linux/fs/erofs/data.c
Linus Torvalds dcb9f48667 Changes since last update:
- Convert metadata APIs to byte offsets;
 
  - Avoid allocating DEFLATE streams unnecessarily;
 
  - Some erofs_show_options() cleanup.
 -----BEGIN PGP SIGNATURE-----
 
 iQJFBAABCgAvFiEEQ0A6bDUS9Y+83NPFUXZn5Zlu5qoFAmZQmHARHHhpYW5nQGtl
 cm5lbC5vcmcACgkQUXZn5Zlu5qrGnhAAnvOifMYekIgY/W0PSGSe85XtXps5vBjo
 rixZ/vNAl8NrLgzHY5lX+4dbENywEULzdxYAgF4VN9eKNGyuZ4oCBmYStoGueQ41
 N1oq36O/CVJDCOLkFUwjD6GpHngjJR3xiU8DRrhKdPZJeYXVEJwZB4KOOymorkO0
 Xn9SPrF/GC4YDWJL901RKT8p6gyRNWiWJ/+hwDAxfmCSuzW2uRNnBLeXNvjqj4Z3
 u5WEaFSlNRlLWnZPcHy8O3t/XAPkhvTN+C5+YeaePWyHc5WYOM9mWt8VLOFQb60K
 l+q/cnWXw+8NNbxnuccWVJfEb6zUJmZ5/yTm+Ndutrpk5dFSPb6DjZo5/K36dGls
 r02XysW+Jl24wBIFkYRHild2WT+gSqo/zyIDsSt/DF+DhpqmnIqAASx4yJenw7ib
 BNV4m4gQflLrORKpVmsKyHrm5GuHsTWsGc51iX1uqsdfDgN79mFgR1taBAZw162P
 pPeWuD6XYE+eT+t5nggnXqmZ5qatEhTFkYDjUzSq4ZQfyZnRG8Tl6zbBuyVhaxsO
 zH1rAmwtI6x+ehHI46Kurh8HT6UrB0CNM6RokYKr6JWVzIdFPPMVKkxcq2KozTPf
 CBu+Whh/WGFROM8JT2KGCnuz2ZBUZXDtNBJmW+ZnA+z9b7xZ1f31nio4vKKdZU+R
 swpnV+0q9cs=
 =qDDl
 -----END PGP SIGNATURE-----

Merge tag 'erofs-for-6.10-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs

Pull more erofs updates from Gao Xiang:
 "The main ones are metadata API conversion to byte offsets by Al Viro.

  Another patch gets rid of unnecessary memory allocation out of DEFLATE
  decompressor. The remaining one is a trivial cleanup.

   - Convert metadata APIs to byte offsets

   - Avoid allocating DEFLATE streams unnecessarily

   - Some erofs_show_options() cleanup"

* tag 'erofs-for-6.10-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs:
  erofs: avoid allocating DEFLATE streams before mounting
  z_erofs_pcluster_begin(): don't bother with rounding position down
  erofs: don't round offset down for erofs_read_metabuf()
  erofs: don't align offset for erofs_read_metabuf() (simple cases)
  erofs: mechanically convert erofs_read_metabuf() to offsets
  erofs: clean up erofs_show_options()
2024-05-24 09:31:50 -07:00

