linux-loongson/fs/erofs/dir.c
Chao Yu df0ce6cefa erofs: support to readahead dirent blocks in erofs_readdir()
This patch supports to readahead more blocks in erofs_readdir(), it can
enhance readdir performance in large direcotry.

readdir test in a large directory which contains 12000 sub-files.

		files_per_second
Before:		926385.54
After:		2380435.562

Meanwhile, let's introduces a new sysfs entry to control readahead
bytes to provide more flexible policy for readahead of readdir().
- location: /sys/fs/erofs/<disk>/dir_ra_bytes
- default value: 16384
- disable readahead: set the value to 0

Signed-off-by: Chao Yu <chao@kernel.org>
Reviewed-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Link: https://lore.kernel.org/r/20250721021352.2495371-1-chao@kernel.org
[ Gao Xiang: minor styling adjustment. ]
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
2025-07-24 19:44:08 +08:00

127 lines
3.4 KiB
C

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2017-2018 HUAWEI, Inc.
* https://www.huawei.com/
* Copyright (C) 2022, Alibaba Cloud
*/
#include "internal.h"
static int erofs_fill_dentries(struct inode *dir, struct dir_context *ctx,
void *dentry_blk, struct erofs_dirent *de,
unsigned int nameoff0, unsigned int maxsize)
{
const struct erofs_dirent *end = dentry_blk + nameoff0;
while (de < end) {
unsigned char d_type = fs_ftype_to_dtype(de->file_type);
unsigned int nameoff = le16_to_cpu(de->nameoff);
const char *de_name = (char *)dentry_blk + nameoff;
unsigned int de_namelen;
/* the last dirent in the block? */
if (de + 1 >= end)
de_namelen = strnlen(de_name, maxsize - nameoff);
else
de_namelen = le16_to_cpu(de[1].nameoff) - nameoff;
/* a corrupted entry is found */
if (nameoff + de_namelen > maxsize ||
de_namelen > EROFS_NAME_LEN) {
erofs_err(dir->i_sb, "bogus dirent @ nid %llu",
EROFS_I(dir)->nid);
DBG_BUGON(1);
return -EFSCORRUPTED;
}
if (!dir_emit(ctx, de_name, de_namelen,
erofs_nid_to_ino64(EROFS_SB(dir->i_sb),
le64_to_cpu(de->nid)), d_type))
return 1;
++de;
ctx->pos += sizeof(struct erofs_dirent);
}
return 0;
}
static int erofs_readdir(struct file *f, struct dir_context *ctx)
{
struct inode *dir = file_inode(f);
struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
struct super_block *sb = dir->i_sb;
struct file_ra_state *ra = &f->f_ra;
unsigned long bsz = sb->s_blocksize;
unsigned int ofs = erofs_blkoff(sb, ctx->pos);
pgoff_t ra_pages = DIV_ROUND_UP_POW2(
EROFS_I_SB(dir)->dir_ra_bytes, PAGE_SIZE);
pgoff_t nr_pages = DIV_ROUND_UP_POW2(dir->i_size, PAGE_SIZE);
int err = 0;
bool initial = true;
buf.mapping = dir->i_mapping;
while (ctx->pos < dir->i_size) {
erofs_off_t dbstart = ctx->pos - ofs;
struct erofs_dirent *de;
unsigned int nameoff, maxsize;
if (fatal_signal_pending(current)) {
err = -ERESTARTSYS;
break;
}
/* readahead blocks to enhance performance for large directories */
if (ra_pages) {
pgoff_t idx = DIV_ROUND_UP_POW2(ctx->pos, PAGE_SIZE);
pgoff_t pages = min(nr_pages - idx, ra_pages);
if (pages > 1 && !ra_has_index(ra, idx))
page_cache_sync_readahead(dir->i_mapping, ra,
f, idx, pages);
}
de = erofs_bread(&buf, dbstart, true);
if (IS_ERR(de)) {
erofs_err(sb, "failed to readdir of logical block %llu of nid %llu",
erofs_blknr(sb, dbstart), EROFS_I(dir)->nid);
err = PTR_ERR(de);
break;
}
nameoff = le16_to_cpu(de->nameoff);
if (nameoff < sizeof(struct erofs_dirent) || nameoff >= bsz) {
erofs_err(sb, "invalid de[0].nameoff %u @ nid %llu",
nameoff, EROFS_I(dir)->nid);
err = -EFSCORRUPTED;
break;
}
maxsize = min_t(unsigned int, dir->i_size - dbstart, bsz);
/* search dirents at the arbitrary position */
if (initial) {
initial = false;
ofs = roundup(ofs, sizeof(struct erofs_dirent));
ctx->pos = dbstart + ofs;
}
err = erofs_fill_dentries(dir, ctx, de, (void *)de + ofs,
nameoff, maxsize);
if (err)
break;
ctx->pos = dbstart + maxsize;
ofs = 0;
cond_resched();
}
erofs_put_metabuf(&buf);
if (EROFS_I(dir)->dot_omitted && ctx->pos == dir->i_size) {
if (!dir_emit_dot(f, ctx))
return 0;
++ctx->pos;
}
return err < 0 ? err : 0;
}
const struct file_operations erofs_dir_fops = {
.llseek = generic_file_llseek,
.read = generic_read_dir,
.iterate_shared = erofs_readdir,
};