos = 0; - err = logfs_delete_dd(inode, pos); - iput(inode); - if (err) { - super->s_rename_dir = ino; - super->s_rename_pos = pos; - goto fail; - } - } - return 0; -fail: - LOGFS_BUG(sb); - return -EIO; -} - -const struct inode_operations logfs_dir_iops = { - .create = logfs_create, - .link = logfs_link, - .lookup = logfs_lookup, - .mkdir = logfs_mkdir, - .mknod = logfs_mknod, - .rename = logfs_rename, - .rmdir = logfs_rmdir, - .symlink = logfs_symlink, - .unlink = logfs_unlink, -}; -const struct file_operations logfs_dir_fops = { - .fsync = logfs_fsync, - .unlocked_ioctl = logfs_ioctl, - .iterate_shared = logfs_readdir, - .read = generic_read_dir, - .llseek = generic_file_llseek, -}; diff --git a/fs/logfs/file.c b/fs/logfs/file.c deleted file mode 100644 index 1db04930ad57..000000000000 --- a/fs/logfs/file.c +++ /dev/null @@ -1,285 +0,0 @@ -/* - * fs/logfs/file.c - prepare_write, commit_write and friends - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - */ -#include "logfs.h" -#include <linux/sched.h> -#include <linux/writeback.h> - -static int logfs_write_begin(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned flags, - struct page **pagep, void **fsdata) -{ - struct inode *inode = mapping->host; - struct page *page; - pgoff_t index = pos >> PAGE_SHIFT; - - page = grab_cache_page_write_begin(mapping, index, flags); - if (!page) - return -ENOMEM; - *pagep = page; - - if ((len == PAGE_SIZE) || PageUptodate(page)) - return 0; - if ((pos & PAGE_MASK) >= i_size_read(inode)) { - unsigned start = pos & (PAGE_SIZE - 1); - unsigned end = start + len; - - /* Reading beyond i_size is simple: memset to zero */ - zero_user_segments(page, 0, start, end, PAGE_SIZE); - return 0; - } - return logfs_readpage_nolock(page); -} - -static int logfs_write_end(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, struct page *page, - void *fsdata) -{ - struct inode *inode = mapping->host; - pgoff_t index = page->index; - unsigned start = pos & (PAGE_SIZE - 1); - unsigned end = start + copied; - int ret = 0; - - BUG_ON(PAGE_SIZE != inode->i_sb->s_blocksize); - BUG_ON(page->index > I3_BLOCKS); - - if (copied < len) { - /* - * Short write of a non-initialized paged. Just tell userspace - * to retry the entire page. - */ - if (!PageUptodate(page)) { - copied = 0; - goto out; - } - } - if (copied == 0) - goto out; /* FIXME: do we need to update inode? */ - - if (i_size_read(inode) < (index << PAGE_SHIFT) + end) { - i_size_write(inode, (index << PAGE_SHIFT) + end); - mark_inode_dirty_sync(inode); - } - - SetPageUptodate(page); - if (!PageDirty(page)) { - if (!get_page_reserve(inode, page)) - __set_page_dirty_nobuffers(page); - else - ret = logfs_write_buf(inode, page, WF_LOCK); - } -out: - unlock_page(page); - put_page(page); - return ret ? ret : copied; -} - -int logfs_readpage(struct file *file, struct page *page) -{ - int ret; - - ret = logfs_readpage_nolock(page); - unlock_page(page); - return ret; -} - -/* Clear the page's dirty flag in the radix tree. */ -/* TODO: mucking with PageWriteback is silly. Add a generic function to clear - * the dirty bit from the radix tree for filesystems that don't have to wait - * for page writeback to finish (i.e. any compressing filesystem). - */ -static void clear_radix_tree_dirty(struct page *page) -{ - BUG_ON(PagePrivate(page) || page->private); - set_page_writeback(page); - end_page_writeback(page); -} - -static int __logfs_writepage(struct page *page) -{ - struct inode *inode = page->mapping->host; - int err; - - err = logfs_write_buf(inode, page, WF_LOCK); - if (err) - set_page_dirty(page); - else - clear_radix_tree_dirty(page); - unlock_page(page); - return err; -} - -static int logfs_writepage(struct page *page, struct writeback_control *wbc) -{ - struct inode *inode = page->mapping->host; - loff_t i_size = i_size_read(inode); - pgoff_t end_index = i_size >> PAGE_SHIFT; - unsigned offset; - u64 bix; - level_t level; - - log_file("logfs_writepage(%lx, %lx, %p)\n", inode->i_ino, page->index, - page); - - logfs_unpack_index(page->index, &bix, &level); - - /* Indirect blocks are never truncated */ - if (level != 0) - return __logfs_writepage(page); - - /* - * TODO: everything below is a near-verbatim copy of nobh_writepage(). - * The relevant bits should be factored out after logfs is merged. - */ - - /* Is the page fully inside i_size? */ - if (bix < end_index) - return __logfs_writepage(page); - - /* Is the page fully outside i_size? (truncate in progress) */ - offset = i_size & (PAGE_SIZE-1); - if (bix > end_index || offset == 0) { - unlock_page(page); - return 0; /* don't care */ - } - - /* - * The page straddles i_size. It must be zeroed out on each and every - * writepage invokation because it may be mmapped. "A file is mapped - * in multiples of the page size. For a file that is not a multiple of - * the page size, the remaining memory is zeroed when mapped, and - * writes to that region are not written out to the file." - */ - zero_user_segment(page, offset, PAGE_SIZE); - return __logfs_writepage(page); -} - -static void logfs_invalidatepage(struct page *page, unsigned int offset, - unsigned int length) -{ - struct logfs_block *block = logfs_block(page); - - if (block->reserved_bytes) { - struct super_block *sb = page->mapping->host->i_sb; - struct logfs_super *super = logfs_super(sb); - - super->s_dirty_pages -= block->reserved_bytes; - block->ops->free_block(sb, block); - BUG_ON(bitmap_weight(block->alias_map, LOGFS_BLOCK_FACTOR)); - } else - move_page_to_btree(page); - BUG_ON(PagePrivate(page) || page->private); -} - -static int logfs_releasepage(struct page *page, gfp_t only_xfs_uses_this) -{ - return 0; /* None of these are easy to release */ -} - - -long logfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg) -{ - struct inode *inode = file_inode(file); - struct logfs_inode *li = logfs_inode(inode); - unsigned int oldflags, flags; - int err; - - switch (cmd) { - case FS_IOC_GETFLAGS: - flags = li->li_flags & LOGFS_FL_USER_VISIBLE; - return put_user(flags, (int __user *)arg); - case FS_IOC_SETFLAGS: - if (IS_RDONLY(inode)) - return -EROFS; - - if (!inode_owner_or_capable(inode)) - return -EACCES; - - err = get_user(flags, (int __user *)arg); - if (err) - return err; - - inode_lock(inode); - oldflags = li->li_flags; - flags &= LOGFS_FL_USER_MODIFIABLE; - flags |= oldflags & ~LOGFS_FL_USER_MODIFIABLE; - li->li_flags = flags; - inode_unlock(inode); - - inode->i_ctime = current_time(inode); - mark_inode_dirty_sync(inode); - return 0; - - default: - return -ENOTTY; - } -} - -int logfs_fsync(struct file *file, loff_t start, loff_t end, int datasync) -{ - struct super_block *sb = file->f_mapping->host->i_sb; - struct inode *inode = file->f_mapping->host; - int ret; - - ret = filemap_write_and_wait_range(inode->i_mapping, start, end); - if (ret) - return ret; - - inode_lock(inode); - logfs_get_wblocks(sb, NULL, WF_LOCK); - logfs_write_anchor(sb); - logfs_put_wblocks(sb, NULL, WF_LOCK); - inode_unlock(inode); - - return 0; -} - -static int logfs_setattr(struct dentry *dentry, struct iattr *attr) -{ - struct inode *inode = d_inode(dentry); - int err = 0; - - err = setattr_prepare(dentry, attr); - if (err) - return err; - - if (attr->ia_valid & ATTR_SIZE) { - err = logfs_truncate(inode, attr->ia_size); - if (err) - return err; - } - - setattr_copy(inode, attr); - mark_inode_dirty(inode); - return 0; -} - -const struct inode_operations logfs_reg_iops = { - .setattr = logfs_setattr, -}; - -const struct file_operations logfs_reg_fops = { - .read_iter = generic_file_read_iter, - .write_iter = generic_file_write_iter, - .fsync = logfs_fsync, - .unlocked_ioctl = logfs_ioctl, - .llseek = generic_file_llseek, - .mmap = generic_file_readonly_mmap, - .open = generic_file_open, -}; - -const struct address_space_operations logfs_reg_aops = { - .invalidatepage = logfs_invalidatepage, - .readpage = logfs_readpage, - .releasepage = logfs_releasepage, - .set_page_dirty = __set_page_dirty_nobuffers, - .writepage = logfs_writepage, - .writepages = generic_writepages, - .write_begin = logfs_write_begin, - .write_end = logfs_write_end, -}; diff --git a/fs/logfs/gc.c b/fs/logfs/gc.c deleted file mode 100644 index d4efb061bdc5..000000000000 --- a/fs/logfs/gc.c +++ /dev/null @@ -1,732 +0,0 @@ -/* - * fs/logfs/gc.c - garbage collection code - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - */ -#include "logfs.h" -#include <linux/sched.h> -#include <linux/slab.h> - -/* - * Wear leveling needs to kick in when the difference between low erase - * counts and high erase counts gets too big. A good value for "too big" - * may be somewhat below 10% of maximum erase count for the device. - * Why not 397, to pick a nice round number with no specific meaning? :) - * - * WL_RATELIMIT is the minimum time between two wear level events. A huge - * number of segments may fulfil the requirements for wear leveling at the - * same time. If that happens we don't want to cause a latency from hell, - * but just gently pick one segment every so often and minimize overhead. - */ -#define WL_DELTA 397 -#define WL_RATELIMIT 100 -#define MAX_OBJ_ALIASES 2600 -#define SCAN_RATIO 512 /* number of scanned segments per gc'd segment */ -#define LIST_SIZE 64 /* base size of candidate lists */ -#define SCAN_ROUNDS 128 /* maximum number of complete medium scans */ -#define SCAN_ROUNDS_HIGH 4 /* maximum number of higher-level scans */ - -static int no_free_segments(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - - return super->s_free_list.count; -} - -/* journal has distance -1, top-most ifile layer distance 0 */ -static u8 root_distance(struct super_block *sb, gc_level_t __gc_level) -{ - struct logfs_super *super = logfs_super(sb); - u8 gc_level = (__force u8)__gc_level; - - switch (gc_level) { - case 0: /* fall through */ - case 1: /* fall through */ - case 2: /* fall through */ - case 3: - /* file data or indirect blocks */ - return super->s_ifile_levels + super->s_iblock_levels - gc_level; - case 6: /* fall through */ - case 7: /* fall through */ - case 8: /* fall through */ - case 9: - /* inode file data or indirect blocks */ - return super->s_ifile_levels - (gc_level - 6); - default: - printk(KERN_ERR"LOGFS: segment of unknown level %x found\n", - gc_level); - WARN_ON(1); - return super->s_ifile_levels + super->s_iblock_levels; - } -} - -static int segment_is_reserved(struct super_block *sb, u32 segno) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_area *area; - void *reserved; - int i; - - /* Some segments are reserved. Just pretend they were all valid */ - reserved = btree_lookup32(&super->s_reserved_segments, segno); - if (reserved) - return 1; - - /* Currently open segments */ - for_each_area(i) { - area = super->s_area[i]; - if (area->a_is_open && area->a_segno == segno) - return 1; - } - - return 0; -} - -static void logfs_mark_segment_bad(struct super_block *sb, u32 segno) -{ - BUG(); -} - -/* - * Returns the bytes consumed by valid objects in this segment. Object headers - * are counted, the segment header is not. - */ -static u32 logfs_valid_bytes(struct super_block *sb, u32 segno, u32 *ec, - gc_level_t *gc_level) -{ - struct logfs_segment_entry se; - u32 ec_level; - - logfs_get_segment_entry(sb, segno, &se); - if (se.ec_level == cpu_to_be32(BADSEG) || - se.valid == cpu_to_be32(RESERVED)) - return RESERVED; - - ec_level = be32_to_cpu(se.ec_level); - *ec = ec_level >> 4; - *gc_level = GC_LEVEL(ec_level & 0xf); - return be32_to_cpu(se.valid); -} - -static void logfs_cleanse_block(struct super_block *sb, u64 ofs, u64 ino, - u64 bix, gc_level_t gc_level) -{ - struct inode *inode; - int err, cookie; - - inode = logfs_safe_iget(sb, ino, &cookie); - err = logfs_rewrite_block(inode, bix, ofs, gc_level, 0); - BUG_ON(err); - logfs_safe_iput(inode, cookie); -} - -static u32 logfs_gc_segment(struct super_block *sb, u32 segno) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_segment_header sh; - struct logfs_object_header oh; - u64 ofs, ino, bix; - u32 seg_ofs, logical_segno, cleaned = 0; - int err, len, valid; - gc_level_t gc_level; - - LOGFS_BUG_ON(segment_is_reserved(sb, segno), sb); - - btree_insert32(&super->s_reserved_segments, segno, (void *)1, GFP_NOFS); - err = wbuf_read(sb, dev_ofs(sb, segno, 0), sizeof(sh), &sh); - BUG_ON(err); - gc_level = GC_LEVEL(sh.level); - logical_segno = be32_to_cpu(sh.segno); - if (sh.crc != logfs_crc32(&sh, sizeof(sh), 4)) { - logfs_mark_segment_bad(sb, segno); - cleaned = -1; - goto out; - } - - for (seg_ofs = LOGFS_SEGMENT_HEADERSIZE; - seg_ofs + sizeof(oh) < super->s_segsize; ) { - ofs = dev_ofs(sb, logical_segno, seg_ofs); - err = wbuf_read(sb, dev_ofs(sb, segno, seg_ofs), sizeof(oh), - &oh); - BUG_ON(err); - - if (!memchr_inv(&oh, 0xff, sizeof(oh))) - break; - - if (oh.crc != logfs_crc32(&oh, sizeof(oh) - 4, 4)) { - logfs_mark_segment_bad(sb, segno); - cleaned = super->s_segsize - 1; - goto out; - } - - ino = be64_to_cpu(oh.ino); - bix = be64_to_cpu(oh.bix); - len = sizeof(oh) + be16_to_cpu(oh.len); - valid = logfs_is_valid_block(sb, ofs, ino, bix, gc_level); - if (valid == 1) { - logfs_cleanse_block(sb, ofs, ino, bix, gc_level); - cleaned += len; - } else if (valid == 2) { - /* Will be invalid upon journal commit */ - cleaned += len; - } - seg_ofs += len; - } -out: - btree_remove32(&super->s_reserved_segments, segno); - return cleaned; -} - -static struct gc_candidate *add_list(struct gc_candidate *cand, - struct candidate_list *list) -{ - struct rb_node **p = &list->rb_tree.rb_node; - struct rb_node *parent = NULL; - struct gc_candidate *cur; - int comp; - - cand->list = list; - while (*p) { - parent = *p; - cur = rb_entry(parent, struct gc_candidate, rb_node); - - if (list->sort_by_ec) - comp = cand->erase_count < cur->erase_count; - else - comp = cand->valid < cur->valid; - - if (comp) - p = &parent->rb_left; - else - p = &parent->rb_right; - } - rb_link_node(&cand->rb_node, parent, p); - rb_insert_color(&cand->rb_node, &list->rb_tree); - - if (list->count <= list->maxcount) { - list->count++; - return NULL; - } - cand = rb_entry(rb_last(&list->rb_tree), struct gc_candidate, rb_node); - rb_erase(&cand->rb_node, &list->rb_tree); - cand->list = NULL; - return cand; -} - -static void remove_from_list(struct gc_candidate *cand) -{ - struct candidate_list *list = cand->list; - - rb_erase(&cand->rb_node, &list->rb_tree); - list->count--; -} - -static void free_candidate(struct super_block *sb, struct gc_candidate *cand) -{ - struct logfs_super *super = logfs_super(sb); - - btree_remove32(&super->s_cand_tree, cand->segno); - kfree(cand); -} - -u32 get_best_cand(struct super_block *sb, struct candidate_list *list, u32 *ec) -{ - struct gc_candidate *cand; - u32 segno; - - BUG_ON(list->count == 0); - - cand = rb_entry(rb_first(&list->rb_tree), struct gc_candidate, rb_node); - remove_from_list(cand); - segno = cand->segno; - if (ec) - *ec = cand->erase_count; - free_candidate(sb, cand); - return segno; -} - -/* - * We have several lists to manage segments with. The reserve_list is used to - * deal with bad blocks. We try to keep the best (lowest ec) segments on this - * list. - * The free_list contains free segments for normal usage. It usually gets the - * second pick after the reserve_list. But when the free_list is running short - * it is more important to keep the free_list full than to keep a reserve. - * - * Segments that are not free are put onto a per-level low_list. If we have - * to run garbage collection, we pick a candidate from there. All segments on - * those lists should have at least some free space so GC will make progress. - * - * And last we have the ec_list, which is used to pick segments for wear - * leveling. - * - * If all appropriate lists are full, we simply free the candidate and forget - * about that segment for a while. We have better candidates for each purpose. - */ -static void __add_candidate(struct super_block *sb, struct gc_candidate *cand) -{ - struct logfs_super *super = logfs_super(sb); - u32 full = super->s_segsize - LOGFS_SEGMENT_RESERVE; - - if (cand->valid == 0) { - /* 100% free segments */ - log_gc_noisy("add reserve segment %x (ec %x) at %llx\n", - cand->segno, cand->erase_count, - dev_ofs(sb, cand->segno, 0)); - cand = add_list(cand, &super->s_reserve_list); - if (cand) { - log_gc_noisy("add free segment %x (ec %x) at %llx\n", - cand->segno, cand->erase_count, - dev_ofs(sb, cand->segno, 0)); - cand = add_list(cand, &super->s_free_list); - } - } else { - /* good candidates for Garbage Collection */ - if (cand->valid < full) - cand = add_list(cand, &super->s_low_list[cand->dist]); - /* good candidates for wear leveling, - * segments that were recently written get ignored */ - if (cand) - cand = add_list(cand, &super->s_ec_list); - } - if (cand) - free_candidate(sb, cand); -} - -static int add_candidate(struct super_block *sb, u32 segno, u32 valid, u32 ec, - u8 dist) -{ - struct logfs_super *super = logfs_super(sb); - struct gc_candidate *cand; - - cand = kmalloc(sizeof(*cand), GFP_NOFS); - if (!cand) - return -ENOMEM; - - cand->segno = segno; - cand->valid = valid; - cand->erase_count = ec; - cand->dist = dist; - - btree_insert32(&super->s_cand_tree, segno, cand, GFP_NOFS); - __add_candidate(sb, cand); - return 0; -} - -static void remove_segment_from_lists(struct super_block *sb, u32 segno) -{ - struct logfs_super *super = logfs_super(sb); - struct gc_candidate *cand; - - cand = btree_lookup32(&super->s_cand_tree, segno); - if (cand) { - remove_from_list(cand); - free_candidate(sb, cand); - } -} - -static void scan_segment(struct super_block *sb, u32 segno) -{ - u32 valid, ec = 0; - gc_level_t gc_level = 0; - u8 dist; - - if (segment_is_reserved(sb, segno)) - return; - - remove_segment_from_lists(sb, segno); - valid = logfs_valid_bytes(sb, segno, &ec, &gc_level); - if (valid == RESERVED) - return; - - dist = root_distance(sb, gc_level); - add_candidate(sb, segno, valid, ec, dist); -} - -static struct gc_candidate *first_in_list(struct candidate_list *list) -{ - if (list->count == 0) - return NULL; - return rb_entry(rb_first(&list->rb_tree), struct gc_candidate, rb_node); -} - -/* - * Find the best segment for garbage collection. Main criterion is - * the segment requiring the least effort to clean. Secondary - * criterion is to GC on the lowest level available. - * - * So we search the least effort segment on the lowest level first, - * then move up and pick another segment iff is requires significantly - * less effort. Hence the LOGFS_MAX_OBJECTSIZE in the comparison. - */ -static struct gc_candidate *get_candidate(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int i, max_dist; - struct gc_candidate *cand = NULL, *this; - - max_dist = min(no_free_segments(sb), LOGFS_NO_AREAS - 1); - - for (i = max_dist; i >= 0; i--) { - this = first_in_list(&super->s_low_list[i]); - if (!this) - continue; - if (!cand) - cand = this; - if (this->valid + LOGFS_MAX_OBJECTSIZE <= cand->valid) - cand = this; - } - return cand; -} - -static int __logfs_gc_once(struct super_block *sb, struct gc_candidate *cand) -{ - struct logfs_super *super = logfs_super(sb); - gc_level_t gc_level; - u32 cleaned, valid, segno, ec; - u8 dist; - - if (!cand) { - log_gc("GC attempted, but no candidate found\n"); - return 0; - } - - segno = cand->segno; - dist = cand->dist; - valid = logfs_valid_bytes(sb, segno, &ec, &gc_level); - free_candidate(sb, cand); - log_gc("GC segment #%02x at %llx, %x required, %x free, %x valid, %llx free\n", - segno, (u64)segno << super->s_segshift, - dist, no_free_segments(sb), valid, - super->s_free_bytes); - cleaned = logfs_gc_segment(sb, segno); - log_gc("GC segment #%02x complete - now %x valid\n", segno, - valid - cleaned); - BUG_ON(cleaned != valid); - return 1; -} - -static int logfs_gc_once(struct super_block *sb) -{ - struct gc_candidate *cand; - - cand = get_candidate(sb); - if (cand) - remove_from_list(cand); - return __logfs_gc_once(sb, cand); -} - -/* returns 1 if a wrap occurs, 0 otherwise */ -static int logfs_scan_some(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - u32 segno; - int i, ret = 0; - - segno = super->s_sweeper; - for (i = SCAN_RATIO; i > 0; i--) { - segno++; - if (segno >= super->s_no_segs) { - segno = 0; - ret = 1; - /* Break out of the loop. We want to read a single - * block from the segment size on next invocation if - * SCAN_RATIO is set to match block size - */ - break; - } - - scan_segment(sb, segno); - } - super->s_sweeper = segno; - return ret; -} - -/* - * In principle, this function should loop forever, looking for GC candidates - * and moving data. LogFS is designed in such a way that this loop is - * guaranteed to terminate. - * - * Limiting the loop to some iterations serves purely to catch cases when - * these guarantees have failed. An actual endless loop is an obvious bug - * and should be reported as such. - */ -static void __logfs_gc_pass(struct super_block *sb, int target) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_block *block; - int round, progress, last_progress = 0; - - /* - * Doing too many changes to the segfile at once would result - * in a large number of aliases. Write the journal before - * things get out of hand. - */ - if (super->s_shadow_tree.no_shadowed_segments >= MAX_OBJ_ALIASES) - logfs_write_anchor(sb); - - if (no_free_segments(sb) >= target && - super->s_no_object_aliases < MAX_OBJ_ALIASES) - return; - - log_gc("__logfs_gc_pass(%x)\n", target); - for (round = 0; round < SCAN_ROUNDS; ) { - if (no_free_segments(sb) >= target) - goto write_alias; - - /* Sync in-memory state with on-medium state in case they - * diverged */ - logfs_write_anchor(sb); - round += logfs_scan_some(sb); - if (no_free_segments(sb) >= target) - goto write_alias; - progress = logfs_gc_once(sb); - if (progress) - last_progress = round; - else if (round - last_progress > 2) - break; - continue; - - /* - * The goto logic is nasty, I just don't know a better way to - * code it. GC is supposed to ensure two things: - * 1. Enough free segments are available. - * 2. The number of aliases is bounded. - * When 1. is achieved, we take a look at 2. and write back - * some alias-containing blocks, if necessary. However, after - * each such write we need to go back to 1., as writes can - * consume free segments. - */ -write_alias: - if (super->s_no_object_aliases < MAX_OBJ_ALIASES) - return; - if (list_empty(&super->s_object_alias)) { - /* All aliases are still in btree */ - return; - } - log_gc("Write back one alias\n"); - block = list_entry(super->s_object_alias.next, - struct logfs_block, alias_list); - block->ops->write_block(block); - /* - * To round off the nasty goto logic, we reset round here. It - * is a safety-net for GC not making any progress and limited - * to something reasonably small. If incremented it for every - * single alias, the loop could terminate rather quickly. - */ - round = 0; - } - LOGFS_BUG(sb); -} - -static int wl_ratelimit(struct super_block *sb, u64 *next_event) -{ - struct logfs_super *super = logfs_super(sb); - - if (*next_event < super->s_gec) { - *next_event = super->s_gec + WL_RATELIMIT; - return 0; - } - return 1; -} - -static void logfs_wl_pass(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct gc_candidate *wl_cand, *free_cand; - - if (wl_ratelimit(sb, &super->s_wl_gec_ostore)) - return; - - wl_cand = first_in_list(&super->s_ec_list); - if (!wl_cand) - return; - free_cand = first_in_list(&super->s_free_list); - if (!free_cand) - return; - - if (wl_cand->erase_count < free_cand->erase_count + WL_DELTA) { - remove_from_list(wl_cand); - __logfs_gc_once(sb, wl_cand); - } -} - -/* - * The journal needs wear leveling as well. But moving the journal is an - * expensive operation so we try to avoid it as much as possible. And if we - * have to do it, we move the whole journal, not individual segments. - * - * Ratelimiting is not strictly necessary here, it mainly serves to avoid the - * calculations. First we check whether moving the journal would be a - * significant improvement. That means that a) the current journal segments - * have more wear than the future journal segments and b) the current journal - * segments have more wear than normal ostore segments. - * Rationale for b) is that we don't have to move the journal if it is aging - * less than the ostore, even if the reserve segments age even less (they are - * excluded from wear leveling, after all). - * Next we check that the superblocks have less wear than the journal. Since - * moving the journal requires writing the superblocks, we have to protect the - * superblocks even more than the journal. - * - * Also we double the acceptable wear difference, compared to ostore wear - * leveling. Journal data is read and rewritten rapidly, comparatively. So - * soft errors have much less time to accumulate and we allow the journal to - * be a bit worse than the ostore. - */ -static void logfs_journal_wl_pass(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct gc_candidate *cand; - u32 min_journal_ec = -1, max_reserve_ec = 0; - int i; - - if (wl_ratelimit(sb, &super->s_wl_gec_journal)) - return; - - if (super->s_reserve_list.count < super->s_no_journal_segs) { - /* Reserve is not full enough to move complete journal */ - return; - } - - journal_for_each(i) - if (super->s_journal_seg[i]) - min_journal_ec = min(min_journal_ec, - super->s_journal_ec[i]); - cand = rb_entry(rb_first(&super->s_free_list.rb_tree), - struct gc_candidate, rb_node); - max_reserve_ec = cand->erase_count; - for (i = 0; i < 2; i++) { - struct logfs_segment_entry se; - u32 segno = seg_no(sb, super->s_sb_ofs[i]); - u32 ec; - - logfs_get_segment_entry(sb, segno, &se); - ec = be32_to_cpu(se.ec_level) >> 4; - max_reserve_ec = max(max_reserve_ec, ec); - } - - if (min_journal_ec > max_reserve_ec + 2 * WL_DELTA) { - do_logfs_journal_wl_pass(sb); - } -} - -void logfs_gc_pass(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - - //BUG_ON(mutex_trylock(&logfs_super(sb)->s_w_mutex)); - /* Write journal before free space is getting saturated with dirty - * objects. - */ - if (super->s_dirty_used_bytes + super->s_dirty_free_bytes - + LOGFS_MAX_OBJECTSIZE >= super->s_free_bytes) - logfs_write_anchor(sb); - __logfs_gc_pass(sb, super->s_total_levels); - logfs_wl_pass(sb); - logfs_journal_wl_pass(sb); -} - -static int check_area(struct super_block *sb, int i) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_area *area = super->s_area[i]; - gc_level_t gc_level; - u32 cleaned, valid, ec; - u32 segno = area->a_segno; - u64 ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes); - - if (!area->a_is_open) - return 0; - - if (super->s_devops->can_write_buf(sb, ofs) == 0) - return 0; - - printk(KERN_INFO"LogFS: Possibly incomplete write at %llx\n", ofs); - /* - * The device cannot write back the write buffer. Most likely the - * wbuf was already written out and the system crashed at some point - * before the journal commit happened. In that case we wouldn't have - * to do anything. But if the crash happened before the wbuf was - * written out correctly, we must GC this segment. So assume the - * worst and always do the GC run. - */ - area->a_is_open = 0; - valid = logfs_valid_bytes(sb, segno, &ec, &gc_level); - cleaned = logfs_gc_segment(sb, segno); - if (cleaned != valid) - return -EIO; - return 0; -} - -int logfs_check_areas(struct super_block *sb) -{ - int i, err; - - for_each_area(i) { - err = check_area(sb, i); - if (err) - return err; - } - return 0; -} - -static void logfs_init_candlist(struct candidate_list *list, int maxcount, - int sort_by_ec) -{ - list->count = 0; - list->maxcount = maxcount; - list->sort_by_ec = sort_by_ec; - list->rb_tree = RB_ROOT; -} - -int logfs_init_gc(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int i; - - btree_init_mempool32(&super->s_cand_tree, super->s_btree_pool); - logfs_init_candlist(&super->s_free_list, LIST_SIZE + SCAN_RATIO, 1); - logfs_init_candlist(&super->s_reserve_list, - super->s_bad_seg_reserve, 1); - for_each_area(i) - logfs_init_candlist(&super->s_low_list[i], LIST_SIZE, 0); - logfs_init_candlist(&super->s_ec_list, LIST_SIZE, 1); - return 0; -} - -static void logfs_cleanup_list(struct super_block *sb, - struct candidate_list *list) -{ - struct gc_candidate *cand; - - while (list->count) { - cand = rb_entry(list->rb_tree.rb_node, struct gc_candidate, - rb_node); - remove_from_list(cand); - free_candidate(sb, cand); - } - BUG_ON(list->rb_tree.rb_node); -} - -void logfs_cleanup_gc(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int i; - - if (!super->s_free_list.count) - return; - - /* - * FIXME: The btree may still contain a single empty node. So we - * call the grim visitor to clean up that mess. Btree code should - * do it for us, really. - */ - btree_grim_visitor32(&super->s_cand_tree, 0, NULL); - logfs_cleanup_list(sb, &super->s_free_list); - logfs_cleanup_list(sb, &super->s_reserve_list); - for_each_area(i) - logfs_cleanup_list(sb, &super->s_low_list[i]); - logfs_cleanup_list(sb, &super->s_ec_list); -} diff --git a/fs/logfs/inode.c b/fs/logfs/inode.c deleted file mode 100644 index f440a1525da8..000000000000 --- a/fs/logfs/inode.c +++ /dev/null @@ -1,428 +0,0 @@ -/* - * fs/logfs/inode.c - inode handling code - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - */ -#include "logfs.h" -#include <linux/slab.h> -#include <linux/writeback.h> -#include <linux/backing-dev.h> - -/* - * How soon to reuse old inode numbers? LogFS doesn't store deleted inodes - * on the medium. It therefore also lacks a method to store the previous - * generation number for deleted inodes. Instead a single generation number - * is stored which will be used for new inodes. Being just a 32bit counter, - * this can obvious wrap relatively quickly. So we only reuse inodes if we - * know that a fair number of inodes can be created before we have to increment - * the generation again - effectively adding some bits to the counter. - * But being too aggressive here means we keep a very large and very sparse - * inode file, wasting space on indirect blocks. - * So what is a good value? Beats me. 64k seems moderately bad on both - * fronts, so let's use that for now... - * - * NFS sucks, as everyone already knows. - */ -#define INOS_PER_WRAP (0x10000) - -/* - * Logfs' requirement to read inodes for garbage collection makes life a bit - * harder. GC may have to read inodes that are in I_FREEING state, when they - * are being written out - and waiting for GC to make progress, naturally. - * - * So we cannot just call iget() or some variant of it, but first have to check - * whether the inode in question might be in I_FREEING state. Therefore we - * maintain our own per-sb list of "almost deleted" inodes and check against - * that list first. Normally this should be at most 1-2 entries long. - * - * Also, inodes have logfs-specific reference counting on top of what the vfs - * does. When .destroy_inode is called, normally the reference count will drop - * to zero and the inode gets deleted. But if GC accessed the inode, its - * refcount will remain nonzero and final deletion will have to wait. - * - * As a result we have two sets of functions to get/put inodes: - * logfs_safe_iget/logfs_safe_iput - safe to call from GC context - * logfs_iget/iput - normal version - */ -static struct kmem_cache *logfs_inode_cache; - -static DEFINE_SPINLOCK(logfs_inode_lock); - -static void logfs_inode_setops(struct inode *inode) -{ - switch (inode->i_mode & S_IFMT) { - case S_IFDIR: - inode->i_op = &logfs_dir_iops; - inode->i_fop = &logfs_dir_fops; - inode->i_mapping->a_ops = &logfs_reg_aops; - break; - case S_IFREG: - inode->i_op = &logfs_reg_iops; - inode->i_fop = &logfs_reg_fops; - inode->i_mapping->a_ops = &logfs_reg_aops; - break; - case S_IFLNK: - inode->i_op = &page_symlink_inode_operations; - inode_nohighmem(inode); - inode->i_mapping->a_ops = &logfs_reg_aops; - break; - case S_IFSOCK: /* fall through */ - case S_IFBLK: /* fall through */ - case S_IFCHR: /* fall through */ - case S_IFIFO: - init_special_inode(inode, inode->i_mode, inode->i_rdev); - break; - default: - BUG(); - } -} - -static struct inode *__logfs_iget(struct super_block *sb, ino_t ino) -{ - struct inode *inode = iget_locked(sb, ino); - int err; - - if (!inode) - return ERR_PTR(-ENOMEM); - if (!