450 lines
11 KiB
C

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2017-2018 HUAWEI, Inc.
* https://www.huawei.com/
* Copyright (C) 2021, Alibaba Cloud
*/
#include "internal.h"
#include <linux/sched/mm.h>
#include <trace/events/erofs.h>
void erofs_unmap_metabuf(struct erofs_buf *buf)
{
if (buf->kmap_type == EROFS_KMAP)
kunmap_local(buf->base);
buf->base = NULL;
buf->kmap_type = EROFS_NO_KMAP;
}
void erofs_put_metabuf(struct erofs_buf *buf)
{
if (!buf->page)
return;
erofs_unmap_metabuf(buf);
put_page(buf->page);
buf->page = NULL;
}
/*
* Derive the block size from inode->i_blkbits to make compatible with
* anonymous inode in fscache mode.
*/
void *erofs_bread(struct erofs_buf *buf, erofs_off_t offset,
enum erofs_kmap_type type)
{
pgoff_t index = offset >> PAGE_SHIFT;
struct page *page = buf->page;
struct folio *folio;
unsigned int nofs_flag;
if (!page || page->index != index) {
erofs_put_metabuf(buf);
nofs_flag = memalloc_nofs_save();
folio = read_cache_folio(buf->mapping, index, NULL, NULL);
memalloc_nofs_restore(nofs_flag);
if (IS_ERR(folio))
return folio;
/* should already be PageUptodate, no need to lock page */
page = folio_file_page(folio, index);
buf->page = page;
}
if (buf->kmap_type == EROFS_NO_KMAP) {
if (type == EROFS_KMAP)
buf->base = kmap_local_page(page);
buf->kmap_type = type;
} else if (buf->kmap_type != type) {
DBG_BUGON(1);
return ERR_PTR(-EFAULT);
}
if (type == EROFS_NO_KMAP)
return NULL;
return buf->base + (offset & ~PAGE_MASK);
}
void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb)
{
if (erofs_is_fscache_mode(sb))
buf->mapping = EROFS_SB(sb)->s_fscache->inode->i_mapping;
else
buf->mapping = sb->s_bdev->bd_mapping;
}
void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb,
erofs_off_t offset, enum erofs_kmap_type type)
{
erofs_init_metabuf(buf, sb);
return erofs_bread(buf, offset, type);
}
static int erofs_map_blocks_flatmode(struct inode *inode,
struct erofs_map_blocks *map)
{
erofs_blk_t nblocks, lastblk;
u64 offset = map->m_la;
struct erofs_inode *vi = EROFS_I(inode);
struct super_block *sb = inode->i_sb;
bool tailendpacking = (vi->datalayout == EROFS_INODE_FLAT_INLINE);
nblocks = erofs_iblks(inode);
lastblk = nblocks - tailendpacking;
/* there is no hole in flatmode */
map->m_flags = EROFS_MAP_MAPPED;
if (offset < erofs_pos(sb, lastblk)) {
map->m_pa = erofs_pos(sb, vi->raw_blkaddr) + map->m_la;
map->m_plen = erofs_pos(sb, lastblk) - offset;
} else if (tailendpacking) {
map->m_pa = erofs_iloc(inode) + vi->inode_isize +
vi->xattr_isize + erofs_blkoff(sb, offset);
map->m_plen = inode->i_size - offset;
/* inline data should be located in the same meta block */
if (erofs_blkoff(sb, map->m_pa) + map->m_plen > sb->s_blocksize) {
erofs_err(sb, "inline data cross block boundary @ nid %llu",
vi->nid);
DBG_BUGON(1);
return -EFSCORRUPTED;
}
map->m_flags |= EROFS_MAP_META;
} else {
erofs_err(sb, "internal error @ nid: %llu (size %llu), m_la 0x%llx",
vi->nid, inode->i_size, map->m_la);
DBG_BUGON(1);
return -EIO;
}
return 0;
}
int erofs_map_blocks(struct inode *inode, struct erofs_map_blocks *map)
{
struct super_block *sb = inode->i_sb;
struct erofs_inode *vi = EROFS_I(inode);
struct erofs_inode_chunk_index *idx;
struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
u64 chunknr;
unsigned int unit;
erofs_off_t pos;
void *kaddr;
int err = 0;
trace_erofs_map_blocks_enter(inode, map, 0);
map->m_deviceid = 0;
if (map->m_la >= inode->i_size) {
/* leave out-of-bound access unmapped */
map->m_flags = 0;
map->m_plen = 0;
goto out;
}
if (vi->datalayout != EROFS_INODE_CHUNK_BASED) {
err = erofs_map_blocks_flatmode(inode, map);
goto out;
