@@ -6,6 +6,8 @@ config BTRFS_FS
select ZLIB_DEFLATE
select LZO_COMPRESS
select LZO_DECOMPRESS
+ select LZ4_COMPRESS
+ select LZ4_DECOMPRESS
select RAID6_PQ
select XOR_BLOCKS
select SRCU
@@ -6,7 +6,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
transaction.o inode.o file.o tree-defrag.o \
extent_map.o sysfs.o struct-funcs.o xattr.o ordered-data.o \
extent_io.o volumes.o async-thread.o ioctl.o locking.o orphan.o \
- export.o tree-log.o free-space-cache.o zlib.o lzo.o \
+ export.o tree-log.o free-space-cache.o zlib.o lzo.o lz4.o \
compression.o delayed-ref.o relocation.o delayed-inode.o scrub.o \
reada.o backref.o ulist.o qgroup.o send.o dev-replace.o raid56.o \
uuid-tree.o props.o hash.o free-space-tree.o
@@ -755,6 +755,7 @@ static struct {
static const struct btrfs_compress_op * const btrfs_compress_op[] = {
&btrfs_zlib_compress,
&btrfs_lzo_compress,
+ &btrfs_lz4_compress,
};
void __init btrfs_init_compress(void)
@@ -79,5 +79,6 @@ struct btrfs_compress_op {
extern const struct btrfs_compress_op btrfs_zlib_compress;
extern const struct btrfs_compress_op btrfs_lzo_compress;
+extern const struct btrfs_compress_op btrfs_lz4_compress;
#endif
@@ -528,6 +528,7 @@ struct btrfs_super_block {
#define BTRFS_FEATURE_INCOMPAT_RAID56 (1ULL << 7)
#define BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA (1ULL << 8)
#define BTRFS_FEATURE_INCOMPAT_NO_HOLES (1ULL << 9)
+#define BTRFS_FEATURE_INCOMPAT_COMPRESS_LZ4 (1ULL << 10)
#define BTRFS_FEATURE_COMPAT_SUPP 0ULL
#define BTRFS_FEATURE_COMPAT_SAFE_SET 0ULL
@@ -545,6 +546,7 @@ struct btrfs_super_block {
BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS | \
BTRFS_FEATURE_INCOMPAT_BIG_METADATA | \
BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO | \
+ BTRFS_FEATURE_INCOMPAT_COMPRESS_LZ4 | \
BTRFS_FEATURE_INCOMPAT_RAID56 | \
BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF | \
BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA | \
@@ -719,8 +721,9 @@ enum btrfs_compression_type {
BTRFS_COMPRESS_NONE = 0,
BTRFS_COMPRESS_ZLIB = 1,
BTRFS_COMPRESS_LZO = 2,
- BTRFS_COMPRESS_TYPES = 2,
- BTRFS_COMPRESS_LAST = 3,
+ BTRFS_COMPRESS_LZ4 = 3,
+ BTRFS_COMPRESS_TYPES = 3,
+ BTRFS_COMPRESS_LAST = 4,
};
struct btrfs_inode_item {
@@ -2766,6 +2766,8 @@ int open_ctree(struct super_block *sb,
features |= BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF;
if (tree_root->fs_info->compress_type == BTRFS_COMPRESS_LZO)
features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
+ else if (tree_root->fs_info->compress_type == BTRFS_COMPRESS_LZ4)
+ features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZ4;
if (features & BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA)
printk(KERN_INFO "BTRFS: has skinny extents\n");
@@ -325,6 +325,8 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
if (root->fs_info->compress_type == BTRFS_COMPRESS_LZO)
comp = "lzo";
+ else if (root->fs_info->compress_type == BTRFS_COMPRESS_LZ4)
+ comp = "lz4";
else
comp = "zlib";
ret = btrfs_set_prop(inode, "btrfs.compression",
@@ -1455,6 +1457,8 @@ int btrfs_defrag_file(struct inode *inode, struct file *file,
if (range->compress_type == BTRFS_COMPRESS_LZO) {
btrfs_set_fs_incompat(root->fs_info, COMPRESS_LZO);
+ } else if (range->compress_type == BTRFS_COMPRESS_LZ4) {
+ btrfs_set_fs_incompat(root->fs_info, COMPRESS_LZ4);
}
ret = defrag_count;
new file mode 100644
@@ -0,0 +1,445 @@
+/*
+ * Copyright (C) 2008 Oracle. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+#include <linux/err.h>