(inode->i_state & I_NEW)) - return inode; - - err = logfs_read_inode(inode); - if (err || inode->i_nlink == 0) { - /* inode->i_nlink == 0 can be true when called from - * block validator */ - /* set i_nlink to 0 to prevent caching */ - clear_nlink(inode); - logfs_inode(inode)->li_flags |= LOGFS_IF_ZOMBIE; - iget_failed(inode); - if (!err) - err = -ENOENT; - return ERR_PTR(err); - } - - logfs_inode_setops(inode); - unlock_new_inode(inode); - return inode; -} - -struct inode *logfs_iget(struct super_block *sb, ino_t ino) -{ - BUG_ON(ino == LOGFS_INO_MASTER); - BUG_ON(ino == LOGFS_INO_SEGFILE); - return __logfs_iget(sb, ino); -} - -/* - * is_cached is set to 1 if we hand out a cached inode, 0 otherwise. - * this allows logfs_iput to do the right thing later - */ -struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *is_cached) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_inode *li; - - if (ino == LOGFS_INO_MASTER) - return super->s_master_inode; - if (ino == LOGFS_INO_SEGFILE) - return super->s_segfile_inode; - - spin_lock(&logfs_inode_lock); - list_for_each_entry(li, &super->s_freeing_list, li_freeing_list) - if (li->vfs_inode.i_ino == ino) { - li->li_refcount++; - spin_unlock(&logfs_inode_lock); - *is_cached = 1; - return &li->vfs_inode; - } - spin_unlock(&logfs_inode_lock); - - *is_cached = 0; - return __logfs_iget(sb, ino); -} - -static void logfs_i_callback(struct rcu_head *head) -{ - struct inode *inode = container_of(head, struct inode, i_rcu); - kmem_cache_free(logfs_inode_cache, logfs_inode(inode)); -} - -static void __logfs_destroy_inode(struct inode *inode) -{ - struct logfs_inode *li = logfs_inode(inode); - - BUG_ON(li->li_block); - list_del(&li->li_freeing_list); - call_rcu(&inode->i_rcu, logfs_i_callback); -} - -static void __logfs_destroy_meta_inode(struct inode *inode) -{ - struct logfs_inode *li = logfs_inode(inode); - BUG_ON(li->li_block); - call_rcu(&inode->i_rcu, logfs_i_callback); -} - -static void logfs_destroy_inode(struct inode *inode) -{ - struct logfs_inode *li = logfs_inode(inode); - - if (inode->i_ino < LOGFS_RESERVED_INOS) { - /* - * The reserved inodes are never destroyed unless we are in - * unmont path. - */ - __logfs_destroy_meta_inode(inode); - return; - } - - BUG_ON(list_empty(&li->li_freeing_list)); - spin_lock(&logfs_inode_lock); - li->li_refcount--; - if (li->li_refcount == 0) - __logfs_destroy_inode(inode); - spin_unlock(&logfs_inode_lock); -} - -void logfs_safe_iput(struct inode *inode, int is_cached) -{ - if (inode->i_ino == LOGFS_INO_MASTER) - return; - if (inode->i_ino == LOGFS_INO_SEGFILE) - return; - - if (is_cached) { - logfs_destroy_inode(inode); - return; - } - - iput(inode); -} - -static void logfs_init_inode(struct super_block *sb, struct inode *inode) -{ - struct logfs_inode *li = logfs_inode(inode); - int i; - - li->li_flags = 0; - li->li_height = 0; - li->li_used_bytes = 0; - li->li_block = NULL; - i_uid_write(inode, 0); - i_gid_write(inode, 0); - inode->i_size = 0; - inode->i_blocks = 0; - inode->i_ctime = current_time(inode); - inode->i_mtime = current_time(inode); - li->li_refcount = 1; - INIT_LIST_HEAD(&li->li_freeing_list); - - for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) - li->li_data[i] = 0; - - return; -} - -static struct inode *logfs_alloc_inode(struct super_block *sb) -{ - struct logfs_inode *li; - - li = kmem_cache_alloc(logfs_inode_cache, GFP_NOFS); - if (!li) - return NULL; - logfs_init_inode(sb, &li->vfs_inode); - return &li->vfs_inode; -} - -/* - * In logfs inodes are written to an inode file. The inode file, like any - * other file, is managed with a inode. The inode file's inode, aka master - * inode, requires special handling in several respects. First, it cannot be - * written to the inode file, so it is stored in the journal instead. - * - * Secondly, this inode cannot be written back and destroyed before all other - * inodes have been written. The ordering is important. Linux' VFS is happily - * unaware of the ordering constraint and would ordinarily destroy the master - * inode at umount time while other inodes are still in use and dirty. Not - * good. - * - * So logfs makes sure the master inode is not written until all other inodes - * have been destroyed. Sadly, this method has another side-effect. The VFS - * will notice one remaining inode and print a frightening warning message. - * Worse, it is impossible to judge whether such a warning was caused by the - * master inode or any other inodes have leaked as well. - * - * Our attempt of solving this is with logfs_new_meta_inode() below. Its - * purpose is to create a new inode that will not trigger the warning if such - * an inode is still in use. An ugly hack, no doubt. Suggections for - * improvement are welcome. - * - * AV: that's what ->put_super() is for... - */ -struct inode *logfs_new_meta_inode(struct super_block *sb, u64 ino) -{ - struct inode *inode; - - inode = new_inode(sb); - if (!inode) - return ERR_PTR(-ENOMEM); - - inode->i_mode = S_IFREG; - inode->i_ino = ino; - inode->i_data.a_ops = &logfs_reg_aops; - mapping_set_gfp_mask(&inode->i_data, GFP_NOFS); - - return inode; -} - -struct inode *logfs_read_meta_inode(struct super_block *sb, u64 ino) -{ - struct inode *inode; - int err; - - inode = logfs_new_meta_inode(sb, ino); - if (IS_ERR(inode)) - return inode; - - err = logfs_read_inode(inode); - if (err) { - iput(inode); - return ERR_PTR(err); - } - logfs_inode_setops(inode); - return inode; -} - -static int logfs_write_inode(struct inode *inode, struct writeback_control *wbc) -{ - int ret; - long flags = WF_LOCK; - - /* Can only happen if creat() failed. Safe to skip. */ - if (logfs_inode(inode)->li_flags & LOGFS_IF_STILLBORN) - return 0; - - ret = __logfs_write_inode(inode, NULL, flags); - LOGFS_BUG_ON(ret, inode->i_sb); - return ret; -} - -/* called with inode->i_lock held */ -static int logfs_drop_inode(struct inode *inode) -{ - struct logfs_super *super = logfs_super(inode->i_sb); - struct logfs_inode *li = logfs_inode(inode); - - spin_lock(&logfs_inode_lock); - list_move(&li->li_freeing_list, &super->s_freeing_list); - spin_unlock(&logfs_inode_lock); - return generic_drop_inode(inode); -} - -static void logfs_set_ino_generation(struct super_block *sb, - struct inode *inode) -{ - struct logfs_super *super = logfs_super(sb); - u64 ino; - - mutex_lock(&super->s_journal_mutex); - ino = logfs_seek_hole(super->s_master_inode, super->s_last_ino + 1); - super->s_last_ino = ino; - super->s_inos_till_wrap--; - if (super->s_inos_till_wrap < 0) { - super->s_last_ino = LOGFS_RESERVED_INOS; - super->s_generation++; - super->s_inos_till_wrap = INOS_PER_WRAP; - } - inode->i_ino = ino; - inode->i_generation = super->s_generation; - mutex_unlock(&super->s_journal_mutex); -} - -struct inode *logfs_new_inode(struct inode *dir, umode_t mode) -{ - struct super_block *sb = dir->i_sb; - struct inode *inode; - - inode = new_inode(sb); - if (!inode) - return ERR_PTR(-ENOMEM); - - logfs_init_inode(sb, inode); - - /* inherit parent flags */ - logfs_inode(inode)->li_flags |= - logfs_inode(dir)->li_flags & LOGFS_FL_INHERITED; - - inode->i_mode = mode; - logfs_set_ino_generation(sb, inode); - - inode_init_owner(inode, dir, mode); - logfs_inode_setops(inode); - insert_inode_hash(inode); - - return inode; -} - -static void logfs_init_once(void *_li) -{ - struct logfs_inode *li = _li; - int i; - - li->li_flags = 0; - li->li_used_bytes = 0; - li->li_refcount = 1; - for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) - li->li_data[i] = 0; - inode_init_once(&li->vfs_inode); -} - -static int logfs_sync_fs(struct super_block *sb, int wait) -{ - logfs_get_wblocks(sb, NULL, WF_LOCK); - logfs_write_anchor(sb); - logfs_put_wblocks(sb, NULL, WF_LOCK); - return 0; -} - -static void logfs_put_super(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - /* kill the meta-inodes */ - iput(super->s_segfile_inode); - iput(super->s_master_inode); - iput(super->s_mapping_inode); -} - -const struct super_operations logfs_super_operations = { - .alloc_inode = logfs_alloc_inode, - .destroy_inode = logfs_destroy_inode, - .evict_inode = logfs_evict_inode, - .drop_inode = logfs_drop_inode, - .put_super = logfs_put_super, - .write_inode = logfs_write_inode, - .statfs = logfs_statfs, - .sync_fs = logfs_sync_fs, -}; - -int logfs_init_inode_cache(void) -{ - logfs_inode_cache = kmem_cache_create("logfs_inode_cache", - sizeof(struct logfs_inode), 0, - SLAB_RECLAIM_ACCOUNT|SLAB_ACCOUNT, - logfs_init_once); - if (!logfs_inode_cache) - return -ENOMEM; - return 0; -} - -void logfs_destroy_inode_cache(void) -{ - /* - * Make sure all delayed rcu free inodes are flushed before we - * destroy cache. - */ - rcu_barrier(); - kmem_cache_destroy(logfs_inode_cache); -} diff --git a/fs/logfs/journal.c b/fs/logfs/journal.c deleted file mode 100644 index 2a09b8d73989..000000000000 --- a/fs/logfs/journal.c +++ /dev/null @@ -1,894 +0,0 @@ -/* - * fs/logfs/journal.c - journal handling code - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - */ -#include "logfs.h" -#include <linux/slab.h> - -static void logfs_calc_free(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - u64 reserve, no_segs = super->s_no_segs; - s64 free; - int i; - - /* superblock segments */ - no_segs -= 2; - super->s_no_journal_segs = 0; - /* journal */ - journal_for_each(i) - if (super->s_journal_seg[i]) { - no_segs--; - super->s_no_journal_segs++; - } - - /* open segments plus one extra per level for GC */ - no_segs -= 2 * super->s_total_levels; - - free = no_segs * (super->s_segsize - LOGFS_SEGMENT_RESERVE); - free -= super->s_used_bytes; - /* just a bit extra */ - free -= super->s_total_levels * 4096; - - /* Bad blocks are 'paid' for with speed reserve - the filesystem - * simply gets slower as bad blocks accumulate. Until the bad blocks - * exceed the speed reserve - then the filesystem gets smaller. - */ - reserve = super->s_bad_segments + super->s_bad_seg_reserve; - reserve *= super->s_segsize - LOGFS_SEGMENT_RESERVE; - reserve = max(reserve, super->s_speed_reserve); - free -= reserve; - if (free < 0) - free = 0; - - super->s_free_bytes = free; -} - -static void reserve_sb_and_journal(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct btree_head32 *head = &super->s_reserved_segments; - int i, err; - - err = btree_insert32(head, seg_no(sb, super->s_sb_ofs[0]), (void *)1, - GFP_KERNEL); - BUG_ON(err); - - err = btree_insert32(head, seg_no(sb, super->s_sb_ofs[1]), (void *)1, - GFP_KERNEL); - BUG_ON(err); - - journal_for_each(i) { - if (!super->s_journal_seg[i]) - continue; - err = btree_insert32(head, super->s_journal_seg[i], (void *)1, - GFP_KERNEL); - BUG_ON(err); - } -} - -static void read_dynsb(struct super_block *sb, - struct logfs_je_dynsb *dynsb) -{ - struct logfs_super *super = logfs_super(sb); - - super->s_gec = be64_to_cpu(dynsb->ds_gec); - super->s_sweeper = be64_to_cpu(dynsb->ds_sweeper); - super->s_victim_ino = be64_to_cpu(dynsb->ds_victim_ino); - super->s_rename_dir = be64_to_cpu(dynsb->ds_rename_dir); - super->s_rename_pos = be64_to_cpu(dynsb->ds_rename_pos); - super->s_used_bytes = be64_to_cpu(dynsb->ds_used_bytes); - super->s_generation = be32_to_cpu(dynsb->ds_generation); -} - -static void read_anchor(struct super_block *sb, - struct logfs_je_anchor *da) -{ - struct logfs_super *super = logfs_super(sb); - struct inode *inode = super->s_master_inode; - struct logfs_inode *li = logfs_inode(inode); - int i; - - super->s_last_ino = be64_to_cpu(da->da_last_ino); - li->li_flags = 0; - li->li_height = da->da_height; - i_size_write(inode, be64_to_cpu(da->da_size)); - li->li_used_bytes = be64_to_cpu(da->da_used_bytes); - - for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) - li->li_data[i] = be64_to_cpu(da->da_data[i]); -} - -static void read_erasecount(struct super_block *sb, - struct logfs_je_journal_ec *ec) -{ - struct logfs_super *super = logfs_super(sb); - int i; - - journal_for_each(i) - super->s_journal_ec[i] = be32_to_cpu(ec->ec[i]); -} - -static int read_area(struct super_block *sb, struct logfs_je_area *a) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_area *area = super->s_area[a->gc_level]; - u64 ofs; - u32 writemask = ~(super->s_writesize - 1); - - if (a->gc_level >= LOGFS_NO_AREAS) - return -EIO; - if (a->vim != VIM_DEFAULT) - return -EIO; /* TODO: close area and continue */ - - area->a_used_bytes = be32_to_cpu(a->used_bytes); - area->a_written_bytes = area->a_used_bytes & writemask; - area->a_segno = be32_to_cpu(a->segno); - if (area->a_segno) - area->a_is_open = 1; - - ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes); - if (super->s_writesize > 1) - return logfs_buf_recover(area, ofs, a + 1, super->s_writesize); - else - return logfs_buf_recover(area, ofs, NULL, 0); -} - -static void *unpack(void *from, void *to) -{ - struct logfs_journal_header *jh = from; - void *data = from + sizeof(struct logfs_journal_header); - int err; - size_t inlen, outlen; - - inlen = be16_to_cpu(jh->h_len); - outlen = be16_to_cpu(jh->h_datalen); - - if (jh->h_compr == COMPR_NONE) - memcpy(to, data, inlen); - else { - err = logfs_uncompress(data, to, inlen, outlen); - BUG_ON(err); - } - return to; -} - -static int __read_je_header(struct super_block *sb, u64 ofs, - struct logfs_journal_header *jh) -{ - struct logfs_super *super = logfs_super(sb); - size_t bufsize = max_t(size_t, sb->s_blocksize, super->s_writesize) - + MAX_JOURNAL_HEADER; - u16 type, len, datalen; - int err; - - /* read header only */ - err = wbuf_read(sb, ofs, sizeof(*jh), jh); - if (err) - return err; - type = be16_to_cpu(jh->h_type); - len = be16_to_cpu(jh->h_len); - datalen = be16_to_cpu(jh->h_datalen); - if (len > sb->s_blocksize) - return -EIO; - if ((type < JE_FIRST) || (type > JE_LAST)) - return -EIO; - if (datalen > bufsize) - return -EIO; - return 0; -} - -static int __read_je_payload(struct super_block *sb, u64 ofs, - struct logfs_journal_header *jh) -{ - u16 len; - int err; - - len = be16_to_cpu(jh->h_len); - err = wbuf_read(sb, ofs + sizeof(*jh), len, jh + 1); - if (err) - return err; - if (jh->h_crc != logfs_crc32(jh, len + sizeof(*jh), 4)) { - /* Old code was confused. It forgot about the header length - * and stopped calculating the crc 16 bytes before the end - * of data - ick! - * FIXME: Remove this hack once the old code is fixed. - */ - if (jh->h_crc == logfs_crc32(jh, len, 4)) - WARN_ON_ONCE(1); - else - return -EIO; - } - return 0; -} - -/* - * jh needs to be large enough to hold the complete entry, not just the header - */ -static int __read_je(struct super_block *sb, u64 ofs, - struct logfs_journal_header *jh) -{ - int err; - - err = __read_je_header(sb, ofs, jh); - if (err) - return err; - return __read_je_payload(sb, ofs, jh); -} - -static int read_je(struct super_block *sb, u64 ofs) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_journal_header *jh = super->s_compressed_je; - void *scratch = super->s_je; - u16 type, datalen; - int err; - - err = __read_je(sb, ofs, jh); - if (err) - return err; - type = be16_to_cpu(jh->h_type); - datalen = be16_to_cpu(jh->h_datalen); - - switch (type) { - case JE_DYNSB: - read_dynsb(sb, unpack(jh, scratch)); - break; - case JE_ANCHOR: - read_anchor(sb, unpack(jh, scratch)); - break; - case JE_ERASECOUNT: - read_erasecount(sb, unpack(jh, scratch)); - break; - case JE_AREA: - err = read_area(sb, unpack(jh, scratch)); - break; - case JE_OBJ_ALIAS: - err = logfs_load_object_aliases(sb, unpack(jh, scratch), - datalen); - break; - default: - WARN_ON_ONCE(1); - return -EIO; - } - return err; -} - -static int logfs_read_segment(struct super_block *sb, u32 segno) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_journal_header *jh = super->s_compressed_je; - u64 ofs, seg_ofs = dev_ofs(sb, segno, 0); - u32 h_ofs, last_ofs = 0; - u16 len, datalen, last_len = 0; - int i, err; - - /* search for most recent commit */ - for (h_ofs = 0; h_ofs < super->s_segsize; h_ofs += sizeof(*jh)) { - ofs = seg_ofs + h_ofs; - err = __read_je_header(sb, ofs, jh); - if (err) - continue; - if (jh->h_type != cpu_to_be16(JE_COMMIT)) - continue; - err = __read_je_payload(sb, ofs, jh); - if (err) - continue; - len = be16_to_cpu(jh->h_len); - datalen = be16_to_cpu(jh->h_datalen); - if ((datalen > sizeof(super->s_je_array)) || - (datalen % sizeof(__be64))) - continue; - last_ofs = h_ofs; - last_len = datalen; - h_ofs += ALIGN(len, sizeof(*jh)) - sizeof(*jh); - } - /* read commit */ - if (last_ofs == 0) - return -ENOENT; - ofs = seg_ofs + last_ofs; - log_journal("Read commit from %llx\n", ofs); - err = __read_je(sb, ofs, jh); - BUG_ON(err); /* We should have caught it in the scan loop already */ - if (err) - return err; - /* uncompress */ - unpack(jh, super->s_je_array); - super->s_no_je = last_len / sizeof(__be64); - /* iterate over array */ - for (i = 0; i < super->s_no_je; i++) { - err = read_je(sb, be64_to_cpu(super->s_je_array[i])); - if (err) - return err; - } - super->s_journal_area->a_segno = segno; - return 0; -} - -static u64 read_gec(struct super_block *sb, u32 segno) -{ - struct logfs_segment_header sh; - __be32 crc; - int err; - - if (!segno) - return 0; - err = wbuf_read(sb, dev_ofs(sb, segno, 0), sizeof(sh), &sh); - if (err) - return 0; - crc = logfs_crc32(&sh, sizeof(sh), 4); - if (crc != sh.crc) { - WARN_ON(sh.gec != cpu_to_be64(0xffffffffffffffffull)); - /* Most likely it was just erased */ - return 0; - } - return be64_to_cpu(sh.gec); -} - -static int logfs_read_journal(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - u64 gec[LOGFS_JOURNAL_SEGS], max; - u32 segno; - int i, max_i; - - max = 0; - max_i = -1; - journal_for_each(i) { - segno = super->s_journal_seg[i]; - gec[i] = read_gec(sb, super->s_journal_seg[i]); - if (gec[i] > max) { - max = gec[i]; - max_i = i; - } - } - if (max_i == -1) - return -EIO; - /* FIXME: Try older segments in case of error */ - return logfs_read_segment(sb, super->s_journal_seg[max_i]); -} - -/* - * First search the current segment (outer loop), then pick the next segment - * in the array, skipping any zero entries (inner loop). - */ -static void journal_get_free_segment(struct logfs_area *area) -{ - struct logfs_super *super = logfs_super(area->a_sb); - int i; - - journal_for_each(i) { - if (area->a_segno != super->s_journal_seg[i]) - continue; - - do { - i++; - if (i == LOGFS_JOURNAL_SEGS) - i = 0; - } while (!super->s_journal_seg[i]); - - area->a_segno = super->s_journal_seg[i]; - area->a_erase_count = ++(super->s_journal_ec[i]); - log_journal("Journal now at %x (ec %x)\n", area->a_segno, - area->a_erase_count); - return; - } - BUG(); -} - -static void journal_get_erase_count(struct logfs_area *area) -{ - /* erase count is stored globally and incremented in - * journal_get_free_segment() - nothing to do here */ -} - -static int journal_erase_segment(struct logfs_area *area) -{ - struct super_block *sb = area->a_sb; - union { - struct logfs_segment_header sh; - unsigned char c[ALIGN(sizeof(struct logfs_segment_header), 16)]; - } u; - u64 ofs; - int err; - - err = logfs_erase_segment(sb, area->a_segno, 1); - if (err) - return err; - - memset(&u, 0, sizeof(u)); - u.sh.pad = 0; - u.sh.type = SEG_JOURNAL; - u.sh.level = 0; - u.sh.segno = cpu_to_be32(area->a_segno); - u.sh.ec = cpu_to_be32(area->a_erase_count); - u.sh.gec = cpu_to_be64(logfs_super(sb)->s_gec); - u.sh.crc = logfs_crc32(&u.sh, sizeof(u.sh), 4); - - /* This causes a bug in segment.c. Not yet. */ - //logfs_set_segment_erased(sb, area->a_segno, area->a_erase_count, 0); - - ofs = dev_ofs(sb, area->a_segno, 0); - area->a_used_bytes = sizeof(u); - logfs_buf_write(area, ofs, &u, sizeof(u)); - return 0; -} - -static size_t __logfs_write_header(struct logfs_super *super, - struct logfs_journal_header *jh, size_t len, size_t datalen, - u16 type, u8 compr) -{ - jh->h_len = cpu_to_be16(len); - jh->h_type = cpu_to_be16(type); - jh->h_datalen = cpu_to_be16(datalen); - jh->h_compr = compr; - jh->h_pad[0] = 'H'; - jh->h_pad[1] = 'E'; - jh->h_pad[2] = 'A'; - jh->h_pad[3] = 'D'; - jh->h_pad[4] = 'R'; - jh->h_crc = logfs_crc32(jh, len + sizeof(*jh), 4); - return ALIGN(len, 16) + sizeof(*jh); -} - -static size_t logfs_write_header(struct logfs_super *super, - struct logfs_journal_header *jh, size_t datalen, u16 type) -{ - size_t len = datalen; - - return __logfs_write_header(super, jh, len, datalen, type, COMPR_NONE); -} - -static inline size_t logfs_journal_erasecount_size(struct logfs_super *super) -{ - return LOGFS_JOURNAL_SEGS * sizeof(__be32); -} - -static void *logfs_write_erasecount(struct super_block *sb, void *_ec, - u16 *type, size_t *len) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_je_journal_ec *ec = _ec; - int i; - - journal_for_each(i) - ec->ec[i] = cpu_to_be32(super->s_journal_ec[i]); - *type = JE_ERASECOUNT; - *len = logfs_journal_erasecount_size(super); - return ec; -} - -static void account_shadow(void *_shadow, unsigned long _sb, u64 ignore, - size_t ignore2) -{ - struct logfs_shadow *shadow = _shadow; - struct super_block *sb = (void *)_sb; - struct logfs_super *super = logfs_super(sb); - - /* consume new space */ - super->s_free_bytes -= shadow->new_len; - super->s_used_bytes += shadow->new_len; - super->s_dirty_used_bytes -= shadow->new_len; - - /* free up old space */ - super->s_free_bytes += shadow->old_len; - super->s_used_bytes -= shadow->old_len; - super->s_dirty_free_bytes -= shadow->old_len; - - logfs_set_segment_used(sb, shadow->old_ofs, -shadow->old_len); - logfs_set_segment_used(sb, shadow->new_ofs, shadow->new_len); - - log_journal("account_shadow(%llx, %llx, %x) %llx->%llx %x->%x\n", - shadow->ino, shadow->bix, shadow->gc_level, - shadow->old_ofs, shadow->new_ofs, - shadow->old_len, shadow->new_len); - mempool_free(shadow, super->s_shadow_pool); -} - -static void account_shadows(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct inode *inode = super->s_master_inode; - struct logfs_inode *li = logfs_inode(inode); - struct shadow_tree *tree = &super->s_shadow_tree; - - btree_grim_visitor64(&tree->new, (unsigned long)sb, account_shadow); - btree_grim_visitor64(&tree->old, (unsigned long)sb, account_shadow); - btree_grim_visitor32(&tree->segment_map, 0, NULL); - tree->no_shadowed_segments = 0; - - if (li->li_block) { - /* - * We never actually use the structure, when attached to the - * master inode. But it is easier to always free it here than - * to have checks in several places elsewhere when allocating - * it. - */ - li->li_block->ops->free_block(sb, li->li_block); - } - BUG_ON((s64)li->li_used_bytes < 0); -} - -static void *__logfs_write_anchor(struct super_block *sb, void *_da, - u16 *type, size_t *len) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_je_anchor *da = _da; - struct inode *inode = super->s_master_inode; - struct logfs_inode *li = logfs_inode(inode); - int i; - - da->da_height = li->li_height; - da->da_last_ino = cpu_to_be64(super->s_last_ino); - da->da_size = cpu_to_be64(i_size_read(inode)); - da->da_used_bytes = cpu_to_be64(li->li_used_bytes); - for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) - da->da_data[i] = cpu_to_be64(li->li_data[i]); - *type = JE_ANCHOR; - *len = sizeof(*da); - return da; -} - -static void *logfs_write_dynsb(struct super_block *sb, void *_dynsb, - u16 *type, size_t *len) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_je_dynsb *dynsb = _dynsb; - - dynsb->ds_gec = cpu_to_be64(super->s_gec); - dynsb->ds_sweeper = cpu_to_be64(super->s_sweeper); - dynsb->ds_victim_ino = cpu_to_be64(super->s_victim_ino); - dynsb->ds_rename_dir = cpu_to_be64(super->s_rename_dir); - dynsb->ds_rename_pos = cpu_to_be64(super->s_rename_pos); - dynsb->ds_used_bytes = cpu_to_be64(super->s_used_bytes); - dynsb->ds_generation = cpu_to_be32(super->s_generation); - *type = JE_DYNSB; - *len = sizeof(*dynsb); - return dynsb; -} - -static void write_wbuf(struct super_block *sb, struct logfs_area *area, - void *wbuf) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - u64 ofs; - pgoff_t index; - int page_ofs; - struct page *page; - - ofs = dev_ofs(sb, area->a_segno, - area->a_used_bytes & ~(super->s_writesize - 1)); - index = ofs >> PAGE_SHIFT; - page_ofs = ofs & (PAGE_SIZE - 1); - - page = find_or_create_page(mapping, index, GFP_NOFS); - BUG_ON(!page); - memcpy(wbuf, page_address(page) + page_ofs, super->s_writesize); - unlock_page(page); -} - -static void *logfs_write_area(struct super_block *sb, void *_a, - u16 *type, size_t *len) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_area *area = super->s_area[super->s_sum_index]; - struct logfs_je_area *a = _a; - - a->vim = VIM_DEFAULT; - a->gc_level = super->s_sum_index; - a->used_bytes = cpu_to_be32(area->a_used_bytes); - a->segno = cpu_to_be32(area->a_segno); - if (super->s_writesize > 1) - write_wbuf(sb, area, a + 1); - - *type = JE_AREA; - *len = sizeof(*a) + super->s_writesize; - return a; -} - -static void *logfs_write_commit(struct super_block *sb, void *h, - u16 *type, size_t *len) -{ - struct logfs_super *super = logfs_super(sb); - - *type = JE_COMMIT; - *len = super->s_no_je * sizeof(__be64); - return super->s_je_array; -} - -static size_t __logfs_write_je(struct super_block *sb, void *buf, u16 type, - size_t len) -{ - struct logfs_super *super = logfs_super(sb); - void *header = super->s_compressed_je; - void *data = header + sizeof(struct logfs_journal_header); - ssize_t compr_len, pad_len; - u8 compr = COMPR_ZLIB; - - if (len == 0) - return logfs_write_header(super, header, 0, type); - - compr_len = logfs_compress(buf, data, len, sb->s_blocksize); - if (compr_len < 0 || type == JE_ANCHOR) { - memcpy(data, buf, len); - compr_len = len; - compr = COMPR_NONE; - } - - pad_len = ALIGN(compr_len, 16); - memset(data + compr_len, 0, pad_len - compr_len); - - return __logfs_write_header(super, header, compr_len, len, type, compr); -} - -static s64 logfs_get_free_bytes(struct logfs_area *area, size_t *bytes, - int must_pad) -{ - u32 writesize = logfs_super(area->a_sb)->s_writesize; - s32 ofs; - int ret; - - ret = logfs_open_area(area, *bytes); - if (ret) - return -EAGAIN; - - ofs = area->a_used_bytes; - area->a_used_bytes += *bytes; - - if (must_pad) { - area->a_used_bytes = ALIGN(area->a_used_bytes, writesize); - *bytes = area->a_used_bytes - ofs; - } - - return dev_ofs(area->a_sb, area->a_segno, ofs); -} - -static int logfs_write_je_buf(struct super_block *sb, void *buf, u16 type, - size_t buf_len) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_area *area = super->s_journal_area; - struct logfs_journal_header *jh = super->s_compressed_je; - size_t len; - int must_pad = 0; - s64 ofs; - - len = __logfs_write_je(sb, buf, type, buf_len); - if (jh->h_type == cpu_to_be16(JE_COMMIT)) - must_pad = 1; - - ofs = logfs_get_free_bytes(area, &len, must_pad); - if (ofs < 0) - return ofs; - logfs_buf_write(area, ofs, super->s_compressed_je, len); - BUG_ON(super->s_no_je >= MAX_JOURNAL_ENTRIES); - super->s_je_array[super->s_no_je++] = cpu_to_be64(ofs); - return 0; -} - -static int logfs_write_je(struct super_block *sb, - void* (*write)(struct super_block *sb, void *scratch, - u16 *type, size_t *len)) -{ - void *buf; - size_t len; - u16 type; - - buf = write(sb, logfs_super(sb)->s_je, &type, &len); - return logfs_write_je_buf(sb, buf, type, len); -} - -int write_alias_journal(struct super_block *sb, u64 ino, u64 bix, - level_t level, int child_no, __be64 val) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_obj_alias *oa = super->s_je; - int err = 0, fill = super->s_je_fill; - - log_aliases("logfs_write_obj_aliases #%x(%llx, %llx, %x, %x) %llx\n", - fill, ino, bix, level, child_no, be64_to_cpu(val)); - oa[fill].ino = cpu_to_be64(ino); - oa[fill].bix = cpu_to_be64(bix); - oa[fill].val = val; - oa[fill].level = (__force u8)level; - oa[fill].child_no = cpu_to_be16(child_no); - fill++; - if (fill >= sb->s_blocksize / sizeof(*oa)) { - err = logfs_write_je_buf(sb, oa, JE_OBJ_ALIAS, sb->s_blocksize); - fill = 0; - } - - super->s_je_fill = fill; - return err; -} - -static int logfs_write_obj_aliases(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int err; - - log_journal("logfs_write_obj_aliases: %d aliases to write\n", - super->s_no_object_aliases); - super->s_je_fill = 0; - err = logfs_write_obj_aliases_pagecache(sb); - if (err) - return err; - - if (super->s_je_fill) - err = logfs_write_je_buf(sb, super->s_je, JE_OBJ_ALIAS, - super->s_je_fill - * sizeof(struct logfs_obj_alias)); - return err; -} - -/* - * Write all journal entries. The goto logic ensures that all journal entries - * are written whenever a new segment is used. It is ugly and potentially a - * bit wasteful, but robustness is more important. With this we can *always* - * erase all journal segments except the one containing the most recent commit. - */ -void logfs_write_anchor(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_area *area = super->s_journal_area; - int i, err; - - if (!