linux-stable/fs/nilfs2/segbuf.c

478 lines
12 KiB
C
Raw Permalink Normal View History

// SPDX-License-Identifier: GPL-2.0+
/*
* NILFS segment buffer
*
* Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation.
*
* Written by Ryusuke Konishi.
*
*/
#include <linux/buffer_head.h>
#include <linux/writeback.h>
#include <linux/crc32.h>
#include <linux/backing-dev.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include "page.h"
#include "segbuf.h"
struct nilfs_write_info {
struct the_nilfs *nilfs;
struct bio *bio;
int start, end; /* The region to be submitted */
int rest_blocks;
int max_pages;
int nr_vecs;
sector_t blocknr;
};
static int nilfs_segbuf_write(struct nilfs_segment_buffer *segbuf,
struct the_nilfs *nilfs);
static int nilfs_segbuf_wait(struct nilfs_segment_buffer *segbuf);
struct nilfs_segment_buffer *nilfs_segbuf_new(struct super_block *sb)
{
struct nilfs_segment_buffer *segbuf;
segbuf = kmem_cache_alloc(nilfs_segbuf_cachep, GFP_NOFS);
if (unlikely(!segbuf))
return NULL;
segbuf->sb_super = sb;
INIT_LIST_HEAD(&segbuf->sb_list);
INIT_LIST_HEAD(&segbuf->sb_segsum_buffers);
INIT_LIST_HEAD(&segbuf->sb_payload_buffers);
segbuf->sb_super_root = NULL;
init_completion(&segbuf->sb_bio_event);
atomic_set(&segbuf->sb_err, 0);
segbuf->sb_nbio = 0;
return segbuf;
}
void nilfs_segbuf_free(struct nilfs_segment_buffer *segbuf)
{
kmem_cache_free(nilfs_segbuf_cachep, segbuf);
}
void nilfs_segbuf_map(struct nilfs_segment_buffer *segbuf, __u64 segnum,
unsigned long offset, struct the_nilfs *nilfs)
{
segbuf->sb_segnum = segnum;
nilfs_get_segment_range(nilfs, segnum, &segbuf->sb_fseg_start,
&segbuf->sb_fseg_end);
segbuf->sb_pseg_start = segbuf->sb_fseg_start + offset;
segbuf->sb_rest_blocks =
segbuf->sb_fseg_end - segbuf->sb_pseg_start + 1;
}
/**
* nilfs_segbuf_map_cont - map a new log behind a given log
* @segbuf: new segment buffer
* @prev: segment buffer containing a log to be continued
*/
void nilfs_segbuf_map_cont(struct nilfs_segment_buffer *segbuf,
struct nilfs_segment_buffer *prev)
{
segbuf->sb_segnum = prev->sb_segnum;
segbuf->sb_fseg_start = prev->sb_fseg_start;
segbuf->sb_fseg_end = prev->sb_fseg_end;
segbuf->sb_pseg_start = prev->sb_pseg_start + prev->sb_sum.nblocks;
segbuf->sb_rest_blocks =
segbuf->sb_fseg_end - segbuf->sb_pseg_start + 1;
}
void nilfs_segbuf_set_next_segnum(struct nilfs_segment_buffer *segbuf,
__u64 nextnum, struct the_nilfs *nilfs)
{
segbuf->sb_nextnum = nextnum;
segbuf->sb_sum.next = nilfs_get_segment_start_blocknr(nilfs, nextnum);
}
int nilfs_segbuf_extend_segsum(struct nilfs_segment_buffer *segbuf)
{
struct buffer_head *bh;
bh = sb_getblk(segbuf->sb_super,
segbuf->sb_pseg_start + segbuf->sb_sum.nsumblk);
if (unlikely(!bh))
return -ENOMEM;
nilfs2: fix buffer corruption due to concurrent device reads As a result of analysis of a syzbot report, it turned out that in three cases where nilfs2 allocates block device buffers directly via sb_getblk, concurrent reads to the device can corrupt the allocated buffers. Nilfs2 uses sb_getblk for segment summary blocks, that make up a log header, and the super root block, that is the trailer, and when moving and writing the second super block after fs resize. In any of these, since the uptodate flag is not set when storing metadata to be written in the allocated buffers, the stored metadata will be overwritten if a device read of the same block occurs concurrently before the write. This causes metadata corruption and misbehavior in the log write itself, causing warnings in