+#include <linux/sched.h>
+#include <linux/pagemap.h>
+#include <linux/bio.h>
+#include <linux/lz4.h>
+#include "compression.h"
+
+#define LZ4_LEN 4
+
+struct workspace {
+ void *mem;
+ void *buf; /* where decompressed data goes */
+ void *cbuf; /* where compressed data goes */
+ struct list_head list;
+};
+
+static void lz4_free_workspace(struct list_head *ws)
+{
+ struct workspace *workspace = list_entry(ws, struct workspace, list);
+
+ vfree(workspace->buf);
+ vfree(workspace->cbuf);
+ vfree(workspace->mem);
+ kfree(workspace);
+}
+
+static struct list_head *lz4_alloc_workspace(void)
+{
+ struct workspace *workspace;
+
+ workspace = kzalloc(sizeof(*workspace), GFP_NOFS);
+ if (!workspace)
+ return ERR_PTR(-ENOMEM);
+
+ workspace->mem = vmalloc(LZ4_MEM_COMPRESS);
+ workspace->buf = vmalloc(lz4_compressbound(PAGE_CACHE_SIZE));
+ workspace->cbuf = vmalloc(lz4_compressbound(PAGE_CACHE_SIZE));
+ if (!workspace->mem || !workspace->buf || !workspace->cbuf)
+ goto fail;
+
+ INIT_LIST_HEAD(&workspace->list);
+
+ return &workspace->list;
+fail:
+ lz4_free_workspace(&workspace->list);
+ return ERR_PTR(-ENOMEM);
+}
+
+static inline void write_compress_length(char *buf, size_t len)
+{
+ __le32 dlen;
+
+ dlen = cpu_to_le32(len);
+ memcpy(buf, &dlen, LZ4_LEN);
+}
+
+static inline size_t read_compress_length(char *buf)
+{
+ __le32 dlen;
+
+ memcpy(&dlen, buf, LZ4_LEN);
+ return le32_to_cpu(dlen);
+}
+
+static int lz4_compress_pages(struct list_head *ws,
+ struct address_space *mapping,
+ u64 start, unsigned long len,
+ struct page **pages,
+ unsigned long nr_dest_pages,
+ unsigned long *out_pages,
+ unsigned long *total_in,
+ unsigned long *total_out,
+ unsigned long max_out)
+{
+ struct workspace *workspace = list_entry(ws, struct workspace, list);
+ int ret = 0;
+ char *data_in;
+ char *cpage_out;
+ int nr_pages = 0;
+ struct page *in_page = NULL;
+ struct page *out_page = NULL;
+ unsigned long bytes_left;
+
+ size_t in_len;
+ size_t out_len;
+ char *buf;
+ unsigned long tot_in = 0;
+ unsigned long tot_out = 0;
+ unsigned long pg_bytes_left;
+ unsigned long out_offset;
+ unsigned long bytes;
+
+ *out_pages = 0;
+ *total_out = 0;
+ *total_in = 0;
+
+ in_page = find_get_page(mapping, start >> PAGE_CACHE_SHIFT);
+ data_in = kmap(in_page);
+
+ /*
+ * store the size of all chunks of compressed data in
+ * the first 4 bytes
+ */
+ out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+ if (!out_page) {
+ ret = -ENOMEM;
+ goto out;
+ }
+ cpage_out = kmap(out_page);
+ out_offset = LZ4_LEN;
+ tot_out = LZ4_LEN;
+ pages[0] = out_page;
+ nr_pages = 1;
+ pg_bytes_left = PAGE_CACHE_SIZE - LZ4_LEN;
+
+ /* compress at most one page of data each time */
+ in_len = min(len, PAGE_CACHE_SIZE);
+ while (tot_in < len) {
+ ret = lz4_compress(data_in, in_len, workspace->cbuf,
+ &out_len, workspace->mem);
+ if (ret) {
+ pr_debug("BTRFS: deflate in loop returned %d\n",
+ ret);
+ ret = -EIO;
+ goto out;
+ }
+
+ /* store the size of this chunk of compressed data */
+ write_compress_length(cpage_out + out_offset, out_len);
+ tot_out += LZ4_LEN;
+ out_offset += LZ4_LEN;
+ pg_bytes_left -= LZ4_LEN;
+
+ tot_in += in_len;
+ tot_out += out_len;
+
+ /* copy bytes from the working buffer into the pages */
+ buf = workspace->cbuf;
+ while (out_len) {
+ bytes = min_t(unsigned long, pg_bytes_left, out_len);
+
+ memcpy(cpage_out + out_offset, buf, bytes);
+
+ out_len -= bytes;
+ pg_bytes_left -= bytes;
+ buf += bytes;
+ out_offset += bytes;
+
+ /*
+ * we need another page for writing out.