(super->s_flags & LOGFS_SB_FLAG_DIRTY)) - return; - super->s_flags &= ~LOGFS_SB_FLAG_DIRTY; - - BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN); - mutex_lock(&super->s_journal_mutex); - - /* Do this first or suffer corruption */ - logfs_sync_segments(sb); - account_shadows(sb); - -again: - super->s_no_je = 0; - for_each_area(i) { - if (!super->s_area[i]->a_is_open) - continue; - super->s_sum_index = i; - err = logfs_write_je(sb, logfs_write_area); - if (err) - goto again; - } - err = logfs_write_obj_aliases(sb); - if (err) - goto again; - err = logfs_write_je(sb, logfs_write_erasecount); - if (err) - goto again; - err = logfs_write_je(sb, __logfs_write_anchor); - if (err) - goto again; - err = logfs_write_je(sb, logfs_write_dynsb); - if (err) - goto again; - /* - * Order is imperative. First we sync all writes, including the - * non-committed journal writes. Then we write the final commit and - * sync the current journal segment. - * There is a theoretical bug here. Syncing the journal segment will - * write a number of journal entries and the final commit. All these - * are written in a single operation. If the device layer writes the - * data back-to-front, the commit will precede the other journal - * entries, leaving a race window. - * Two fixes are possible. Preferred is to fix the device layer to - * ensure writes happen front-to-back. Alternatively we can insert - * another logfs_sync_area() super->s_devops->sync() combo before - * writing the commit. - */ - /* - * On another subject, super->s_devops->sync is usually not necessary. - * Unless called from sys_sync or friends, a barrier would suffice. - */ - super->s_devops->sync(sb); - err = logfs_write_je(sb, logfs_write_commit); - if (err) - goto again; - log_journal("Write commit to %llx\n", - be64_to_cpu(super->s_je_array[super->s_no_je - 1])); - logfs_sync_area(area); - BUG_ON(area->a_used_bytes != area->a_written_bytes); - super->s_devops->sync(sb); - - mutex_unlock(&super->s_journal_mutex); - return; -} - -void do_logfs_journal_wl_pass(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_area *area = super->s_journal_area; - struct btree_head32 *head = &super->s_reserved_segments; - u32 segno, ec; - int i, err; - - log_journal("Journal requires wear-leveling.\n"); - /* Drop old segments */ - journal_for_each(i) - if (super->s_journal_seg[i]) { - btree_remove32(head, super->s_journal_seg[i]); - logfs_set_segment_unreserved(sb, - super->s_journal_seg[i], - super->s_journal_ec[i]); - super->s_journal_seg[i] = 0; - super->s_journal_ec[i] = 0; - } - /* Get new segments */ - for (i = 0; i < super->s_no_journal_segs; i++) { - segno = get_best_cand(sb, &super->s_reserve_list, &ec); - super->s_journal_seg[i] = segno; - super->s_journal_ec[i] = ec; - logfs_set_segment_reserved(sb, segno); - err = btree_insert32(head, segno, (void *)1, GFP_NOFS); - BUG_ON(err); /* mempool should prevent this */ - err = logfs_erase_segment(sb, segno, 1); - BUG_ON(err); /* FIXME: remount-ro would be nicer */ - } - /* Manually move journal_area */ - freeseg(sb, area->a_segno); - area->a_segno = super->s_journal_seg[0]; - area->a_is_open = 0; - area->a_used_bytes = 0; - /* Write journal */ - logfs_write_anchor(sb); - /* Write superblocks */ - err = logfs_write_sb(sb); - BUG_ON(err); -} - -static const struct logfs_area_ops journal_area_ops = { - .get_free_segment = journal_get_free_segment, - .get_erase_count = journal_get_erase_count, - .erase_segment = journal_erase_segment, -}; - -int logfs_init_journal(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - size_t bufsize = max_t(size_t, sb->s_blocksize, super->s_writesize) - + MAX_JOURNAL_HEADER; - int ret = -ENOMEM; - - mutex_init(&super->s_journal_mutex); - btree_init_mempool32(&super->s_reserved_segments, super->s_btree_pool); - - super->s_je = kzalloc(bufsize, GFP_KERNEL); - if (!super->s_je) - return ret; - - super->s_compressed_je = kzalloc(bufsize, GFP_KERNEL); - if (!super->s_compressed_je) - return ret; - - super->s_master_inode = logfs_new_meta_inode(sb, LOGFS_INO_MASTER); - if (IS_ERR(super->s_master_inode)) - return PTR_ERR(super->s_master_inode); - - ret = logfs_read_journal(sb); - if (ret) - return -EIO; - - reserve_sb_and_journal(sb); - logfs_calc_free(sb); - - super->s_journal_area->a_ops = &journal_area_ops; - return 0; -} - -void logfs_cleanup_journal(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - - btree_grim_visitor32(&super->s_reserved_segments, 0, NULL); - - kfree(super->s_compressed_je); - kfree(super->s_je); -} diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h deleted file mode 100644 index 11209eefbe59..000000000000 --- a/fs/logfs/logfs.h +++ /dev/null @@ -1,733 +0,0 @@ -/* - * fs/logfs/logfs.h - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - * - * Private header for logfs. - */ -#ifndef FS_LOGFS_LOGFS_H -#define FS_LOGFS_LOGFS_H - -#include <linux/types.h> -#include <linux/btree.h> -#include <linux/crc32.h> -#include <linux/fs.h> -#include <linux/kernel.h> -#include <linux/mempool.h> -#include <linux/pagemap.h> -#include <linux/mtd/mtd.h> -#include "logfs_abi.h" - -#define LOGFS_DEBUG_SUPER (0x0001) -#define LOGFS_DEBUG_SEGMENT (0x0002) -#define LOGFS_DEBUG_JOURNAL (0x0004) -#define LOGFS_DEBUG_DIR (0x0008) -#define LOGFS_DEBUG_FILE (0x0010) -#define LOGFS_DEBUG_INODE (0x0020) -#define LOGFS_DEBUG_READWRITE (0x0040) -#define LOGFS_DEBUG_GC (0x0080) -#define LOGFS_DEBUG_GC_NOISY (0x0100) -#define LOGFS_DEBUG_ALIASES (0x0200) -#define LOGFS_DEBUG_BLOCKMOVE (0x0400) -#define LOGFS_DEBUG_ALL (0xffffffff) - -#define LOGFS_DEBUG (0x01) -/* - * To enable specific log messages, simply define LOGFS_DEBUG to match any - * or all of the above. - */ -#ifndef LOGFS_DEBUG -#define LOGFS_DEBUG (0) -#endif - -#define log_cond(cond, fmt, arg...) do { \ - if (cond) \ - printk(KERN_DEBUG fmt, ##arg); \ -} while (0) - -#define log_super(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_SUPER, fmt, ##arg) -#define log_segment(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_SEGMENT, fmt, ##arg) -#define log_journal(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_JOURNAL, fmt, ##arg) -#define log_dir(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_DIR, fmt, ##arg) -#define log_file(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_FILE, fmt, ##arg) -#define log_inode(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_INODE, fmt, ##arg) -#define log_readwrite(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_READWRITE, fmt, ##arg) -#define log_gc(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_GC, fmt, ##arg) -#define log_gc_noisy(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_GC_NOISY, fmt, ##arg) -#define log_aliases(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_ALIASES, fmt, ##arg) -#define log_blockmove(fmt, arg...) \ - log_cond(LOGFS_DEBUG & LOGFS_DEBUG_BLOCKMOVE, fmt, ##arg) - -#define PG_pre_locked PG_owner_priv_1 -#define PagePreLocked(page) test_bit(PG_pre_locked, &(page)->flags) -#define SetPagePreLocked(page) set_bit(PG_pre_locked, &(page)->flags) -#define ClearPagePreLocked(page) clear_bit(PG_pre_locked, &(page)->flags) - -/* FIXME: This should really be somewhere in the 64bit area. */ -#define LOGFS_LINK_MAX (1<<30) - -/* Read-only filesystem */ -#define LOGFS_SB_FLAG_RO 0x0001 -#define LOGFS_SB_FLAG_DIRTY 0x0002 -#define LOGFS_SB_FLAG_OBJ_ALIAS 0x0004 -#define LOGFS_SB_FLAG_SHUTDOWN 0x0008 - -/* Write Control Flags */ -#define WF_LOCK 0x01 /* take write lock */ -#define WF_WRITE 0x02 /* write block */ -#define WF_DELETE 0x04 /* delete old block */ - -typedef u8 __bitwise level_t; -typedef u8 __bitwise gc_level_t; - -#define LEVEL(level) ((__force level_t)(level)) -#define GC_LEVEL(gc_level) ((__force gc_level_t)(gc_level)) - -#define SUBLEVEL(level) ( (void)((level) == LEVEL(1)), \ - (__force level_t)((__force u8)(level) - 1) ) - -/** - * struct logfs_area - area management information - * - * @a_sb: the superblock this area belongs to - * @a_is_open: 1 if the area is currently open, else 0 - * @a_segno: segment number of area - * @a_written_bytes: number of bytes already written back - * @a_used_bytes: number of used bytes - * @a_ops: area operations (either journal or ostore) - * @a_erase_count: erase count - * @a_level: GC level - */ -struct logfs_area { /* a segment open for writing */ - struct super_block *a_sb; - int a_is_open; - u32 a_segno; - u32 a_written_bytes; - u32 a_used_bytes; - const struct logfs_area_ops *a_ops; - u32 a_erase_count; - gc_level_t a_level; -}; - -/** - * struct logfs_area_ops - area operations - * - * @get_free_segment: fill area->ofs with the offset of a free segment - * @get_erase_count: fill area->erase_count (needs area->ofs) - * @erase_segment: erase and setup segment - */ -struct logfs_area_ops { - void (*get_free_segment)(struct logfs_area *area); - void (*get_erase_count)(struct logfs_area *area); - int (*erase_segment)(struct logfs_area *area); -}; - -struct logfs_super; /* forward */ -/** - * struct logfs_device_ops - device access operations - * - * @readpage: read one page (mm page) - * @writeseg: write one segment. may be a partial segment - * @erase: erase one segment - * @read: read from the device - * @erase: erase part of the device - * @can_write_buf: decide whether wbuf can be written to ofs - */ -struct logfs_device_ops { - struct page *(*find_first_sb)(struct super_block *sb, u64 *ofs); - struct page *(*find_last_sb)(struct super_block *sb, u64 *ofs); - int (*write_sb)(struct super_block *sb, struct page *page); - int (*readpage)(void *_sb, struct page *page); - void (*writeseg)(struct super_block *sb, u64 ofs, size_t len); - int (*erase)(struct super_block *sb, loff_t ofs, size_t len, - int ensure_write); - int (*can_write_buf)(struct super_block *sb, u64 ofs); - void (*sync)(struct super_block *sb); - void (*put_device)(struct logfs_super *s); -}; - -/** - * struct candidate_list - list of similar candidates - */ -struct candidate_list { - struct rb_root rb_tree; - int count; - int maxcount; - int sort_by_ec; -}; - -/** - * struct gc_candidate - "candidate" segment to be garbage collected next - * - * @list: list (either free of low) - * @segno: segment number - * @valid: number of valid bytes - * @erase_count: erase count of segment - * @dist: distance from tree root - * - * Candidates can be on two lists. The free list contains electees rather - * than candidates - segments that no longer contain any valid data. The - * low list contains candidates to be picked for GC. It should be kept - * short. It is not required to always pick a perfect candidate. In the - * worst case GC will have to move more data than absolutely necessary. - */ -struct gc_candidate { - struct rb_node rb_node; - struct candidate_list *list; - u32 segno; - u32 valid; - u32 erase_count; - u8 dist; -}; - -/** - * struct logfs_journal_entry - temporary structure used during journal scan - * - * @used: - * @version: normalized version - * @len: length - * @offset: offset - */ -struct logfs_journal_entry { - int used; - s16 version; - u16 len; - u16 datalen; - u64 offset; -}; - -enum transaction_state { - CREATE_1 = 1, - CREATE_2, - UNLINK_1, - UNLINK_2, - CROSS_RENAME_1, - CROSS_RENAME_2, - TARGET_RENAME_1, - TARGET_RENAME_2, - TARGET_RENAME_3 -}; - -/** - * struct logfs_transaction - essential fields to support atomic dirops - * - * @ino: target inode - * @dir: inode of directory containing dentry - * @pos: pos of dentry in directory - */ -struct logfs_transaction { - enum transaction_state state; - u64 ino; - u64 dir; - u64 pos; -}; - -/** - * struct logfs_shadow - old block in the shadow of a not-yet-committed new one - * @old_ofs: offset of old block on medium - * @new_ofs: offset of new block on medium - * @ino: inode number - * @bix: block index - * @old_len: size of old block, including header - * @new_len: size of new block, including header - * @level: block level - */ -struct logfs_shadow { - u64 old_ofs; - u64 new_ofs; - u64 ino; - u64 bix; - int old_len; - int new_len; - gc_level_t gc_level; -}; - -/** - * struct shadow_tree - * @new: shadows where old_ofs==0, indexed by new_ofs - * @old: shadows where old_ofs!=0, indexed by old_ofs - * @segment_map: bitfield of segments containing shadows - * @no_shadowed_segment: number of segments containing shadows - */ -struct shadow_tree { - struct btree_head64 new; - struct btree_head64 old; - struct btree_head32 segment_map; - int no_shadowed_segments; -}; - -struct object_alias_item { - struct list_head list; - __be64 val; - int child_no; -}; - -/** - * struct logfs_block - contains any block state - * @type: indirect block or inode - * @full: number of fully populated children - * @partial: number of partially populated children - * - * Most blocks are directly represented by page cache pages. But when a block - * becomes dirty, is part of a transaction, contains aliases or is otherwise - * special, a struct logfs_block is allocated to track the additional state. - * Inodes are very similar to indirect blocks, so they can also get one of - * these structures added when appropriate. - */ -#define BLOCK_INDIRECT 1 /* Indirect block */ -#define BLOCK_INODE 2 /* Inode */ -struct logfs_block_ops; -struct logfs_block { - struct list_head alias_list; - struct list_head item_list; - struct super_block *sb; - u64 ino; - u64 bix; - level_t level; - struct page *page; - struct inode *inode; - struct logfs_transaction *ta; - unsigned long alias_map[LOGFS_BLOCK_FACTOR / BITS_PER_LONG]; - const struct logfs_block_ops *ops; - int full; - int partial; - int reserved_bytes; -}; - -typedef int write_alias_t(struct super_block *sb, u64 ino, u64 bix, - level_t level, int child_no, __be64 val); -struct logfs_block_ops { - void (*write_block)(struct logfs_block *block); - void (*free_block)(struct super_block *sb, struct logfs_block*block); - int (*write_alias)(struct super_block *sb, - struct logfs_block *block, - write_alias_t *write_one_alias); -}; - -#define MAX_JOURNAL_ENTRIES 256 - -struct logfs_super { - struct mtd_info *s_mtd; /* underlying device */ - struct block_device *s_bdev; /* underlying device */ - const struct logfs_device_ops *s_devops;/* device access */ - struct inode *s_master_inode; /* inode file */ - struct inode *s_segfile_inode; /* segment file */ - struct inode *s_mapping_inode; /* device mapping */ - atomic_t s_pending_writes; /* outstanting bios */ - long s_flags; - mempool_t *s_btree_pool; /* for btree nodes */ - mempool_t *s_alias_pool; /* aliases in segment.c */ - u64 s_feature_incompat; - u64 s_feature_ro_compat; - u64 s_feature_compat; - u64 s_feature_flags; - u64 s_sb_ofs[2]; - struct page *s_erase_page; /* for dev_bdev.c */ - /* alias.c fields */ - struct btree_head32 s_segment_alias; /* remapped segments */ - int s_no_object_aliases; - struct list_head s_object_alias; /* remapped objects */ - struct btree_head128 s_object_alias_tree; /* remapped objects */ - struct mutex s_object_alias_mutex; - /* dir.c fields */ - struct mutex s_dirop_mutex; /* for creat/unlink/rename */ - u64 s_victim_ino; /* used for atomic dir-ops */ - u64 s_rename_dir; /* source directory ino */ - u64 s_rename_pos; /* position of source dd */ - /* gc.c fields */ - long s_segsize; /* size of a segment */ - int s_segshift; /* log2 of segment size */ - long s_segmask; /* 1 << s_segshift - 1 */ - long s_no_segs; /* segments on device */ - long s_no_journal_segs; /* segments used for journal */ - long s_no_blocks; /* blocks per segment */ - long s_writesize; /* minimum write size */ - int s_writeshift; /* log2 of write size */ - u64 s_size; /* filesystem size */ - struct logfs_area *s_area[LOGFS_NO_AREAS]; /* open segment array */ - u64 s_gec; /* global erase count */ - u64 s_wl_gec_ostore; /* time of last wl event */ - u64 s_wl_gec_journal; /* time of last wl event */ - u64 s_sweeper; /* current sweeper pos */ - u8 s_ifile_levels; /* max level of ifile */ - u8 s_iblock_levels; /* max level of regular files */ - u8 s_data_levels; /* # of segments to leaf block*/ - u8 s_total_levels; /* sum of above three */ - struct btree_head32 s_cand_tree; /* all candidates */ - struct candidate_list s_free_list; /* 100% free segments */ - struct candidate_list s_reserve_list; /* Bad segment reserve */ - struct candidate_list s_low_list[LOGFS_NO_AREAS];/* good candidates */ - struct candidate_list s_ec_list; /* wear level candidates */ - struct btree_head32 s_reserved_segments;/* sb, journal, bad, etc. */ - /* inode.c fields */ - u64 s_last_ino; /* highest ino used */ - long s_inos_till_wrap; - u32 s_generation; /* i_generation for new files */ - struct list_head s_freeing_list; /* inodes being freed */ - /* journal.c fields */ - struct mutex s_journal_mutex; - void *s_je; /* journal entry to compress */ - void *s_compressed_je; /* block to write to journal */ - u32 s_journal_seg[LOGFS_JOURNAL_SEGS]; /* journal segments */ - u32 s_journal_ec[LOGFS_JOURNAL_SEGS]; /* journal erasecounts */ - u64 s_last_version; - struct logfs_area *s_journal_area; /* open journal segment */ - __be64 s_je_array[MAX_JOURNAL_ENTRIES]; - int s_no_je; - - int s_sum_index; /* for the 12 summaries */ - struct shadow_tree s_shadow_tree; - int s_je_fill; /* index of current je */ - /* readwrite.c fields */ - struct mutex s_write_mutex; - int s_lock_count; - mempool_t *s_block_pool; /* struct logfs_block pool */ - mempool_t *s_shadow_pool; /* struct logfs_shadow pool */ - struct list_head s_writeback_list; /* writeback pages */ - /* - * Space accounting: - * - s_used_bytes specifies space used to store valid data objects. - * - s_dirty_used_bytes is space used to store non-committed data - * objects. Those objects have already been written themselves, - * but they don't become valid until all indirect blocks up to the - * journal have been written as well. - * - s_dirty_free_bytes is space used to store the old copy of a - * replaced object, as long as the replacement is non-committed. - * In other words, it is the amount of space freed when all dirty - * blocks are written back. - * - s_free_bytes is the amount of free space available for any - * purpose. - * - s_root_reserve is the amount of free space available only to - * the root user. Non-privileged users can no longer write once - * this watermark has been reached. - * - s_speed_reserve is space which remains unused to speed up - * garbage collection performance. - * - s_dirty_pages is the space reserved for currently dirty pages. - * It is a pessimistic estimate, so some/most will get freed on - * page writeback. - * - * s_used_bytes + s_free_bytes + s_speed_reserve = total usable size - */ - u64 s_free_bytes; - u64 s_used_bytes; - u64 s_dirty_free_bytes; - u64 s_dirty_used_bytes; - u64 s_root_reserve; - u64 s_speed_reserve; - u64 s_dirty_pages; - /* Bad block handling: - * - s_bad_seg_reserve is a number of segments usually kept - * free. When encountering bad blocks, the affected segment's data - * is _temporarily_ moved to a reserved segment. - * - s_bad_segments is the number of known bad segments. - */ - u32 s_bad_seg_reserve; - u32 s_bad_segments; -}; - -/** - * struct logfs_inode - in-memory inode - * - * @vfs_inode: struct inode - * @li_data: data pointers - * @li_used_bytes: number of used bytes - * @li_freeing_list: used to track inodes currently being freed - * @li_flags: inode flags - * @li_refcount: number of internal (GC-induced) references - */ -struct logfs_inode { - struct inode vfs_inode; - u64 li_data[LOGFS_EMBEDDED_FIELDS]; - u64 li_used_bytes; - struct list_head li_freeing_list; - struct logfs_block *li_block; - u32 li_flags; - u8 li_height; - int li_refcount; -}; - -#define journal_for_each(__i) for (__i = 0; __i < LOGFS_JOURNAL_SEGS; __i++) -#define for_each_area(__i) for (__i = 0; __i < LOGFS_NO_AREAS; __i++) -#define for_each_area_down(__i) for (__i = LOGFS_NO_AREAS - 1; __i >= 0; __i--) - -/* compr.c */ -int logfs_compress(void *in, void *out, size_t inlen, size_t outlen); -int logfs_uncompress(void *in, void *out, size_t inlen, size_t outlen); -int __init logfs_compr_init(void); -void logfs_compr_exit(void); - -/* dev_bdev.c */ -#ifdef CONFIG_BLOCK -int logfs_get_sb_bdev(struct logfs_super *s, - struct file_system_type *type, - const char *devname); -#else -static inline int logfs_get_sb_bdev(struct logfs_super *s, - struct file_system_type *type, - const char *devname) -{ - return -ENODEV; -} -#endif - -/* dev_mtd.c */ -#if IS_ENABLED(CONFIG_MTD) -int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr); -#else -static inline int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr) -{ - return -ENODEV; -} -#endif - -/* dir.c */ -extern const struct inode_operations logfs_dir_iops; -extern const struct file_operations logfs_dir_fops; -int logfs_replay_journal(struct super_block *sb); - -/* file.c */ -extern const struct inode_operations logfs_reg_iops; -extern const struct file_operations logfs_reg_fops; -extern const struct address_space_operations logfs_reg_aops; -int logfs_readpage(struct file *file, struct page *page); -long logfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg); -int logfs_fsync(struct file *file, loff_t start, loff_t end, int datasync); - -/* gc.c */ -u32 get_best_cand(struct super_block *sb, struct candidate_list *list, u32 *ec); -void logfs_gc_pass(struct super_block *sb); -int logfs_check_areas(struct super_block *sb); -int logfs_init_gc(struct super_block *sb); -void logfs_cleanup_gc(struct super_block *sb); - -/* inode.c */ -extern const struct super_operations logfs_super_operations; -struct inode *logfs_iget(struct super_block *sb, ino_t ino); -struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *cookie); -void logfs_safe_iput(struct inode *inode, int cookie); -struct inode *logfs_new_inode(struct inode *dir, umode_t mode); -struct inode *logfs_new_meta_inode(struct super_block *sb, u64 ino); -struct inode *logfs_read_meta_inode(struct super_block *sb, u64 ino); -int logfs_init_inode_cache(void); -void logfs_destroy_inode_cache(void); -void logfs_set_blocks(struct inode *inode, u64 no); -/* these logically belong into inode.c but actually reside in readwrite.c */ -int logfs_read_inode(struct inode *inode); -int __logfs_write_inode(struct inode *inode, struct page *, long flags); -void logfs_evict_inode(struct inode *inode); - -/* journal.c */ -void logfs_write_anchor(struct super_block *sb); -int logfs_init_journal(struct super_block *sb); -void logfs_cleanup_journal(struct super_block *sb); -int write_alias_journal(struct super_block *sb, u64 ino, u64 bix, - level_t level, int child_no, __be64 val); -void do_logfs_journal_wl_pass(struct super_block *sb); - -/* readwrite.c */ -pgoff_t logfs_pack_index(u64 bix, level_t level); -void logfs_unpack_index(pgoff_t index, u64 *bix, level_t *level); -int logfs_inode_write(struct inode *inode, const void *buf, size_t count, - loff_t bix, long flags, struct shadow_tree *shadow_tree); -int logfs_readpage_nolock(struct page *page); -int logfs_write_buf(struct inode *inode, struct page *page, long flags); -int logfs_delete(struct inode *inode, pgoff_t index, - struct shadow_tree *shadow_tree); -int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs, - gc_level_t gc_level, long flags); -int logfs_is_valid_block(struct super_block *sb, u64 ofs, u64 ino, u64 bix, - gc_level_t gc_level); -int logfs_truncate(struct inode *inode, u64 size); -u64 logfs_seek_hole(struct inode *inode, u64 bix); -u64 logfs_seek_data(struct inode *inode, u64 bix); -int logfs_open_segfile(struct super_block *sb); -int logfs_init_rw(struct super_block *sb); -void logfs_cleanup_rw(struct super_block *sb); -void logfs_add_transaction(struct inode *inode, struct logfs_transaction *ta); -void logfs_del_transaction(struct inode *inode, struct logfs_transaction *ta); -void logfs_write_block(struct logfs_block *block, long flags); -int logfs_write_obj_aliases_pagecache(struct super_block *sb); -void logfs_get_segment_entry(struct super_block *sb, u32 segno, - struct logfs_segment_entry *se); -void logfs_set_segment_used(struct super_block *sb, u64 ofs, int increment); -void logfs_set_segment_erased(struct super_block *sb, u32 segno, u32 ec, - gc_level_t gc_level); -void logfs_set_segment_reserved(struct super_block *sb, u32 segno); -void logfs_set_segment_unreserved(struct super_block *sb, u32 segno, u32 ec); -struct logfs_block *__alloc_block(struct super_block *sb, - u64 ino, u64 bix, level_t level); -void __free_block(struct super_block *sb, struct logfs_block *block); -void btree_write_block(struct logfs_block *block); -void initialize_block_counters(struct page *page, struct logfs_block *block, - __be64 *array, int page_is_empty); -int logfs_exist_block(struct inode *inode, u64 bix); -int get_page_reserve(struct inode *inode, struct page *page); -void logfs_get_wblocks(struct super_block *sb, struct page *page, int lock); -void logfs_put_wblocks(struct super_block *sb, struct page *page, int lock); -extern const struct logfs_block_ops indirect_block_ops; - -/* segment.c */ -int logfs_erase_segment(struct super_block *sb, u32 ofs, int ensure_erase); -int wbuf_read(struct super_block *sb, u64 ofs, size_t len, void *buf); -int logfs_segment_read(struct inode *inode, struct page *page, u64 ofs, u64 bix, - level_t level); -int logfs_segment_write(struct inode *inode, struct page *page, - struct logfs_shadow *shadow); -int logfs_segment_delete(struct inode *inode, struct logfs_shadow *shadow); -int logfs_load_object_aliases(struct super_block *sb, - struct logfs_obj_alias *oa, int count); -void move_page_to_btree(struct page *page); -int logfs_init_mapping(struct super_block *sb); -void logfs_sync_area(struct logfs_area *area); -void logfs_sync_segments(struct super_block *sb); -void freeseg(struct super_block *sb, u32 segno); -void free_areas(struct super_block *sb); - -/* area handling */ -int logfs_init_areas(struct super_block *sb); -void logfs_cleanup_areas(struct super_block *sb); -int logfs_open_area(struct logfs_area *area, size_t bytes); -int __logfs_buf_write(struct logfs_area *area, u64 ofs, void *buf, size_t len, - int use_filler); - -static inline int logfs_buf_write(struct logfs_area *area, u64 ofs, - void *buf, size_t len) -{ - return __logfs_buf_write(area, ofs, buf, len, 0); -} - -static inline int logfs_buf_recover(struct logfs_area *area, u64 ofs, - void *buf, size_t len) -{ - return __logfs_buf_write(area, ofs, buf, len, 1); -} - -/* super.c */ -struct page *emergency_read_begin(struct address_space *mapping, pgoff_t index); -void emergency_read_end(struct page *page); -void logfs_crash_dump(struct super_block *sb); -int logfs_statfs(struct dentry *dentry, struct kstatfs *stats); -int logfs_check_ds(struct logfs_disk_super *ds); -int logfs_write_sb(struct super_block *sb); - -static inline struct logfs_super *logfs_super(struct super_block *sb) -{ - return sb->s_fs_info; -} - -static inline struct logfs_inode *logfs_inode(struct inode *inode) -{ - return container_of(inode, struct logfs_inode, vfs_inode); -} - -static inline void logfs_set_ro(struct super_block *sb) -{ - logfs_super(sb)->s_flags |= LOGFS_SB_FLAG_RO; -} - -#define LOGFS_BUG(sb) do { \ - struct super_block *__sb = sb; \ - logfs_crash_dump(__sb); \ - logfs_super(__sb)->s_flags |= LOGFS_SB_FLAG_RO; \ - BUG(); \ -} while (0) - -#define LOGFS_BUG_ON(condition, sb) \ - do { if (unlikely(condition)) LOGFS_BUG((sb)); } while (0) - -static inline __be32 logfs_crc32(void *data, size_t len, size_t skip) -{ - return cpu_to_be32(crc32(~0, data+skip, len-skip)); -} - -static inline u8 logfs_type(struct inode *inode) -{ - return (inode->i_mode >> 12) & 15; -} - -static inline pgoff_t logfs_index(struct super_block *sb, u64 pos) -{ - return pos >> sb->s_blocksize_bits; -} - -static inline u64 dev_ofs(struct super_block *sb, u32 segno, u32 ofs) -{ - return ((u64)segno << logfs_super(sb)->s_segshift) + ofs; -} - -static inline u32 seg_no(struct super_block *sb, u64 ofs) -{ - return ofs >> logfs_super(sb)->s_segshift; -} - -static inline u32 seg_ofs(struct super_block *sb, u64 ofs) -{ - return ofs & logfs_super(sb)->s_segmask; -} - -static inline u64 seg_align(struct super_block *sb, u64 ofs) -{ - return ofs & ~logfs_super(sb)->s_segmask; -} - -static inline struct logfs_block *logfs_block(struct page *page) -{ - return (void *)page->private; -} - -static inline level_t shrink_level(gc_level_t __level) -{ - u8 level = (__force u8)__level; - - if (level >= LOGFS_MAX_LEVELS) - level -= LOGFS_MAX_LEVELS; - return (__force level_t)level; -} - -static inline gc_level_t expand_level(u64 ino, level_t __level) -{ - u8 level = (__force u8)__level; - - if (ino == LOGFS_INO_MASTER) { - /* ifile has separate areas */ - level += LOGFS_MAX_LEVELS; - } - return (__force gc_level_t)level; -} - -static inline int logfs_block_shift(struct super_block *sb, level_t level) -{ - level = shrink_level((__force gc_level_t)level); - return (__force int)level * (sb->s_blocksize_bits - 3); -} - -static inline u64 logfs_block_mask(struct super_block *sb, level_t level) -{ - return ~0ull << logfs_block_shift(sb, level); -} - -static inline struct logfs_area *get_area(struct super_block *sb, - gc_level_t gc_level) -{ - return logfs_super(sb)->s_area[(__force u8)gc_level]; -} - -static inline void logfs_mempool_destroy(mempool_t *pool) -{ - if (pool) - mempool_destroy(pool); -} - -#endif diff --git a/fs/logfs/logfs_abi.h b/fs/logfs/logfs_abi.h deleted file mode 100644 index ae960519c54a..000000000000 --- a/fs/logfs/logfs_abi.h +++ /dev/null @@ -1,629 +0,0 @@ -/* - * fs/logfs/logfs_abi.h - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - * - * Public header for logfs. - */ -#ifndef FS_LOGFS_LOGFS_ABI_H -#define FS_LOGFS_LOGFS_ABI_H - -/* For out-of-kernel compiles */ -#ifndef BUILD_BUG_ON -#define BUILD_BUG_ON(condition) /**/ -#endif - -#define SIZE_CHECK(type, size) \ -static inline void check_##type(void) \ -{ \ - BUILD_BUG_ON(sizeof(struct type) != (size)); \ -} - -/* - * Throughout the logfs code, we're constantly dealing with blocks at - * various positions or offsets. To remove confusion, we stricly - * distinguish between a "position" - the logical position within a - * file and an "offset" - the physical location within the device. - * - * Any usage of the term offset for a logical location or position for - * a physical one is a bug and should get fixed. - */ - -/* - * Block are allocated in one of several segments depending on their - * level. The following levels are used: - * 0 - regular data block - * 1 - i1 indirect blocks - * 2 - i2 indirect blocks - * 3 - i3 indirect blocks - * 4 - i4 indirect blocks - * 5 - i5 indirect blocks - * 6 - ifile data blocks - * 7 - ifile i1 indirect blocks - * 8 - ifile i2 indirect blocks - * 9 - ifile i3 indirect blocks - * 10 - ifile i4 indirect blocks - * 11 - ifile i5 indirect blocks - * Potential levels to be used in the future: - * 12 - gc recycled blocks, long-lived data - * 13 - replacement blocks, short-lived data - * - * Levels 1-11 are necessary for robust gc operations and help separate - * short-lived metadata from longer-lived file data. In the future, - * file data should get separated into several segments based on simple - * heuristics. Old data recycled during gc operation is expected to be - * long-lived. New data is of uncertain life expectancy. New data - * used to replace older blocks in existing files is expected to be - * short-lived. - */ - - -/* Magic numbers. 