nilfs_btree_assign() as reported. Fix these issues by setting an uptodate flag on the buffer head on the first or before modifying each buffer obtained with sb_getblk, and clearing the flag on failure. When setting the uptodate flag, the lock_buffer/unlock_buffer pair is used to perform necessary exclusive control, and the buffer is filled to ensure that uninitialized bytes are not mixed into the data read from others. As for buffers for segment summary blocks, they are filled incrementally, so if the uptodate flag was unset on their allocation, set the flag and zero fill the buffer once at that point. Also, regarding the superblock move routine, the starting point of the memset call to zerofill the block is incorrectly specified, which can cause a buffer overflow on file systems with block sizes greater than 4KiB. In addition, if the superblock is moved within a large block, it is necessary to assume the possibility that the data in the superblock will be destroyed by zero-filling before copying. So fix these potential issues as well. Link: https://lkml.kernel.org/r/20230609035732.20426-1-konishi.ryusuke@gmail.com Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com> Reported-by: syzbot+31837fe952932efc8fb9@syzkaller.appspotmail.com Closes: https://lkml.kernel.org/r/00000000000030000a05e981f475@google.com Tested-by: Ryusuke Konishi <konishi.ryusuke@gmail.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-06-09 03:57:32 +00:00
lock_buffer(bh);
if (!buffer_uptodate(bh)) {
memset(bh->b_data, 0, bh->b_size);
set_buffer_uptodate(bh);
}
unlock_buffer(bh);
nilfs_segbuf_add_segsum_buffer(segbuf, bh);
return 0;
}
int nilfs_segbuf_extend_payload(struct nilfs_segment_buffer *segbuf,
struct buffer_head **bhp)
{
struct buffer_head *bh;
bh = sb_getblk(segbuf->sb_super,
segbuf->sb_pseg_start + segbuf->sb_sum.nblocks);
if (unlikely(!bh))
return -ENOMEM;
nilfs_segbuf_add_payload_buffer(segbuf, bh);
*bhp = bh;
return 0;
}
int nilfs_segbuf_reset(struct nilfs_segment_buffer *segbuf, unsigned int flags,
time64_t ctime, __u64 cno)
{
int err;
segbuf->sb_sum.nblocks = segbuf->sb_sum.nsumblk = 0;
err = nilfs_segbuf_extend_segsum(segbuf);
if (unlikely(err))
return err;
segbuf->sb_sum.flags = flags;
segbuf->sb_sum.sumbytes = sizeof(struct nilfs_segment_summary);
segbuf->sb_sum.nfinfo = segbuf->sb_sum.nfileblk = 0;
segbuf->sb_sum.ctime = ctime;
segbuf->sb_sum.cno = cno;
return 0;
}
/*
* Setup segment summary
*/
void nilfs_segbuf_fill_in_segsum(struct nilfs_segment_buffer *segbuf)
{
struct nilfs_segment_summary *raw_sum;
struct buffer_head *bh_sum;
bh_sum = list_entry(segbuf->sb_segsum_buffers.next,
struct buffer_head, b_assoc_buffers);
raw_sum = (struct nilfs_segment_summary *)bh_sum->b_data;
raw_sum->ss_magic = cpu_to_le32(NILFS_SEGSUM_MAGIC);
raw_sum->ss_bytes = cpu_to_le16(sizeof(*raw_sum));
raw_sum->ss_flags = cpu_to_le16(segbuf->sb_sum.flags);
raw_sum->ss_seq = cpu_to_le64(segbuf->sb_sum.seg_seq);
raw_sum->ss_create = cpu_to_le64(segbuf->sb_sum.ctime);
raw_sum->ss_next = cpu_to_le64(segbuf->sb_sum.next);
raw_sum->ss_nblocks = cpu_to_le32(segbuf->sb_sum.nblocks);
raw_sum->ss_nfinfo = cpu_to_le32(segbuf->sb_sum.nfinfo);
raw_sum->ss_sumbytes = cpu_to_le32(segbuf->sb_sum.sumbytes);
raw_sum->ss_pad = 0;
raw_sum->ss_cno = cpu_to_le64(segbuf->sb_sum.cno);
}
/*
* CRC calculation routines
*/
static void
nilfs_segbuf_fill_in_segsum_crc(struct nilfs_segment_buffer *segbuf, u32 seed)
{
struct buffer_head *bh;
struct nilfs_segment_summary *raw_sum;
unsigned long size, bytes = segbuf->sb_sum.sumbytes;