+ *
+ * Note if there's less than 4 bytes left, we just
+ * skip to a new page.
+ */
+ if ((out_len == 0 && pg_bytes_left < LZ4_LEN) ||
+ pg_bytes_left == 0) {
+ if (pg_bytes_left) {
+ memset(cpage_out + out_offset, 0,
+ pg_bytes_left);
+ tot_out += pg_bytes_left;
+ }
+
+ /* we're done, don't allocate new page */
+ if (out_len == 0 && tot_in >= len)
+ break;
+
+ kunmap(out_page);
+ if (nr_pages == nr_dest_pages) {
+ out_page = NULL;
+ ret = -E2BIG;
+ goto out;
+ }
+
+ out_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
+ if (!out_page) {
+ ret = -ENOMEM;
+ goto out;
+ }
+ cpage_out = kmap(out_page);
+ pages[nr_pages++] = out_page;
+
+ pg_bytes_left = PAGE_CACHE_SIZE;
+ out_offset = 0;
+ }
+ }
+
+ /* we're making it bigger, give up */
+ if (tot_in > 8192 && tot_in < tot_out) {
+ ret = -E2BIG;
+ goto out;
+ }
+
+ /* we're all done */
+ if (tot_in >= len)
+ break;
+
+ if (tot_out > max_out)
+ break;
+
+ bytes_left = len - tot_in;
+ kunmap(in_page);
+ page_cache_release(in_page);
+
+ start += PAGE_CACHE_SIZE;
+ in_page = find_get_page(mapping, start >> PAGE_CACHE_SHIFT);
+ data_in = kmap(in_page);
+ in_len = min(bytes_left, PAGE_CACHE_SIZE);
+ }
+
+ if (tot_out > tot_in)
+ goto out;
+
+ /* store the size of all chunks of compressed data */
+ cpage_out = kmap(pages[0]);
+ write_compress_length(cpage_out, tot_out);
+
+ kunmap(pages[0]);
+
+ ret = 0;
+ *total_out = tot_out;
+ *total_in = tot_in;
+out:
+ *out_pages = nr_pages;
+ if (out_page)
+ kunmap(out_page);
+
+ if (in_page) {
+ kunmap(in_page);
+ page_cache_release(in_page);
+ }
+
+ return ret;
+}
+
+static int lz4_decompress_biovec(struct list_head *ws,
+ struct page **pages_in,
+ u64 disk_start,
+ struct bio_vec *bvec,
+ int vcnt,
+ size_t srclen)
+{
+ struct workspace *workspace = list_entry(ws, struct workspace, list);
+ int ret = 0, ret2;
+ char *data_in;
+ unsigned long page_in_index = 0;
+ unsigned long page_out_index = 0;
+ unsigned long total_pages_in = DIV_ROUND_UP(srclen, PAGE_CACHE_SIZE);
+ unsigned long buf_start;
+ unsigned long buf_offset = 0;
+ unsigned long bytes;
+ unsigned long working_bytes;
+ unsigned long pg_offset;
+
+ size_t in_len;
+ size_t out_len;
+ unsigned long in_offset;
+ unsigned long in_page_bytes_left;
+ unsigned long tot_in;
+ unsigned long tot_out;
+ unsigned long tot_len;
+ char *buf;
+ bool may_late_unmap, need_unmap;
+
+ data_in = kmap(pages_in[0]);
+ tot_len = read_compress_length(data_in);
+
+ tot_in = LZ4_LEN;
+ in_offset = LZ4_LEN;
+ tot_len = min_t(size_t, srclen, tot_len);
+ in_page_bytes_left = PAGE_CACHE_SIZE - LZ4_LEN;
+
+ tot_out = 0;
+ pg_offset = 0;
+
+ while (tot_in < tot_len) {
+ in_len = read_compress_length(data_in + in_offset);
+ in_page_bytes_left -= LZ4_LEN;
+ in_offset += LZ4_LEN;
+ tot_in += LZ4_LEN;
+
+ tot_in += in_len;
+ working_bytes = in_len;
+ may_late_unmap = need_unmap = false;
+