64bit for superblock, 32bit for statfs f_type */ -#define LOGFS_MAGIC 0x7a3a8e5cb9d5bf67ull -#define LOGFS_MAGIC_U32 0xc97e8168u - -/* - * Various blocksize related macros. Blocksize is currently fixed at 4KiB. - * Sooner or later that should become configurable and the macros replaced - * by something superblock-dependent. Pointers in indirect blocks are and - * will remain 64bit. - * - * LOGFS_BLOCKSIZE - self-explaining - * LOGFS_BLOCK_FACTOR - number of pointers per indirect block - * LOGFS_BLOCK_BITS - log2 of LOGFS_BLOCK_FACTOR, used for shifts - */ -#define LOGFS_BLOCKSIZE (4096ull) -#define LOGFS_BLOCK_FACTOR (LOGFS_BLOCKSIZE / sizeof(u64)) -#define LOGFS_BLOCK_BITS (9) - -/* - * Number of blocks at various levels of indirection. There are 16 direct - * block pointers plus a single indirect pointer. - */ -#define I0_BLOCKS (16) -#define I1_BLOCKS LOGFS_BLOCK_FACTOR -#define I2_BLOCKS (LOGFS_BLOCK_FACTOR * I1_BLOCKS) -#define I3_BLOCKS (LOGFS_BLOCK_FACTOR * I2_BLOCKS) -#define I4_BLOCKS (LOGFS_BLOCK_FACTOR * I3_BLOCKS) -#define I5_BLOCKS (LOGFS_BLOCK_FACTOR * I4_BLOCKS) - -#define INDIRECT_INDEX I0_BLOCKS -#define LOGFS_EMBEDDED_FIELDS (I0_BLOCKS + 1) - -/* - * Sizes at which files require another level of indirection. Files smaller - * than LOGFS_EMBEDDED_SIZE can be completely stored in the inode itself, - * similar like ext2 fast symlinks. - * - * Data at a position smaller than LOGFS_I0_SIZE is accessed through the - * direct pointers, else through the 1x indirect pointer and so forth. - */ -#define LOGFS_EMBEDDED_SIZE (LOGFS_EMBEDDED_FIELDS * sizeof(u64)) -#define LOGFS_I0_SIZE (I0_BLOCKS * LOGFS_BLOCKSIZE) -#define LOGFS_I1_SIZE (I1_BLOCKS * LOGFS_BLOCKSIZE) -#define LOGFS_I2_SIZE (I2_BLOCKS * LOGFS_BLOCKSIZE) -#define LOGFS_I3_SIZE (I3_BLOCKS * LOGFS_BLOCKSIZE) -#define LOGFS_I4_SIZE (I4_BLOCKS * LOGFS_BLOCKSIZE) -#define LOGFS_I5_SIZE (I5_BLOCKS * LOGFS_BLOCKSIZE) - -/* - * Each indirect block pointer must have this flag set, if all block pointers - * behind it are set, i.e. there is no hole hidden in the shadow of this - * indirect block pointer. - */ -#define LOGFS_FULLY_POPULATED (1ULL << 63) -#define pure_ofs(ofs) (ofs & ~LOGFS_FULLY_POPULATED) - -/* - * LogFS needs to separate data into levels. Each level is defined as the - * maximal possible distance from the master inode (inode of the inode file). - * Data blocks reside on level 0, 1x indirect block on level 1, etc. - * Inodes reside on level 6, indirect blocks for the inode file on levels 7-11. - * This effort is necessary to guarantee garbage collection to always make - * progress. - * - * LOGFS_MAX_INDIRECT is the maximal indirection through indirect blocks, - * LOGFS_MAX_LEVELS is one more for the actual data level of a file. It is - * the maximal number of levels for one file. - * LOGFS_NO_AREAS is twice that, as the inode file and regular files are - * effectively stacked on top of each other. - */ -#define LOGFS_MAX_INDIRECT (5) -#define LOGFS_MAX_LEVELS (LOGFS_MAX_INDIRECT + 1) -#define LOGFS_NO_AREAS (2 * LOGFS_MAX_LEVELS) - -/* Maximum size of filenames */ -#define LOGFS_MAX_NAMELEN (255) - -/* Number of segments in the primary journal. */ -#define LOGFS_JOURNAL_SEGS (16) - -/* Maximum number of free/erased/etc. segments in journal entries */ -#define MAX_CACHED_SEGS (64) - - -/* - * LOGFS_OBJECT_HEADERSIZE is the size of a single header in the object store, - * LOGFS_MAX_OBJECTSIZE the size of the largest possible object, including - * its header, - * LOGFS_SEGMENT_RESERVE is the amount of space reserved for each segment for - * its segment header and the padded space at the end when no further objects - * fit. - */ -#define LOGFS_OBJECT_HEADERSIZE (0x1c) -#define LOGFS_SEGMENT_HEADERSIZE (0x18) -#define LOGFS_MAX_OBJECTSIZE (LOGFS_OBJECT_HEADERSIZE + LOGFS_BLOCKSIZE) -#define LOGFS_SEGMENT_RESERVE \ - (LOGFS_SEGMENT_HEADERSIZE + LOGFS_MAX_OBJECTSIZE - 1) - -/* - * Segment types: - * SEG_SUPER - Data or indirect block - * SEG_JOURNAL - Inode - * SEG_OSTORE - Dentry - */ -enum { - SEG_SUPER = 0x01, - SEG_JOURNAL = 0x02, - SEG_OSTORE = 0x03, -}; - -/** - * struct logfs_segment_header - per-segment header in the ostore - * - * @crc: crc32 of header (there is no data) - * @pad: unused, must be 0 - * @type: segment type, see above - * @level: GC level for all objects in this segment - * @segno: segment number - * @ec: erase count for this segment - * @gec: global erase count at time of writing - */ -struct logfs_segment_header { - __be32 crc; - __be16 pad; - __u8 type; - __u8 level; - __be32 segno; - __be32 ec; - __be64 gec; -}; - -SIZE_CHECK(logfs_segment_header, LOGFS_SEGMENT_HEADERSIZE); - -#define LOGFS_FEATURES_INCOMPAT (0ull) -#define LOGFS_FEATURES_RO_COMPAT (0ull) -#define LOGFS_FEATURES_COMPAT (0ull) - -/** - * struct logfs_disk_super - on-medium superblock - * - * @ds_magic: magic number, must equal LOGFS_MAGIC - * @ds_crc: crc32 of structure starting with the next field - * @ds_ifile_levels: maximum number of levels for ifile - * @ds_iblock_levels: maximum number of levels for regular files - * @ds_data_levels: number of separate levels for data - * @pad0: reserved, must be 0 - * @ds_feature_incompat: incompatible filesystem features - * @ds_feature_ro_compat: read-only compatible filesystem features - * @ds_feature_compat: compatible filesystem features - * @ds_flags: flags - * @ds_segment_shift: log2 of segment size - * @ds_block_shift: log2 of block size - * @ds_write_shift: log2 of write size - * @pad1: reserved, must be 0 - * @ds_journal_seg: segments used by primary journal - * @ds_root_reserve: bytes reserved for the superuser - * @ds_speed_reserve: bytes reserved to speed up GC - * @ds_bad_seg_reserve: number of segments reserved to handle bad blocks - * @pad2: reserved, must be 0 - * @pad3: reserved, must be 0 - * - * Contains only read-only fields. Read-write fields like the amount of used - * space is tracked in the dynamic superblock, which is stored in the journal. - */ -struct logfs_disk_super { - struct logfs_segment_header ds_sh; - __be64 ds_magic; - - __be32 ds_crc; - __u8 ds_ifile_levels; - __u8 ds_iblock_levels; - __u8 ds_data_levels; - __u8 ds_segment_shift; - __u8 ds_block_shift; - __u8 ds_write_shift; - __u8 pad0[6]; - - __be64 ds_filesystem_size; - __be32 ds_segment_size; - __be32 ds_bad_seg_reserve; - - __be64 ds_feature_incompat; - __be64 ds_feature_ro_compat; - - __be64 ds_feature_compat; - __be64 ds_feature_flags; - - __be64 ds_root_reserve; - __be64 ds_speed_reserve; - - __be32 ds_journal_seg[LOGFS_JOURNAL_SEGS]; - - __be64 ds_super_ofs[2]; - __be64 pad3[8]; -}; - -SIZE_CHECK(logfs_disk_super, 256); - -/* - * Object types: - * OBJ_BLOCK - Data or indirect block - * OBJ_INODE - Inode - * OBJ_DENTRY - Dentry - */ -enum { - OBJ_BLOCK = 0x04, - OBJ_INODE = 0x05, - OBJ_DENTRY = 0x06, -}; - -/** - * struct logfs_object_header - per-object header in the ostore - * - * @crc: crc32 of header, excluding data_crc - * @len: length of data - * @type: object type, see above - * @compr: compression type - * @ino: inode number - * @bix: block index - * @data_crc: crc32 of payload - */ -struct logfs_object_header { - __be32 crc; - __be16 len; - __u8 type; - __u8 compr; - __be64 ino; - __be64 bix; - __be32 data_crc; -} __attribute__((packed)); - -SIZE_CHECK(logfs_object_header, LOGFS_OBJECT_HEADERSIZE); - -/* - * Reserved inode numbers: - * LOGFS_INO_MASTER - master inode (for inode file) - * LOGFS_INO_ROOT - root directory - * LOGFS_INO_SEGFILE - per-segment used bytes and erase count - */ -enum { - LOGFS_INO_MAPPING = 0x00, - LOGFS_INO_MASTER = 0x01, - LOGFS_INO_ROOT = 0x02, - LOGFS_INO_SEGFILE = 0x03, - LOGFS_RESERVED_INOS = 0x10, -}; - -/* - * Inode flags. High bits should never be written to the medium. They are - * reserved for in-memory usage. - * Low bits should either remain in sync with the corresponding FS_*_FL or - * reuse slots that obviously don't make sense for logfs. - * - * LOGFS_IF_DIRTY Inode must be written back - * LOGFS_IF_ZOMBIE Inode has been deleted - * LOGFS_IF_STILLBORN -ENOSPC happened when creating inode - */ -#define LOGFS_IF_COMPRESSED 0x00000004 /* == FS_COMPR_FL */ -#define LOGFS_IF_DIRTY 0x20000000 -#define LOGFS_IF_ZOMBIE 0x40000000 -#define LOGFS_IF_STILLBORN 0x80000000 - -/* Flags available to chattr */ -#define LOGFS_FL_USER_VISIBLE (LOGFS_IF_COMPRESSED) -#define LOGFS_FL_USER_MODIFIABLE (LOGFS_IF_COMPRESSED) -/* Flags inherited from parent directory on file/directory creation */ -#define LOGFS_FL_INHERITED (LOGFS_IF_COMPRESSED) - -/** - * struct logfs_disk_inode - on-medium inode - * - * @di_mode: file mode - * @di_pad: reserved, must be 0 - * @di_flags: inode flags, see above - * @di_uid: user id - * @di_gid: group id - * @di_ctime: change time - * @di_mtime: modify time - * @di_refcount: reference count (aka nlink or link count) - * @di_generation: inode generation, for nfs - * @di_used_bytes: number of bytes used - * @di_size: file size - * @di_data: data pointers - */ -struct logfs_disk_inode { - __be16 di_mode; - __u8 di_height; - __u8 di_pad; - __be32 di_flags; - __be32 di_uid; - __be32 di_gid; - - __be64 di_ctime; - __be64 di_mtime; - - __be64 di_atime; - __be32 di_refcount; - __be32 di_generation; - - __be64 di_used_bytes; - __be64 di_size; - - __be64 di_data[LOGFS_EMBEDDED_FIELDS]; -}; - -SIZE_CHECK(logfs_disk_inode, 200); - -#define INODE_POINTER_OFS \ - (offsetof(struct logfs_disk_inode, di_data) / sizeof(__be64)) -#define INODE_USED_OFS \ - (offsetof(struct logfs_disk_inode, di_used_bytes) / sizeof(__be64)) -#define INODE_SIZE_OFS \ - (offsetof(struct logfs_disk_inode, di_size) / sizeof(__be64)) -#define INODE_HEIGHT_OFS (0) - -/** - * struct logfs_disk_dentry - on-medium dentry structure - * - * @ino: inode number - * @namelen: length of file name - * @type: file type, identical to bits 12..15 of mode - * @name: file name - */ -/* FIXME: add 6 bytes of padding to remove the __packed */ -struct logfs_disk_dentry { - __be64 ino; - __be16 namelen; - __u8 type; - __u8 name[LOGFS_MAX_NAMELEN]; -} __attribute__((packed)); - -SIZE_CHECK(logfs_disk_dentry, 266); - -#define RESERVED 0xffffffff -#define BADSEG 0xffffffff -/** - * struct logfs_segment_entry - segment file entry - * - * @ec_level: erase count and level - * @valid: number of valid bytes - * - * Segment file contains one entry for every segment. ec_level contains the - * erasecount in the upper 28 bits and the level in the lower 4 bits. An - * ec_level of BADSEG (-1) identifies bad segments. valid contains the number - * of valid bytes or RESERVED (-1 again) if the segment is used for either the - * superblock or the journal, or when the segment is bad. - */ -struct logfs_segment_entry { - __be32 ec_level; - __be32 valid; -}; - -SIZE_CHECK(logfs_segment_entry, 8); - -/** - * struct logfs_journal_header - header for journal entries (JEs) - * - * @h_crc: crc32 of journal entry - * @h_len: length of compressed journal entry, - * not including header - * @h_datalen: length of uncompressed data - * @h_type: JE type - * @h_compr: compression type - * @h_pad: reserved - */ -struct logfs_journal_header { - __be32 h_crc; - __be16 h_len; - __be16 h_datalen; - __be16 h_type; - __u8 h_compr; - __u8 h_pad[5]; -}; - -SIZE_CHECK(logfs_journal_header, 16); - -/* - * Life expectency of data. - * VIM_DEFAULT - default vim - * VIM_SEGFILE - for segment file only - very short-living - * VIM_GC - GC'd data - likely long-living - */ -enum logfs_vim { - VIM_DEFAULT = 0, - VIM_SEGFILE = 1, -}; - -/** - * struct logfs_je_area - wbuf header - * - * @segno: segment number of area - * @used_bytes: number of bytes already used - * @gc_level: GC level - * @vim: life expectancy of data - * - * "Areas" are segments currently being used for writing. There is at least - * one area per GC level. Several may be used to separate long-living from - * short-living data. If an area with unknown vim is encountered, it can - * simply be closed. - * The write buffer immediately follow this header. - */ -struct logfs_je_area { - __be32 segno; - __be32 used_bytes; - __u8 gc_level; - __u8 vim; -} __attribute__((packed)); - -SIZE_CHECK(logfs_je_area, 10); - -#define MAX_JOURNAL_HEADER \ - (sizeof(struct logfs_journal_header) + sizeof(struct logfs_je_area)) - -/** - * struct logfs_je_dynsb - dynamic superblock - * - * @ds_gec: global erase count - * @ds_sweeper: current position of GC "sweeper" - * @ds_rename_dir: source directory ino (see dir.c documentation) - * @ds_rename_pos: position of source dd (see dir.c documentation) - * @ds_victim_ino: victims of incomplete dir operation (see dir.c) - * @ds_victim_ino: parent inode of victim (see dir.c) - * @ds_used_bytes: number of used bytes - */ -struct logfs_je_dynsb { - __be64 ds_gec; - __be64 ds_sweeper; - - __be64 ds_rename_dir; - __be64 ds_rename_pos; - - __be64 ds_victim_ino; - __be64 ds_victim_parent; /* XXX */ - - __be64 ds_used_bytes; - __be32 ds_generation; - __be32 pad; -}; - -SIZE_CHECK(logfs_je_dynsb, 64); - -/** - * struct logfs_je_anchor - anchor of filesystem tree, aka master inode - * - * @da_size: size of inode file - * @da_last_ino: last created inode - * @da_used_bytes: number of bytes used - * @da_data: data pointers - */ -struct logfs_je_anchor { - __be64 da_size; - __be64 da_last_ino; - - __be64 da_used_bytes; - u8 da_height; - u8 pad[7]; - - __be64 da_data[LOGFS_EMBEDDED_FIELDS]; -}; - -SIZE_CHECK(logfs_je_anchor, 168); - -/** - * struct logfs_je_spillout - spillout entry (from 1st to 2nd journal) - * - * @so_segment: segments used for 2nd journal - * - * Length of the array is given by h_len field in the header. - */ -struct logfs_je_spillout { - __be64 so_segment[0]; -}; - -SIZE_CHECK(logfs_je_spillout, 0); - -/** - * struct logfs_je_journal_ec - erase counts for all journal segments - * - * @ec: erase count - * - * Length of the array is given by h_len field in the header. - */ -struct logfs_je_journal_ec { - __be32 ec[0]; -}; - -SIZE_CHECK(logfs_je_journal_ec, 0); - -/** - * struct logfs_je_free_segments - list of free segmetns with erase count - */ -struct logfs_je_free_segments { - __be32 segno; - __be32 ec; -}; - -SIZE_CHECK(logfs_je_free_segments, 8); - -/** - * struct logfs_seg_alias - list of segment aliases - */ -struct logfs_seg_alias { - __be32 old_segno; - __be32 new_segno; -}; - -SIZE_CHECK(logfs_seg_alias, 8); - -/** - * struct logfs_obj_alias - list of object aliases - */ -struct logfs_obj_alias { - __be64 ino; - __be64 bix; - __be64 val; - u8 level; - u8 pad[5]; - __be16 child_no; -}; - -SIZE_CHECK(logfs_obj_alias, 32); - -/** - * Compression types. - * - * COMPR_NONE - uncompressed - * COMPR_ZLIB - compressed with zlib - */ -enum { - COMPR_NONE = 0, - COMPR_ZLIB = 1, -}; - -/* - * Journal entries come in groups of 16. First group contains unique - * entries, next groups contain one entry per level - * - * JE_FIRST - smallest possible journal entry number - * - * JEG_BASE - base group, containing unique entries - * JE_COMMIT - commit entry, validates all previous entries - * JE_DYNSB - dynamic superblock, anything that ought to be in the - * superblock but cannot because it is read-write data - * JE_ANCHOR - anchor aka master inode aka inode file's inode - * JE_ERASECOUNT erasecounts for all journal segments - * JE_SPILLOUT - unused - * JE_SEG_ALIAS - aliases segments - * JE_AREA - area description - * - * JE_LAST - largest possible journal entry number - */ -enum { - JE_FIRST = 0x01, - - JEG_BASE = 0x00, - JE_COMMIT = 0x02, - JE_DYNSB = 0x03, - JE_ANCHOR = 0x04, - JE_ERASECOUNT = 0x05, - JE_SPILLOUT = 0x06, - JE_OBJ_ALIAS = 0x0d, - JE_AREA = 0x0e, - - JE_LAST = 0x0e, -}; - -#endif diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c deleted file mode 100644 index bf19bf4a243f..000000000000 --- a/fs/logfs/readwrite.c +++ /dev/null @@ -1,2298 +0,0 @@ -/* - * fs/logfs/readwrite.c - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - * - * - * Actually contains five sets of very similar functions: - * read read blocks from a file - * seek_hole find next hole - * seek_data find next data block - * valid check whether a block still belongs to a file - * write write blocks to a file - * delete delete a block (for directories and ifile) - * rewrite move existing blocks of a file to a new location (gc helper) - * truncate truncate a file - */ -#include "logfs.h" -#include <linux/sched.h> -#include <linux/slab.h> - -static u64 adjust_bix(u64 bix, level_t level) -{ - switch (level) { - case 0: - return bix; - case LEVEL(1): - return max_t(u64, bix, I0_BLOCKS); - case LEVEL(2): - return max_t(u64, bix, I1_BLOCKS); - case LEVEL(3): - return max_t(u64, bix, I2_BLOCKS); - case LEVEL(4): - return max_t(u64, bix, I3_BLOCKS); - case LEVEL(5): - return max_t(u64, bix, I4_BLOCKS); - default: - WARN_ON(1); - return bix; - } -} - -static inline u64 maxbix(u8 height) -{ - return 1ULL << (LOGFS_BLOCK_BITS * height); -} - -/** - * The inode address space is cut in two halves. Lower half belongs to data - * pages, upper half to indirect blocks. If the high bit (INDIRECT_BIT) is - * set, the actual block index (bix) and level can be derived from the page - * index. - * - * The lowest three bits of the block index are set to 0 after packing and - * unpacking. Since the lowest n bits (9 for 4KiB blocksize) are ignored - * anyway this is harmless. - */ -#define ARCH_SHIFT (BITS_PER_LONG - 32) -#define INDIRECT_BIT (0x80000000UL << ARCH_SHIFT) -#define LEVEL_SHIFT (28 + ARCH_SHIFT) -static inline pgoff_t first_indirect_block(void) -{ - return INDIRECT_BIT | (1ULL << LEVEL_SHIFT); -} - -pgoff_t logfs_pack_index(u64 bix, level_t level) -{ - pgoff_t index; - - BUG_ON(bix >= INDIRECT_BIT); - if (level == 0) - return bix; - - index = INDIRECT_BIT; - index |= (__force long)level << LEVEL_SHIFT; - index |= bix >> ((__force u8)level * LOGFS_BLOCK_BITS); - return index; -} - -void logfs_unpack_index(pgoff_t index, u64 *bix, level_t *level) -{ - u8 __level; - - if (!(index & INDIRECT_BIT)) { - *bix = index; - *level = 0; - return; - } - - __level = (index & ~INDIRECT_BIT) >> LEVEL_SHIFT; - *level = LEVEL(__level); - *bix = (index << (__level * LOGFS_BLOCK_BITS)) & ~INDIRECT_BIT; - *bix = adjust_bix(*bix, *level); - return; -} -#undef ARCH_SHIFT -#undef INDIRECT_BIT -#undef LEVEL_SHIFT - -/* - * Time is stored as nanoseconds since the epoch. - */ -static struct timespec be64_to_timespec(__be64 betime) -{ - return ns_to_timespec(be64_to_cpu(betime)); -} - -static __be64 timespec_to_be64(struct timespec tsp) -{ - return cpu_to_be64((u64)tsp.tv_sec * NSEC_PER_SEC + tsp.tv_nsec); -} - -static void logfs_disk_to_inode(struct logfs_disk_inode *di, struct inode*inode) -{ - struct logfs_inode *li = logfs_inode(inode); - int i; - - inode->i_mode = be16_to_cpu(di->di_mode); - li->li_height = di->di_height; - li->li_flags = be32_to_cpu(di->di_flags); - i_uid_write(inode, be32_to_cpu(di->di_uid)); - i_gid_write(inode, be32_to_cpu(di->di_gid)); - inode->i_size = be64_to_cpu(di->di_size); - logfs_set_blocks(inode, be64_to_cpu(di->di_used_bytes)); - inode->i_atime = be64_to_timespec(di->di_atime); - inode->i_ctime = be64_to_timespec(di->di_ctime); - inode->i_mtime = be64_to_timespec(di->di_mtime); - set_nlink(inode, be32_to_cpu(di->di_refcount)); - inode->i_generation = be32_to_cpu(di->di_generation); - - switch (inode->i_mode & S_IFMT) { - case S_IFSOCK: /* fall through */ - case S_IFBLK: /* fall through */ - case S_IFCHR: /* fall through */ - case S_IFIFO: - inode->i_rdev = be64_to_cpu(di->di_data[0]); - break; - case S_IFDIR: /* fall through */ - case S_IFREG: /* fall through */ - case S_IFLNK: - for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) - li->li_data[i] = be64_to_cpu(di->di_data[i]); - break; - default: - BUG(); - } -} - -static void logfs_inode_to_disk(struct inode *inode, struct logfs_disk_inode*di) -{ - struct logfs_inode *li = logfs_inode(inode); - int i; - - di->di_mode = cpu_to_be16(inode->i_mode); - di->di_height = li->li_height; - di->di_pad = 0; - di->di_flags = cpu_to_be32(li->li_flags); - di->di_uid = cpu_to_be32(i_uid_read(inode)); - di->di_gid = cpu_to_be32(i_gid_read(inode)); - di->di_size = cpu_to_be64(i_size_read(inode)); - di->di_used_bytes = cpu_to_be64(li->li_used_bytes); - di->di_atime = timespec_to_be64(inode->i_atime); - di->di_ctime = timespec_to_be64(inode->i_ctime); - di->di_mtime = timespec_to_be64(inode->i_mtime); - di->di_refcount = cpu_to_be32(inode->i_nlink); - di->di_generation = cpu_to_be32(inode->i_generation); - - switch (inode->i_mode & S_IFMT) { - case S_IFSOCK: /* fall through */ - case S_IFBLK: /* fall through */ - case S_IFCHR: /* fall through */ - case S_IFIFO: - di->di_data[0] = cpu_to_be64(inode->i_rdev); - break; - case S_IFDIR: /* fall through */ - case S_IFREG: /* fall through */ - case S_IFLNK: - for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) - di->di_data[i] = cpu_to_be64(li->li_data[i]); - break; - default: - BUG(); - } -} - -static void __logfs_set_blocks(struct inode *inode) -{ - struct super_block *sb = inode->i_sb; - struct logfs_inode *li = logfs_inode(inode); - - inode->i_blocks = ULONG_MAX; - if (li->li_used_bytes >> sb->s_blocksize_bits < ULONG_MAX) - inode->i_blocks = ALIGN(li->li_used_bytes, 512) >> 9; -} - -void logfs_set_blocks(struct inode *inode, u64 bytes) -{ - struct logfs_inode *li = logfs_inode(inode); - - li->li_used_bytes = bytes; - __logfs_set_blocks(inode); -} - -static void prelock_page(struct super_block *sb, struct page *page, int lock) -{ - struct logfs_super *super = logfs_super(sb); - - BUG_ON(!PageLocked(page)); - if (lock) { - BUG_ON(PagePreLocked(page)); - SetPagePreLocked(page); - } else { - /* We are in GC path. */ - if (PagePreLocked(page)) - super->s_lock_count++; - else - SetPagePreLocked(page); - } -} - -static void preunlock_page(struct super_block *sb, struct page *page, int lock) -{ - struct logfs_super *super = logfs_super(sb); - - BUG_ON(!PageLocked(page)); - if (lock) - ClearPagePreLocked(page); - else { - /* We are in GC path. */ - BUG_ON(!PagePreLocked(page)); - if (super->s_lock_count) - super->s_lock_count--; - else - ClearPagePreLocked(page); - } -} - -/* - * Logfs is prone to an AB-BA deadlock where one task tries to acquire - * s_write_mutex with a locked page and GC tries to get that page while holding - * s_write_mutex. - * To solve this issue logfs will ignore the page lock iff the page in question - * is waiting for s_write_mutex. We annotate this fact by setting PG_pre_locked - * in addition to PG_locked. - */ -void logfs_get_wblocks(struct super_block *sb, struct page *page, int lock) -{ - struct logfs_super *super = logfs_super(sb); - - if (page) - prelock_page(sb, page, lock); - - if (lock) { - mutex_lock(&super->s_write_mutex); - logfs_gc_pass(sb); - /* FIXME: We also have to check for shadowed space - * and mempool fill grade */ - } -} - -void logfs_put_wblocks(struct super_block *sb, struct page *page, int lock) -{ - struct logfs_super *super = logfs_super(sb); - - if (page) - preunlock_page(sb, page, lock); - /* Order matters - we must clear PG_pre_locked before releasing - * s_write_mutex or we could race against another task. */ - if (lock) - mutex_unlock(&super->s_write_mutex); -} - -static struct page *logfs_get_read_page(struct inode *inode, u64 bix, - level_t level) -{ - return find_or_create_page(inode->i_mapping, - logfs_pack_index(bix, level), GFP_NOFS); -} - -static void logfs_put_read_page(struct page *page) -{ - unlock_page(page); - put_page(page); -} - -static void logfs_lock_write_page(struct page *page) -{ - int loop = 0; - - while (unlikely(!trylock_page(page))) { - if (loop++ > 0x1000) { - /* Has been observed once so far... */ - printk(KERN_ERR "stack at %p\n", &loop); - BUG(); - } - if (PagePreLocked(page)) { - /* Holder of page lock is waiting for us, it - * is safe to use this page. */ - break; - } - /* Some other process has this page locked and has - * nothing to do with us. Wait for it to finish. - */ - schedule(); - } - BUG_ON(!PageLocked(page)); -} - -static struct page *logfs_get_write_page(struct inode *inode, u64 bix, - level_t level) -{ - struct address_space *mapping = inode->i_mapping; - pgoff_t index = logfs_pack_index(bix, level); - struct page *page; - int err; - -repeat: - page = find_get_page(mapping, index); - if (!page) { - page = __page_cache_alloc(GFP_NOFS); - if (!page) - return NULL; - err = add_to_page_cache_lru(page, mapping, index, GFP_NOFS); - if (unlikely(err)) { - put_page(page); - if (err == -EEXIST) - goto repeat; - return NULL; - } - } else logfs_lock_write_page(page); - BUG_ON(!PageLocked(page)); - return page; -} - -static void logfs_unlock_write_page(struct page *page) -{ - if (!PagePreLocked(page)) - unlock_page(page); -} - -static void logfs_put_write_page(struct page *page) -{ - logfs_unlock_write_page(page); - put_page(page); -} - -static struct page *logfs_get_page(struct inode *inode, u64 bix, level_t level, - int rw) -{ - if (rw == READ) - return logfs_get_read_page(inode, bix, level); - else - return logfs_get_write_page(inode, bix, level); -} - -static void logfs_put_page(struct page *page, int rw) -{ - if (rw == READ) - logfs_put_read_page(page); - else - logfs_put_write_page(page); -} - -static unsigned long __get_bits(u64 val, int skip, int no) -{ - u64 ret = val; - - ret >>= skip * no; - ret <<= 64 - no; - ret >>= 64 - no; - return ret; -} - -static unsigned long get_bits(u64 val, level_t skip) -{ - return __get_bits(val, (__force int)skip, LOGFS_BLOCK_BITS); -} - -static inline void init_shadow_tree(struct super_block *sb, - struct shadow_tree *tree) -{ - struct logfs_super *super = logfs_super(sb); - - btree_init_mempool64(&tree->new, super->s_btree_pool); - btree_init_mempool64(&tree->old, super->s_btree_pool); -} - -static void indirect_write_block(struct logfs_block *block) -{ - struct page *page; - struct inode *inode; - int ret; - - page = block->page; - inode = page->mapping->host; - logfs_lock_write_page(page); - ret = logfs_write_buf(inode, page, 0); - logfs_unlock_write_page(page); - /* - * This needs some rework. Unless you want your filesystem to run - * completely synchronously (you don't), the filesystem will always - * report writes as 'successful' before the actual work has been - * done. The actual work gets done here and this is where any errors - * will show up. And there isn't much we can do about it, really. - * - * Some attempts to fix the errors (move from bad blocks, retry io,...) - * have already been done, so anything left should be either a broken - * device or a bug somewhere in logfs itself. Being relatively new, - * the odds currently favor a bug, so for now the line below isn't - * entirely tasteles. - */ - BUG_ON(ret); -} - -static void inode_write_block(struct logfs_block *block) -{ - struct inode *inode; - int ret; - - inode = block->inode; - if (inode->i_ino == LOGFS_INO_MASTER) - logfs_write_anchor(inode->i_sb); - else { - ret = __logfs_write_inode(inode, NULL, 0); - /* see indirect_write_block comment */ - BUG_ON(ret); - } -} - -/* - * This silences a false, yet annoying gcc warning. I hate it when my editor - * jumps into bitops.h each time I recompile this file. - * TODO: Complain to gcc folks about this and upgrade compiler. - */ -static unsigned long fnb(const unsigned long *addr, - unsigned long size, unsigned long offset) -{ - return find_next_bit(addr, size, offset); -} - -static __be64 inode_val0(struct inode *inode) -{ - struct logfs_inode *li = logfs_inode(inode); - u64 val; - - /* - * Explicit shifting generates good code, but must match the format - * of the structure. Add some paranoia just in case. - */ - BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_mode) != 0); - BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_height) != 2); - BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_flags) != 4); - - val = (u64)inode->i_mode << 48 | - (u64)li->li_height << 40 | - (u64)li->li_flags; - return cpu_to_be64(val); -} - -static int inode_write_alias(struct super_block *sb, - struct logfs_block *block, write_alias_t *write_one_alias) -{ - struct inode *inode = block->inode; - struct logfs_inode *li = logfs_inode(inode); - unsigned long pos; - u64 ino , bix; - __be64 val; - level_t level; - int err; - - for (pos = 0; ; pos++) { - pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos); - if (pos >= LOGFS_EMBEDDED_FIELDS + INODE_POINTER_OFS) - return 0; - - switch (pos) { - case INODE_HEIGHT_OFS: - val = inode_val0(inode); - break; - case INODE_USED_OFS: - val = cpu_to_be64(li->li_used_bytes); - break; - case INODE_SIZE_OFS: - val = cpu_to_be64(i_size_read(inode)); - break; - case INODE_POINTER_OFS ... INODE_POINTER_OFS + LOGFS_EMBEDDED_FIELDS - 1: - val = cpu_to_be64(li->li_data[pos - INODE_POINTER_OFS]); - break; - default: - BUG(); - } - - ino = LOGFS_INO_MASTER; - bix = inode->i_ino; - level = LEVEL(0); - err = write_one_alias(sb, ino, bix, level, pos, val); - if (err) - return err; - } -} - -static int indirect_write_alias(struct super_block *sb, - struct logfs_block *block, write_alias_t *write_one_alias) -{ - unsigned long pos; - struct page *page = block->page; - u64 ino , bix; - __be64 *child, val; - level_t level; - int err; - - for (pos = 0; ; pos++) { - pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos); - if (pos >= LOGFS_BLOCK_FACTOR) - return 0; - - ino = page->mapping->host->i_ino; - logfs_unpack_index(page->index, &bix, &level); - child = kmap_atomic(page); - val = child[pos]; - kunmap_atomic(child); - err = write_one_alias(sb, ino, bix, level, pos, val); - if (err) - return err; - } -} - -int logfs_write_obj_aliases_pagecache(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_block *block; - int err; - - list_for_each_entry(block, &super->s_object_alias, alias_list) { - err = block->ops->write_alias(sb, block, write_alias_journal); - if (err) - return err; - } - return 0; -} - -void __free_block(struct super_block *sb, struct logfs_block *block) -{ - BUG_ON(!list_empty(&block->item_list)); - list_del(&block->alias_list); - mempool_free(block, logfs_super(sb)->s_block_pool); -} - -static void inode_free_block(struct super_block *sb, struct logfs_block *block) -{ - struct inode *inode = block->inode; - - logfs_inode(inode)->li_block = NULL; - __free_block(sb, block); -} - -static void indirect_free_block(struct super_block *sb, - struct logfs_block *block) -{ - struct page *page = block->page; - - if (PagePrivate(page)) { - ClearPagePrivate(page); - put_page(page); - set_page_private(page, 0); - } - __free_block(sb, block); -} - - -static const struct logfs_block_ops inode_block_ops = { - .write_block = inode_write_block, - .free_block = inode_free_block, - .write_alias = inode_write_alias, -}; - -const struct logfs_block_ops indirect_block_ops = { - .write_block = indirect_write_block, - .free_block = indirect_free_block, - .write_alias = indirect_write_alias, -}; - -struct logfs_block *__alloc_block(struct super_block *sb, - u64 ino, u64 bix, level_t level) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_block *block; - - block = mempool_alloc(super->s_block_pool, GFP_NOFS); - memset(block, 0, sizeof(*block)); - INIT_LIST_HEAD(&block->alias_list); - INIT_LIST_HEAD(&block->item_list); - block->sb = sb; - block->ino = ino; - block->bix = bix; - block->level = level; - return block; -} - -static void alloc_inode_block(struct inode *inode) -{ - struct logfs_inode *li = logfs_inode(inode); - struct logfs_block *block; - - if (li->li_block) - return; - - block = __alloc_block(inode->i_sb, LOGFS_INO_MASTER, inode->i_ino, 0); - block->inode = inode; - li->li_block = block; - block->ops = &inode_block_ops; -} - -void initialize_block_counters(struct page *page, struct logfs_block *block, - __be64 *array, int page_is_empty) -{ - u64 ptr; - int i, start; - - block->partial = 0; - block->full = 0; - start = 0; - if (page->index < first_indirect_block()) { - /* Counters are pointless on level 0 */ - return; - } - if (page->index == first_indirect_block()) { - /* Skip unused pointers */ - start = I0_BLOCKS; - block->full = I0_BLOCKS; - } - if (!page_is_empty) { - for (i = start; i < LOGFS_BLOCK_FACTOR; i++) { - ptr = be64_to_cpu(array[i]); - if (ptr) - block->partial++; - if (ptr & LOGFS_FULLY_POPULATED) - block->full++; - } - } -} - -static void alloc_data_block(struct inode *inode, struct page *page) -{ - struct logfs_block *block; - u64 bix; - level_t level; - - if (PagePrivate(page)) - return; - - logfs_unpack_index(page->index, &bix, &level); - block = __alloc_block(inode->i_sb, inode->i_ino, bix, level); - block->page = page; - - SetPagePrivate(page); - get_page(page); - set_page_private(page, (unsigned long) block); - - block->ops = &indirect_block_ops; -} - -static void alloc_indirect_block(struct inode *inode, struct page *page, - int page_is_empty) -{ - struct logfs_block *block; - __be64 *array; - - if (PagePrivate(page)) - return; - - alloc_data_block(inode, page); - - block = logfs_block(page); - array = kmap_atomic(page); - initialize_block_counters(page, block, array, page_is_empty); - kunmap_atomic(array); -} - -static void block_set_pointer(struct page *page, int index, u64 ptr) -{ - struct logfs_block *block = logfs_block(page); - __be64 *array; - u64 oldptr; - - BUG_ON(!block); - array = kmap_atomic(page); - oldptr = be64_to_cpu(array[index]); - array[index] = cpu_to_be64(ptr); - kunmap_atomic(array); - SetPageUptodate(page); - - block->full += !!(ptr & LOGFS_FULLY_POPULATED) - - !!(oldptr & LOGFS_FULLY_POPULATED); - block->partial += !!ptr - !!oldptr; -} - -static u64 block_get_pointer(struct page *page, int index) -{ - __be64 *block; - u64 ptr; - - block = kmap_atomic(page); - ptr = be64_to_cpu(block[index]); - kunmap_atomic(block); - return ptr; -} - -static int logfs_read_empty(struct page *page) -{ - zero_user_segment(page, 0, PAGE_SIZE); - return 0; -} - -static int logfs_read_direct(struct inode *inode, struct page *page) -{ - struct logfs_inode *li = logfs_inode(inode); - pgoff_t index = page->index; - u64 block; - - block = li->li_data[index]; - if (!block) - return logfs_read_empty(page); - - return logfs_segment_read(inode, page, block, index, 0); -} - -static int logfs_read_loop(struct inode *inode, struct page *page, - int rw_context) -{ - struct logfs_inode *li = logfs_inode(inode); - u64 bix, bofs = li->li_data[INDIRECT_INDEX]; - level_t level, target_level; - int ret; - struct page *ipage; - - logfs_unpack_index(page->index, &bix, &target_level); - if (!bofs) - return logfs_read_empty(page); - - if (bix >= maxbix(li->li_height)) - return logfs_read_empty(page); - - for (level = LEVEL(li->li_height); - (__force u8)level > (__force u8)target_level; - level = SUBLEVEL(level)){ - ipage = logfs_get_page(inode, bix, level, rw_context); - if (!ipage) - return -ENOMEM; - - ret = logfs_segment_read(inode, ipage, bofs, bix, level); - if (ret) { - logfs_put_read_page(ipage); - return ret; - } - - bofs = block_get_pointer(ipage, get_bits(bix, SUBLEVEL(level))); - logfs_put_page(ipage, rw_context); - if (!bofs) - return logfs_read_empty(page); - } - - return logfs_segment_read(inode, page, bofs, bix, 0); -} - -static int logfs_read_block(struct inode *inode, struct page *page, - int rw_context) -{ - pgoff_t index = page->index; - - if (index < I0_BLOCKS) - return logfs_read_direct(inode, page); - return logfs_read_loop(inode, page, rw_context); -} - -static int logfs_exist_loop(struct inode *inode, u64 bix) -{ - struct logfs_inode *li = logfs_inode(inode); - u64 bofs = li->li_data[INDIRECT_INDEX]; - level_t level; - int ret; - struct page *ipage; - - if (!bofs) - return 0; - if (bix >= maxbix(li->li_height)) - return 0; - - for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)) { - ipage = logfs_get_read_page(inode, bix, level); - if (!ipage) - return -ENOMEM; - - ret = logfs_segment_read(inode, ipage, bofs, bix, level); - if (ret) { - logfs_put_read_page(ipage); - return ret; - } - - bofs = block_get_pointer(ipage, get_bits(bix, SUBLEVEL(level))); - logfs_put_read_page(ipage); - if (!bofs) - return 0; - } - - return 1; -} - -int logfs_exist_block(struct inode *inode, u64 bix) -{ - struct logfs_inode *li = logfs_inode(inode); - - if (bix < I0_BLOCKS) - return !!li->li_data[bix]; - return logfs_exist_loop(inode, bix); -} - -static u64 seek_holedata_direct(struct inode *inode, u64 bix, int data) -{ - struct logfs_inode *li = logfs_inode(inode); - - for (; bix < I0_BLOCKS; bix++) - if (data ^ (li->li_data[bix] == 0)) - return bix; - return I0_BLOCKS; -} - -static u64 seek_holedata_loop(struct inode *inode, u64 bix, int data) -{ - struct logfs_inode *li = logfs_inode(inode); - __be64 *rblock; - u64 increment, bofs = li->li_data[INDIRECT_INDEX]; - level_t level; - int ret, slot; - struct page *page; - - BUG_ON(!bofs); - - for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)) { - increment = 1 << (LOGFS_BLOCK_BITS * ((__force u8)level-1)); - page = logfs_get_read_page(inode, bix, level); - if (!page) - return bix; - - ret = logfs_segment_read(inode, page, bofs, bix, level); - if (ret) { - logfs_put_read_page(page); - return bix; - } - - slot = get_bits(bix, SUBLEVEL(level)); - rblock = kmap_atomic(page); - while (slot < LOGFS_BLOCK_FACTOR) { - if (data && (rblock[slot] != 0)) - break; - if (!data && !(be64_to_cpu(rblock[slot]) & LOGFS_FULLY_POPULATED)) - break; - slot++; - bix += increment; - bix &= ~(increment - 1); - } - if (slot >= LOGFS_BLOCK_FACTOR) { - kunmap_atomic(rblock); - logfs_put_read_page(page); - return bix; - } - bofs = be64_to_cpu(rblock[slot]); - kunmap_atomic(rblock); - logfs_put_read_page(page); - if (!bofs) { - BUG_ON(data); - return bix; - } - } - return bix; -} - -/** - * logfs_seek_hole - find next hole starting at a given block index - * @inode: inode to search in - * @bix: block index to start searching - * - * Returns next hole. If the file doesn't contain any further holes, the - * block address next to eof is returned instead. - */ -u64 logfs_seek_hole(struct inode *inode, u64 bix) -{ - struct logfs_inode *li = logfs_inode(inode); - - if (bix < I0_BLOCKS) { - bix = seek_holedata_direct(inode, bix, 0); - if (bix < I0_BLOCKS) - return bix; - } - - if (!li->li_data[INDIRECT_INDEX]) - return bix; - else if (li->li_data[INDIRECT_INDEX] & LOGFS_FULLY_POPULATED) - bix = maxbix(li->li_height); - else if (bix >= maxbix(li->li_height)) - return bix; - else { - bix = seek_holedata_loop(inode, bix, 0); - if (bix < maxbix(li->li_height)) - return bix; - /* Should not happen anymore. But if some port writes semi- - * corrupt images (as this one used to) we might run into it. - */ - WARN_ON_ONCE(bix == maxbix(li->li_height)); - } - - return bix; -} - -static u64 __logfs_seek_data(struct inode *inode, u64 bix) -{ - struct logfs_inode *li = logfs_inode(inode); - - if (bix < I0_BLOCKS) { - bix = seek_holedata_direct(inode, bix, 1); - if (bix < I0_BLOCKS) - return bix; - } - - if (bix < maxbix(li->li_height)) { - if (!li->li_data[INDIRECT_INDEX]) - bix = maxbix(li->li_height); - else - return seek_holedata_loop(inode, bix, 1); - } - - return bix; -} - -/** - * logfs_seek_data - find next data block after a given block index - * @inode: inode to search in - * @bix: block index to start searching - * - * Returns next data block. If the file doesn't contain any further data - * blocks, the last block in the file is returned instead. - */ -u64 logfs_seek_data(struct inode *inode, u64 bix) -{ - struct super_block *sb = inode->i_sb; - u64 ret, end; - - ret = __logfs_seek_data(inode, bix); - end = i_size_read(inode) >> sb->s_blocksize_bits; - if (ret >= end) - ret = max(bix, end); - return ret; -} - -static int logfs_is_valid_direct(struct logfs_inode *li, u64 bix, u64 ofs) -{ - return pure_ofs(li->li_data[bix]) == ofs; -} - -static int __logfs_is_valid_loop(struct inode *inode, u64 bix, - u64 ofs, u64 bofs) -{ - struct logfs_inode *li = logfs_inode(inode); - level_t level; - int ret; - struct page *page; - - for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)){ - page = logfs_get_write_page(inode, bix, level); - BUG_ON(!page); - - ret = logfs_segment_read(inode, page, bofs, bix, level); - if (ret) { - logfs_put_write_page(page); - return 0; - } - - bofs = block_get_pointer(page, get_bits(bix, SUBLEVEL(level))); - logfs_put_write_page(page); - if (!bofs) - return 0; - - if (pure_ofs(bofs) == ofs) - return 1; - } - return 0; -} - -static int logfs_is_valid_loop(struct inode *inode, u64 bix, u64 ofs) -{ - struct logfs_inode *li = logfs_inode(inode); - u64 bofs = li->li_data[INDIRECT_INDEX]; - - if (!bofs) - return 0; - - if (bix >= maxbix(li->li_height)) - return 0; - - if (pure_ofs(bofs) == ofs) - return 1; - - return __logfs_is_valid_loop(inode, bix, ofs, bofs); -} - -static int __logfs_is_valid_block(struct inode *inode, u64 bix, u64 ofs) -{ - struct logfs_inode *li = logfs_inode(inode); - - if ((inode->i_nlink == 0) && atomic_read(&inode->i_count) == 1) - return 0; - - if (bix < I0_BLOCKS) - return logfs_is_valid_direct(li, bix, ofs); - return logfs_is_valid_loop(inode, bix, ofs); -} - -/** - * logfs_is_valid_block - check whether this block is still valid - * - * @sb: superblock - * @ofs: block physical offset - * @ino: block inode number - * @bix: block index - * @gc_level: block level - * - * Returns 0 if the block is invalid, 1 if it is valid and 2 if it will - * become invalid once the journal is written. - */ -int logfs_is_valid_block(struct super_block *sb, u64 ofs, u64 ino, u64 bix, - gc_level_t gc_level) -{ - struct logfs_super *super = logfs_super(sb); - struct inode *inode; - int ret, cookie; - - /* Umount closes a segment with free blocks remaining. Those - * blocks are by definition invalid. */ - if (ino == -1) - return 0; - - LOGFS_BUG_ON((u64)(u_long)ino != ino, sb); - - inode = logfs_safe_iget(sb, ino, &cookie); - if (IS_ERR(inode)) - goto invalid; - - ret = __logfs_is_valid_block(inode, bix, ofs); - logfs_safe_iput(inode, cookie); - if (ret) - return ret; - -invalid: - /* Block is nominally invalid, but may still sit in the shadow tree, - * waiting for a journal commit. - */ - if (btree_lookup64(&super->s_shadow_tree.old, ofs)) - return 2; - return 0; -} - -int logfs_readpage_nolock(struct page *page) -{ - struct inode *inode = page->mapping->host; - int ret = -EIO; - - ret = logfs_read_block(inode, page, READ); - - if (ret) { - ClearPageUptodate(page); - SetPageError(page); - } else { - SetPageUptodate(page); - ClearPageError(page); - } - flush_dcache_page(page); - - return ret; -} - -static int logfs_reserve_bytes(struct inode *inode, int bytes) -{ - struct logfs_super *super = logfs_super(inode->i_sb); - u64 available = super->s_free_bytes + super->s_dirty_free_bytes - - super->s_dirty_used_bytes - super->s_dirty_pages; - - if (!bytes) - return 0; - - if (available < bytes) - return -ENOSPC; - - if (available < bytes + super->s_root_reserve && - !capable(CAP_SYS_RESOURCE)) - return -ENOSPC; - - return 0; -} - -int get_page_reserve(struct inode *inode, struct page *page) -{ - struct logfs_super *super = logfs_super(inode->i_sb); - struct logfs_block *block = logfs_block(page); - int ret; - - if (block && block->reserved_bytes) - return 0; - - logfs_get_wblocks(inode->i_sb, page, WF_LOCK); - while ((ret = logfs_reserve_bytes(inode, 6 * LOGFS_MAX_OBJECTSIZE)) && - !list_empty(&super->s_writeback_list)) { - block = list_entry(super->s_writeback_list.next, - struct logfs_block, alias_list); - block->ops->write_block(block); - } - if (!ret) { - alloc_data_block(inode, page); - block = logfs_block(page); - block->reserved_bytes += 6 * LOGFS_MAX_OBJECTSIZE; - super->s_dirty_pages += 6 * LOGFS_MAX_OBJECTSIZE; - list_move_tail(&block->alias_list, &super->s_writeback_list); - } - logfs_put_wblocks(inode->i_sb, page, WF_LOCK); - return ret; -} - -/* - * We are protected by write lock. Push victims up to superblock level - * and release transaction when appropriate. - */ -/* FIXME: This is currently called from the wrong spots. */ -static void logfs_handle_transaction(struct inode *inode, - struct logfs_transaction *ta) -{ - struct logfs_super *super = logfs_super(inode->i_sb); - - if (!ta) - return; - logfs_inode(inode)->li_block->ta = NULL; - - if (inode->i_ino != LOGFS_INO_MASTER) { - BUG(); /* FIXME: Yes, this needs more thought */ - /* just remember the transaction until inode is written */ - //BUG_ON(logfs_inode(inode)->li_transaction); - //logfs_inode(inode)->li_transaction = ta; - return; - } - - switch (ta->state) { - case CREATE_1: /* fall through */ - case UNLINK_1: - BUG_ON(super->s_victim_ino); - super->s_victim_ino = ta->ino; - break; - case CREATE_2: /* fall through */ - case UNLINK_2: - BUG_ON(super->s_victim_ino != ta->ino); - super->s_victim_ino = 0; - /* transaction ends here - free it */ - kfree(ta); - break; - case CROSS_RENAME_1: - BUG_ON(super->s_rename_dir); - BUG_ON(super->s_rename_pos); - super->s_rename_dir = ta->dir; - super->s_rename_pos = ta->pos; - break; - case CROSS_RENAME_2: - BUG_ON(super->s_rename_dir != ta->dir); - BUG_ON(super->s_rename_pos != ta->pos); - super->s_rename_dir = 0; - super->s_rename_pos = 0; - kfree(ta); - break; - case TARGET_RENAME_1: - BUG_ON(super->s_rename_dir); - BUG_ON(super->s_rename_pos); - BUG_ON(super->s_victim_ino); - super->s_rename_dir = ta->dir; - super->s_rename_pos = ta->pos; - super->s_victim_ino = ta->ino; - break; - case TARGET_RENAME_2: - BUG_ON(super->s_rename_dir != ta->dir); - BUG_ON(super->s_rename_pos != ta->pos); - BUG_ON(super->s_victim_ino != ta->ino); - super->s_rename_dir = 0; - super->s_rename_pos = 0; - break; - case TARGET_RENAME_3: - BUG_ON(super->s_rename_dir); - BUG_ON(super->s_rename_pos); - BUG_ON(super->s_victim_ino != ta->ino); - super->s_victim_ino = 0; - kfree(ta); - break; - default: - BUG(); - } -} - -/* - * Not strictly a reservation, but rather a check that we still have enough - * space to satisfy the write. - */ -static int logfs_reserve_blocks(struct inode *inode, int blocks) -{ - return logfs_reserve_bytes(inode, blocks * LOGFS_MAX_OBJECTSIZE); -} - -struct write_control { - u64 ofs; - long flags; -}; - -static struct logfs_shadow *alloc_shadow(struct inode *inode, u64 bix, - level_t level, u64 old_ofs) -{ - struct logfs_super *super = logfs_super(inode->i_sb); - struct logfs_shadow *shadow; - - shadow = mempool_alloc(super->s_shadow_pool, GFP_NOFS); - memset(shadow, 0, sizeof(*shadow)); - shadow->ino = inode->i_ino; - shadow->bix = bix; - shadow->gc_level = expand_level(inode->i_ino, level); - shadow->old_ofs = old_ofs & ~LOGFS_FULLY_POPULATED; - return shadow; -} - -static void free_shadow(struct inode *inode, struct logfs_shadow *shadow) -{ - struct logfs_super *super = logfs_super(inode->i_sb); - - mempool_free(shadow, super->s_shadow_pool); -} - -static void mark_segment(struct shadow_tree *tree, u32 segno) -{ - int err; - - if (!btree_lookup32(&tree->segment_map, segno)) { - err = btree_insert32(&tree->segment_map, segno, (void *)1, - GFP_NOFS); - BUG_ON(err); - tree->no_shadowed_segments++; - } -} - -/** - * fill_shadow_tree - Propagate shadow tree changes due to a write - * @inode: Inode owning the page - * @page: Struct page that was written - * @shadow: Shadow for the current write - * - * Writes in logfs can result in two semi-valid objects. The old object - * is still valid as long as it can be reached by following pointers on - * the medium. Only when writes propagate all the way up to the journal - * has the new object safely replaced the old one. - * - * To handle this problem, a struct logfs_shadow is used to represent - * every single write. It is attached to the indirect block, which is - * marked dirty. When the indirect block is written, its shadows are - * handed up to the next indirect block (or inode). Untimately they - * will reach the master inode and be freed upon journal commit. - * - * This function handles a single step in the propagation. It adds the - * shadow for the current write to the tree, along with any shadows in - * the page's tree, in case it was an indirect block. If a page is - * written, the inode parameter is left NULL, if an inode is written, - * the page parameter is left NULL. - */ -static void fill_shadow_tree(struct inode *inode, struct page *page, - struct logfs_shadow *shadow) -{ - struct logfs_super *super = logfs_super(inode->i_sb); - struct logfs_block *block = logfs_block(page); - struct shadow_tree *tree = &super->s_shadow_tree; - - if (PagePrivate(page)) { - if (block->alias_map) - super->s_no_object_aliases -= bitmap_weight( - block->alias_map, LOGFS_BLOCK_FACTOR); - logfs_handle_transaction(inode, block->ta); - block->ops->free_block(inode->i_sb, block); - } - if (shadow) { - if (shadow->old_ofs) - btree_insert64(&tree->old, shadow->old_ofs, shadow, - GFP_NOFS); - else - btree_insert64(&tree->new, shadow->new_ofs, shadow, - GFP_NOFS); - - super->s_dirty_used_bytes += shadow->new_len; - super->s_dirty_free_bytes += shadow->old_len; - mark_segment(tree, shadow->old_ofs >> super->s_segshift); - mark_segment(tree, shadow->new_ofs >> super->s_segshift); - } -} - -static void logfs_set_alias(struct super_block *sb, struct logfs_block *block, - long child_no) -{ - struct logfs_super *super = logfs_super(sb); - - if (block->inode && block->inode->i_ino == LOGFS_INO_MASTER) { - /* Aliases in the master inode are pointless. */ - return; - } - - if (!test_bit(child_no, block->alias_map)) { - set_bit(child_no, block->alias_map); - super->s_no_object_aliases++; - } - list_move_tail(&block->alias_list, &super->s_object_alias); -} - -/* - * Object aliases can and often do change the size and occupied space of a - * file. So not only do we have to change the pointers, we also have to - * change inode->i_size and li->li_used_bytes. Which is done by setting - * another two object aliases for the inode itself. - */ -static void set_iused(struct inode *inode, struct logfs_shadow *shadow) -{ - struct logfs_inode *li = logfs_inode(inode); - - if (shadow->new_len == shadow->old_len) - return; - - alloc_inode_block(inode); - li->li_used_bytes += shadow->new_len - shadow->old_len; - __logfs_set_blocks(inode); - logfs_set_alias(inode->i_sb, li->li_block, INODE_USED_OFS); - logfs_set_alias(inode->i_sb, li->li_block, INODE_SIZE_OFS); -} - -static int logfs_write_i0(struct inode *inode, struct page *page, - struct write_control *wc) -{ - struct logfs_shadow *shadow; - u64 bix; - level_t level; - int full, err = 0; - - logfs_unpack_index(page->index, &bix, &level); - if (wc->ofs == 0) - if (logfs_reserve_blocks(inode, 1)) - return -ENOSPC; - - shadow = alloc_shadow(inode, bix, level, wc->ofs); - if (wc->flags & WF_WRITE) - err = logfs_segment_write(inode, page, shadow); - if (wc->flags & WF_DELETE) - logfs_segment_delete(inode, shadow); - if (err) { - free_shadow(inode, shadow); - return err; - } - - set_iused(inode, shadow); - full = 1; - if (level != 0) { - alloc_indirect_block(inode, page, 0); - full = logfs_block(page)->full == LOGFS_BLOCK_FACTOR; - } - fill_shadow_tree(inode, page, shadow); - wc->ofs = shadow->new_ofs; - if (wc->ofs && full) - wc->ofs |= LOGFS_FULLY_POPULATED; - return 0; -} - -static int logfs_write_direct(struct inode *inode, struct page *page, - long flags) -{ - struct logfs_inode *li = logfs_inode(inode); - struct write_control wc = { - .ofs = li->li_data[page->index], - .flags = flags, - }; - int err; - - alloc_inode_block(inode); - - err = logfs_write_i0(inode, page, &wc); - if (err) - return err; - - li->li_data[page->index] = wc.ofs; - logfs_set_alias(inode->i_sb, li->li_block, - page->index + INODE_POINTER_OFS); - return 0; -} - -static int ptr_change(u64 ofs, struct page *page) -{ - struct logfs_block *block = logfs_block(page); - int empty0, empty1, full0, full1; - - empty0 = ofs == 0; - empty1 = block->partial == 0; - if (empty0 != empty1) - return 1; - - /* The !! is necessary to shrink result to int */ - full0 = !!