}
if (vi->chunkformat & EROFS_CHUNK_FORMAT_INDEXES)
unit = sizeof(*idx); /* chunk index */
else
unit = EROFS_BLOCK_MAP_ENTRY_SIZE; /* block map */
chunknr = map->m_la >> vi->chunkbits;
pos = ALIGN(erofs_iloc(inode) + vi->inode_isize +
vi->xattr_isize, unit) + unit * chunknr;
kaddr = erofs_read_metabuf(&buf, sb, pos, EROFS_KMAP);
if (IS_ERR(kaddr)) {
err = PTR_ERR(kaddr);
goto out;
}
map->m_la = chunknr << vi->chunkbits;
map->m_plen = min_t(erofs_off_t, 1UL << vi->chunkbits,
round_up(inode->i_size - map->m_la, sb->s_blocksize));
/* handle block map */
if (!(vi->chunkformat & EROFS_CHUNK_FORMAT_INDEXES)) {
__le32 *blkaddr = kaddr;
if (le32_to_cpu(*blkaddr) == EROFS_NULL_ADDR) {
map->m_flags = 0;
} else {
map->m_pa = erofs_pos(sb, le32_to_cpu(*blkaddr));
map->m_flags = EROFS_MAP_MAPPED;
}
goto out_unlock;
}
/* parse chunk indexes */
idx = kaddr;
switch (le32_to_cpu(idx->blkaddr)) {
case EROFS_NULL_ADDR:
map->m_flags = 0;
break;
default:
map->m_deviceid = le16_to_cpu(idx->device_id) &
EROFS_SB(sb)->device_id_mask;
map->m_pa = erofs_pos(sb, le32_to_cpu(idx->blkaddr));
map->m_flags = EROFS_MAP_MAPPED;
break;
}
out_unlock:
erofs_put_metabuf(&buf);
out:
if (!err)
map->m_llen = map->m_plen;
trace_erofs_map_blocks_exit(inode, map, 0, err);
return err;
}
int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *map)
{
struct erofs_dev_context *devs = EROFS_SB(sb)->devs;
struct erofs_device_info *dif;
int id;
map->m_bdev = sb->s_bdev;
map->m_daxdev = EROFS_SB(sb)->dax_dev;
map->m_dax_part_off = EROFS_SB(sb)->dax_part_off;
map->m_fscache = EROFS_SB(sb)->s_fscache;
if (map->m_deviceid) {
down_read(&devs->rwsem);
dif = idr_find(&devs->tree, map->m_deviceid - 1);
if (!dif) {
up_read(&devs->rwsem);
return -ENODEV;
}
if (devs->flatdev) {
map->m_pa += erofs_pos(sb, dif->mapped_blkaddr);
up_read(&devs->rwsem);
return 0;
}
map->m_bdev = dif->bdev_file ? file_bdev(dif->bdev_file) : NULL;
map->m_daxdev = dif->dax_dev;
map->m_dax_part_off = dif->dax_part_off;
map->m_fscache = dif->fscache;
up_read(&devs->rwsem);
} else if (devs->extra_devices && !devs->flatdev) {
down_read(&devs->rwsem);
idr_for_each_entry(&devs->tree, dif, id) {
erofs_off_t startoff, length;
if (!dif->mapped_blkaddr)
continue;
startoff = erofs_pos(sb, dif->mapped_blkaddr);
length = erofs_pos(sb, dif->blocks);
if (map->m_pa >= startoff &&
map->m_pa < startoff + length) {
map->m_pa -= startoff;
map->m_bdev = dif->bdev_file ?
file_bdev(dif->bdev_file) : NULL;
map->m_daxdev = dif->dax_dev;
map->m_dax_part_off = dif->dax_part_off;
map->m_fscache = dif->fscache;
break;
}
}
up_read(&devs->rwsem);
}
return 0;
}
static int erofs_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
unsigned int flags, struct iomap *iomap, struct iomap *srcmap)
{
int ret;
struct super_block *sb = inode->i_sb;
struct erofs_map_blocks map;
struct erofs_map_dev mdev;
map.m_la = offset;
map.m_llen = length;
ret = erofs_map_blocks(inode, &map);
if (ret < 0)
return ret;
mdev = (struct erofs_map_dev) {
.m_deviceid = map.m_deviceid,
.m_pa = map.m_pa,
};
ret = erofs_map_dev(sb, &mdev);
if (ret)
return ret;
iomap->offset = map.m_la;
if (flags & IOMAP_DAX)
iomap->dax_dev = mdev.m_daxdev;
else
iomap->bdev = mdev.m_bdev;
iomap->length = map.m_llen;
iomap->flags = 0;
iomap->private = NULL;
if (!(map.m_flags & EROFS_MAP_MAPPED)) {
iomap->type = IOMAP_HOLE;
iomap->addr = IOMAP_NULL_ADDR;
if (!iomap->length)
iomap->length = length;
return 0;
}
if (map.m_flags & EROFS_MAP_META) {
void *ptr;
struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
iomap->type = IOMAP_INLINE;
ptr = erofs_read_metabuf(&buf, sb, mdev.m_pa, EROFS_KMAP);
if (IS_ERR(ptr))