+ /* fast path: avoid using the working buffer */
+ if (in_page_bytes_left >= in_len) {
+ buf = data_in + in_offset;
+ bytes = in_len;
+ may_late_unmap = true;
+ goto cont;
+ }
+
+ /* copy bytes from the pages into the working buffer */
+ buf = workspace->cbuf;
+ buf_offset = 0;
+ while (working_bytes) {
+ bytes = min(working_bytes, in_page_bytes_left);
+
+ memcpy(buf + buf_offset, data_in + in_offset, bytes);
+ buf_offset += bytes;
+cont:
+ working_bytes -= bytes;
+ in_page_bytes_left -= bytes;
+ in_offset += bytes;
+
+ /* check if we need to pick another page */
+ if ((working_bytes == 0 && in_page_bytes_left < LZ4_LEN)
+ || in_page_bytes_left == 0) {
+ tot_in += in_page_bytes_left;
+
+ if (working_bytes == 0 && tot_in >= tot_len)
+ break;
+
+ if (page_in_index + 1 >= total_pages_in) {
+ ret = -EIO;
+ goto done;
+ }
+
+ if (may_late_unmap)
+ need_unmap = true;
+ else
+ kunmap(pages_in[page_in_index]);
+
+ data_in = kmap(pages_in[++page_in_index]);
+
+ in_page_bytes_left = PAGE_CACHE_SIZE;
+ in_offset = 0;
+ }
+ }
+
+ out_len = lz4_compressbound(PAGE_CACHE_SIZE);
+ ret = lz4_decompress_unknownoutputsize(buf, in_len,
+ workspace->buf,
+ &out_len);
+ if (need_unmap)
+ kunmap(pages_in[page_in_index - 1]);
+ if (ret) {
+ pr_warn("BTRFS: decompress failed\n");
+ ret = -EIO;
+ break;
+ }
+
+ buf_start = tot_out;
+ tot_out += out_len;
+
+ ret2 = btrfs_decompress_buf2page(workspace->buf, buf_start,
+ tot_out, disk_start,
+ bvec, vcnt,
+ &page_out_index, &pg_offset);
+ if (ret2 == 0)
+ break;
+ }
+done:
+ kunmap(pages_in[page_in_index]);
+ if (!ret)
+ btrfs_clear_biovec_end(bvec, vcnt, page_out_index, pg_offset);
+ return ret;
+}
+
+static int lz4_decompress_btrfs(struct list_head *ws, unsigned char *data_in,
+ struct page *dest_page,
+ unsigned long start_byte,
+ size_t srclen, size_t destlen)
+{
+ struct workspace *workspace = list_entry(ws, struct workspace, list);
+ size_t in_len;
+ size_t out_len;
+ size_t tot_len;
+ int ret = 0;
+ char *kaddr;
+ unsigned long bytes;
+
+ WARN_ON(srclen < LZ4_LEN);
+
+ tot_len = read_compress_length(data_in);
+ data_in += LZ4_LEN;
+
+ in_len = read_compress_length(data_in);
+ data_in += LZ4_LEN;
+
+ out_len = PAGE_CACHE_SIZE;
+ ret = lz4_decompress_unknownoutputsize(data_in, in_len,
+ workspace->buf, &out_len);
+ if (ret) {
+ pr_warn("BTRFS: decompress failed!\n");
+ ret = -EIO;
+ goto out;
+ }
+
+ if (out_len < start_byte) {
+ ret = -EIO;
+ goto out;
+ }
+
+ /*
+ * the caller is already checking against PAGE_SIZE, but lets
+ * move this check closer to the memcpy/memset
+ */
+ destlen = min_t(unsigned long, destlen, PAGE_SIZE);
+ bytes = min_t(unsigned long, destlen, out_len - start_byte);
+
+ kaddr = kmap_atomic(dest_page);
+ memcpy(kaddr, workspace->buf + start_byte, bytes);
+
+ /*
+ * btrfs_getblock is doing a zero on the tail of the page too,
+ * but this will cover anything missing from the decompressed
+ * data.