(ofs & LOGFS_FULLY_POPULATED); - full1 = block->full == LOGFS_BLOCK_FACTOR; - if (full0 != full1) - return 1; - return 0; -} - -static int __logfs_write_rec(struct inode *inode, struct page *page, - struct write_control *this_wc, - pgoff_t bix, level_t target_level, level_t level) -{ - int ret, page_empty = 0; - int child_no = get_bits(bix, SUBLEVEL(level)); - struct page *ipage; - struct write_control child_wc = { - .flags = this_wc->flags, - }; - - ipage = logfs_get_write_page(inode, bix, level); - if (!ipage) - return -ENOMEM; - - if (this_wc->ofs) { - ret = logfs_segment_read(inode, ipage, this_wc->ofs, bix, level); - if (ret) - goto out; - } else if (!PageUptodate(ipage)) { - page_empty = 1; - logfs_read_empty(ipage); - } - - child_wc.ofs = block_get_pointer(ipage, child_no); - - if ((__force u8)level-1 > (__force u8)target_level) - ret = __logfs_write_rec(inode, page, &child_wc, bix, - target_level, SUBLEVEL(level)); - else - ret = logfs_write_i0(inode, page, &child_wc); - - if (ret) - goto out; - - alloc_indirect_block(inode, ipage, page_empty); - block_set_pointer(ipage, child_no, child_wc.ofs); - /* FIXME: first condition seems superfluous */ - if (child_wc.ofs || logfs_block(ipage)->partial) - this_wc->flags |= WF_WRITE; - /* the condition on this_wc->ofs ensures that we won't consume extra - * space for indirect blocks in the future, which we cannot reserve */ - if (!this_wc->ofs || ptr_change(this_wc->ofs, ipage)) - ret = logfs_write_i0(inode, ipage, this_wc); - else - logfs_set_alias(inode->i_sb, logfs_block(ipage), child_no); -out: - logfs_put_write_page(ipage); - return ret; -} - -static int logfs_write_rec(struct inode *inode, struct page *page, - pgoff_t bix, level_t target_level, long flags) -{ - struct logfs_inode *li = logfs_inode(inode); - struct write_control wc = { - .ofs = li->li_data[INDIRECT_INDEX], - .flags = flags, - }; - int ret; - - alloc_inode_block(inode); - - if (li->li_height > (__force u8)target_level) - ret = __logfs_write_rec(inode, page, &wc, bix, target_level, - LEVEL(li->li_height)); - else - ret = logfs_write_i0(inode, page, &wc); - if (ret) - return ret; - - if (li->li_data[INDIRECT_INDEX] != wc.ofs) { - li->li_data[INDIRECT_INDEX] = wc.ofs; - logfs_set_alias(inode->i_sb, li->li_block, - INDIRECT_INDEX + INODE_POINTER_OFS); - } - return ret; -} - -void logfs_add_transaction(struct inode *inode, struct logfs_transaction *ta) -{ - alloc_inode_block(inode); - logfs_inode(inode)->li_block->ta = ta; -} - -void logfs_del_transaction(struct inode *inode, struct logfs_transaction *ta) -{ - struct logfs_block *block = logfs_inode(inode)->li_block; - - if (block && block->ta) - block->ta = NULL; -} - -static int grow_inode(struct inode *inode, u64 bix, level_t level) -{ - struct logfs_inode *li = logfs_inode(inode); - u8 height = (__force u8)level; - struct page *page; - struct write_control wc = { - .flags = WF_WRITE, - }; - int err; - - BUG_ON(height > 5 || li->li_height > 5); - while (height > li->li_height || bix >= maxbix(li->li_height)) { - page = logfs_get_write_page(inode, I0_BLOCKS + 1, - LEVEL(li->li_height + 1)); - if (!page) - return -ENOMEM; - logfs_read_empty(page); - alloc_indirect_block(inode, page, 1); - block_set_pointer(page, 0, li->li_data[INDIRECT_INDEX]); - err = logfs_write_i0(inode, page, &wc); - logfs_put_write_page(page); - if (err) - return err; - li->li_data[INDIRECT_INDEX] = wc.ofs; - wc.ofs = 0; - li->li_height++; - logfs_set_alias(inode->i_sb, li->li_block, INODE_HEIGHT_OFS); - } - return 0; -} - -static int __logfs_write_buf(struct inode *inode, struct page *page, long flags) -{ - struct logfs_super *super = logfs_super(inode->i_sb); - pgoff_t index = page->index; - u64 bix; - level_t level; - int err; - - flags |= WF_WRITE | WF_DELETE; - inode->i_ctime = inode->i_mtime = current_time(inode); - - logfs_unpack_index(index, &bix, &level); - if (logfs_block(page) && logfs_block(page)->reserved_bytes) - super->s_dirty_pages -= logfs_block(page)->reserved_bytes; - - if (index < I0_BLOCKS) - return logfs_write_direct(inode, page, flags); - - bix = adjust_bix(bix, level); - err = grow_inode(inode, bix, level); - if (err) - return err; - return logfs_write_rec(inode, page, bix, level, flags); -} - -int logfs_write_buf(struct inode *inode, struct page *page, long flags) -{ - struct super_block *sb = inode->i_sb; - int ret; - - logfs_get_wblocks(sb, page, flags & WF_LOCK); - ret = __logfs_write_buf(inode, page, flags); - logfs_put_wblocks(sb, page, flags & WF_LOCK); - return ret; -} - -static int __logfs_delete(struct inode *inode, struct page *page) -{ - long flags = WF_DELETE; - int err; - - inode->i_ctime = inode->i_mtime = current_time(inode); - - if (page->index < I0_BLOCKS) - return logfs_write_direct(inode, page, flags); - err = grow_inode(inode, page->index, 0); - if (err) - return err; - return logfs_write_rec(inode, page, page->index, 0, flags); -} - -int logfs_delete(struct inode *inode, pgoff_t index, - struct shadow_tree *shadow_tree) -{ - struct super_block *sb = inode->i_sb; - struct page *page; - int ret; - - page = logfs_get_read_page(inode, index, 0); - if (!page) - return -ENOMEM; - - logfs_get_wblocks(sb, page, 1); - ret = __logfs_delete(inode, page); - logfs_put_wblocks(sb, page, 1); - - logfs_put_read_page(page); - - return ret; -} - -int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs, - gc_level_t gc_level, long flags) -{ - level_t level = shrink_level(gc_level); - struct page *page; - int err; - - page = logfs_get_write_page(inode, bix, level); - if (!page) - return -ENOMEM; - - err = logfs_segment_read(inode, page, ofs, bix, level); - if (!err) { - if (level != 0) - alloc_indirect_block(inode, page, 0); - err = logfs_write_buf(inode, page, flags); - if (!err && shrink_level(gc_level) == 0) { - /* Rewrite cannot mark the inode dirty but has to - * write it immediately. - * Q: Can't we just create an alias for the inode - * instead? And if not, why not? - */ - if (inode->i_ino == LOGFS_INO_MASTER) - logfs_write_anchor(inode->i_sb); - else { - err = __logfs_write_inode(inode, page, flags); - } - } - } - logfs_put_write_page(page); - return err; -} - -static int truncate_data_block(struct inode *inode, struct page *page, - u64 ofs, struct logfs_shadow *shadow, u64 size) -{ - loff_t pageofs = page->index << inode->i_sb->s_blocksize_bits; - u64 bix; - level_t level; - int err; - - /* Does truncation happen within this page? */ - if (size <= pageofs || size - pageofs >= PAGE_SIZE) - return 0; - - logfs_unpack_index(page->index, &bix, &level); - BUG_ON(level != 0); - - err = logfs_segment_read(inode, page, ofs, bix, level); - if (err) - return err; - - zero_user_segment(page, size - pageofs, PAGE_SIZE); - return logfs_segment_write(inode, page, shadow); -} - -static int logfs_truncate_i0(struct inode *inode, struct page *page, - struct write_control *wc, u64 size) -{ - struct logfs_shadow *shadow; - u64 bix; - level_t level; - int err = 0; - - logfs_unpack_index(page->index, &bix, &level); - BUG_ON(level != 0); - shadow = alloc_shadow(inode, bix, level, wc->ofs); - - err = truncate_data_block(inode, page, wc->ofs, shadow, size); - if (err) { - free_shadow(inode, shadow); - return err; - } - - logfs_segment_delete(inode, shadow); - set_iused(inode, shadow); - fill_shadow_tree(inode, page, shadow); - wc->ofs = shadow->new_ofs; - return 0; -} - -static int logfs_truncate_direct(struct inode *inode, u64 size) -{ - struct logfs_inode *li = logfs_inode(inode); - struct write_control wc; - struct page *page; - int e; - int err; - - alloc_inode_block(inode); - - for (e = I0_BLOCKS - 1; e >= 0; e--) { - if (size > (e+1) * LOGFS_BLOCKSIZE) - break; - - wc.ofs = li->li_data[e]; - if (!wc.ofs) - continue; - - page = logfs_get_write_page(inode, e, 0); - if (!page) - return -ENOMEM; - err = logfs_segment_read(inode, page, wc.ofs, e, 0); - if (err) { - logfs_put_write_page(page); - return err; - } - err = logfs_truncate_i0(inode, page, &wc, size); - logfs_put_write_page(page); - if (err) - return err; - - li->li_data[e] = wc.ofs; - } - return 0; -} - -/* FIXME: these need to become per-sb once we support different blocksizes */ -static u64 __logfs_step[] = { - 1, - I1_BLOCKS, - I2_BLOCKS, - I3_BLOCKS, -}; - -static u64 __logfs_start_index[] = { - I0_BLOCKS, - I1_BLOCKS, - I2_BLOCKS, - I3_BLOCKS -}; - -static inline u64 logfs_step(level_t level) -{ - return __logfs_step[(__force u8)level]; -} - -static inline u64 logfs_factor(u8 level) -{ - return __logfs_step[level] * LOGFS_BLOCKSIZE; -} - -static inline u64 logfs_start_index(level_t level) -{ - return __logfs_start_index[(__force u8)level]; -} - -static void logfs_unpack_raw_index(pgoff_t index, u64 *bix, level_t *level) -{ - logfs_unpack_index(index, bix, level); - if (*bix <= logfs_start_index(SUBLEVEL(*level))) - *bix = 0; -} - -static int __logfs_truncate_rec(struct inode *inode, struct page *ipage, - struct write_control *this_wc, u64 size) -{ - int truncate_happened = 0; - int e, err = 0; - u64 bix, child_bix, next_bix; - level_t level; - struct page *page; - struct write_control child_wc = { /* FIXME: flags */ }; - - logfs_unpack_raw_index(ipage->index, &bix, &level); - err = logfs_segment_read(inode, ipage, this_wc->ofs, bix, level); - if (err) - return err; - - for (e = LOGFS_BLOCK_FACTOR - 1; e >= 0; e--) { - child_bix = bix + e * logfs_step(SUBLEVEL(level)); - next_bix = child_bix + logfs_step(SUBLEVEL(level)); - if (size > next_bix * LOGFS_BLOCKSIZE) - break; - - child_wc.ofs = pure_ofs(block_get_pointer(ipage, e)); - if (!child_wc.ofs) - continue; - - page = logfs_get_write_page(inode, child_bix, SUBLEVEL(level)); - if (!page) - return -ENOMEM; - - if ((__force u8)level > 1) - err = __logfs_truncate_rec(inode, page, &child_wc, size); - else - err = logfs_truncate_i0(inode, page, &child_wc, size); - logfs_put_write_page(page); - if (err) - return err; - - truncate_happened = 1; - alloc_indirect_block(inode, ipage, 0); - block_set_pointer(ipage, e, child_wc.ofs); - } - - if (!truncate_happened) { - printk("ineffectual truncate (%lx, %lx, %llx)\n", inode->i_ino, ipage->index, size); - return 0; - } - - this_wc->flags = WF_DELETE; - if (logfs_block(ipage)->partial) - this_wc->flags |= WF_WRITE; - - return logfs_write_i0(inode, ipage, this_wc); -} - -static int logfs_truncate_rec(struct inode *inode, u64 size) -{ - struct logfs_inode *li = logfs_inode(inode); - struct write_control wc = { - .ofs = li->li_data[INDIRECT_INDEX], - }; - struct page *page; - int err; - - alloc_inode_block(inode); - - if (!wc.ofs) - return 0; - - page = logfs_get_write_page(inode, 0, LEVEL(li->li_height)); - if (!page) - return -ENOMEM; - - err = __logfs_truncate_rec(inode, page, &wc, size); - logfs_put_write_page(page); - if (err) - return err; - - if (li->li_data[INDIRECT_INDEX] != wc.ofs) - li->li_data[INDIRECT_INDEX] = wc.ofs; - return 0; -} - -static int __logfs_truncate(struct inode *inode, u64 size) -{ - int ret; - - if (size >= logfs_factor(logfs_inode(inode)->li_height)) - return 0; - - ret = logfs_truncate_rec(inode, size); - if (ret) - return ret; - - return logfs_truncate_direct(inode, size); -} - -/* - * Truncate, by changing the segment file, can consume a fair amount - * of resources. So back off from time to time and do some GC. - * 8 or 2048 blocks should be well within safety limits even if - * every single block resided in a different segment. - */ -#define TRUNCATE_STEP (8 * 1024 * 1024) -int logfs_truncate(struct inode *inode, u64 target) -{ - struct super_block *sb = inode->i_sb; - u64 size = i_size_read(inode); - int err = 0; - - size = ALIGN(size, TRUNCATE_STEP); - while (size > target) { - if (size > TRUNCATE_STEP) - size -= TRUNCATE_STEP; - else - size = 0; - if (size < target) - size = target; - - logfs_get_wblocks(sb, NULL, 1); - err = __logfs_truncate(inode, size); - if (!err) - err = __logfs_write_inode(inode, NULL, 0); - logfs_put_wblocks(sb, NULL, 1); - } - - if (!err) { - err = inode_newsize_ok(inode, target); - if (err) - goto out; - - truncate_setsize(inode, target); - } - - out: - /* I don't trust error recovery yet. */ - WARN_ON(err); - return err; -} - -static void move_page_to_inode(struct inode *inode, struct page *page) -{ - struct logfs_inode *li = logfs_inode(inode); - struct logfs_block *block = logfs_block(page); - - if (!block) - return; - - log_blockmove("move_page_to_inode(%llx, %llx, %x)\n", - block->ino, block->bix, block->level); - BUG_ON(li->li_block); - block->ops = &inode_block_ops; - block->inode = inode; - li->li_block = block; - - block->page = NULL; - if (PagePrivate(page)) { - ClearPagePrivate(page); - put_page(page); - set_page_private(page, 0); - } -} - -static void move_inode_to_page(struct page *page, struct inode *inode) -{ - struct logfs_inode *li = logfs_inode(inode); - struct logfs_block *block = li->li_block; - - if (!block) - return; - - log_blockmove("move_inode_to_page(%llx, %llx, %x)\n", - block->ino, block->bix, block->level); - BUG_ON(PagePrivate(page)); - block->ops = &indirect_block_ops; - block->page = page; - - if (!PagePrivate(page)) { - SetPagePrivate(page); - get_page(page); - set_page_private(page, (unsigned long) block); - } - - block->inode = NULL; - li->li_block = NULL; -} - -int logfs_read_inode(struct inode *inode) -{ - struct super_block *sb = inode->i_sb; - struct logfs_super *super = logfs_super(sb); - struct inode *master_inode = super->s_master_inode; - struct page *page; - struct logfs_disk_inode *di; - u64 ino = inode->i_ino; - - if (ino << sb->s_blocksize_bits > i_size_read(master_inode)) - return -ENODATA; - if (!logfs_exist_block(master_inode, ino)) - return -ENODATA; - - page = read_cache_page(master_inode->i_mapping, ino, - (filler_t *)logfs_readpage, NULL); - if (IS_ERR(page)) - return PTR_ERR(page); - - di = kmap_atomic(page); - logfs_disk_to_inode(di, inode); - kunmap_atomic(di); - move_page_to_inode(inode, page); - put_page(page); - return 0; -} - -/* Caller must logfs_put_write_page(page); */ -static struct page *inode_to_page(struct inode *inode) -{ - struct inode *master_inode = logfs_super(inode->i_sb)->s_master_inode; - struct logfs_disk_inode *di; - struct page *page; - - BUG_ON(inode->i_ino == LOGFS_INO_MASTER); - - page = logfs_get_write_page(master_inode, inode->i_ino, 0); - if (!page) - return NULL; - - di = kmap_atomic(page); - logfs_inode_to_disk(inode, di); - kunmap_atomic(di); - move_inode_to_page(page, inode); - return page; -} - -static int do_write_inode(struct inode *inode) -{ - struct super_block *sb = inode->i_sb; - struct inode *master_inode = logfs_super(sb)->s_master_inode; - loff_t size = (inode->i_ino + 1) << inode->i_sb->s_blocksize_bits; - struct page *page; - int err; - - BUG_ON(inode->i_ino == LOGFS_INO_MASTER); - /* FIXME: lock inode */ - - if (i_size_read(master_inode) < size) - i_size_write(master_inode, size); - - /* TODO: Tell vfs this inode is clean now */ - - page = inode_to_page(inode); - if (!page) - return -ENOMEM; - - /* FIXME: transaction is part of logfs_block now. Is that enough? */ - err = logfs_write_buf(master_inode, page, 0); - if (err) - move_page_to_inode(inode, page); - - logfs_put_write_page(page); - return err; -} - -static void logfs_mod_segment_entry(struct super_block *sb, u32 segno, - int write, - void (*change_se)(struct logfs_segment_entry *, long), - long arg) -{ - struct logfs_super *super = logfs_super(sb); - struct inode *inode; - struct page *page; - struct logfs_segment_entry *se; - pgoff_t page_no; - int child_no; - - page_no = segno >> (sb->s_blocksize_bits - 3); - child_no = segno & ((sb->s_blocksize >> 3) - 1); - - inode = super->s_segfile_inode; - page = logfs_get_write_page(inode, page_no, 0); - BUG_ON(!page); /* FIXME: We need some reserve page for this case */ - if (!PageUptodate(page)) - logfs_read_block(inode, page, WRITE); - - if (write) - alloc_indirect_block(inode, page, 0); - se = kmap_atomic(page); - change_se(se + child_no, arg); - if (write) { - logfs_set_alias(sb, logfs_block(page), child_no); - BUG_ON((int)be32_to_cpu(se[child_no].valid) > super->s_segsize); - } - kunmap_atomic(se); - - logfs_put_write_page(page); -} - -static void __get_segment_entry(struct logfs_segment_entry *se, long _target) -{ - struct logfs_segment_entry *target = (void *)_target; - - *target = *se; -} - -void logfs_get_segment_entry(struct super_block *sb, u32 segno, - struct logfs_segment_entry *se) -{ - logfs_mod_segment_entry(sb, segno, 0, __get_segment_entry, (long)se); -} - -static void __set_segment_used(struct logfs_segment_entry *se, long increment) -{ - u32 valid; - - valid = be32_to_cpu(se->valid); - valid += increment; - se->valid = cpu_to_be32(valid); -} - -void logfs_set_segment_used(struct super_block *sb, u64 ofs, int increment) -{ - struct logfs_super *super = logfs_super(sb); - u32 segno = ofs >> super->s_segshift; - - if (!increment) - return; - - logfs_mod_segment_entry(sb, segno, 1, __set_segment_used, increment); -} - -static void __set_segment_erased(struct logfs_segment_entry *se, long ec_level) -{ - se->ec_level = cpu_to_be32(ec_level); -} - -void logfs_set_segment_erased(struct super_block *sb, u32 segno, u32 ec, - gc_level_t gc_level) -{ - u32 ec_level = ec << 4 | (__force u8)gc_level; - - logfs_mod_segment_entry(sb, segno, 1, __set_segment_erased, ec_level); -} - -static void __set_segment_reserved(struct logfs_segment_entry *se, long ignore) -{ - se->valid = cpu_to_be32(RESERVED); -} - -void logfs_set_segment_reserved(struct super_block *sb, u32 segno) -{ - logfs_mod_segment_entry(sb, segno, 1, __set_segment_reserved, 0); -} - -static void __set_segment_unreserved(struct logfs_segment_entry *se, - long ec_level) -{ - se->valid = 0; - se->ec_level = cpu_to_be32(ec_level); -} - -void logfs_set_segment_unreserved(struct super_block *sb, u32 segno, u32 ec) -{ - u32 ec_level = ec << 4; - - logfs_mod_segment_entry(sb, segno, 1, __set_segment_unreserved, - ec_level); -} - -int __logfs_write_inode(struct inode *inode, struct page *page, long flags) -{ - struct super_block *sb = inode->i_sb; - int ret; - - logfs_get_wblocks(sb, page, flags & WF_LOCK); - ret = do_write_inode(inode); - logfs_put_wblocks(sb, page, flags & WF_LOCK); - return ret; -} - -static int do_delete_inode(struct inode *inode) -{ - struct super_block *sb = inode->i_sb; - struct inode *master_inode = logfs_super(sb)->s_master_inode; - struct page *page; - int ret; - - page = logfs_get_write_page(master_inode, inode->i_ino, 0); - if (!page) - return -ENOMEM; - - move_inode_to_page(page, inode); - - logfs_get_wblocks(sb, page, 1); - ret = __logfs_delete(master_inode, page); - logfs_put_wblocks(sb, page, 1); - - logfs_put_write_page(page); - return ret; -} - -/* - * ZOMBIE inodes have already been deleted before and should remain dead, - * if it weren't for valid checking. No need to kill them again here. - */ -void logfs_evict_inode(struct inode *inode) -{ - struct super_block *sb = inode->i_sb; - struct logfs_inode *li = logfs_inode(inode); - struct logfs_block *block = li->li_block; - struct page *page; - - if (!inode->i_nlink) { - if (!(li->li_flags & LOGFS_IF_ZOMBIE)) { - li->li_flags |= LOGFS_IF_ZOMBIE; - if (i_size_read(inode) > 0) - logfs_truncate(inode, 0); - do_delete_inode(inode); - } - } - truncate_inode_pages_final(&inode->i_data); - clear_inode(inode); - - /* Cheaper version of write_inode. All changes are concealed in - * aliases, which are moved back. No write to the medium happens. - */ - /* Only deleted files may be dirty at this point */ - BUG_ON(inode->i_state & I_DIRTY && inode->i_nlink); - if (!block) - return; - if ((logfs_super(sb)->s_flags & LOGFS_SB_FLAG_SHUTDOWN)) { - block->ops->free_block(inode->i_sb, block); - return; - } - - page = inode_to_page(inode); - BUG_ON(!page); /* FIXME: Use emergency page */ - logfs_put_write_page(page); -} - -void btree_write_block(struct logfs_block *block) -{ - struct inode *inode; - struct page *page; - int err, cookie; - - inode = logfs_safe_iget(block->sb, block->ino, &cookie); - page = logfs_get_write_page(inode, block->bix, block->level); - - err = logfs_readpage_nolock(page); - BUG_ON(err); - BUG_ON(!PagePrivate(page)); - BUG_ON(logfs_block(page) != block); - err = __logfs_write_buf(inode, page, 0); - BUG_ON(err); - BUG_ON(PagePrivate(page) || page->private); - - logfs_put_write_page(page); - logfs_safe_iput(inode, cookie); -} - -/** - * logfs_inode_write - write inode or dentry objects - * - * @inode: parent inode (ifile or directory) - * @buf: object to write (inode or dentry) - * @count: object size - * @bix: block index - * @flags: write flags - * @shadow_tree: shadow below this inode - * - * FIXME: All caller of this put a 200-300 byte variable on the stack, - * only to call here and do a memcpy from that stack variable. A good - * example of wasted performance and stack space. - */ -int logfs_inode_write(struct inode *inode, const void *buf, size_t count, - loff_t bix, long flags, struct shadow_tree *shadow_tree) -{ - loff_t pos = bix << inode->i_sb->s_blocksize_bits; - int err; - struct page *page; - void *pagebuf; - - BUG_ON(pos & (LOGFS_BLOCKSIZE-1)); - BUG_ON(count > LOGFS_BLOCKSIZE); - page = logfs_get_write_page(inode, bix, 0); - if (!page) - return -ENOMEM; - - pagebuf = kmap_atomic(page); - memcpy(pagebuf, buf, count); - flush_dcache_page(page); - kunmap_atomic(pagebuf); - - if (i_size_read(inode) < pos + LOGFS_BLOCKSIZE) - i_size_write(inode, pos + LOGFS_BLOCKSIZE); - - err = logfs_write_buf(inode, page, flags); - logfs_put_write_page(page); - return err; -} - -int logfs_open_segfile(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct inode *inode; - - inode = logfs_read_meta_inode(sb, LOGFS_INO_SEGFILE); - if (IS_ERR(inode)) - return PTR_ERR(inode); - super->s_segfile_inode = inode; - return 0; -} - -int logfs_init_rw(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int min_fill = 3 * super->s_no_blocks; - - INIT_LIST_HEAD(&super->s_object_alias); - INIT_LIST_HEAD(&super->s_writeback_list); - mutex_init(&super->s_write_mutex); - super->s_block_pool = mempool_create_kmalloc_pool(min_fill, - sizeof(struct logfs_block)); - super->s_shadow_pool = mempool_create_kmalloc_pool(min_fill, - sizeof(struct logfs_shadow)); - return 0; -} - -void logfs_cleanup_rw(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - - logfs_mempool_destroy(super->s_block_pool); - logfs_mempool_destroy(super->s_shadow_pool); -} diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c deleted file mode 100644 index 1efd6055f4b0..000000000000 --- a/fs/logfs/segment.c +++ /dev/null @@ -1,961 +0,0 @@ -/* - * fs/logfs/segment.c - Handling the Object Store - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - * - * Object store or ostore makes up the complete device with exception of - * the superblock and journal areas. Apart from its own metadata it stores - * three kinds of objects: inodes, dentries and blocks, both data and indirect. - */ -#include "logfs.h" -#include <linux/slab.h> - -static int logfs_mark_segment_bad(struct super_block *sb, u32 segno) -{ - struct logfs_super *super = logfs_super(sb); - struct btree_head32 *head = &super->s_reserved_segments; - int err; - - err = btree_insert32(head, segno, (void *)1, GFP_NOFS); - if (err) - return err; - logfs_super(sb)->s_bad_segments++; - /* FIXME: write to journal */ - return 0; -} - -int logfs_erase_segment(struct super_block *sb, u32 segno, int ensure_erase) -{ - struct logfs_super *super = logfs_super(sb); - - super->s_gec++; - - return super->s_devops->erase(sb, (u64)segno << super->s_segshift, - super->s_segsize, ensure_erase); -} - -static s64 logfs_get_free_bytes(struct logfs_area *area, size_t bytes) -{ - s32 ofs; - - logfs_open_area(area, bytes); - - ofs = area->a_used_bytes; - area->a_used_bytes += bytes; - BUG_ON(area->a_used_bytes >= logfs_super(area->a_sb)->s_segsize); - - return dev_ofs(area->a_sb, area->a_segno, ofs); -} - -static struct page *get_mapping_page(struct super_block *sb, pgoff_t index, - int use_filler) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - filler_t *filler = super->s_devops->readpage; - struct page *page; - - BUG_ON(mapping_gfp_constraint(mapping, __GFP_FS)); - if (use_filler) - page = read_cache_page(mapping, index, filler, sb); - else { - page = find_or_create_page(mapping, index, GFP_NOFS); - if (page) - unlock_page(page); - } - return page; -} - -int __logfs_buf_write(struct logfs_area *area, u64 ofs, void *buf, size_t len, - int use_filler) -{ - pgoff_t index = ofs >> PAGE_SHIFT; - struct page *page; - long offset = ofs & (PAGE_SIZE-1); - long copylen; - - /* Only logfs_wbuf_recover may use len==0 */ - BUG_ON(!len && !use_filler); - do { - copylen = min((ulong)len, PAGE_SIZE - offset); - - page = get_mapping_page(area->a_sb, index, use_filler); - if (IS_ERR(page)) - return PTR_ERR(page); - BUG_ON(!page); /* FIXME: reserve a pool */ - SetPageUptodate(page); - memcpy(page_address(page) + offset, buf, copylen); - - if (!PagePrivate(page)) { - SetPagePrivate(page); - get_page(page); - } - put_page(page); - - buf += copylen; - len -= copylen; - offset = 0; - index++; - } while (len); - return 0; -} - -static void pad_partial_page(struct logfs_area *area) -{ - struct super_block *sb = area->a_sb; - struct page *page; - u64 ofs = dev_ofs(sb, area->a_segno, area->a_used_bytes); - pgoff_t index = ofs >> PAGE_SHIFT; - long offset = ofs & (PAGE_SIZE-1); - u32 len = PAGE_SIZE - offset; - - if (len % PAGE_SIZE) { - page = get_mapping_page(sb, index, 0); - BUG_ON(!page); /* FIXME: reserve a pool */ - memset(page_address(page) + offset, 0xff, len); - if (!PagePrivate(page)) { - SetPagePrivate(page); - get_page(page); - } - put_page(page); - } -} - -static void pad_full_pages(struct logfs_area *area) -{ - struct super_block *sb = area->a_sb; - struct logfs_super *super = logfs_super(sb); - u64 ofs = dev_ofs(sb, area->a_segno, area->a_used_bytes); - u32 len = super->s_segsize - area->a_used_bytes; - pgoff_t index = PAGE_ALIGN(ofs) >> PAGE_SHIFT; - pgoff_t no_indizes = len >> PAGE_SHIFT; - struct page *page; - - while (no_indizes) { - page = get_mapping_page(sb, index, 0); - BUG_ON(!page); /* FIXME: reserve a pool */ - SetPageUptodate(page); - memset(page_address(page), 0xff, PAGE_SIZE); - if (!PagePrivate(page)) { - SetPagePrivate(page); - get_page(page); - } - put_page(page); - index++; - no_indizes--; - } -} - -/* - * bdev_writeseg will write full pages. Memset the tail to prevent data leaks. - * Also make sure we allocate (and memset) all pages for final writeout. - */ -static void pad_wbuf(struct logfs_area *area, int final) -{ - pad_partial_page(area); - if (final) - pad_full_pages(area); -} - -/* - * We have to be careful with the alias tree. Since lookup is done by bix, - * it needs to be normalized, so 14, 15, 16, etc. all match when dealing with - * indirect blocks. So always use it through accessor functions. - */ -static void *alias_tree_lookup(struct super_block *sb, u64 ino, u64 bix, - level_t level) -{ - struct btree_head128 *head = &logfs_super(sb)->s_object_alias_tree; - pgoff_t index = logfs_pack_index(bix, level); - - return btree_lookup128(head, ino, index); -} - -static int alias_tree_insert(struct super_block *sb, u64 ino, u64 bix, - level_t level, void *val) -{ - struct btree_head128 *head = &logfs_super(sb)->s_object_alias_tree; - pgoff_t index = logfs_pack_index(bix, level); - - return btree_insert128(head, ino, index, val, GFP_NOFS); -} - -static int btree_write_alias(struct super_block *sb, struct logfs_block *block, - write_alias_t *write_one_alias) -{ - struct object_alias_item *item; - int err; - - list_for_each_entry(item, &block->item_list, list) { - err = write_alias_journal(sb, block->ino, block->bix, - block->level, item->child_no, item->val); - if (err) - return err; - } - return 0; -} - -static const struct logfs_block_ops btree_block_ops = { - .write_block = btree_write_block, - .free_block = __free_block, - .write_alias = btree_write_alias, -}; - -int logfs_load_object_aliases(struct super_block *sb, - struct logfs_obj_alias *oa, int count) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_block *block; - struct object_alias_item *item; - u64 ino, bix; - level_t level; - int i, err; - - super->s_flags |= LOGFS_SB_FLAG_OBJ_ALIAS; - count /= sizeof(*oa); - for (i = 0; i < count; i++) { - item = mempool_alloc(super->s_alias_pool, GFP_NOFS); - if (!item) - return -ENOMEM; - memset(item, 0, sizeof(*item)); - - super->s_no_object_aliases++; - item->val = oa[i].val; - item->child_no = be16_to_cpu(oa[i].child_no); - - ino = be64_to_cpu(oa[i].ino); - bix = be64_to_cpu(oa[i].bix); - level = LEVEL(oa[i].level); - - log_aliases("logfs_load_object_aliases(%llx, %llx, %x, %x) %llx\n", - ino, bix, level, item->child_no, - be64_to_cpu(item->val)); - block = alias_tree_lookup(sb, ino, bix, level); - if (!block) { - block = __alloc_block(sb, ino, bix, level); - block->ops = &btree_block_ops; - err = alias_tree_insert(sb, ino, bix, level, block); - BUG_ON(err); /* mempool empty */ - } - if (test_and_set_bit(item->child_no, block->alias_map)) { - printk(KERN_ERR"LogFS: Alias collision detected\n"); - return -EIO; - } - list_move_tail(&block->alias_list, &super->s_object_alias); - list_add(&item->list, &block->item_list); - } - return 0; -} - -static void kill_alias(void *_block, unsigned long ignore0, - u64 ignore1, u64 ignore2, size_t ignore3) -{ - struct logfs_block *block = _block; - struct super_block *sb = block->sb; - struct logfs_super *super = logfs_super(sb); - struct object_alias_item *item; - - while (!list_empty(&block->item_list)) { - item = list_entry(block->item_list.next, typeof(*item), list); - list_del(&item->list); - mempool_free(item, super->s_alias_pool); - } - block->ops->free_block(sb, block); -} - -static int obj_type(struct inode *inode, level_t level) -{ - if (level == 0) { - if (S_ISDIR(inode->i_mode)) - return OBJ_DENTRY; - if (inode->i_ino == LOGFS_INO_MASTER) - return OBJ_INODE; - } - return OBJ_BLOCK; -} - -static int obj_len(struct super_block *sb, int obj_type) -{ - switch (obj_type) { - case OBJ_DENTRY: - return sizeof(struct logfs_disk_dentry); - case OBJ_INODE: - return sizeof(struct logfs_disk_inode); - case OBJ_BLOCK: - return sb->s_blocksize; - default: - BUG(); - } -} - -static int __logfs_segment_write(struct inode *inode, void *buf, - struct logfs_shadow *shadow, int type, int len, int compr) -{ - struct logfs_area *area; - struct super_block *sb = inode->i_sb; - s64 ofs; - struct logfs_object_header h; - int acc_len; - - if (shadow->gc_level == 0) - acc_len = len; - else - acc_len = obj_len(sb, type); - - area = get_area(sb, shadow->gc_level); - ofs = logfs_get_free_bytes(area, len + LOGFS_OBJECT_HEADERSIZE); - LOGFS_BUG_ON(ofs <= 0, sb); - /* - * Order is important. logfs_get_free_bytes(), by modifying the - * segment file, may modify the content of the very page we're about - * to write now. Which is fine, as long as the calculated crc and - * written data still match. So do the modifications _before_ - * calculating the crc. - */ - - h.len = cpu_to_be16(len); - h.type = type; - h.compr = compr; - h.ino = cpu_to_be64(inode->i_ino); - h.bix = cpu_to_be64(shadow->bix); - h.crc = logfs_crc32(&h, sizeof(h) - 4, 4); - h.data_crc = logfs_crc32(buf, len, 0); - - logfs_buf_write(area, ofs, &h, sizeof(h)); - logfs_buf_write(area, ofs + LOGFS_OBJECT_HEADERSIZE, buf, len); - - shadow->new_ofs = ofs; - shadow->new_len = acc_len + LOGFS_OBJECT_HEADERSIZE; - - return 0; -} - -static s64 logfs_segment_write_compress(struct inode *inode, void *buf, - struct logfs_shadow *shadow, int type, int len) -{ - struct super_block *sb = inode->i_sb; - void *compressor_buf = logfs_super(sb)->s_compressed_je; - ssize_t compr_len; - int ret; - - mutex_lock(&logfs_super(sb)->s_journal_mutex); - compr_len = logfs_compress(buf, compressor_buf, len, len); - - if (compr_len >= 0) { - ret = __logfs_segment_write(inode, compressor_buf, shadow, - type, compr_len, COMPR_ZLIB); - } else { - ret = __logfs_segment_write(inode, buf, shadow, type, len, - COMPR_NONE); - } - mutex_unlock(&logfs_super(sb)->s_journal_mutex); - return ret; -} - -/** - * logfs_segment_write - write data block to object store - * @inode: inode containing data - * - * Returns an errno or zero. - */ -int logfs_segment_write(struct inode *inode, struct page *page, - struct logfs_shadow *shadow) -{ - struct super_block *sb = inode->i_sb; - struct logfs_super *super = logfs_super(sb); - int do_compress, type, len; - int ret; - void *buf; - - super->s_flags |= LOGFS_SB_FLAG_DIRTY; - BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN); - do_compress = logfs_inode(inode)->li_flags & LOGFS_IF_COMPRESSED; - if (shadow->gc_level != 0) { - /* temporarily disable compression for indirect blocks */ - do_compress = 0; - } - - type = obj_type(inode, shrink_level(shadow->gc_level)); - len = obj_len(sb, type); - buf = kmap(page); - if (do_compress) - ret = logfs_segment_write_compress(inode, buf, shadow, type, - len); - else - ret = __logfs_segment_write(inode, buf, shadow, type, len, - COMPR_NONE); - kunmap(page); - - log_segment("logfs_segment_write(%llx, %llx, %x) %llx->%llx %x->%x\n", - shadow->ino, shadow->bix, shadow->gc_level, - shadow->old_ofs, shadow->new_ofs, - shadow->old_len, shadow->new_len); - /* this BUG_ON did catch a locking bug. useful */ - BUG_ON(!