u32 crc;
bh = list_entry(segbuf->sb_segsum_buffers.next, struct buffer_head,
b_assoc_buffers);
raw_sum = (struct nilfs_segment_summary *)bh->b_data;
size = min_t(unsigned long, bytes, bh->b_size);
crc = crc32_le(seed,
(unsigned char *)raw_sum +
sizeof(raw_sum->ss_datasum) + sizeof(raw_sum->ss_sumsum),
size - (sizeof(raw_sum->ss_datasum) +
sizeof(raw_sum->ss_sumsum)));
list_for_each_entry_continue(bh, &segbuf->sb_segsum_buffers,
b_assoc_buffers) {
bytes -= size;
size = min_t(unsigned long, bytes, bh->b_size);
crc = crc32_le(crc, bh->b_data, size);
}
raw_sum->ss_sumsum = cpu_to_le32(crc);
}
static void nilfs_segbuf_fill_in_data_crc(struct nilfs_segment_buffer *segbuf,
u32 seed)
{
struct buffer_head *bh;
struct nilfs_segment_summary *raw_sum;
void *kaddr;
u32 crc;
bh = list_entry(segbuf->sb_segsum_buffers.next, struct buffer_head,
b_assoc_buffers);
raw_sum = (struct nilfs_segment_summary *)bh->b_data;
crc = crc32_le(seed,
(unsigned char *)raw_sum + sizeof(raw_sum->ss_datasum),
bh->b_size - sizeof(raw_sum->ss_datasum));
list_for_each_entry_continue(bh, &segbuf->sb_segsum_buffers,
b_assoc_buffers) {
crc = crc32_le(crc, bh->b_data, bh->b_size);
}
list_for_each_entry(bh, &segbuf->sb_payload_buffers, b_assoc_buffers) {
kaddr = kmap_local_page(bh->b_page);
crc = crc32_le(crc, kaddr + bh_offset(bh), bh->b_size);
kunmap_local(kaddr);
}
raw_sum->ss_datasum = cpu_to_le32(crc);
}
static void
nilfs_segbuf_fill_in_super_root_crc(struct nilfs_segment_buffer *segbuf,
u32 seed)
{
struct nilfs_super_root *raw_sr;
struct the_nilfs *nilfs = segbuf->sb_super->s_fs_info;
unsigned int srsize;
u32 crc;
raw_sr = (struct nilfs_super_root *)segbuf->sb_super_root->b_data;
srsize = NILFS_SR_BYTES(nilfs->ns_inode_size);
crc = crc32_le(seed,
(unsigned char *)raw_sr + sizeof(raw_sr->sr_sum),
srsize - sizeof(raw_sr->sr_sum));
raw_sr->sr_sum = cpu_to_le32(crc);
}
static void nilfs_release_buffers(struct list_head *list)
{
struct buffer_head *bh, *n;
list_for_each_entry_safe(bh, n, list, b_assoc_buffers) {
list_del_init(&bh->b_assoc_buffers);
brelse(bh);
}
}
static void nilfs_segbuf_clear(struct nilfs_segment_buffer *segbuf)
{
nilfs_release_buffers(&segbuf->sb_segsum_buffers);
nilfs_release_buffers(&segbuf->sb_payload_buffers);
segbuf->sb_super_root = NULL;
}
/*
* Iterators for segment buffers
*/
void nilfs_clear_logs(struct list_head *logs)
{
struct nilfs_segment_buffer *segbuf;
list_for_each_entry(segbuf, logs, sb_list)
nilfs_segbuf_clear(segbuf);
}
void nilfs_truncate_logs(struct list_head *logs,
struct nilfs_segment_buffer *last)
{
struct nilfs_segment_buffer *n, *segbuf;
segbuf = list_prepare_entry(last, logs, sb_list);
list_for_each_entry_safe_continue(segbuf, n, logs, sb_list) {
list_del_init(&segbuf->sb_list);
nilfs_segbuf_clear(segbuf);
nilfs_segbuf_free(segbuf);
}
}
int nilfs_write_logs(struct list_head *logs, struct the_nilfs *nilfs)
{
struct nilfs_segment_buffer *segbuf;
int ret = 0;
list_for_each_entry(segbuf, logs, sb_list) {
ret = nilfs_segbuf_write(segbuf, nilfs);
if (ret)
break;
}
return ret;
}
int nilfs_wait_on_logs(struct list_head *logs)
{
struct nilfs_segment_buffer *segbuf;
int err, ret = 0;
list_for_each_entry(segbuf, logs, sb_list) {
err = nilfs_segbuf_wait(segbuf);
if (err && !ret)
ret = err;
}
return ret;
}
/**
* nilfs_add_checksums_on_logs - add checksums on the logs
* @logs: list of segment buffers storing target logs
* @seed: checksum seed value
*/
void nilfs_add_checksums_on_logs(struct list_head *logs, u32 seed)
{