+ */
+ if (bytes < destlen)
+ memset(kaddr + bytes, 0, destlen - bytes);
+ kunmap_atomic(kaddr);
+out:
+ return ret;
+}
+
+const struct btrfs_compress_op btrfs_lz4_compress = {
+ .alloc_workspace = lz4_alloc_workspace,
+ .free_workspace = lz4_free_workspace,
+ .compress_pages = lz4_compress_pages,
+ .decompress_biovec = lz4_decompress_biovec,
+ .decompress = lz4_decompress_btrfs,
+};
@@ -401,6 +401,8 @@ static int prop_compression_apply(struct inode *inode,
if (!strncmp("lzo", value, len))
type = BTRFS_COMPRESS_LZO;
+ else if (!strncmp("lz4", value, len))
+ type = BTRFS_COMPRESS_LZ4;
else if (!strncmp("zlib", value, len))
type = BTRFS_COMPRESS_ZLIB;
else
@@ -420,6 +422,8 @@ static const char *prop_compression_extract(struct inode *inode)
return "zlib";
case BTRFS_COMPRESS_LZO:
return "lzo";
+ case BTRFS_COMPRESS_LZ4:
+ return "lz4";
}
return NULL;
@@ -477,6 +477,13 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
btrfs_clear_opt(info->mount_opt, NODATACOW);
btrfs_clear_opt(info->mount_opt, NODATASUM);
btrfs_set_fs_incompat(info, COMPRESS_LZO);
+ } else if (strcmp(args[0].from, "lz4") == 0) {
+ compress_type = "lz4";
+ info->compress_type = BTRFS_COMPRESS_LZ4;
+ btrfs_set_opt(info->mount_opt, COMPRESS);
+ btrfs_clear_opt(info->mount_opt, NODATACOW);
+ btrfs_clear_opt(info->mount_opt, NODATASUM);
+ btrfs_set_fs_incompat(info, COMPRESS_LZ4);
} else if (strncmp(args[0].from, "no", 2) == 0) {
compress_type = "no";
btrfs_clear_opt(info->mount_opt, COMPRESS);
@@ -1172,6 +1179,8 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
if (btrfs_test_opt(root, COMPRESS)) {
if (info->compress_type == BTRFS_COMPRESS_ZLIB)
compress_type = "zlib";
+ else if (info->compress_type == BTRFS_COMPRESS_LZ4)
+ compress_type = "lz4";
else
compress_type = "lzo";
if (btrfs_test_opt(root, FORCE_COMPRESS))
@@ -197,6 +197,7 @@ BTRFS_FEAT_ATTR_INCOMPAT(mixed_backref, MIXED_BACKREF);
BTRFS_FEAT_ATTR_INCOMPAT(default_subvol, DEFAULT_SUBVOL);
BTRFS_FEAT_ATTR_INCOMPAT(mixed_groups, MIXED_GROUPS);
BTRFS_FEAT_ATTR_INCOMPAT(compress_lzo, COMPRESS_LZO);
+BTRFS_FEAT_ATTR_INCOMPAT(compress_lz4, COMPRESS_LZ4);
BTRFS_FEAT_ATTR_INCOMPAT(big_metadata, BIG_METADATA);
BTRFS_FEAT_ATTR_INCOMPAT(extended_iref, EXTENDED_IREF);
BTRFS_FEAT_ATTR_INCOMPAT(raid56, RAID56);
@@ -208,6 +209,7 @@ static struct attribute *btrfs_supported_feature_attrs[] = {
BTRFS_FEAT_ATTR_PTR(default_subvol),
BTRFS_FEAT_ATTR_PTR(mixed_groups),
BTRFS_FEAT_ATTR_PTR(compress_lzo),
+ BTRFS_FEAT_ATTR_PTR(compress_lz4),
BTRFS_FEAT_ATTR_PTR(big_metadata),
BTRFS_FEAT_ATTR_PTR(extended_iref),
BTRFS_FEAT_ATTR_PTR(raid56),
Like lzo in btrfs, this patch added lz4 compression support so that users can have a third option. Usage: mount -t btrfs -o compress=lz4 dev /mnt Signed-off-by: Geliang Tang <geliangtang@163.com> --- fs/btrfs/Kconfig | 2 + fs/btrfs/Makefile | 2 +- fs/btrfs/compression.c | 1 + fs/btrfs/compression.h | 1 + fs/btrfs/ctree.h | 7 +- fs/btrfs/disk-io.c | 2 + fs/btrfs/ioctl.c | 4 + fs/btrfs/lz4.c | 445 +++++++++++++++++++++++++++++++++++++++++++++++++ fs/btrfs/props.c | 4 + fs/btrfs/super.c | 9 + fs/btrfs/sysfs.c | 2 + 11 files changed, 476 insertions(+), 3 deletions(-) create mode 100644 fs/btrfs/lz4.c