return PTR_ERR(ptr);
iomap->inline_data = ptr;
iomap->private = buf.base;
} else {
iomap->type = IOMAP_MAPPED;
iomap->addr = mdev.m_pa;
if (flags & IOMAP_DAX)
iomap->addr += mdev.m_dax_part_off;
}
return 0;
}
static int erofs_iomap_end(struct inode *inode, loff_t pos, loff_t length,
ssize_t written, unsigned int flags, struct iomap *iomap)
{
void *ptr = iomap->private;
if (ptr) {
struct erofs_buf buf = {
.page = kmap_to_page(ptr),
.base = ptr,
.kmap_type = EROFS_KMAP,
};
DBG_BUGON(iomap->type != IOMAP_INLINE);
erofs_put_metabuf(&buf);
} else {
DBG_BUGON(iomap->type == IOMAP_INLINE);
}
return written;
}
static const struct iomap_ops erofs_iomap_ops = {
.iomap_begin = erofs_iomap_begin,
.iomap_end = erofs_iomap_end,
};
int erofs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
u64 start, u64 len)
{
if (erofs_inode_is_data_compressed(EROFS_I(inode)->datalayout)) {
#ifdef CONFIG_EROFS_FS_ZIP
return iomap_fiemap(inode, fieinfo, start, len,
&z_erofs_iomap_report_ops);
#else
return -EOPNOTSUPP;
#endif
}
return iomap_fiemap(inode, fieinfo, start, len, &erofs_iomap_ops);
}
/*
* since we dont have write or truncate flows, so no inode
* locking needs to be held at the moment.
*/
static int erofs_read_folio(struct file *file, struct folio *folio)
{
return iomap_read_folio(folio, &erofs_iomap_ops);
}
static void erofs_readahead(struct readahead_control *rac)
{
return iomap_readahead(rac, &erofs_iomap_ops);
}
static sector_t erofs_bmap(struct address_space *mapping, sector_t block)
{
return iomap_bmap(mapping, block, &erofs_iomap_ops);
}
static ssize_t erofs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
struct inode *inode = file_inode(iocb->ki_filp);
/* no need taking (shared) inode lock since it's a ro filesystem */
if (!iov_iter_count(to))
return 0;
#ifdef CONFIG_FS_DAX
if (IS_DAX(inode))
return dax_iomap_rw(iocb, to, &erofs_iomap_ops);
#endif
if (iocb->ki_flags & IOCB_DIRECT) {
struct block_device *bdev = inode->i_sb->s_bdev;
unsigned int blksize_mask;
if (bdev)
blksize_mask = bdev_logical_block_size(bdev) - 1;
else
blksize_mask = i_blocksize(inode) - 1;
if ((iocb->ki_pos | iov_iter_count(to) |
iov_iter_alignment(to)) & blksize_mask)
return -EINVAL;
return iomap_dio_rw(iocb, to, &erofs_iomap_ops,
NULL, 0, NULL, 0);
}
return filemap_read(iocb, to, 0);
}
/* for uncompressed (aligned) files and raw access for other files */
const struct address_space_operations erofs_raw_access_aops = {
.read_folio = erofs_read_folio,
.readahead = erofs_readahead,
.bmap = erofs_bmap,
.direct_IO = noop_direct_IO,
.release_folio = iomap_release_folio,
.invalidate_folio = iomap_invalidate_folio,
};
#ifdef CONFIG_FS_DAX
static vm_fault_t erofs_dax_huge_fault(struct vm_fault *vmf,
unsigned int order)
{
return dax_iomap_fault(vmf, order, NULL, NULL, &erofs_iomap_ops);
}
static vm_fault_t erofs_dax_fault(struct vm_fault *vmf)
{
return erofs_dax_huge_fault(vmf, 0);
}
static const struct vm_operations_struct erofs_dax_vm_ops = {
.fault = erofs_dax_fault,
.huge_fault = erofs_dax_huge_fault,
};
static int erofs_file_mmap(struct file *file, struct vm_area_struct *vma)
{
if (!IS_DAX(file_inode(file)))
return generic_file_readonly_mmap(file, vma);
if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE))
return -EINVAL;
vma->vm_ops = &erofs_dax_vm_ops;
vm_flags_set(vma, VM_HUGEPAGE);
return 0;
}
#else
#define erofs_file_mmap generic_file_readonly_mmap
#endif
const struct file_operations erofs_file_fops = {
.llseek = generic_file_llseek,
.read_iter = erofs_file_read_iter,
.mmap = erofs_file_mmap,
.get_unmapped_area = thp_get_unmapped_area,
.splice_read = filemap_splice_read,
};