struct nilfs_segment_buffer *segbuf;
list_for_each_entry(segbuf, logs, sb_list) {
if (segbuf->sb_super_root)
nilfs_segbuf_fill_in_super_root_crc(segbuf, seed);
nilfs_segbuf_fill_in_segsum_crc(segbuf, seed);
nilfs_segbuf_fill_in_data_crc(segbuf, seed);
}
}
/*
* BIO operations
*/
static void nilfs_end_bio_write(struct bio *bio)
{
struct nilfs_segment_buffer *segbuf = bio->bi_private;
if (bio->bi_status)
atomic_inc(&segbuf->sb_err);
bio_put(bio);
complete(&segbuf->sb_bio_event);
}
static int nilfs_segbuf_submit_bio(struct nilfs_segment_buffer *segbuf,
struct nilfs_write_info *wi)
{
struct bio *bio = wi->bio;
bio->bi_end_io = nilfs_end_bio_write;
bio->bi_private = segbuf;
submit_bio(bio);
segbuf->sb_nbio++;
wi->bio = NULL;
wi->rest_blocks -= wi->end - wi->start;
wi->nr_vecs = min(wi->max_pages, wi->rest_blocks);
wi->start = wi->end;
return 0;
}
static void nilfs_segbuf_prepare_write(struct nilfs_segment_buffer *segbuf,
struct nilfs_write_info *wi)
{
wi->bio = NULL;
wi->rest_blocks = segbuf->sb_sum.nblocks;
wi->max_pages = BIO_MAX_VECS;
wi->nr_vecs = min(wi->max_pages, wi->rest_blocks);
wi->start = wi->end = 0;
wi->blocknr = segbuf->sb_pseg_start;
}
static int nilfs_segbuf_submit_bh(struct nilfs_segment_buffer *segbuf,
struct nilfs_write_info *wi,
struct buffer_head *bh)
{
int len, err;
BUG_ON(wi->nr_vecs <= 0);
repeat:
if (!wi->bio) {
wi->bio = bio_alloc(wi->nilfs->ns_bdev, wi->nr_vecs,
REQ_OP_WRITE, GFP_NOIO);
wi->bio->bi_iter.bi_sector = (wi->blocknr + wi->end) <<
(wi->nilfs->ns_blocksize_bits - 9);
}
len = bio_add_page(wi->bio, bh->b_page, bh->b_size, bh_offset(bh));
if (len == bh->b_size) {
wi->end++;
return 0;
}
/* bio is FULL */
err = nilfs_segbuf_submit_bio(segbuf, wi);
/* never submit current bh */
if (likely(!err))
goto repeat;
return err;
}
/**
* nilfs_segbuf_write - submit write requests of a log
* @segbuf: buffer storing a log to be written
* @nilfs: nilfs object
*
* Return Value: On Success, 0 is returned. On Error, one of the following
* negative error code is returned.
*
* %-EIO - I/O error
*
* %-ENOMEM - Insufficient memory available.
*/
static int nilfs_segbuf_write(struct nilfs_segment_buffer *segbuf,
struct the_nilfs *nilfs)
{
struct nilfs_write_info wi;
struct buffer_head *bh;
int res = 0;
wi.nilfs = nilfs;
nilfs_segbuf_prepare_write(segbuf, &wi);
list_for_each_entry(bh, &segbuf->sb_segsum_buffers, b_assoc_buffers) {
res = nilfs_segbuf_submit_bh(segbuf, &wi, bh);
if (unlikely(res))
goto failed_bio;
}
list_for_each_entry(bh, &segbuf->sb_payload_buffers, b_assoc_buffers) {
res = nilfs_segbuf_submit_bh(segbuf, &wi, bh);
if (unlikely(res))
goto failed_bio;
}
if (wi.bio) {
/*
* Last BIO is always sent through the following
* submission.
*/
wi.bio->bi_opf |= REQ_SYNC;
res = nilfs_segbuf_submit_bio(segbuf, &wi);
}
failed_bio:
return res;
}
/**
* nilfs_segbuf_wait - wait for completion of requested BIOs
* @segbuf: segment buffer
*
* Return Value: On Success, 0 is returned. On Error, one of the following
* negative error code is returned.
*
* %-EIO - I/O error
*/
static int nilfs_segbuf_wait(struct nilfs_segment_buffer *segbuf)
{
int err = 0;
if (!segbuf->sb_nbio)
return 0;
do {
wait_for_completion(&segbuf->sb_bio_event);
} while (--segbuf->sb_nbio > 0);
if (unlikely(atomic_read(&segbuf->sb_err) > 0)) {
nilfs_err(segbuf->sb_super,
"I/O error writing log (start-blocknr=%llu, block-count=%lu) in segment %llu",
(unsigned long long)segbuf->sb_pseg_start,
segbuf->sb_sum.nblocks,
(unsigned long long)segbuf->sb_segnum);
err = -EIO;
}
return err;
}