(shadow->new_ofs & (super->s_segsize - 1))); - return ret; -} - -int wbuf_read(struct super_block *sb, u64 ofs, size_t len, void *buf) -{ - pgoff_t index = ofs >> PAGE_SHIFT; - struct page *page; - long offset = ofs & (PAGE_SIZE-1); - long copylen; - - while (len) { - copylen = min((ulong)len, PAGE_SIZE - offset); - - page = get_mapping_page(sb, index, 1); - if (IS_ERR(page)) - return PTR_ERR(page); - memcpy(buf, page_address(page) + offset, copylen); - put_page(page); - - buf += copylen; - len -= copylen; - offset = 0; - index++; - } - return 0; -} - -/* - * The "position" of indirect blocks is ambiguous. It can be the position - * of any data block somewhere behind this indirect block. So we need to - * normalize the positions through logfs_block_mask() before comparing. - */ -static int check_pos(struct super_block *sb, u64 pos1, u64 pos2, level_t level) -{ - return (pos1 & logfs_block_mask(sb, level)) != - (pos2 & logfs_block_mask(sb, level)); -} - -#if 0 -static int read_seg_header(struct super_block *sb, u64 ofs, - struct logfs_segment_header *sh) -{ - __be32 crc; - int err; - - err = wbuf_read(sb, ofs, sizeof(*sh), sh); - if (err) - return err; - crc = logfs_crc32(sh, sizeof(*sh), 4); - if (crc != sh->crc) { - printk(KERN_ERR"LOGFS: header crc error at %llx: expected %x, " - "got %x\n", ofs, be32_to_cpu(sh->crc), - be32_to_cpu(crc)); - return -EIO; - } - return 0; -} -#endif - -static int read_obj_header(struct super_block *sb, u64 ofs, - struct logfs_object_header *oh) -{ - __be32 crc; - int err; - - err = wbuf_read(sb, ofs, sizeof(*oh), oh); - if (err) - return err; - crc = logfs_crc32(oh, sizeof(*oh) - 4, 4); - if (crc != oh->crc) { - printk(KERN_ERR"LOGFS: header crc error at %llx: expected %x, " - "got %x\n", ofs, be32_to_cpu(oh->crc), - be32_to_cpu(crc)); - return -EIO; - } - return 0; -} - -static void move_btree_to_page(struct inode *inode, struct page *page, - __be64 *data) -{ - struct super_block *sb = inode->i_sb; - struct logfs_super *super = logfs_super(sb); - struct btree_head128 *head = &super->s_object_alias_tree; - struct logfs_block *block; - struct object_alias_item *item, *next; - - if (!(super->s_flags & LOGFS_SB_FLAG_OBJ_ALIAS)) - return; - - block = btree_remove128(head, inode->i_ino, page->index); - if (!block) - return; - - log_blockmove("move_btree_to_page(%llx, %llx, %x)\n", - block->ino, block->bix, block->level); - list_for_each_entry_safe(item, next, &block->item_list, list) { - data[item->child_no] = item->val; - list_del(&item->list); - mempool_free(item, super->s_alias_pool); - } - block->page = page; - - if (!PagePrivate(page)) { - SetPagePrivate(page); - get_page(page); - set_page_private(page, (unsigned long) block); - } - block->ops = &indirect_block_ops; - initialize_block_counters(page, block, data, 0); -} - -/* - * This silences a false, yet annoying gcc warning. I hate it when my editor - * jumps into bitops.h each time I recompile this file. - * TODO: Complain to gcc folks about this and upgrade compiler. - */ -static unsigned long fnb(const unsigned long *addr, - unsigned long size, unsigned long offset) -{ - return find_next_bit(addr, size, offset); -} - -void move_page_to_btree(struct page *page) -{ - struct logfs_block *block = logfs_block(page); - struct super_block *sb = block->sb; - struct logfs_super *super = logfs_super(sb); - struct object_alias_item *item; - unsigned long pos; - __be64 *child; - int err; - - if (super->s_flags & LOGFS_SB_FLAG_SHUTDOWN) { - block->ops->free_block(sb, block); - return; - } - log_blockmove("move_page_to_btree(%llx, %llx, %x)\n", - block->ino, block->bix, block->level); - super->s_flags |= LOGFS_SB_FLAG_OBJ_ALIAS; - - for (pos = 0; ; pos++) { - pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos); - if (pos >= LOGFS_BLOCK_FACTOR) - break; - - item = mempool_alloc(super->s_alias_pool, GFP_NOFS); - BUG_ON(!item); /* mempool empty */ - memset(item, 0, sizeof(*item)); - - child = kmap_atomic(page); - item->val = child[pos]; - kunmap_atomic(child); - item->child_no = pos; - list_add(&item->list, &block->item_list); - } - block->page = NULL; - - if (PagePrivate(page)) { - ClearPagePrivate(page); - put_page(page); - set_page_private(page, 0); - } - block->ops = &btree_block_ops; - err = alias_tree_insert(block->sb, block->ino, block->bix, block->level, - block); - BUG_ON(err); /* mempool empty */ - ClearPageUptodate(page); -} - -static int __logfs_segment_read(struct inode *inode, void *buf, - u64 ofs, u64 bix, level_t level) -{ - struct super_block *sb = inode->i_sb; - void *compressor_buf = logfs_super(sb)->s_compressed_je; - struct logfs_object_header oh; - __be32 crc; - u16 len; - int err, block_len; - - block_len = obj_len(sb, obj_type(inode, level)); - err = read_obj_header(sb, ofs, &oh); - if (err) - goto out_err; - - err = -EIO; - if (be64_to_cpu(oh.ino) != inode->i_ino - || check_pos(sb, be64_to_cpu(oh.bix), bix, level)) { - printk(KERN_ERR"LOGFS: (ino, bix) don't match at %llx: " - "expected (%lx, %llx), got (%llx, %llx)\n", - ofs, inode->i_ino, bix, - be64_to_cpu(oh.ino), be64_to_cpu(oh.bix)); - goto out_err; - } - - len = be16_to_cpu(oh.len); - - switch (oh.compr) { - case COMPR_NONE: - err = wbuf_read(sb, ofs + LOGFS_OBJECT_HEADERSIZE, len, buf); - if (err) - goto out_err; - crc = logfs_crc32(buf, len, 0); - if (crc != oh.data_crc) { - printk(KERN_ERR"LOGFS: uncompressed data crc error at " - "%llx: expected %x, got %x\n", ofs, - be32_to_cpu(oh.data_crc), - be32_to_cpu(crc)); - goto out_err; - } - break; - case COMPR_ZLIB: - mutex_lock(&logfs_super(sb)->s_journal_mutex); - err = wbuf_read(sb, ofs + LOGFS_OBJECT_HEADERSIZE, len, - compressor_buf); - if (err) { - mutex_unlock(&logfs_super(sb)->s_journal_mutex); - goto out_err; - } - crc = logfs_crc32(compressor_buf, len, 0); - if (crc != oh.data_crc) { - printk(KERN_ERR"LOGFS: compressed data crc error at " - "%llx: expected %x, got %x\n", ofs, - be32_to_cpu(oh.data_crc), - be32_to_cpu(crc)); - mutex_unlock(&logfs_super(sb)->s_journal_mutex); - goto out_err; - } - err = logfs_uncompress(compressor_buf, buf, len, block_len); - mutex_unlock(&logfs_super(sb)->s_journal_mutex); - if (err) { - printk(KERN_ERR"LOGFS: uncompress error at %llx\n", ofs); - goto out_err; - } - break; - default: - LOGFS_BUG(sb); - err = -EIO; - goto out_err; - } - return 0; - -out_err: - logfs_set_ro(sb); - printk(KERN_ERR"LOGFS: device is read-only now\n"); - LOGFS_BUG(sb); - return err; -} - -/** - * logfs_segment_read - read data block from object store - * @inode: inode containing data - * @buf: data buffer - * @ofs: physical data offset - * @bix: block index - * @level: block level - * - * Returns 0 on success or a negative errno. - */ -int logfs_segment_read(struct inode *inode, struct page *page, - u64 ofs, u64 bix, level_t level) -{ - int err; - void *buf; - - if (PageUptodate(page)) - return 0; - - ofs &= ~LOGFS_FULLY_POPULATED; - - buf = kmap(page); - err = __logfs_segment_read(inode, buf, ofs, bix, level); - if (!err) { - move_btree_to_page(inode, page, buf); - SetPageUptodate(page); - } - kunmap(page); - log_segment("logfs_segment_read(%lx, %llx, %x) %llx (%d)\n", - inode->i_ino, bix, level, ofs, err); - return err; -} - -int logfs_segment_delete(struct inode *inode, struct logfs_shadow *shadow) -{ - struct super_block *sb = inode->i_sb; - struct logfs_super *super = logfs_super(sb); - struct logfs_object_header h; - u16 len; - int err; - - super->s_flags |= LOGFS_SB_FLAG_DIRTY; - BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN); - BUG_ON(shadow->old_ofs & LOGFS_FULLY_POPULATED); - if (!shadow->old_ofs) - return 0; - - log_segment("logfs_segment_delete(%llx, %llx, %x) %llx->%llx %x->%x\n", - shadow->ino, shadow->bix, shadow->gc_level, - shadow->old_ofs, shadow->new_ofs, - shadow->old_len, shadow->new_len); - err = read_obj_header(sb, shadow->old_ofs, &h); - LOGFS_BUG_ON(err, sb); - LOGFS_BUG_ON(be64_to_cpu(h.ino) != inode->i_ino, sb); - LOGFS_BUG_ON(check_pos(sb, shadow->bix, be64_to_cpu(h.bix), - shrink_level(shadow->gc_level)), sb); - - if (shadow->gc_level == 0) - len = be16_to_cpu(h.len); - else - len = obj_len(sb, h.type); - shadow->old_len = len + sizeof(h); - return 0; -} - -void freeseg(struct super_block *sb, u32 segno) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping = super->s_mapping_inode->i_mapping; - struct page *page; - u64 ofs, start, end; - - start = dev_ofs(sb, segno, 0); - end = dev_ofs(sb, segno + 1, 0); - for (ofs = start; ofs < end; ofs += PAGE_SIZE) { - page = find_get_page(mapping, ofs >> PAGE_SHIFT); - if (!page) - continue; - if (PagePrivate(page)) { - ClearPagePrivate(page); - put_page(page); - } - put_page(page); - } -} - -int logfs_open_area(struct logfs_area *area, size_t bytes) -{ - struct super_block *sb = area->a_sb; - struct logfs_super *super = logfs_super(sb); - int err, closed = 0; - - if (area->a_is_open && area->a_used_bytes + bytes <= super->s_segsize) - return 0; - - if (area->a_is_open) { - u64 ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes); - u32 len = super->s_segsize - area->a_written_bytes; - - log_gc("logfs_close_area(%x)\n", area->a_segno); - pad_wbuf(area, 1); - super->s_devops->writeseg(area->a_sb, ofs, len); - freeseg(sb, area->a_segno); - closed = 1; - } - - area->a_used_bytes = 0; - area->a_written_bytes = 0; -again: - area->a_ops->get_free_segment(area); - area->a_ops->get_erase_count(area); - - log_gc("logfs_open_area(%x, %x)\n", area->a_segno, area->a_level); - err = area->a_ops->erase_segment(area); - if (err) { - printk(KERN_WARNING "LogFS: Error erasing segment %x\n", - area->a_segno); - logfs_mark_segment_bad(sb, area->a_segno); - goto again; - } - area->a_is_open = 1; - return closed; -} - -void logfs_sync_area(struct logfs_area *area) -{ - struct super_block *sb = area->a_sb; - struct logfs_super *super = logfs_super(sb); - u64 ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes); - u32 len = (area->a_used_bytes - area->a_written_bytes); - - if (super->s_writesize) - len &= ~(super->s_writesize - 1); - if (len == 0) - return; - pad_wbuf(area, 0); - super->s_devops->writeseg(sb, ofs, len); - area->a_written_bytes += len; -} - -void logfs_sync_segments(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int i; - - for_each_area(i) - logfs_sync_area(super->s_area[i]); -} - -/* - * Pick a free segment to be used for this area. Effectively takes a - * candidate from the free list (not really a candidate anymore). - */ -static void ostore_get_free_segment(struct logfs_area *area) -{ - struct super_block *sb = area->a_sb; - struct logfs_super *super = logfs_super(sb); - - if (super->s_free_list.count == 0) { - printk(KERN_ERR"LOGFS: ran out of free segments\n"); - LOGFS_BUG(sb); - } - - area->a_segno = get_best_cand(sb, &super->s_free_list, NULL); -} - -static void ostore_get_erase_count(struct logfs_area *area) -{ - struct logfs_segment_entry se; - u32 ec_level; - - logfs_get_segment_entry(area->a_sb, area->a_segno, &se); - BUG_ON(se.ec_level == cpu_to_be32(BADSEG) || - se.valid == cpu_to_be32(RESERVED)); - - ec_level = be32_to_cpu(se.ec_level); - area->a_erase_count = (ec_level >> 4) + 1; -} - -static int ostore_erase_segment(struct logfs_area *area) -{ - struct super_block *sb = area->a_sb; - struct logfs_segment_header sh; - u64 ofs; - int err; - - err = logfs_erase_segment(sb, area->a_segno, 0); - if (err) - return err; - - sh.pad = 0; - sh.type = SEG_OSTORE; - sh.level = (__force u8)area->a_level; - sh.segno = cpu_to_be32(area->a_segno); - sh.ec = cpu_to_be32(area->a_erase_count); - sh.gec = cpu_to_be64(logfs_super(sb)->s_gec); - sh.crc = logfs_crc32(&sh, sizeof(sh), 4); - - logfs_set_segment_erased(sb, area->a_segno, area->a_erase_count, - area->a_level); - - ofs = dev_ofs(sb, area->a_segno, 0); - area->a_used_bytes = sizeof(sh); - logfs_buf_write(area, ofs, &sh, sizeof(sh)); - return 0; -} - -static const struct logfs_area_ops ostore_area_ops = { - .get_free_segment = ostore_get_free_segment, - .get_erase_count = ostore_get_erase_count, - .erase_segment = ostore_erase_segment, -}; - -static void free_area(struct logfs_area *area) -{ - if (area) - freeseg(area->a_sb, area->a_segno); - kfree(area); -} - -void free_areas(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int i; - - for_each_area(i) - free_area(super->s_area[i]); - free_area(super->s_journal_area); -} - -static struct logfs_area *alloc_area(struct super_block *sb) -{ - struct logfs_area *area; - - area = kzalloc(sizeof(*area), GFP_KERNEL); - if (!area) - return NULL; - - area->a_sb = sb; - return area; -} - -static void map_invalidatepage(struct page *page, unsigned int o, - unsigned int l) -{ - return; -} - -static int map_releasepage(struct page *page, gfp_t g) -{ - /* Don't release these pages */ - return 0; -} - -static const struct address_space_operations mapping_aops = { - .invalidatepage = map_invalidatepage, - .releasepage = map_releasepage, - .set_page_dirty = __set_page_dirty_nobuffers, -}; - -int logfs_init_mapping(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct address_space *mapping; - struct inode *inode; - - inode = logfs_new_meta_inode(sb, LOGFS_INO_MAPPING); - if (IS_ERR(inode)) - return PTR_ERR(inode); - super->s_mapping_inode = inode; - mapping = inode->i_mapping; - mapping->a_ops = &mapping_aops; - /* Would it be possible to use __GFP_HIGHMEM as well? */ - mapping_set_gfp_mask(mapping, GFP_NOFS); - return 0; -} - -int logfs_init_areas(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int i = -1; - - super->s_alias_pool = mempool_create_kmalloc_pool(600, - sizeof(struct object_alias_item)); - if (!super->s_alias_pool) - return -ENOMEM; - - super->s_journal_area = alloc_area(sb); - if (!super->s_journal_area) - goto err; - - for_each_area(i) { - super->s_area[i] = alloc_area(sb); - if (!super->s_area[i]) - goto err; - super->s_area[i]->a_level = GC_LEVEL(i); - super->s_area[i]->a_ops = &ostore_area_ops; - } - btree_init_mempool128(&super->s_object_alias_tree, - super->s_btree_pool); - return 0; - -err: - for (i--; i >= 0; i--) - free_area(super->s_area[i]); - free_area(super->s_journal_area); - logfs_mempool_destroy(super->s_alias_pool); - return -ENOMEM; -} - -void logfs_cleanup_areas(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - - btree_grim_visitor128(&super->s_object_alias_tree, 0, kill_alias); -} diff --git a/fs/logfs/super.c b/fs/logfs/super.c deleted file mode 100644 index 5751082dba52..000000000000 --- a/fs/logfs/super.c +++ /dev/null @@ -1,653 +0,0 @@ -/* - * fs/logfs/super.c - * - * As should be obvious for Linux kernel code, license is GPLv2 - * - * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> - * - * Generally contains mount/umount code and also serves as a dump area for - * any functions that don't fit elsewhere and neither justify a file of their - * own. - */ -#include "logfs.h" -#include <linux/bio.h> -#include <linux/slab.h> -#include <linux/blkdev.h> -#include <linux/module.h> -#include <linux/mtd/mtd.h> -#include <linux/statfs.h> -#include <linux/buffer_head.h> - -static DEFINE_MUTEX(emergency_mutex); -static struct page *emergency_page; - -struct page *emergency_read_begin(struct address_space *mapping, pgoff_t index) -{ - filler_t *filler = (filler_t *)mapping->a_ops->readpage; - struct page *page; - int err; - - page = read_cache_page(mapping, index, filler, NULL); - if (page) - return page; - - /* No more pages available, switch to emergency page */ - printk(KERN_INFO"Logfs: Using emergency page\n"); - mutex_lock(&emergency_mutex); - err = filler(NULL, emergency_page); - if (err) { - mutex_unlock(&emergency_mutex); - printk(KERN_EMERG"Logfs: Error reading emergency page\n"); - return ERR_PTR(err); - } - return emergency_page; -} - -void emergency_read_end(struct page *page) -{ - if (page == emergency_page) - mutex_unlock(&emergency_mutex); - else - put_page(page); -} - -static void dump_segfile(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_segment_entry se; - u32 segno; - - for (segno = 0; segno < super->s_no_segs; segno++) { - logfs_get_segment_entry(sb, segno, &se); - printk("%3x: %6x %8x", segno, be32_to_cpu(se.ec_level), - be32_to_cpu(se.valid)); - if (++segno < super->s_no_segs) { - logfs_get_segment_entry(sb, segno, &se); - printk(" %6x %8x", be32_to_cpu(se.ec_level), - be32_to_cpu(se.valid)); - } - if (++segno < super->s_no_segs) { - logfs_get_segment_entry(sb, segno, &se); - printk(" %6x %8x", be32_to_cpu(se.ec_level), - be32_to_cpu(se.valid)); - } - if (++segno < super->s_no_segs) { - logfs_get_segment_entry(sb, segno, &se); - printk(" %6x %8x", be32_to_cpu(se.ec_level), - be32_to_cpu(se.valid)); - } - printk("\n"); - } -} - -/* - * logfs_crash_dump - dump debug information to device - * - * The LogFS superblock only occupies part of a segment. This function will - * write as much debug information as it can gather into the spare space. - */ -void logfs_crash_dump(struct super_block *sb) -{ - dump_segfile(sb); -} - -/* - * FIXME: There should be a reserve for root, similar to ext2. - */ -int logfs_statfs(struct dentry *dentry, struct kstatfs *stats) -{ - struct super_block *sb = dentry->d_sb; - struct logfs_super *super = logfs_super(sb); - - stats->f_type = LOGFS_MAGIC_U32; - stats->f_bsize = sb->s_blocksize; - stats->f_blocks = super->s_size >> LOGFS_BLOCK_BITS >> 3; - stats->f_bfree = super->s_free_bytes >> sb->s_blocksize_bits; - stats->f_bavail = super->s_free_bytes >> sb->s_blocksize_bits; - stats->f_files = 0; - stats->f_ffree = 0; - stats->f_namelen = LOGFS_MAX_NAMELEN; - return 0; -} - -static int logfs_sb_set(struct super_block *sb, void *_super) -{ - struct logfs_super *super = _super; - - sb->s_fs_info = super; - sb->s_mtd = super->s_mtd; - sb->s_bdev = super->s_bdev; -#ifdef CONFIG_BLOCK - if (sb->s_bdev) - sb->s_bdi = &bdev_get_queue(sb->s_bdev)->backing_dev_info; -#endif -#ifdef CONFIG_MTD - if (sb->s_mtd) - sb->s_bdi = sb->s_mtd->backing_dev_info; -#endif - return 0; -} - -static int logfs_sb_test(struct super_block *sb, void *_super) -{ - struct logfs_super *super = _super; - struct mtd_info *mtd = super->s_mtd; - - if (mtd && sb->s_mtd == mtd) - return 1; - if (super->s_bdev && sb->s_bdev == super->s_bdev) - return 1; - return 0; -} - -static void set_segment_header(struct logfs_segment_header *sh, u8 type, - u8 level, u32 segno, u32 ec) -{ - sh->pad = 0; - sh->type = type; - sh->level = level; - sh->segno = cpu_to_be32(segno); - sh->ec = cpu_to_be32(ec); - sh->gec = cpu_to_be64(segno); - sh->crc = logfs_crc32(sh, LOGFS_SEGMENT_HEADERSIZE, 4); -} - -static void logfs_write_ds(struct super_block *sb, struct logfs_disk_super *ds, - u32 segno, u32 ec) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_segment_header *sh = &ds->ds_sh; - int i; - - memset(ds, 0, sizeof(*ds)); - set_segment_header(sh, SEG_SUPER, 0, segno, ec); - - ds->ds_ifile_levels = super->s_ifile_levels; - ds->ds_iblock_levels = super->s_iblock_levels; - ds->ds_data_levels = super->s_data_levels; /* XXX: Remove */ - ds->ds_segment_shift = super->s_segshift; - ds->ds_block_shift = sb->s_blocksize_bits; - ds->ds_write_shift = super->s_writeshift; - ds->ds_filesystem_size = cpu_to_be64(super->s_size); - ds->ds_segment_size = cpu_to_be32(super->s_segsize); - ds->ds_bad_seg_reserve = cpu_to_be32(super->s_bad_seg_reserve); - ds->ds_feature_incompat = cpu_to_be64(super->s_feature_incompat); - ds->ds_feature_ro_compat= cpu_to_be64(super->s_feature_ro_compat); - ds->ds_feature_compat = cpu_to_be64(super->s_feature_compat); - ds->ds_feature_flags = cpu_to_be64(super->s_feature_flags); - ds->ds_root_reserve = cpu_to_be64(super->s_root_reserve); - ds->ds_speed_reserve = cpu_to_be64(super->s_speed_reserve); - journal_for_each(i) - ds->ds_journal_seg[i] = cpu_to_be32(super->s_journal_seg[i]); - ds->ds_magic = cpu_to_be64(LOGFS_MAGIC); - ds->ds_crc = logfs_crc32(ds, sizeof(*ds), - LOGFS_SEGMENT_HEADERSIZE + 12); -} - -static int write_one_sb(struct super_block *sb, - struct page *(*find_sb)(struct super_block *sb, u64 *ofs)) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_disk_super *ds; - struct logfs_segment_entry se; - struct page *page; - u64 ofs; - u32 ec, segno; - int err; - - page = find_sb(sb, &ofs); - if (!page) - return -EIO; - ds = page_address(page); - segno = seg_no(sb, ofs); - logfs_get_segment_entry(sb, segno, &se); - ec = be32_to_cpu(se.ec_level) >> 4; - ec++; - logfs_set_segment_erased(sb, segno, ec, 0); - logfs_write_ds(sb, ds, segno, ec); - err = super->s_devops->write_sb(sb, page); - put_page(page); - return err; -} - -int logfs_write_sb(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - int err; - - /* First superblock */ - err = write_one_sb(sb, super->s_devops->find_first_sb); - if (err) - return err; - - /* Last superblock */ - err = write_one_sb(sb, super->s_devops->find_last_sb); - if (err) - return err; - return 0; -} - -static int ds_cmp(const void *ds0, const void *ds1) -{ - size_t len = sizeof(struct logfs_disk_super); - - /* We know the segment headers differ, so ignore them */ - len -= LOGFS_SEGMENT_HEADERSIZE; - ds0 += LOGFS_SEGMENT_HEADERSIZE; - ds1 += LOGFS_SEGMENT_HEADERSIZE; - return memcmp(ds0, ds1, len); -} - -static int logfs_recover_sb(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct logfs_disk_super _ds0, *ds0 = &_ds0; - struct logfs_disk_super _ds1, *ds1 = &_ds1; - int err, valid0, valid1; - - /* read first superblock */ - err = wbuf_read(sb, super->s_sb_ofs[0], sizeof(*ds0), ds0); - if (err) - return err; - /* read last superblock */ - err = wbuf_read(sb, super->s_sb_ofs[1], sizeof(*ds1), ds1); - if (err) - return err; - valid0 = logfs_check_ds(ds0) == 0; - valid1 = logfs_check_ds(ds1) == 0; - - if (!valid0 && valid1) { - printk(KERN_INFO"First superblock is invalid - fixing.\n"); - return write_one_sb(sb, super->s_devops->find_first_sb); - } - if (valid0 && !valid1) { - printk(KERN_INFO"Last superblock is invalid - fixing.\n"); - return write_one_sb(sb, super->s_devops->find_last_sb); - } - if (valid0 && valid1 && ds_cmp(ds0, ds1)) { - printk(KERN_INFO"Superblocks don't match - fixing.\n"); - return logfs_write_sb(sb); - } - /* If neither is valid now, something's wrong. Didn't we properly - * check them before?!? */ - BUG_ON(!valid0 && !valid1); - return 0; -} - -static int logfs_make_writeable(struct super_block *sb) -{ - int err; - - err = logfs_open_segfile(sb); - if (err) - return err; - - /* Repair any broken superblock copies */ - err = logfs_recover_sb(sb); - if (err) - return err; - - /* Check areas for trailing unaccounted data */ - err = logfs_check_areas(sb); - if (err) - return err; - - /* Do one GC pass before any data gets dirtied */ - logfs_gc_pass(sb); - - /* after all initializations are done, replay the journal - * for rw-mounts, if necessary */ - err = logfs_replay_journal(sb); - if (err) - return err; - - return 0; -} - -static int logfs_get_sb_final(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct inode *rootdir; - int err; - - /* root dir */ - rootdir = logfs_iget(sb, LOGFS_INO_ROOT); - if (IS_ERR(rootdir)) - goto fail; - - sb->s_root = d_make_root(rootdir); - if (!sb->s_root) - goto fail; - - /* at that point we know that ->put_super() will be called */ - super->s_erase_page = alloc_pages(GFP_KERNEL, 0); - if (!super->s_erase_page) - return -ENOMEM; - memset(page_address(super->s_erase_page), 0xFF, PAGE_SIZE); - - /* FIXME: check for read-only mounts */ - err = logfs_make_writeable(sb); - if (err) { - __free_page(super->s_erase_page); - return err; - } - - log_super("LogFS: Finished mounting\n"); - return 0; - -fail: - iput(super->s_master_inode); - iput(super->s_segfile_inode); - iput(super->s_mapping_inode); - return -EIO; -} - -int logfs_check_ds(struct logfs_disk_super *ds) -{ - struct logfs_segment_header *sh = &ds->ds_sh; - - if (ds->ds_magic != cpu_to_be64(LOGFS_MAGIC)) - return -EINVAL; - if (sh->crc != logfs_crc32(sh, LOGFS_SEGMENT_HEADERSIZE, 4)) - return -EINVAL; - if (ds->ds_crc != logfs_crc32(ds, sizeof(*ds), - LOGFS_SEGMENT_HEADERSIZE + 12)) - return -EINVAL; - return 0; -} - -static struct page *find_super_block(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct page *first, *last; - - first = super->s_devops->find_first_sb(sb, &super->s_sb_ofs[0]); - if (!first || IS_ERR(first)) - return NULL; - last = super->s_devops->find_last_sb(sb, &super->s_sb_ofs[1]); - if (!last || IS_ERR(last)) { - put_page(first); - return NULL; - } - - if (!logfs_check_ds(page_address(first))) { - put_page(last); - return first; - } - - /* First one didn't work, try the second superblock */ - if (!logfs_check_ds(page_address(last))) { - put_page(first); - return last; - } - - /* Neither worked, sorry folks */ - put_page(first); - put_page(last); - return NULL; -} - -static int __logfs_read_sb(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - struct page *page; - struct logfs_disk_super *ds; - int i; - - page = find_super_block(sb); - if (!page) - return -EINVAL; - - ds = page_address(page); - super->s_size = be64_to_cpu(ds->ds_filesystem_size); - super->s_root_reserve = be64_to_cpu(ds->ds_root_reserve); - super->s_speed_reserve = be64_to_cpu(ds->ds_speed_reserve); - super->s_bad_seg_reserve = be32_to_cpu(ds->ds_bad_seg_reserve); - super->s_segsize = 1 << ds->ds_segment_shift; - super->s_segmask = (1 << ds->ds_segment_shift) - 1; - super->s_segshift = ds->ds_segment_shift; - sb->s_blocksize = 1 << ds->ds_block_shift; - sb->s_blocksize_bits = ds->ds_block_shift; - super->s_writesize = 1 << ds->ds_write_shift; - super->s_writeshift = ds->ds_write_shift; - super->s_no_segs = super->s_size >> super->s_segshift; - super->s_no_blocks = super->s_segsize >> sb->s_blocksize_bits; - super->s_feature_incompat = be64_to_cpu(ds->ds_feature_incompat); - super->s_feature_ro_compat = be64_to_cpu(ds->ds_feature_ro_compat); - super->s_feature_compat = be64_to_cpu(ds->ds_feature_compat); - super->s_feature_flags = be64_to_cpu(ds->ds_feature_flags); - - journal_for_each(i) - super->s_journal_seg[i] = be32_to_cpu(ds->ds_journal_seg[i]); - - super->s_ifile_levels = ds->ds_ifile_levels; - super->s_iblock_levels = ds->ds_iblock_levels; - super->s_data_levels = ds->ds_data_levels; - super->s_total_levels = super->s_ifile_levels + super->s_iblock_levels - + super->s_data_levels; - put_page(page); - return 0; -} - -static int logfs_read_sb(struct super_block *sb, int read_only) -{ - struct logfs_super *super = logfs_super(sb); - int ret; - - super->s_btree_pool = mempool_create(32, btree_alloc, btree_free, NULL); - if (!super->s_btree_pool) - return -ENOMEM; - - btree_init_mempool64(&super->s_shadow_tree.new, super->s_btree_pool); - btree_init_mempool64(&super->s_shadow_tree.old, super->s_btree_pool); - btree_init_mempool32(&super->s_shadow_tree.segment_map, - super->s_btree_pool); - - ret = logfs_init_mapping(sb); - if (ret) - return ret; - - ret = __logfs_read_sb(sb); - if (ret) - return ret; - - if (super->s_feature_incompat & ~LOGFS_FEATURES_INCOMPAT) - return -EIO; - if ((super->s_feature_ro_compat & ~LOGFS_FEATURES_RO_COMPAT) && - !read_only) - return -EIO; - - ret = logfs_init_rw(sb); - if (ret) - return ret; - - ret = logfs_init_areas(sb); - if (ret) - return ret; - - ret = logfs_init_gc(sb); - if (ret) - return ret; - - ret = logfs_init_journal(sb); - if (ret) - return ret; - - return 0; -} - -static void logfs_kill_sb(struct super_block *sb) -{ - struct logfs_super *super = logfs_super(sb); - - log_super("LogFS: Start unmounting\n"); - /* Alias entries slow down mount, so evict as many as possible */ - sync_filesystem(sb); - logfs_write_anchor(sb); - free_areas(sb); - - /* - * From this point on alias entries are simply dropped - and any - * writes to the object store are considered bugs. - */ - log_super("LogFS: Now in shutdown\n"); - generic_shutdown_super(sb); - super->s_flags |= LOGFS_SB_FLAG_SHUTDOWN; - - BUG_ON(super->s_dirty_used_bytes || super->s_dirty_free_bytes); - - logfs_cleanup_gc(sb); - logfs_cleanup_journal(sb); - logfs_cleanup_areas(sb); - logfs_cleanup_rw(sb); - if (super->s_erase_page) - __free_page(super->s_erase_page); - super->s_devops->put_device(super); - logfs_mempool_destroy(super->s_btree_pool); - logfs_mempool_destroy(super->s_alias_pool); - kfree(super); - log_super("LogFS: Finished unmounting\n"); -} - -static struct dentry *logfs_get_sb_device(struct logfs_super *super, - struct file_system_type *type, int flags) -{ - struct super_block *sb; - int err = -ENOMEM; - static int mount_count; - - log_super("LogFS: Start mount %x\n", mount_count++); - - err = -EINVAL; - sb = sget(type, logfs_sb_test, logfs_sb_set, flags | MS_NOATIME, super); - if (IS_ERR(sb)) { - super->s_devops->put_device(super); - kfree(super); - return ERR_CAST(sb); - } - - if (sb->s_root) { - /* Device is already in use */ - super->s_devops->put_device(super); - kfree(super); - return dget(sb->s_root); - } - - /* - * sb->s_maxbytes is limited to 8TB. On 32bit systems, the page cache - * only covers 16TB and the upper 8TB are used for indirect blocks. - * On 64bit system we could bump up the limit, but that would make - * the filesystem incompatible with 32bit systems. - */ - sb->s_maxbytes = (1ull << 43) - 1; - sb->s_max_links = LOGFS_LINK_MAX; - sb->s_op = &logfs_super_operations; - - err = logfs_read_sb(sb, sb->s_flags & MS_RDONLY); - if (err) - goto err1; - - sb->s_flags |= MS_ACTIVE; - err = logfs_get_sb_final(sb); - if (err) { - deactivate_locked_super(sb); - return ERR_PTR(err); - } - return dget(sb->s_root); - -err1: - /* no ->s_root, no ->put_super() */ - iput(super->s_master_inode); - iput(super->s_segfile_inode); - iput(super->s_mapping_inode); - deactivate_locked_super(sb); - return ERR_PTR(err); -} - -static struct dentry *logfs_mount(struct file_system_type *type, int flags, - const char *devname, void *data) -{ - ulong mtdnr; - struct logfs_super *super; - int err; - - super = kzalloc(sizeof(*super), GFP_KERNEL); - if (!super) - return ERR_PTR(-ENOMEM); - - mutex_init(&super->s_dirop_mutex); - mutex_init(&super->s_object_alias_mutex); - INIT_LIST_HEAD(&super->s_freeing_list); - - if (!devname) - err = logfs_get_sb_bdev(super, type, devname); - else if (strncmp(devname, "mtd", 3)) - err = logfs_get_sb_bdev(super, type, devname); - else { - char *garbage; - mtdnr = simple_strtoul(devname+3, &garbage, 0); - if (*garbage) - err = -EINVAL; - else - err = logfs_get_sb_mtd(super, mtdnr); - } - - if (err) { - kfree(super); - return ERR_PTR(err); - } - - return logfs_get_sb_device(super, type, flags); -} - -static struct file_system_type logfs_fs_type = { - .owner = THIS_MODULE, - .name = "logfs", - .mount = logfs_mount, - .kill_sb = logfs_kill_sb, - .fs_flags = FS_REQUIRES_DEV, - -}; -MODULE_ALIAS_FS("logfs"); - -static int __init logfs_init(void) -{ - int ret; - - emergency_page = alloc_pages(GFP_KERNEL, 0); - if (!emergency_page) - return -ENOMEM; - - ret = logfs_compr_init(); - if (ret) - goto out1; - - ret = logfs_init_inode_cache(); - if (ret) - goto out2; - - ret = register_filesystem(&logfs_fs_type); - if (!ret) - return 0; - logfs_destroy_inode_cache(); -out2: - logfs_compr_exit(); -out1: - __free_pages(emergency_page, 0); - return ret; -} - -static void __exit logfs_exit(void) -{ - unregister_filesystem(&logfs_fs_type); - logfs_destroy_inode_cache(); - logfs_compr_exit(); - __free_pages(emergency_page, 0); -} - -module_init(logfs_init); -module_exit(logfs_exit); - -MODULE_LICENSE("GPL v2"); -MODULE_AUTHOR("Joern Engel <joern@logfs.org>"); -MODULE_DESCRIPTION("scalable flash filesystem"); diff --git a/fs/namei.c b/fs/namei.c index 5b4eed221530..092ac5667ec7 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1725,30 +1725,35 @@ static int pick_link(struct nameidata *nd, struct path *link, return 1; } +enum {WALK_FOLLOW = 1, WALK_MORE = 2}; + /* * Do we need to follow links? We _really_ want to be able * to do this check without having to look at inode->i_op, * so we keep a cache of "no, this doesn't need follow_link" * for the common case. */ -static inline int should_follow_link(struct nameidata *nd, struct path *link, - int follow, - struct inode *inode, unsigned seq) +static inline int step_into(struct nameidata *nd, struct path *path, + int flags, struct inode *inode, unsigned seq) { - if (likely(!d_is_symlink(link->dentry))) - return 0; - if (!follow) + if (!(flags & WALK_MORE) && nd->depth) + put_link(nd); + if (likely(!d_is_symlink(path->dentry)) || + !(flags & WALK_FOLLOW || nd->flags & LOOKUP_FOLLOW)) { + /* not a symlink or should not follow */ + path_to_nameidata(path, nd); + nd->inode = inode; + nd->seq = seq; return 0; + } /* make sure that d_is_symlink above matches inode */ if (nd->flags & LOOKUP_RCU) { - if (read_seqcount_retry(&link->dentry->d_seq, seq)) + if (read_seqcount_retry(&path->dentry->d_seq, seq)) return -ECHILD; } - return pick_link(nd, link, inode, seq); + return pick_link(nd, path, inode, seq); } -enum {WALK_GET = 1, WALK_PUT = 2}; - static int walk_component(struct nameidata *nd, int flags) { struct path path; @@ -1762,7 +1767,7 @@ static int walk_component(struct nameidata *nd, int flags) */ if (unlikely(nd->last_type != LAST_NORM)) { err = handle_dots(nd, nd->last_type); - if (flags & WALK_PUT) + if (!(flags & WALK_MORE) && nd->depth) put_link(nd); return err; } @@ -1789,15 +1794,7 @@ static int walk_component(struct nameidata *nd, int flags) inode = d_backing_inode(path.dentry); } - if (flags & WALK_PUT) - put_link(nd); - err = should_follow_link(nd, &path, flags & WALK_GET, inode, seq); - if (unlikely(err)) - return err; - path_to_nameidata(&path, nd); - nd->inode = inode; - nd->seq = seq; - return 0; + return step_into(nd, &path, flags, inode, seq); } /* @@ -2104,9 +2101,10 @@ OK: if (!name) return 0; /* last component of nested symlink */ - err = walk_component(nd, WALK_GET | WALK_PUT); + err = walk_component(nd, WALK_FOLLOW); } else { - err = walk_component(nd, WALK_GET); + /* not the last component */ + err = walk_component(nd, WALK_FOLLOW | WALK_MORE); } if (err < 0) return err; @@ -2248,12 +2246,7 @@ static inline int lookup_last(struct nameidata *nd) nd->flags |= LOOKUP_FOLLOW | LOOKUP_DIRECTORY; nd->flags &= ~LOOKUP_PARENT; - return walk_component(nd, - nd->flags & LOOKUP_FOLLOW - ? nd->depth - ? WALK_PUT | WALK_GET - : WALK_GET - : 0); + return walk_component(nd, 0); } /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */ @@ -2558,28 +2551,9 @@ int user_path_at_empty(int dfd, const char __user *name, unsigned flags, } EXPORT_SYMBOL(user_path_at_empty); -/* - * NB: most callers don't do anything directly with the reference to the - * to struct filename, but the nd->last pointer points into the name string - * allocated by getname. So we must hold the reference to it until all - * path-walking is complete. - */ -static inline struct filename * -user_path_parent(int dfd, const char __user *path, - struct path *parent, - struct qstr *last, - int *type, - unsigned int flags) -{ - /* only LOOKUP_REVAL is allowed in extra flags */ - return filename_parentat(dfd, getname(path), flags & LOOKUP_REVAL, - parent, last, type); -} - /** * mountpoint_last - look up last component for umount * @nd: pathwalk nameidata - currently pointing at parent directory of "last" - * @path: pointer to container for result * * This is a special lookup_last function just for umount. In this case, we * need to resolve the path without doing any revalidation. @@ -2592,23 +2566,20 @@ user_path_parent(int dfd, const char __user *path, * * Returns: * -error: if there was an error during lookup. This includes -ENOENT if the - * lookup found a negative dentry. The nd->path reference will also be - * put in this case. + * lookup found a negative dentry. * - * 0: if we successfully resolved nd->path and found it to not to be a - * symlink that needs to be followed. "path" will also be populated. - * The nd->path reference will also be put. + * 0: if we successfully resolved nd->last and found it to not to be a + * symlink that needs to be followed. * * 1: if we successfully resolved nd->last and found it to be a symlink - * that needs to be followed. "path" will be populated with the path - * to the link, and nd->path will *not* be put. + * that needs to be followed. */ static int -mountpoint_last(struct nameidata *nd, struct path *path) +mountpoint_last(struct nameidata *nd) { int error = 0; - struct dentry *dentry; struct dentry *dir = nd->path.dentry; + struct path path; /* If we're in rcuwalk, drop out of it to handle last component */ if (nd->flags & LOOKUP_RCU) { @@ -2622,37 +2593,28 @@ mountpoint_last(struct nameidata *nd, struct path *path) error = handle_dots(nd, nd->last_type); if (error) return error; - dentry = dget(nd->path.dentry); + path.dentry = dget(nd->path.dentry); } else { - dentry = d_lookup(dir, &nd->last); - if (!dentry) { + path.dentry = d_lookup(dir, &nd->last); + if (!path.dentry) { /* * No cached dentry. Mounted dentries are pinned in the * cache, so that means that this dentry is probably * a symlink or the path doesn't actually point * to a mounted dentry. */ - dentry = lookup_slow(&nd->last, dir, + path.dentry = lookup_slow(&nd->last, dir, nd->flags | LOOKUP_NO_REVAL); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); + if (IS_ERR(path.dentry)) + return PTR_ERR(path.dentry); } } - if (d_is_negative(dentry)) { - dput(dentry); + if (d_is_negative(path.dentry)) { + dput(path.dentry); return -ENOENT; } - if (nd->depth) - put_link(nd); - path->dentry = dentry; - path->mnt = nd->path.mnt; - error = should_follow_link(nd, path, nd->flags & LOOKUP_FOLLOW, - d_backing_inode(dentry), 0); - if (unlikely(error)) - return error; - mntget(path->mnt); - follow_mount(path); - return 0; + path.mnt = nd->path.mnt; + return step_into(nd, &path, 0, d_backing_inode(path.dentry), 0); } /** @@ -2672,13 +2634,19 @@ path_mountpoint(struct nameidata *nd, unsigned flags, struct path *path) if (IS_ERR(s)) return PTR_ERR(s); while (!(err = link_path_walk(s, nd)) && - (err = mountpoint_last(nd, path)) > 0) { + (err = mountpoint_last(nd)) > 0) { s = trailing_symlink(nd); if (IS_ERR(s)) { err = PTR_ERR(s); break; } } + if (!err) { + *path = nd->path; + nd->path.mnt = NULL; + nd->path.dentry = NULL; + follow_mount(path); + } terminate_walk(nd); return err; } @@ -3335,18 +3303,11 @@ static int do_last(struct nameidata *nd, seq = 0; /* out of RCU mode, so the value doesn't matter */ inode = d_backing_inode(path.dentry); finish_lookup: - if (nd->depth) - put_link(nd); - error = should_follow_link(nd, &path, nd->flags & LOOKUP_FOLLOW, - inode, seq); + error = step_into(nd, &path, 0, inode, seq); if (unlikely(error)) return error; - - path_to_nameidata(&path, nd); - nd->inode = inode; - nd->seq = seq; - /* Why this, you ask? _Now_ we might have grown LOOKUP_JUMPED... */ finish_open: + /* Why this, you ask? _Now_ we might have grown LOOKUP_JUMPED... */ error = complete_walk(nd); if (error) return error; @@ -3861,8 +3822,8 @@ static long do_rmdir(int dfd, const char __user *pathname) int type; unsigned int lookup_flags = 0; retry: - name = user_path_parent(dfd, pathname, - &path, &last, &type, lookup_flags); + name = filename_parentat(dfd, getname(pathname), lookup_flags, + &path, &last, &type); if (IS_ERR(name)) return PTR_ERR(name); @@ -3991,8 +3952,8 @@ static long do_unlinkat(int dfd, const char __user *pathname) struct inode *delegated_inode = NULL; unsigned int lookup_flags = 0; retry: - name = user_path_parent(dfd, pathname, - &path, &last, &type, lookup_flags); + name = filename_parentat(dfd, getname(pathname), lookup_flags, + &path, &last, &type); if (IS_ERR(name)) return PTR_ERR(name); @@ -4491,15 +4452,15 @@ SYSCALL_DEFINE5(renameat2, int, olddfd, const char __user *, oldname, target_flags = 0; retry: - from = user_path_parent(olddfd, oldname, - &old_path, &old_last, &old_type, lookup_flags); + from = filename_parentat(olddfd, getname(oldname), lookup_flags, + &old_path, &old_last, &old_type); if (IS_ERR(from)) { error = PTR_ERR(from); goto exit; } - to = user_path_parent(newdfd, newname, - &new_path, &new_last, &new_type, lookup_flags); + to = filename_parentat(newdfd, getname(newname), lookup_flags, + &new_path, &new_last, &new_type); if (IS_ERR(to)) { error = PTR_ERR(to); goto exit1; diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c index dd38ca1f2ecb..83ca77231707 100644 --- a/fs/ncpfs/file.c +++ b/fs/ncpfs/file.c @@ -203,7 +203,7 @@ ncp_file_write_iter(struct kiocb *iocb, struct iov_iter *from) bufsize - (pos % bufsize), iov_iter_count(from)); - if (copy_from_iter(bouncebuffer, to_write, from) != to_write) { + if (!copy_from_iter_full(bouncebuffer, to_write, from)) { errno = -EFAULT; break; } diff --git a/fs/orangefs/devorangefs-req.c b/fs/orangefs/devorangefs-req.c index 516ffb4dc9a0..b0ced669427e 100644 --- a/fs/orangefs/devorangefs-req.c +++ b/fs/orangefs/devorangefs-req.c @@ -355,7 +355,6 @@ static ssize_t orangefs_devreq_write_iter(struct kiocb *iocb, __u64 tag; } head; int total = ret = iov_iter_count(iter); - int n; int downcall_size = sizeof(struct orangefs_downcall_s); int head_size = sizeof(head); @@ -372,8 +371,7 @@ static ssize_t orangefs_devreq_write_iter(struct kiocb *iocb, return -EFAULT; } - n = copy_from_iter(&head, head_size, iter); - if (n < head_size) { + if (!copy_from_iter_full(&head, head_size, iter)) { gossip_err("%s: failed to copy head.\n", __func__); return -EFAULT; } @@ -407,8 +405,7 @@ static ssize_t orangefs_devreq_write_iter(struct kiocb *iocb, return ret; } - n = copy_from_iter(&op->downcall, downcall_size, iter); - if (n != downcall_size) { + if (!copy_from_iter_full(&op->downcall, downcall_size, iter)) { gossip_err("%s: failed to copy downcall.\n", __func__); goto Efault; } @@ -462,10 +459,8 @@ static ssize_t orangefs_devreq_write_iter(struct kiocb *iocb, goto Enomem; } memset(op->downcall.trailer_buf, 0, op->downcall.trailer_size); - n = copy_from_iter(op->downcall.trailer_buf, - op->downcall.trailer_size, - iter); - if (n != op->downcall.trailer_size) { + if (!copy_from_iter_full(op->downcall.trailer_buf, + op->downcall.trailer_size, iter)) { gossip_err("%s: failed to copy trailer.\n", __func__); vfree(op->downcall.trailer_buf); goto Efault; diff --git a/fs/orangefs/file.c b/fs/orangefs/file.c index 02cc6139ec90..e6bbc8083d77 100644 --- a/fs/orangefs/file.c +++ b/fs/orangefs/file.c @@ -724,7 +724,7 @@ static int orangefs_lock(struct file *filp, int cmd, struct file_lock *fl) { int rc = -EINVAL; - if (ORANGEFS_SB(filp->f_inode->i_sb)->flags & ORANGEFS_OPT_LOCAL_LOCK) { + if (ORANGEFS_SB(file_inode(filp)->i_sb)->flags & ORANGEFS_OPT_LOCAL_LOCK) { if (cmd == F_GETLK) { rc = 0; posix_test_lock(filp, fl); diff --git a/fs/orangefs/orangefs-debugfs.c b/fs/orangefs/orangefs-debugfs.c index b5dbc9c6530c..27e75cf28b3a 100644 --- a/fs/orangefs/orangefs-debugfs.c +++ b/fs/orangefs/orangefs-debugfs.c @@ -434,6 +434,7 @@ static ssize_t orangefs_debug_write(struct file *file, char *debug_string; struct orangefs_kernel_op_s *new_op = NULL; struct client_debug_mask c_mask = { NULL, 0, 0 }; + char *s; gossip_debug(GOSSIP_DEBUGFS_DEBUG, "orangefs_debug_write: %pD\n", @@ -521,8 +522,9 @@ static ssize_t orangefs_debug_write(struct file *file, } mutex_lock(&orangefs_debug_lock); - memset(file->f_inode->i_private, 0, ORANGEFS_MAX_DEBUG_STRING_LEN); - sprintf((char *)file->f_inode->i_private, "%s\n", debug_string); + s = file_inode(file)->i_private; + memset(s, 0, ORANGEFS_MAX_DEBUG_STRING_LEN); + sprintf(s, "%s\n", debug_string); mutex_unlock(&orangefs_debug_lock); *ppos += count; diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c index 36795eed40b0..2838bddb1f91 100644 --- a/fs/overlayfs/copy_up.c +++ b/fs/overlayfs/copy_up.c @@ -33,7 +33,7 @@ static int ovl_check_fd(const void *data, struct file *f, unsigned int fd) { const struct dentry *dentry = data; - if (f->f_inode == d_inode(dentry)) + if (file_inode(f) == d_inode(dentry)) pr_warn_ratelimited("overlayfs: Warning: Copying up %pD, but open R/O on fd %u which will cease to be coherent [pid=%d %s]\n", f, fd, current->pid, current->comm); return 0; diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 332e76756f54..ac7fa34db8a7 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2818,12 +2818,12 @@ static inline int skb_add_data(struct sk_buff *skb, if (skb->ip_summed == CHECKSUM_NONE) { __wsum csum = 0; - if (csum_and_copy_from_iter(skb_put(skb, copy), copy, - &csum, from) == copy) { + if (csum_and_copy_from_iter_full(skb_put(skb, copy), copy, + &csum, from)) { skb->csum = csum_block_add(skb->csum, csum, off); return 0; } - } else if (copy_from_iter(skb_put(skb, copy), copy, from) == copy) + } else if (copy_from_iter_full(skb_put(skb, copy), copy, from)) return 0; __skb_trim(skb, off); diff --git a/include/linux/uio.h b/include/linux/uio.h index d5aba1512b8b..804e34c6f981 100644 --- a/include/linux/uio.h +++ b/include/linux/uio.h @@ -89,7 +89,9 @@ size_t copy_page_from_iter(struct page *page, size_t offset, size_t bytes, struct iov_iter *i); size_t copy_to_iter(const void *addr, size_t bytes, struct iov_iter *i); size_t copy_from_iter(void *addr, size_t bytes, struct iov_iter *i); +bool copy_from_iter_full(void *addr, size_t bytes, struct iov_iter *i); size_t copy_from_iter_nocache(void *addr, size_t bytes, struct iov_iter *i); +bool copy_from_iter_full_nocache(void *addr, size_t bytes, struct iov_iter *i); size_t iov_iter_zero(size_t bytes, struct iov_iter *); unsigned long iov_iter_alignment(const struct iov_iter *i); unsigned long iov_iter_gap_alignment(const struct iov_iter *i); @@ -155,6 +157,7 @@ static inline void iov_iter_reexpand(struct iov_iter *i, size_t count) } size_t csum_and_copy_to_iter(const void *addr, size_t bytes, __wsum *csum, struct iov_iter *i); size_t csum_and_copy_from_iter(void *addr, size_t bytes, __wsum *csum, struct iov_iter *i); +bool csum_and_copy_from_iter_full(void *addr, size_t bytes, __wsum *csum, struct iov_iter *i); int import_iovec(int type, const struct iovec __user * uvector, unsigned nr_segs, unsigned fast_segs, diff --git a/include/net/sock.h b/include/net/sock.h index e17aa3de2b4d..282d065e286b 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1836,13 +1836,13 @@ static inline int skb_do_copy_data_nocache(struct sock *sk, struct sk_buff *skb, { if (skb->ip_summed == CHECKSUM_NONE) { __wsum csum = 0; - if (csum_and_copy_from_iter(to, copy, &csum, from) != copy) + if (!csum_and_copy_from_iter_full(to, copy, &csum, from)) return -EFAULT; skb->csum = csum_block_add(skb->csum, csum, offset); } else if (sk->sk_route_caps & NETIF_F_NOCACHE_COPY) { - if (copy_from_iter_nocache(to, copy, from) != copy) + if (!copy_from_iter_full_nocache(to, copy, from)) return -EFAULT; - } else if (copy_from_iter(to, copy, from) != copy) + } else if (!copy_from_iter_full(to, copy, from)) return -EFAULT; return 0; diff --git a/include/net/udplite.h b/include/net/udplite.h index 36097d388219..ea340524f99b 100644 --- a/include/net/udplite.h +++ b/include/net/udplite.h @@ -20,7 +20,7 @@ static __inline__ int udplite_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb) { struct msghdr *msg = from; - return copy_from_iter(to, len, &msg->msg_iter) != len ? -EFAULT : 0; + return copy_from_iter_full(to, len, &msg->msg_iter) ? 0 : -EFAULT; } /* Designate sk as UDP-Lite socket */ diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c index 686e068ec3da..2d7bdcb996b6 100644 --- a/kernel/audit_watch.c +++ b/kernel/audit_watch.c @@ -547,8 +547,8 @@ int audit_exe_compare(struct task_struct *tsk, struct audit_fsnotify_mark *mark) exe_file = get_task_exe_file(tsk); if (!exe_file) return 0; - ino = exe_file->f_inode->i_ino; - dev = exe_file->f_inode->i_sb->s_dev; + ino = file_inode(exe_file)->i_ino; + dev = file_inode(exe_file)->i_sb->s_dev; fput(exe_file); return audit_mark_compare(mark, ino, dev); } diff --git a/kernel/events/core.c b/kernel/events/core.c index faf073d0287f..ab15509fab8c 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -6698,7 +6698,7 @@ static bool perf_addr_filter_match(struct perf_addr_filter *filter, struct file *file, unsigned long offset, unsigned long size) { - if (filter->inode != file->f_inode) + if (filter->inode != file_inode(file)) return false; if (filter->offset > offset + size) diff --git a/kernel/locking/qspinlock_stat.h b/kernel/locking/qspinlock_stat.h index eb0a599fcf58..e852be4851fc 100644 --- a/kernel/locking/qspinlock_stat.h +++ b/kernel/locking/qspinlock_stat.h @@ -108,11 +108,7 @@ static ssize_t qstat_read(struct file *file, char __user *user_buf, /* * Get the counter ID stored in file->f_inode->i_private */ - if (!file->f_inode) { - WARN_ON_ONCE(1); - return -EBADF; - } - counter = (long)(file->f_inode->i_private); + counter = (long)file_inode(file)->i_private; if (counter >= qstat_num) return -EBADF; @@ -177,11 +173,7 @@ static ssize_t qstat_write(struct file *file, const char __user *user_buf, /* * Get the counter ID stored in file->f_inode->i_private */ - if (!file->f_inode) { - WARN_ON_ONCE(1); - return -EBADF; - } - if ((long)(file->f_inode->i_private) != qstat_reset_cnts) + if ((long)file_inode(file)->i_private != qstat_reset_cnts) return count; for_each_possible_cpu(cpu) { diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index bc2e220ed2b0..e2cdd87e7a63 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -733,7 +733,7 @@ static ssize_t devkmsg_write(struct kiocb *iocb, struct iov_iter *from) return -ENOMEM; buf[len] = '\0'; - if (copy_from_iter(buf, len, from) != len) { + if (!copy_from_iter_full(buf, len, from)) { kfree(buf); return -EFAULT; } diff --git a/lib/iov_iter.c b/lib/iov_iter.c index 691a52b634fe..228892dabba6 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -569,6 +569,31 @@ size_t copy_from_iter(void *addr, size_t bytes, struct iov_iter *i) } EXPORT_SYMBOL(copy_from_iter); +bool copy_from_iter_full(void *addr, size_t bytes, struct iov_iter *i) +{ + char *to = addr; + if (unlikely(i->type & ITER_PIPE)) { + WARN_ON(1); + return false; + } + if (unlikely(i->count < bytes)) \ + return false; + + iterate_all_kinds(i, bytes, v, ({ + if (__copy_from_user((to += v.iov_len) - v.iov_len, + v.iov_base, v.iov_len)) + return false; + 0;}), + memcpy_from_page((to += v.bv_len) - v.bv_len, v.bv_page, + v.bv_offset, v.bv_len), + memcpy((to += v.iov_len) - v.iov_len, v.iov_base, v.iov_len) + ) + + iov_iter_advance(i, bytes); + return true; +} +EXPORT_SYMBOL(copy_from_iter_full); + size_t copy_from_iter_nocache(void *addr, size_t bytes, struct iov_iter *i) { char *to = addr; @@ -588,6 +613,30 @@ size_t copy_from_iter_nocache(void *addr, size_t bytes, struct iov_iter *i) } EXPORT_SYMBOL(copy_from_iter_nocache); +bool copy_from_iter_full_nocache(void *addr, size_t bytes, struct iov_iter *i) +{ + char *to = addr; + if (unlikely(i->type & ITER_PIPE)) { + WARN_ON(1); + return false; + } + if (unlikely(i->count < bytes)) \ + return false; + iterate_all_kinds(i, bytes, v, ({ + if (__copy_from_user_nocache((to += v.iov_len) - v.iov_len, + v.iov_base, v.iov_len)) + return false; + 0;}), + memcpy_from_page((to += v.bv_len) - v.bv_len, v.bv_page, + v.bv_offset, v.bv_len), + memcpy((to += v.iov_len) - v.iov_len, v.iov_base, v.iov_len) + ) + + iov_iter_advance(i, bytes); + return true; +} +EXPORT_SYMBOL(copy_from_iter_full_nocache); + size_t copy_page_to_iter(struct page *page, size_t offset, size_t bytes, struct iov_iter *i) { @@ -1009,7 +1058,7 @@ size_t csum_and_copy_from_iter(void *addr, size_t bytes, __wsum *csum, } iterate_and_advance(i, bytes, v, ({ int err = 0; - next = csum_and_copy_from_user(v.iov_base, + next = csum_and_copy_from_user(v.iov_base, (to += v.iov_len) - v.iov_len, v.iov_len, 0, &err); if (!err) { @@ -1038,6 +1087,51 @@ size_t csum_and_copy_from_iter(void *addr, size_t bytes, __wsum *csum, } EXPORT_SYMBOL(csum_and_copy_from_iter); +bool csum_and_copy_from_iter_full(void *addr, size_t bytes, __wsum *csum, + struct iov_iter *i) +{ + char *to = addr; + __wsum sum, next; + size_t off = 0; + sum = *csum; + if (unlikely(i->type & ITER_PIPE)) { + WARN_ON(1); + return false; + } + if (unlikely(i->count < bytes)) + return false; + iterate_all_kinds(i, bytes, v, ({ + int err = 0; + next = csum_and_copy_from_user(v.iov_base, + (to += v.iov_len) - v.iov_len, + v.iov_len, 0, &err); + if (err) + return false; + sum = csum_block_add(sum, next, off); + off += v.iov_len; + 0; + }), ({ + char *p = kmap_atomic(v.bv_page); + next = csum_partial_copy_nocheck(p + v.bv_offset, + (to += v.bv_len) - v.bv_len, + v.bv_len, 0); + kunmap_atomic(p); + sum = csum_block_add(sum, next, off); + off += v.bv_len; + }),({ + next = csum_partial_copy_nocheck(v.iov_base, + (to += v.iov_len) - v.iov_len, + v.iov_len, 0); + sum = csum_block_add(sum, next, off); + off += v.iov_len; + }) + ) + *csum = sum; + iov_iter_advance(i, bytes); + return true; +} +EXPORT_SYMBOL(csum_and_copy_from_iter_full); + size_t csum_and_copy_to_iter(const void *addr, size_t bytes, __wsum *csum, struct iov_iter *i) { @@ -1052,7 +1146,7 @@ size_t csum_and_copy_to_iter(const void *addr, size_t bytes, __wsum *csum, iterate_and_advance(i, bytes, v, ({ int err = 0; next = csum_and_copy_to_user((from += v.iov_len) - v.iov_len, - v.iov_base, + v.iov_base, v.iov_len, 0, &err); if (!err) { sum = csum_block_add(sum, next, off); diff --git a/net/atm/common.c b/net/atm/common.c index 6dc12305799e..a3ca922d307b 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -630,7 +630,7 @@ int vcc_sendmsg(struct socket *sock, struct msghdr *m, size_t size) goto out; skb->dev = NULL; /* for paths shared with net_device interfaces */ ATM_SKB(skb)->atm_options = vcc->atm_options; - if (copy_from_iter(skb_put(skb, size), size, &m->msg_iter) != size) { + if (!copy_from_iter_full(skb_put(skb, size), size, &m->msg_iter)) { kfree_skb(skb); error = -EFAULT; goto out; diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 577f1c01454a..ce0b5dd01953 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -2127,7 +2127,7 @@ static inline int l2cap_skbuff_fromiovec(struct l2cap_chan *chan, struct sk_buff **frag; int sent = 0; - if (copy_from_iter(skb_put(skb, count), count, &msg->msg_iter) != count) + if (!copy_from_iter_full(skb_put(skb, count), count, &msg->msg_iter)) return -EFAULT; sent += count; @@ -2147,8 +2147,8 @@ static inline int l2cap_skbuff_fromiovec(struct l2cap_chan *chan, *frag = tmp; - if (copy_from_iter(skb_put(*frag, count), count, - &msg->msg_iter) != count) + if (!copy_from_iter_full(skb_put(*frag, count), count, + &msg->msg_iter)) return -EFAULT; sent += count; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 9ffc2625cddd..6c9615c90f37 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -826,11 +826,11 @@ ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk struct msghdr *msg = from; if (skb->ip_summed == CHECKSUM_PARTIAL) { - if (copy_from_iter(to, len, &msg->msg_iter) != len) + if (!copy_from_iter_full(to, len, &msg->msg_iter)) return -EFAULT; } else { __wsum csum = 0; - if (csum_and_copy_from_iter(to, len, &csum, &msg->msg_iter) != len) + if (!csum_and_copy_from_iter_full(to, len, &csum, &msg->msg_iter)) return -EFAULT; skb->csum = csum_block_add(skb->csum, csum, odd); } diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index 5b2635e69a92..86cca610f4c2 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -609,15 +609,15 @@ int ping_getfrag(void *from, char *to, fraglen -= sizeof(struct icmphdr); if (fraglen < 0) BUG(); - if (csum_and_copy_from_iter(to + sizeof(struct icmphdr), + if (!csum_and_copy_from_iter_full(to + sizeof(struct icmphdr), fraglen, &pfh->wcheck, - &pfh->msg->msg_iter) != fraglen) + &pfh->msg->msg_iter)) return -EFAULT; } else if (offset < sizeof(struct icmphdr)) { BUG(); } else { - if (csum_and_copy_from_iter(to, fraglen, &pfh->wcheck, - &pfh->msg->msg_iter) != fraglen) + if (!csum_and_copy_from_iter_full(to, fraglen, &pfh->wcheck, + &pfh->msg->msg_iter)) return -EFAULT; } diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 89f2e8c1f4dc..49cd0c70a13a 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2397,14 +2397,11 @@ static int __packet_snd_vnet_parse(struct virtio_net_hdr *vnet_hdr, size_t len) static int packet_snd_vnet_parse(struct msghdr *msg, size_t *len, struct virtio_net_hdr *vnet_hdr) { - int n; - if (*len < sizeof(*vnet_hdr)) return -EINVAL; *len -= sizeof(*vnet_hdr); - n = copy_from_iter(vnet_hdr, sizeof(*vnet_hdr), &msg->msg_iter); - if (n != sizeof(*vnet_hdr)) + if (!copy_from_iter_full(vnet_hdr, sizeof(*vnet_hdr), &msg->msg_iter)) return -EFAULT; return __packet_snd_vnet_parse(vnet_hdr, *len); diff --git a/net/tipc/msg.c b/net/tipc/msg.c index 17201aa8423d..a22be502f1bd 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -268,7 +268,7 @@ int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, __skb_queue_tail(list, skb); skb_copy_to_linear_data(skb, mhdr, mhsz); pktpos = skb->data + mhsz; - if (copy_from_iter(pktpos, dsz, &m->msg_iter) == dsz) + if (copy_from_iter_full(pktpos, dsz, &m->msg_iter)) return dsz; rc = -EFAULT; goto error; @@ -299,7 +299,7 @@ int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, if (drem < pktrem) pktrem = drem; - if (copy_from_iter(pktpos, pktrem, &m->msg_iter) != pktrem) { + if (!copy_from_iter_full(pktpos, pktrem, &m->msg_iter)) { rc = -EFAULT; goto error; } diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index d580ad06b792..f89f1900e58d 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -1074,7 +1074,7 @@ long keyctl_instantiate_key_common(key_serial_t id, } ret = -EFAULT; - if (copy_from_iter(payload, plen, from) != plen) + if (!copy_from_iter_full(payload, plen, from)) goto error2; } diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index 4d90257d03ad..94dc9d406ce3 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -225,7 +225,7 @@ static int smk_bu_credfile(const struct cred *cred, struct file *file, { struct task_smack *tsp = cred->security; struct smack_known *sskp = tsp->smk_task; - struct inode *inode = file->f_inode; + struct inode *inode = file_inode(file); struct inode_smack *isp = inode->i_security; char acc[SMK_NUM_ACCESS_TYPE + 1]; |